Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2013-08-31.

MISSING IMAGE

Material Information

Title:
Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2013-08-31.
Physical Description:
Book
Language:
english
Creator:
Czarnecki,David Mark Ii
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Horticultural Science
Committee Chair:
Deng, Zhanao
Committee Co-Chair:
Clark, David G
Committee Members:
Scott, John W
Chandler, Craig K
Quesenberry, Kenneth H

Subjects

Subjects / Keywords:
Horticultural Science -- Dissertations, Academic -- UF
Genre:
Horticultural Science thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Statement of Responsibility:
by David Mark Ii Czarnecki.
Thesis:
Thesis (Ph.D.)--University of Florida, 2011.
Local:
Adviser: Deng, Zhanao.
Local:
Co-adviser: Clark, David G.
Electronic Access:
INACCESSIBLE UNTIL 2013-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2011
System ID:
UFE0042835:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 GENETIC STERILIZATION AND REPRODUCTIVE BIOLOGY OF LANTANA CAMARA By DAVID MARK CZARNECKI II A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 David Mark Czarnecki II

PAGE 3

3 To my family and friends

PAGE 4

4 ACKNOWLEDGMENTS Completion of this project required the help of numerous individuals over the last few years. I would like to thank my parents for understanding my absence from their lives and my inability to participate in the important events that I have missed because of the commitments associated with my degree My friends over the years have provided me with support, advice, and much needed grounding without which I would not have had the focus to complete my projects. Last ly I would like to thank my professor and advisors, who have given me the guidance and advice to learn and grow as a scientist

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 10 LIST OF FIGURES ................................ ................................ ................................ ........ 12 LIST OF ABBREVIATIONS ................................ ................................ ........................... 13 ABSTRACT ................................ ................................ ................................ ................... 15 CHAPTER 1 I NTRODUCTION ................................ ................................ ................................ .... 17 Overvi ew and Rationale ................................ ................................ .......................... 17 Floriculture Industry and Concerns about Introduced Plant Species ................ 17 Lantana camara ................................ ................................ ............................... 18 Need To Steril ize Lantana camara ................................ ................................ ... 19 Invasive Plants ................................ ................................ ................................ ........ 22 Potential Ecological and Economic Damages ................................ .................. 22 Managing Invasive Plants ................................ ................................ ................. 23 Regulatory Measures ................................ ................................ ....................... 24 Biological Factors Determining Invasive Potential of Plants ................................ ... 26 Seed Production and Dispersal ................................ ................................ ........ 26 Pollen Production and Dispersal ................................ ................................ ...... 26 Vegetative Propagules ................................ ................................ ..................... 27 Assess ing Invasive Potential of Plants ................................ ............................. 27 Genetic Approaches to Reduce or Eliminate Invasive Potential of Plants .............. 28 Traditional Breeding ................................ ................................ ......................... 29 Mutational Breeding ................................ ................................ ......................... 30 Transgen ic Approaches ................................ ................................ .................... 30 Ploidy Manipulation ................................ ................................ .......................... 31 Research Objectives ................................ ................................ ............................... 31 2 MALE FERTILITY OF L ANTANA CAMARA ................................ ........................... 33 Rationale ................................ ................................ ................................ ................. 33 Materials and Meth ods ................................ ................................ ............................ 34 Plant Materials ................................ ................................ ................................ .. 34 Ploidy Analysis ................................ ................................ ................................ 35 Pollen Stain ing ................................ ................................ ................................ 36 Experimental Designs ................................ ................................ ...................... 37 Statistical Analysis ................................ ................................ ............................ 37 Meiotic Observation ................................ ................................ .......................... 38

PAGE 6

6 Results and Discussion ................................ ................................ ........................... 38 Ploidy Analysis ................................ ................................ ................................ 38 Dye Comparison for Lantana Pollen Stainability Determination ....................... 39 Pollen Stainability of L. camara Cultivars and Breeding Lines .......................... 39 Diploids ................................ ................................ ................................ ...... 40 Triploids ................................ ................................ ................................ ..... 41 Tetraploids ................................ ................................ ................................ 41 Pentaploids and hexaploids ................................ ................................ ....... 42 Meiotic Abnormalities ................................ ................................ ....................... 42 Summary ................................ ................................ ................................ ................ 43 3 FEMALE FERTILITY OF L ANTANA CAMARA ................................ ....................... 51 Rationale ................................ ................................ ................................ ................. 51 Materials and Methods ................................ ................................ ............................ 53 Plant Materials ................................ ................................ ................................ .. 53 Assessing Fruit Production ................................ ................................ ............... 53 In sect Damage ................................ ................................ ................................ 54 Evaluating Seed Germination ................................ ................................ ........... 55 Calculating Female Fertility Index ................................ ................................ .... 55 Plant Dry Weight ................................ ................................ .............................. 55 Experimental Design ................................ ................................ ........................ 56 Statistical Analysis ................................ ................................ ............................ 56 Results and Discussion ................................ ................................ ........................... 56 Fruit Production in L. camara ................................ ................................ ........... 56 Seed Germination in L. camara ................................ ................................ ........ 57 Female Fertility Index (FFI) of L. camara ................................ .......................... 58 Fruit Production, Seed Germination and Female Fertilit y Index of Diploid L. camara ................................ ................................ ................................ .......... 58 Fruit Production, Seed Germination and Female Fertility Index of Triploid L. camara ................................ ................................ ................................ .......... 59 Non UFG producing triploids ................................ ................................ ..... 59 UFG producing triploids ................................ ................................ ............. 60 Frui t Production, Seed Germination and Female Fertility Index of Tetraploid L. camara ................................ ................................ ................................ ...... 61 Non UFG producing tetraploids ................................ ................................ 61 UFG producing tetraploids ................................ ................................ ......... 61 Fruit Production, Seed Germination and Female Fertility Index of Pentaploid and Hexaploid L. camara ................................ ................................ .............. 62 Correlation Analysis ................................ ................................ ......................... 63 Principal Component Analysis ................................ ................................ .......... 64 Summary ................................ ................................ ................................ ................ 64 4 OC C UR R ENCE OF UNREDUCED FEMALE GAMETES LEADS TO SEXUAL POLYPLOIDIZATION IN LANTANA CAMARA ................................ ....................... 72 Rationale ................................ ................................ ................................ ................. 72

PAGE 7

7 Materials and Methods ................................ ................................ ............................ 75 Plant Materials ................................ ................................ ................................ .. 75 Pollinatio ns ................................ ................................ ................................ ....... 75 Progeny Growing ................................ ................................ .............................. 76 Ploidy Analysis ................................ ................................ ................................ 77 Results ................................ ................................ ................................ .................... 77 Ploidy Analysis ................................ ................................ ................................ 77 Ploidy analysis of OP progeny ................................ ................................ ... 78 Ploidy analysis of CP progeny ................................ ................................ ... 81 Transmission of UFG Formation ................................ ................................ ...... 82 Effect of 2 n Gamete Formation on Seed Production by Triploids ..................... 82 Discussion ................................ ................................ ................................ .............. 83 Mechanisms of UFG Formation in L. camara ................................ ................... 83 Occurrence of UFGs and Polyploidization in Lantana ................................ ...... 85 UFGs and Seed Set in Polyploids ................................ ................................ .... 87 Effect of 2 n Gamete Formation on Seed Production by Triploids ..................... 88 5 M ULTIPLE MODES OF REPRODUCTION AS REVEALED BY PLOIDY AND MICROSATELLITE M ARKER ANALYSIS OF LANTANA CAMARA ...................... 97 Rationale ................................ ................................ ................................ ................. 97 Materials and Methods ................................ ................................ ............................ 98 Plant Materials ................................ ................................ ................................ .. 98 Pollinations ................................ ................................ ................................ 98 Seed germination ................................ ................................ ....................... 99 Flow cytometry ................................ ................................ ........................... 99 DNA extraction ................................ ................................ ........................... 99 Microsatellite analysis ................................ ................................ .............. 100 Results ................................ ................................ ................................ .................. 100 Ploidy Analysis of OP Progeny ................................ ................................ ....... 100 Ploidy distribution in the OP progeny of diploid L. camara ....................... 101 OP progeny of triploid L. camara ................................ ............................. 102 OP progeny of tetraploid L. camara ................................ ......................... 104 OP progeny of pentaploid and hexaploid L. camara ................................ 106 Ploidy Analysis of CP Progeny ................................ ................................ ....... 106 Summary of Ploidy Analysis Results ................................ .............................. 10 9 SSR Marker Analysis ................................ ................................ ..................... 109 Putative 2 n + n progeny ................................ ................................ ........... 110 Putative 2 n + 0 progeny from controlled pollinations ............................... 111 Putative 2 n + 0 progeny of open pollinated pentaploid and hexaploid ..... 114 Putative 2 n + 0 and n + n progeny of diploids and tetraploids ................. 114 Putative 4 n + 0, 2 n + 2 n and 4 n + n progeny ................................ .......... 116 Putative n + 0 progeny ................................ ................................ ............. 118 Twin progeny ................................ ................................ ........................... 119 Discussion ................................ ................................ ................................ ............ 122 Multiple Modes of Reproduction in L. camara ................................ ................ 122 Formation of UFGs and DUFGs ................................ ............................... 123

PAGE 8

8 Apomixis ................................ ................................ ................................ .. 124 Twin seedlings ................................ ................................ ......................... 124 Implications for Lantana Breeding ................................ ................................ .. 125 Evolutionary and Ecological Significance ................................ ....................... 126 6 H YBRIDIZATION P OTENTIAL BETWEEN C ULTIVATED L ANTANA CAMARA AND L ANTANA DEPRESSA ................................ ................................ ................ 140 Rationale ................................ ................................ ................................ ............... 140 Materials and Methods ................................ ................................ .......................... 143 Plant Materials ................................ ................................ ................................ 143 Hand Pollina tion ................................ ................................ ............................. 143 Pollen Staining ................................ ................................ ............................... 144 Seed Germination ................................ ................................ .......................... 144 Progeny Ploidy Analysis ................................ ................................ ................. 144 Results and Discussion ................................ ................................ ......................... 145 L. c amara as the Pollen Source ................................ ................................ ..... 145 Diploid L. camara cultivars ................................ ................................ ....... 146 Triploid L. camara cultivars/breeding lines ................................ ............... 146 Tetraploid L. camara cultivars/breeding lines ................................ ........... 146 L. camara pentaploids and hexaploids ................................ ..................... 147 L. camara as the Female ................................ ................................ ......... 147 Diploid L. camara ................................ ................................ ..................... 148 Triploid L camara ................................ ................................ .................... 148 Tetraploid L. camara ................................ ................................ ................ 149 Pentaploid and hexaploid L. camara ................................ ........................ 150 Ploidy analysis of progeny from L. camara and L. depressa crosses ...... 150 Summary ................................ ................................ ................................ .............. 152 7 DEVELOPING STERILE T RIPLOIDS IN LANTANA CAMARA ............................ 159 Rationale ................................ ................................ ................................ ............... 159 Materials and Meth ods ................................ ................................ .......................... 160 Parental Plant Materials ................................ ................................ ................. 160 Hand Pollination ................................ ................................ ............................. 161 Progeny Gr owing and Evaluation ................................ ................................ ... 161 Statewide trials of promising triploids ................................ ....................... 162 Evaluating male sterility ................................ ................................ ........... 164 Evaluating female sterility ................................ ................................ ........ 164 Evaluating plant performance ................................ ................................ .. 164 Statistical Analysis ................................ ................................ .......................... 165 Results ................................ ................................ ................................ .................. 165 Triploid Generation ................................ ................................ ......................... 165 Selecting parents for interploid crosses ................................ ................... 165 Effects of parental combination on pollination success ............................ 166 Seasonal effects on pollination success rates ................................ .......... 168 Effects of growing temperatures ................................ .............................. 168

PAGE 9

9 Selection of new triploids ................................ ................................ ......... 169 Male Sterility of Promising Triploid Selections ................................ ................ 169 Pollen stain ing ................................ ................................ .......................... 169 Hand pollination ................................ ................................ ....................... 170 Female sterility of promising triploid selections ................................ ........ 170 Plant performance of promising triploid selections ................................ ... 171 Discussion ................................ ................................ ................................ ............ 175 Breeding L. camara For Sterile Cultivar Release ................................ ........... 175 Optimizing environment ................................ ................................ ........... 175 Opti mizing crosses ................................ ................................ ................... 176 Selecting triploids as candidates for releasing ................................ ......... 177 Summary ................................ ................................ ................................ .............. 178 8 CONCLUSIONS ................................ ................................ ................................ ... 190 Rationale ................................ ................................ ................................ ............... 190 Male and Female Fertility of L. camara ................................ ................................ 190 Multiple Modes of Reproduction in L. camara ................................ ....................... 191 L. camara Lantana depressa ................................ .. 192 Developing Sterile Triploid Selections ................................ ................................ .. 193 Future Opportunities for Lantana Breeding ................................ ........................... 193 APPENDIX A F ULL DATASET OF POLLEN STAINING OF ALL LANTANA LINES STAINED .. 195 B B RANCH CUTTINGS FROM B ALM, F LORIDA TRIAL ................................ ......... 198 LIST OF REFERENCES ................................ ................................ ............................. 199 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 210

PAGE 10

10 LIST OF TABLES Table page 2 1 Analysis of variance table for comparing two vital stains used to determine the pollen stainability ................................ ................................ ......................... 45 2 2 Analysis of variance table for pollen stainability of 32 Lantana camara and two Lantana montevidensis cultivars. ................................ ................................ 45 2 3 Ploidy levels, pollen counts, and stainability of 34 lantana l ines detected with cotton blue. ................................ ................................ ................................ ......... 46 2 4 Difference in pollen stainability among ploidy levels of Lantana spp from two seasons. ................................ ................................ ................................ ............. 47 3 1 Analysis of variance table for fruit production of 32 of L. camara cultivars/breeding lines. ................................ ................................ ...................... 66 3 2 Analysis of variance table for seed germination of 32 lines of L. camara cultivars/breeding lines. ................................ ................................ ...................... 66 3 3 Average seed production and female fertility characteristics of 32 L. camara lines. ................................ ................................ ................................ ................... 67 3 4 Average fruit production, seed germination, and female fertility index of 32 L. camara by ploidy level and unreduced female gamete production. .................... 69 3 5 Correlation of female and male fertility traits and statistical probability .............. 69 4 1 Ploidy level, ancestry, source of plant material, and formation of unreduced female gametes in lantana cultivars and breeding lines. ................................ .... 90 4 2 Expected distribution of ploidy levels in the progeny of diploid and tetraploid lantana . ................................ ................................ ................................ .............. 91 4 3 Distribution of ploidy levels in the progeny from self and open pollination (SP and OP). ................................ ................................ ................................ ............. 92 4 4 Distribution of ploidy levels in the progeny of controlled pollin ations grouped by seed parent. ................................ ................................ ................................ .. 93 4 5 Formation of 2 n female gametes in the first and/or second generation progen ................................ ................................ .. 9 4 4 6 performed in February 2009 in Wimauma, FL. ................................ ................... 95 5 1 Polymerase chain reaction cycles to detect differences in L. camara .............. 130

PAGE 11

11 5 2 Primer sequences for SSR markers used to detect differences between parents and progeny of L. camara ................................ ................................ ... 130 5 3 Distribution of the ploidy level of progeny from field open pollination experiments of L. camara commercial cultivars and breeding lines. ................. 131 5 4 SSR marker analysis and ploidy level of parents and progeny from controlled and open pollinations of L. camara cultivars/b reeding lines ............................ 132 6 1 Fruit set on L. depressa flowers pollinated with L. camara .............................. 154 6 2 Differences among L. camara ploidy levels in causing fruit set on L. depressa flowers and seed germination. ................................ ................................ .......... 155 6 3 Fruit set of L. camara flowers pollinated with L. depressa ............................... 156 6 4 Total seed set and germination rates when L. camara was used as a female for L. depressa pollen over two seasons. ................................ ......................... 157 6 5 Ploidy level distribution of crosses between L. camara and L. depressa ......... 157 7 1 Parental lines used in interploid cross pollinations. ................................ .......... 179 7 2 Pollination success rates for eight interploid crosses between two tetraploids and two diploids for triploid generation. ................................ ............................ 179 7 3 Results of full diallel crosses of four lines of Lantana camara indicating the rates of seed production compared to the rates of pollen stainability. .............. 180 7 4 Pollen stainability of 10 new triploid lines, three commercial cultivars and L. depressa ................................ ................................ ................................ .......... 181 7 5 Fruit set of Lantana depressa when nine Lantana camara triploids and three commercial cultivars were used as a pollen source. ................................ ......... 182 7 6 Fruit production per peduncle of 10 Lantana camara triploids and two commercial cultivars with Lantana depressa ................................ ................... 183 7 7 Plant quality ratings of new triploid lines and commercial cultivars with Lantana depressa . ................................ ................................ ........................... 185 7 8 Flower intensity ratings of new triploid lined and commercial cultivars with Lantana depressa ................................ ................................ ............................ 187 A 1 Pollen stainability of all lines screened in Seasons 1 and 2. ............................. 195 B 1 Quantitative data for potential analysis of ornamental characteristics of L. camara for eva luation and selection. ................................ ................................ 198

PAGE 12

12 LIST OF FIGURES Figure page 2 1 Example of stainable (S) and unstainable (U) Lantana pollen grains detected with cotton blue under a bright field miscroscope ................................ .............. 48 2 2 Linear correlation of pollen stainability (%) of eight L. camara and one L. depressa lines detected by two vital stains ................................ ........................ 48 2 3 Stages of normal meiosis in Lantana camara ................................ .................... 49 2 4 Abnormalities observed in Lantana camara meiosis. ................................ ......... 50 3 1 Pictures of L. camara seed peduncles ................................ ............................... 70 4 1 Summary of observed pathways for polyploid formation in Lantana camara ..... 96 4 2 its progeny. ................................ ................................ ................................ ......... 96 5 1 Identification of twins from germination. ................................ ........................... 137 5 2 Microsatelli te analysis of Lantana camara lines to confirm modes of reproduction. ................................ ................................ ................................ ..... 137 5 3 Multiple modes of reproduction demonstrated by L. camara ........................... 138 5 4 Population interactions of L. camara from controlled and open pollination observations. ................................ ................................ ................................ .... 139 6 1 Polynomial relationship between L. camara pollen stainability and fruit set of L. depressa flowers pollinated with L. camara ................................ ................. 158 7 1 Effect of three growth chamber temperatures on the percent fruit set of Lantana camara ................................ ...... 189

PAGE 13

13 LIST OF ABBREVIATION S APHIS Animal P lant H ealth I nspection S ervice CB Cotton b lue vital stain CP Controlled pollination CMS Cytoplasmic male sterility DNA Deoxyribonucleic acid DUFG Double unreduced female gamete production EBN Endosperm balance number principle FDA Fluorescein diacetate stain FDR First division restitution FFI Female fertility index FI Flower ing intensity FLEEPC Florida E xotic P lant P est C ouncil FPP Fruit per peduncle LSD Least significant difference NASS National A griculture S tatistics S ervice NFREC North Florida Research and Education Center NTSYS Numerical taxonomy system computer program OP Open pollination PCA Principal Component analysis PCR Polymerase ch ain reaction PID Percent insect damage PFP Percent fruiting peduncles PQ Plant q uality RA Risk assessment

PAGE 14

14 SAS Statistical A nalysis S oftware SDR Second division restitution SP Self pollinated SSR Simple sequence repeat UFBL University of Florida breeding l ine UFG U nreduced female gamete UMG Unreduced male gamete UPL Unknown Pittsburg line USDA United States D epartment of A griculture

PAGE 15

15 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy GENETIC STERI LIZATION AND REPRODUCTIVE BIOLOGY OF LANTANA CAMARA By David Mark Czarnecki II August 201 1 Chair: Zhanao Deng Cochair: David Clark Major: Horticultur al Science Lantana camara is an important ornamental and landscape plant. Yet, it is a Category I invasive species that that can hybridize with the Florida native species Lantana depressa Sterile cultivars are needed as a preventive measure to control the invasiveness of L. camara This study sought to identify the primary biological factors controlling L. camara L. camara potential with L. depressa and to develop new sterile L. camara cultivars Male fertility was assessed based on pollen stainability. Pollen stainability varied from 1.8% to 81.1 % among 32 L. camara cultivars. Ploidy level was found to be the most important factor determining L. camara pollen stainability. On average, diploids exhibited the high est pollen stainability, followed by tetraploids, pentaploids, hexaploids, and triploids. L. camara cultivars differed considerably in fruit production, ranging from 0.003 to 7.173 fruit per peduncle. Ploidy level, unreduced female gamete (UFG) productio n, and apomixis played significant roles in determining the fruit production capacity of L. camara Triploids not producing UFGs and apomitic seed were most female sterile.

PAGE 16

16 Ploidy and simple sequence repeat marker analyses showed that L. camara could fo rm three types of female gametes and two types of male gametes and develop seed through fertilization or apomixis, leading to six modes of reproduction and resulting in six types of progeny [ n + n n + 0 (haploidization), 2 n + n (sexual polyploidization), 2 n + 0 (clonal seed), 4 n + n (sexual polyploidization), and 4 n + 0 (apomictic polyploidization) ] UFG and apomictic seed production r estore d the female fertility of triploids in L. camara The two traits were not observed in any of the f ive diploid s but were present in three of the six tetraploids evaluated in this study Pollen stainability of L. camara was the most important factor determining the potential of L. camara as a male parent to hybridize with L. depressa D iploid L. camara cultivars were the most compatible with L. depressa Triploid L. camara with p ollen stainability below 10% showed little potential to cross pollinate L. depressa When L. camara was pollinated with L. depressa ploidy level and mode of reproduction of L. camara were the primary factors determining fruit production. UFG producing triploid L. camara was highly crossable as a female with L. depressa whereas non UFG producing triploid L. camara w as the least crossable with L. depressa Interploid y crosses made b etween diploid and tetraploid cultivars/breeding lines not carrying the UFG production and apomixis traits resulted in four triploids that showed high levels of male and female sterility and performed and flowered well in southern, central, and northern F lorida. These triploids have shown potential to be released as new sterile cultivars.

PAGE 17

17 CHAPTER 1 INTRODUCTION Overview and Rationale Floriculture Industry and Concerns about Introduced Plant Species The f loriculture i ndustry is one of the most important sectors of production agriculture in the United States. According to the most recent USDA report, this industry generated a total wholesale value of $3.83 billion ( www.usda.gov ) [ U.S. Department of Agriculture/National Agricultural Statistics Service (USDA/NASS) 2010 ]. Of t he total wholesale value $2.3 billion were from garden/bedding plants and herbaceous perennials. Nationwide, Florida is the second largest producer of bedding plants and flower crops, generating a wholesale value of $696 million annually. More than 20% o f the bedding plants and flower crops used in the United States are produced in Florida. horticulture industry, which was estimated to generate an economic impact of ~$15.2 billion annually and provide 294,179 jobs (Hodges and Haydu 2006). One distinct feature of the floriculture industry is that it produces and uses producers and consumers are constantly looking for new species and cultivars to introduce. Frequently many new introductions come from other states, countries, or continents (Anderson 2004b). Many introductions are primarily made based on their plant, foliage and flower characteristics, i.e. attribu tes closely associated with ornamental values. A major concern about introduced plant species in the floriculture industry is their potential to become weedy, even invasive. Whe n this happens, the introduced species

PAGE 18

18 may cause significant ecological, envir onmental, and economic damages. This may result in high costs to manage or control the introduced species which may far outweigh the benefits the species have brought to the industry and society (Simberloff 1996). Plant breeding has played a critical role in providing new cultivars to fuel the growth of the floriculture industry (and other agricultural and horticultural industries ) (Anderson 2004a). It has been suggested that plant breeding should help address the invasive species issue facing industry by developing and selecting non invasive cultivars (Anderson 2006). Lantana camara Lantana camara is a member of Verbenaceae L. (1753). It originated in Central and South America and the West Indies (Sanders 2001) but European explorers intro duced and spread it to almost all the tropical colonies by 1900 (Howard 1969). Plants of this species produce attractive flowers year round, attract numerous species of pollinators (including at least 24 species of butterflies Schemske 1976; Goulson an d Derwent 2004), tolerate harsh environmental conditions (droughts, pollution, salts, etc.), and have low maintenance requirements. These attributes make lantana an ideal plant for landscape use (Arnold, 2002; Klingaman, 2000; Mungai, 1999; Starman and L ombardini, 2006; Veracion, 1983). The species has even been evaluated for chemical byproducts and biofuel production (Ghisalberti 2000; Prasha 2007; Sahu and Panda, 1998). Chemical byproducts produced have been useful although limited thus far. Additi onally, lantana is easy to produce commercially. The species has perfect flowers borne on a spicate raceme and is self compatible and will set seed in the presence of

PAGE 19

19 various pollinators ( Czarnecki, 200 9; Heywood 1993; Sanders, 2001) Commercial production of the species is done by vegetative stem cuttings which usually root readily. With high ornamental values, L. camara is a very important floricultural crop in many parts of the world, especially in the southern United States (Be aulieu, 2008; Hammer, 2004; Howard, 1969). According to a survey conducted in 2004, 19% of responding nurseries in Florida gr e w lantana commercially, generating an estimated value of $40 million This one species accounted for over one percent of Florida nursery industry plant sales. In Florida, lantana production provides 288 jobs (Wirth et al. 2004) L. camara is a very important crop in a number of other states in the s outhern United States as well. Lantana breeding has been very active in i ncreasing diversity in lantana plant growth form and flower color It was estimated that over 650 varieties have been developed over the last 200 years As of 1969, there were over 600 named varieties These numbers indicate consumer strong interest i n lantana (Day et al., 2003; Howard, 1969; Hammer, 2004) Need To Sterilize Lantana camara L. camara has naturalized through most of the tropical and sub tropical world (Ramey 1999 ; Sanders 2006) and i t has been considered one of the 100 worst weeds in the world (Lowe et al., 2000). The only factor limit ing the spread of this species seems to be temperature. L. camara does not survive extended periods of sub zero (celcius) temperatures. In the United States, L. camara has been found in 14 contiguous states of the south from North Carolina to California. It also grows in Hawaii, Puerto Rico, and the Virgin Islands (USDA NRCS 2011).

PAGE 20

20 In Florida, L. camara is l isted as a C ategory I invasive plant (FLE PPC 2011). Sanders (1987 a ) list ed 60 locations of hybridization between L. camara and th e native L depressa All three varieties of L. depressa var depressa var sanibelensis and var floridana have been shown to successfully hybridize with L. camara (Sanders, 1987) The vast occurrence of mutualist pollinators increases such hybridiztions (Suehs et al., 2006). Such interspecific hybridizations have resulted in genetic contamination of native species (Anderson and Ascher 1994 ; Anderson 2001) L. camara has been shown to be able to out compete native species in natural habitats. It does this by utilizing resources better and by hindering germination and the growth of other plants (Arora and Kohli, 1993; Duggin and Gentle 1998). The authors also found this species to be highly opportunistic and more likely to infest disturbed sites with increased light availability, including cleared land for agricultural, commercial, and residential purposes. These studies or observations show that L camara can di sturb natural areas, hybridize with native plants, and change the ecology of the environment (Hammer 2004 ; Sanders 1987 a) Thus, L. camara has been considered one of the most invasive, persistent, and noxious weeds worldwide. It has been found to be invasive in over 30 different countries and well known with at least 96 common names (Morton, 1994). The invasiveness of L. camara also may threaten the floriculture industry. Because of its popularity and high demands, many nurse ries are engaged in propagating and producing L. camara as described above. If the invasiveness of L. camara is not effectively controlled and L. camara is subsequently banned from

PAGE 21

21 propagation, production and utilization, a substantial economic impact is expected to occur to the industry (Wirth et al., 2004). The spread of this species is also of concern due to its toxicity to animals. Wagstaff (2008) compiled a summary of over 60 reports of lantana causing illness in buffalo, cattle sheep, and goat s Based on the type of animals already affected, it is likely that other ranging animals are also afflicted. Green fruit of L. camara can be toxic to young humans if consum ed (Wolfson and Solomons 1964 ). A number of biological, physical, and chemical control measures have been used to control L. camara (Cillers and Neser, 1991; Day et al., 2003; Graff, 1986; Graff, 1987; Ramey, 1999). However, in many situations these methods are not feasible due to high costs labor requirements or the fact that infestation sites are inaccessible for treatments (Day et al., 2003). In many situations, biological control is the only viable long term solution to managing L. camara In Australia alone, millions of dollars and numerous projects have gone into searching for potential biological agents and introducing them to control L. camara (Thomas et al., 2006; Zalucki et al., 2007). Despite intensive efforts in many countries, biological control of lantana is only partia lly successful, and results are frequentl y poor or unstable (Broughton, 2000; Zalucki et al., 2007). Several native lantana species have overlapping distributions with L. camara (Castillo et al., 2007; Sanders, 1987a, 1987b, 2001). Consequently, the use of biological agents to control L. camara in Florida and the Caribbean region has been very limited. Prolific seed and pollen production is the greatest determinant for L. camara invasiveness. Sexual sterilization of L. camara could eliminate its potent ial to spread

PAGE 22

22 sexual propagules that may directly cause ecological or physical problems and prevent its cross pollination with native Lantana spp Invasive Plants Potential Ecological and Economic Damages Plant species have been introduced for many purpos es such as ornamental use environmental manipulation, and human or animal consumption. I ntroduced plant species may escape cultivation and invade natural, disturbed, and/or agricultural land. Examples of some invasive species that have altered ecosystem s include climbing y ams k udzu melaleuca, and tree of h eaven ( Miller, 2003; Myers, 1983; Rayamajhi et al., 2002). Invasive plant species may cause substantial ecological, environmental, and/or economic damage. For example, invasive plants can reduc e so il fertility (Wardle 1994) and increase light competition that prevents reproduction and inhibits growth of native species (Weihe and Neely 1997 ). Some invasive s pecies may replace native flora with essentially mono cultures of the invasive species. In Galapagos, L. camara out competes a native plant that serves as the habitat of a native bird, ultimately endangering flora and fauna simultaneously (Cronk and Fulle r, 1995). Plant invaders may also usurp local pollinators of showy flowering native species causing lower seed production, cause gene transfer from native species to the invad ing relatives allowing better adaptation to the invading hybrids further aggrava ting the problem (Brown and Mitchell 2001 ; Kandori, 2009 ). Gleditsch and Carlo (2010) have shown that invasive species may provide additional food sources for local fauna. The effects on local wildlife could become a net positive for the environment or cause some species to become more prolific, thus changing the ecology of an area.

PAGE 23

23 Approximately $34.5 billion is either spent or lost annually to invasive plant species (Pimentel et al. 2005). Groups responsible for control measures could invest small pe rcentages of the total cost to stave off future problems and it would still amount to large sums to support preventive research. The economic impact associated with control and eradication of invasive ornamental species cannot be calculated directly. Sin ce the sale of some invasive species with desirable horticultural traits is still occurring forcing nurserymen to stop their sale may be difficult. It is unknown exactly how much eliminating the sale of economically important species from a nursery will a ffect their bottom line and the employees needed to produce those crops (Perrings et al., 2002; Wirth et al., 2004). It has also been suggested that plant value for the horticultural industry may not only be measured by the value of sales but also the cos t to control escaped species (Wirth et al., 2004). Humans have been the primary culprits of invasive plant introduction s (Mack and Erneberg, 2002; Mack and Lonsdale, 2001). O rnamental horticulturists were found to be responsible for over 2 3 0 woody plant species introduced for the landscape trade in the United States that have now naturalized ( Reichard, 1994; Reichard and Hamilton 199 7 ). A study from Australia showed that 70% of the invasive plant species in Australia come from ornamental introductions (Virtue et al., 2004). It is expected that as human populations increase so will invasive plant species (Foxcroft et al. 2008). Managing Invasive Plant s In the event a species takes hold and becomes of ecological or economic detriment, control measures should be taken. The best strategy for invasive species management seems to be proper screening of the materials before release to prevent the problem.

PAGE 24

24 A review of measures taken to control Chinese tallow in Florida is a three step process: 1) removing m ature plants, 2) revisit to remove seedlings and missed plants, 3) introduce biological controls if possible and continued surveillance. The review of t his s pecies suggests monitoring the area until the seed bank (deposited seed in the soil) is no longer a threat (Jubinsky and Anderson, 1996). This example highlights the difficulty in controlling established species and underscores the importance in prevention with sterilization and screening methods. It seems that controlling the flow of invasive mater ials from the production side may be more practical than post establishment control. This is primarily because plant management regimes are rarely effective and economically feasible to control invasive species (Pimentel 2000). Regulatory Measures To dat e the U.S. has a large number of regulations regarding the importation and use of plant materials. An extensive review of federal regulations by Freeman et al. (2009) describes that plant importation has been a continually evolving process. The initial a ttempt to manage invasive plant s began in 1912. The first legislation was the plant quarantine act which gave the federal government the ability to regulate the import and transfer of plant material across state lines and into the country in general. The oversight of plant material was consolidated within the United States Department of Agriculture (USDA) under the Animal and Plant Health Inspection Service (APHIS). The primary function of APHIS is to regulate and control the movement and management of p lant species to prevent exotic species from causing harm to the United States. Since the formation of APHIS, its role has expanded to enforce more rigorous importation standards. The problem of invasive species became so important that in 2004 the Plant

PAGE 25

25 Protection Act of 2000, which generally condensed the agencies involved in invasive plant management, was amended to allow for the Secretary of Agriculture to provide grants for weed control as appropriated by Congress. A parallel means to control these invasive species was made in 2001, when the National Invasive Species Council was formed by executive order. Similar l egislation has led to recommendations in Australia for plant risk assessment (RA) agencies. ts the costs of preventive RA s would be negligible when compared to the costs of controlling newly released invasive species. Studies of the U.S. specifically (Reichard and Hamilton 1997) indicate models of invasion patterns with recommendation protocols to screen plants for horticultural use. These studies suggest that local or regional screening and selection panels of new materials would be best to determine appropriate and low risk plant introductions (Reichard and White 2001) I n Florida invasive p lants are a highly debated subject. To help guide issues related to invasive species an invasive species management agency has been formed called the Florida E xotic P est P lant C ouncil (FLEPPC) This council designates problem species and works to make r ecommendations for control and management (F L EPPC 2011). The council has divided plants into two categories each of which determines a different level of invasiveness. Category I species are plants that have shown the ability to change the structure or ecology of an environment. Species with the ability to cro ss pollinate with native species are also placed in this group. Category II species have been shown to be abundant in the Florida ecosystem but have not yet met the criteria of

PAGE 26

26 category I species (FLEPPC, 2009). Plants in the second category may be chang ed to the first if damage to the ecosystem is demonstrated. This council has also provided several definitions that are useful when describing invasive species. These definitions include: Exotic (a species introduced to Florida), Native (a species natur ally occurring in Florida), Naturalized Exotic (an exotic species species that is expanding its presence in native plant communities). The council has listed over 140 plant species in categor ies I and II for the state of Florida ( FLEPPC 2009). Keller et al. (2007) suggests that boards such as the FLEPPC w ill cause t he percentage of invasive plants introduced to decrease as better management and predictive practic es are implemented to screen new plant intr oductions. Biological Factors Determining Invasive Potential of Plants Seed Production a nd Dispersal The ability to produce and disperse seed is one of the most critical aspects of a species surviva bility The d egree to which a plant is able to accomplish this goal is one of the main factors determining the invasive potential of a species (Dozier 1999). Thus, seed (and fruit) production and seed germination have been the primary criteria in evaluating exotic sp 2010). Pollen Production and Dispersal The ability of exotic plants to produce viable pollen could cause grave problems to native congeneric species. This ability can allow exotic plan ts to out cross with native gene pool, even endanger the native species (Sanders, 1987a). Cross pollinations may

PAGE 27

27 lead to the development of hybrid populations that can sp read over vast distances (Moody and Les, 2007). The absence of viable male gametes would dramatically hybridization likelihood should be conducted to test the ability o f pollen to cause seed set (Olsen and Ranney, 2006 and Trueblood et al., 2010). In addition, pollen mediated gene flow from exotic species may cause native populations to change dramatically. These changes may cause natives to become increasingly prevale nt or cause detrimental effects to local and native populations (Wright 1931). Research about the effects of this type of invasive behavior needs to be conducted since limited information is available currently (Eastham and Sweet 2002) Vegetative Propagules Vegetative propagation is a trait of concern for plant species that are able to produce rhizomes, stolons, or other vegetative propagules as means of plant spread. Determining the vegetative reproduction potential is critical with clonal invasi ve species (B mov et al., 2003). However, in some cases (for example, invasive watermilfoil), it was not clear to what extent the spread of the invasive was due to hybridization events or the vegetative spread of the plant materials (Moody and Les, 2007 ). This shows the importance of assessing multiple traits that might be associated with invasive potential. Assessing Invasive Potential of Plant s A compilation by Anderson (2006) of major invasive characteristics for screening included flowering abilit y, pollinator attractiveness, non dispersed seed, unattractive fruiting structures for consumption, non vigorous plants, lack of seed germination, and the degree of sterility of the plant affecting male and female gamete formation. However, t he most common methods for assessing plant invasive potential include

PAGE 28

28 pollen viability testing for male reproduction, quantifying seed production and viability through open pollination or controlled crosses for female reproduction and characterizing the ability and amount of v egetative spread of a plant. For different invasive species, the primary biological factors determining the Dozier (1999) focused on the female ferti lity of Ardisia crenata T he main aspects Dozier identified for characterization included survivability in the environment in question and its ability to disperse and produce seed. Wilson and Mecca (2003) investigated the female fertility of Ru ellia tweediana by cataloguing the production of seed and testing the viability of that seed. A similar study evaluating invasive potential was conducted on Nandina domestica (Knox and Wilson, 2004). Studies such as these provide a strong basis for recom mend ing whether a species should be considered potentially invasive Assessment of invasive or potentially invasive species should include any tools available to quantify the relative aggressiveness of a species. These assessments will allow for better r ecommendations for effective management of existing populations (Roush and Radosevich 1985). Genetic Approaches to Reduce or Eliminate Invasive Potential of Plants Over the last one to two decades, significant breeding efforts have been undertaken to st eriliz e some of the most important yet invasive ornamental species (Ranney, 2004). These breeding efforts adopt a variety of approaches, including traditional, mutational, and transformational methods Each method ha s its positive and negative attribute s. Anderson (2006) reviewed the most promising avenues of plant sterilization ; breeding and genetic transformation. The method used is highly species dependant An example of this is Hieracium aurantiacum an old ornamental plant that

PAGE 29

29 has been found to be largely clonal (from apomixis) in its distribution across North America. Traditional breeding would be difficult with this species as the plant produces apomictic seed (Loomis and Fishman 2009) H owever genetic transformation may p rovide a new possibility. The methods for sterilization will also depend on the genetic composition of the species and their ultimate value s For m ost ornamentals it would only be practical to sterilize through traditional breeding manipulating existin g sterility inducing genes, incompatibility systems, ploidy levels etc These methods are less controversial and may have greater market acceptance (Anderson 2006). Traditional Breeding C ytoplasmic male sterility (CMS) has been widely used to create ste rility in plants (Duvick 1959). Other breeding methods that have been used for plant sterility include i ncompatibility systems (Culley and Hardiman 2007), sterility genes that control male or female fertility (Singh, 2003) and the production of intersp ecific or wide hybrids Collectively, these sterilization mechanisms are naturally occurring and breeders can manipulate them accordingly. Caution may need to be taken when using some of these breeding methods. T he Callery pear is a good example. It was considered to be sterile due to self incompatibility and non invasive but now it has become invasive because of the addition of new cultivars These new cultivars are cross compatible with the original Callery pear, and these cro ss compatibilities have allowed invasive behavior to occur (Culley and Hardiman 2007) This case illustrates the importance of understanding the interaction s of cultivars developed through traditional breeding.

PAGE 30

30 As described by Anderson (2006), self inc ompatible non invasive cultivars may regain fertility due to some sort of sterility break down, which may allow a previously safe species to now spread freely by back crossing with the progenitor plants. Some breeding approaches can create sterility, but t hey may not always be the most useful for a given breeding system. For example, generating CMS lines in a species can be done by introducing alien cytoplasm (Peohlman and Sleper, 1995). However, maintaining the CMS lines may not always be economically fe asible (Duvick 1959). A breeding dead end may frequently be encountered when new cultivar release s have both mal e and female sterility. Mutational Breeding The premise of mutation breeding is that adding a mutagen (usually radiation or chemical) to pla nt material will cause irreparable harm to the DNA of the plant and will a ffect genes necessary for sexual reproduction or disrupt some other cytological process needed for fertility (Elkonin and Tsvetova, 2008) In general the problems associated with th is process include difficulties achieving high levels of sterility, stabiliz ing the mutation s and obtaining adequate rate s of successful mutation s Transgenic Approaches Transgenic approaches also show promise as a means to control invasive plants (Park et al. 2002). Potentially genetic t ransform ation could generate very high levels of male and/ or female sterility with good stability This potential has been realized in some model plant species either by inserting a gene that can cause sterility or by silen cing a gene that is necessary for male and / or female reproduction A good example of transgenic sterility can be found in petunia. Dotson et al. ( 1996) introduced a gene that caused male and female sterility. Petunia is a model system in which t issue

PAGE 31

31 culture and transformation protocols have been well defined. However, many ornamental plant species do not have efficient tissue cultural protocols and/or transformation protocols, thus increasing the difficulty of applying transgenic approaches. I n addition, transgenic plants are considered highly controversial, and consumer acceptance may become an issue (Anderson 2006, Eastham and Sweet, 2002) Ploidy Manipulation This approach has been used very successfully in a number of fruit and vegetable crops ( Kihara, 1958 ) It is being explored in numerous ornamental plants A major advantage of this approach is that it is generally in expensive to undertake. There have been some cases in which an invasive plant species has been sterilized through ploi dy manipulation and the new sterile forms ha ve been reintroduced to the market as safe alternatives ( Trueblood et al., 2010 ). For example, C atalpa section Catalpa spp breeding incorporated a hybrid approach of mutation and traditional breeding to generat e higher ploidy levels and the hybridization of close relatives (Olsen and Ranney 2006 ) This project produce d highly sterile plants while simultaneously improv ing the ornamental characteristics of the plants. Ploidy manipulation could be a successful and practical means of improving the acceptability of many crops, especially when it is combined with traditional breeding approaches (Anderson 2006) Research Objectives L. camara is a very important ornament al and landscape crop in the United States, especially in Florida. Yet, it is an introduced naturalized exotic species that has disturbed natural and agricultural lands, hybridized with native plants, and caused

PAGE 32

32 ecological damage. If the invasiveness of L. camara is not effectively controlled and L. camara is subsequently banned from propagation, production and utilization, a substantial economic impact is expected to occur. Genetic sterilization has potential as an economical, preventive measure to cont rol the invasiveness in L. camara This research seeks to 1) identify the primary biological factors controlling L. camara and female fertility, 2) assess L camara L. depressa and 3) develop new sterile lantana lines that can be used to replace existing invasive forms. To fulfill these objectives, numerous experiments were conducted, and they are organized into six chapters for this dissertation.

PAGE 33

33 CHAPTER 2 MALE FERTILITY OF L antana camara Rationale The primary concern of ecologists about L. camara L depressa an extremely rare species in Florida (FLEPPC, 201 1 ; Hammer, 2004). Little information is available regarding the pollen viability of L. camara cultivars currently in commercial production. Previous studies of L. camara of plants. Results from those studies indicate a wide range of pollen viability in L. camara (Raghavan and Arora 1960). The main methods used in lantana pollen viability studies have been vital dye based staining, including aniline blue solution (Spies 1984c ; Sanders 1987 b ). Other authors have examined pollen vi ability but did not provide methods (Raghavan and Arora 1960; Khoshoo and Mahal 1967). Brewbaker (1967) observed that lantana pollen grains were binuclete and might have the potential to germinate on artificial media. Several attempts have been made to germinate lantana pollen on artificial media, but so far all in vain (Khaleel 1972; personal observations from variations of Brewbaker and Kwack 1963 ). The cause(s) of such in vitro germination failures remain to be identified. Thus pollen staining ha s been the primary method used in lantana pollen viability assessment. The resultant pollen stainability is considered as a Understanding the relationships between L. camara ity and ploidy (Spies and Stirton 1982ab; Spies 1984a). According to Raghavan and Arora (1960)

PAGE 34

34 and Khoshoo and Mahal (1967), naturalized diploid L. camara could have 0 90% pollen stainability, while triploids and tetraploids could have 16 47% and 16 91% pollen stainability, respectively. As regard s to chromosomal pairing, Spies and Stirton (1982b) detected that diploid and tetraploid L. camara plants maintained the most normal chromosomal pairings during meiosis. Their data showed that 70 100% of meiotic chromosome configurations in diploids were bivalent s and 39 82% of the chromosome configurations in tetraploids were bivalent s In their observation, 80 90% of pollen mother cells in hexaploids yielded normal meiotic products at telophase II. Triploids and pentaploids examined at the same stage exhibited normal meiotic products in 0 50% and 10 20% of the pollen mother cells re spectively. Their data indicated a strong correlation between normal bivalent chromosome pairing to high levels of pollen viability in diploids and triploids and a moderate level of negative correlation between trivalent and pollen viability in tetraploid s. The current study assess ed the male fertility of important commercial cultivars of L. camara to examine the relationships between ploidy level and male fertility and to determine the potential of triploid generation for reducing L. camara otential. Materials and Methods Plant Materials There were 26 commercial cultivars and six University of Florida breeding lines (UFBL) Two sterile accessions of L movevidensis (Henderson 1969) a relative of L. camara were included as a negative contr ol (Table 2 3) L. camara and L. montividensis plants were propagated by cuttings and grown in plastic containers (15.2 cm in diameter) filled with a commercial soilless mix (Fafard 2P mix, Florida Potting

PAGE 35

35 Soil). They were irrigated through drip lines and fertilized by constant feeding with 150 ppm of a 20 20 20 water soluble fertilizer (Gainesville greenhouses, S eason one) or by incorporating a controlled release fertilizer ( Osmocote 15N 3.9P 10K, 5 6 mon ths release at 21 C; The Scotts Company, Marysville, OH) at 6.51 kg m 3 (Gulf Coast Research and Education Center, Balm, FL; S eason two) All plants were grown inside the greenhouse under natural light. The day and night temperature inside the greenhous e s were set at 29/21 C and was maintained below 32 and above 15 C. Ploidy Analysis Analysis was performed using fully expanded young leaves and the Partec PA I ploidy analyzer and the CyStain UV Ploidy Precise P dye (Partec, Mnster, Germany ). The manufac turer recommended ploidy analysis procedure was followed with minor modifications (supplemented with 2% w/v PVP and 0.01% mercaptoethanol) with dye mixture kept on ice The ploidy level of a progeny was determined by comparing to one or more commercial cultivars (reference cultivars) with known ploidy levels that were included in the same analysis. The ploidy levels of the reference cultivars were confirmed by counting and C overnight and pre treated (3:1 methanol : acetic acid) fluid for 2 d, and stored in 70% ethanol at 4 C. Fixed root tips were hydrolyzed in 1 N HCl at 60 C for 5 to 10 min, squashed in acet 0 carmine on glass slides, and observed under a 100 0 x magnification on Olympus BH 2 microscope.

PAGE 36

36 (2 n =2 x =22) (2 n =4 x =44) Pollen Staining Two experiments were conducted to assess L. camara s pollen stainability: the first one in April and May 2008 ( S eason 1) and the second one in November 2009 ( S eason 2). A com parison of two vital stains was comprised of eight commercial Cajun Pink Pink Caprice and Tangerine and L. depressa var depressa The vital stains used to assess the pollen stainability of L. camara were lactophenol cotton blue (CB) (Eng. Scientific, Inc., Clifton, NJ) and fluorescein diacetate (FDA); (Sigma Aldrich, St. Louis, MO). The commercial cotton blue stain solution containe d phenol, glycerol, lactic acid and anailine blue and was ready for use. Flower clusters (inflorescences) were collected when the clusters each had one or more flowers partially open. Predehiscent anthers were removed from unopened flowers of each clus ter and placed in a 1.5 mL Eppendorf tube with ~100 L of cotton blue stain. Anthers were stained overnight at 65C in a water bath and then rinsed three times with distilled water. Care was taken not to burst anthers while rinsing them. Rinsed anthers were squashed in 50 L of 80:20% glycerol:water on a glass slide. Pollen grains were observed under a bright field microscope (Leica MZ16F or Olympus BH 2) using a 40 0 x magnification objective. Photos of pollen grains were taken using Olympus Q Color 5 (O lympus Corporation of America, Center Valley, PA) or a Kodak Easy Share camera (Eastman Kodak, Rochester, NY) modified to attach to the microscope. Images of pollen grains were later viewed and counted on computer.

PAGE 37

37 For FDA staining, the protocol of Helso p Harrison and Helsop Harrison (1970) was followed with minor modifications. Fresh anthers were collected from unopened flowers and stained overnight in a FDA solution containing 10 6 M FDA and 0.25 M sucrose at room temperature (~22C) in the dark. Stai ned anthers were transferred onto a microscope slide and covered with a coverslip. The slide was gently tapped and pressed to release pollen grains out of anthers. Pollen grains were examined using a fluorescence microscope ( Olympus BX 41 ). Uniformly rou nd, non wrinkled, brightly fluorescing pollen grains were considered viable. Non fluorescing or lightly fluorescing and wrinkled or deformed pollen grains were considered non viable. Experimental Designs Lantana plants from which flowers were collected fo r pollen staining were arranged in the greenhouse following a randomized complete block design with four (Season 1) or three (Season 2) replicates (plants). The experimental unit was single plants propagated by cuttings and grown in plastic containers. Fo r each experimental unit, three flower clusters, approximately 3 4 flowers, and 12 anthers were examined. One (S eason 2) to t wo ( S eason 1) slides were prepared f or each experimental unit; three fields on a slide were randomly selected and photographed All pollen grains in a given field were counted (Figure 2 1) Statistical Analysis Pollen stainability data were analyzed using PROC GLM in SAS for Windows 9.2 (SAS Institute Cary, NC) to determine the signific ance of differences among ploidy levels l antana lines, seasons, and stain ing dyes Data t ransformation of stained pollen (%) was performed using the a rcsine s quare root function. To determine differences among lines and ploidy levels seasons were combined. When differences were

PAGE 38

38 significant, mean separation analysis was performed using the p rocedure in SAS Meiotic Observation Entire i mmature flower clusters 1 20 mm in size, were collected from greenhouse grown plants, fixed in a fresh, cold solution of methanol: acetic acid (3:1, v/v) for at least 24 hours, and kept in 20 C, as described in Brando et al (2009). From each inflorescence 12 15 anthers were removed and placed on a glass slide. Anthers were squashed in a small drop of modified carbol fuchsin stain (Kao 1975) anther debris was carefully removed, and a cover slip was placed on the slide At the time when pollen mother cells microspores or pollen grains were released in the initial squash, the cover slip was then sealed with nail polish and then firmly t apped to distribute cells across the slide The prepared slides were observed under an Olympus BH 2 compound microscope. P hotos of pollen mother cells undergoing meiosis were taken using an Olympus Oly 750 microscope camera and the computer software Imag e Pro 6.2. Lantana cultivars examined for meiosis include Athens Rose Lola Miss Huff New Gold Pink Caprice and Red Bandana Plants were provided for m eiotic work to be conducted by Amanda Herschberger in Athens Georgia at the University of Georgia Results and Discussion Ploidy Analysis A mong the 26 commercial cultivars analyzed for ploidy level ; 13 were triploids, six were tetraploids, three pentaploids, and one was hexaploid. Only three cultivars were diploids. This ploidy level distrib ution is quite different from what w as reported previously from wild collect ed accessions (Spies 1984 b ), where tetraploids were more

PAGE 39

39 common than triploids. This difference may reflect t he increased efforts by commercial and private lantana breeders toward producing triploid lantana cultivars. Considering the scarcity at diploid, pentaploid and hexaploid levels, efforts were made t o identify breeding lines at these ploidy levels so that the effects of ploidy level on lantana male sterility could be assessed more adequately Toward this, two additional diploid, pentaploid, and hexaploid breeding lines were added (LAOP 9, LAOP 30, 629 1, 629 2, 620 1, and 621 4) to the 26 commercial cultivars. Dye Comparison for Lantana Pollen Stainability Determination CB and FDA were used in parallel to stain the pollen of nine Lantana lines T he two staining methods revealed similar percentages of pollen stainability or viability in these cultivars. Statistical analysis indicates that the two staining methods were not sign ificantly different (Table 2 1) and there was no interaction between staining methods and lantana lines used. Further, the pollen stainability by CB and the viability by FDA were highly positively correlated, with an R 2 of 0.94 (p< 0.0001) (Figure 2 2). The results indicate that when stain ing lantana pollen, either dye c ould generate similar results The stains were more effective than aceto carmine and aceto orcein which were also used preliminarily FDA staining requires the use of fluorescence microscopes, thus CB was chosen for the rest of the pollen stainability assessments. Pollen Stainability of L. camara Cultivars and Breeding Lines Significant differences were found among pollen stainability of the cultivars (Table 2 3) I n S eason 1, the lowest pollen stainability was 0.8% ( New Gold ) and the highest was 88.7% (LAOP 9), i.e. a 110 fold difference. In Season 2, the lowest and highest pollen stainabilit ies were 1.7% ( New Gold again) and 76.7% ( Lola ), respectively, and ha d a 45 fold difference.

PAGE 40

40 Most of the cultivars/breeding lines (17) had similar ( 5%) pollen stainability between the two seasons. However, the pollen stainability of nine cultivars ( Cream LAOP 9, LAOP 30, Landmark Peach Sunrise Improved Lucky Red Hot Improved Carlos Radiation Cajun Pink and Spreading Sunset ) fluctuated between seasons ( 10 to 20.5%) Additionally, there were six cultivars whose pollen stainability changed by plus or minus 5 to 10%. These fluctuations resulted in an overall significant difference between the two seasons and significant interactions between cultivar and season (Table 2 3). When data w ere pooled by ploidy level, diploids had the highest pollen stainability, an average of 74.8 % and triploids had the lowest, an average of 9.3% (Table 2 4) Pollen stainability of tetraploids was between those of diploids and triploids, an average of 45.1%. Pentaploids had an average of 34.6% pollen stainability, not statistically significant. Hexaploids had even lower stainability, an average of 18%, significantly different from other ploidy levels. These results showed that ploidy level is an important factor determining L. camara Diploids All diploid cultivars and breeding lines had high pollen stainability, ranging from 54.4% to 88.7%. Pollen stainability of Denholm White and Lola was consistent between seasons ( 4.3% to 8.9%), while that of Cream LAOP 9 and LAOP 30 changed more ( 16.8% to 20.5% ) Overall, Lola and its two open pollinated progeny had similar pollen stainability.

PAGE 41

41 Triploids Based on the average pollen stainability over two seasons, triploids could be divided into three groups. The first group had pollen stainabil ity below 10%, which includes nine cultivars, Landmark Pink Dawn Lemon Drop Miss Huff New Gold New Red Lantana Red Butler Red Spread Lantana Samson and Sunset Among them, New Gold and Miss Huff had the lowest pollen stainability, 1.8 % to 1.9%. Their pollen stainability changed little ( 0.1 to 4.9 in percentage) between the two seasons. The second group had pollen stainability between 10% and 20% and consisted of two cultivars, Lucky Red Hot Im proved and Patriot Fire Wagon Another two cultivars ( Athens Rose and Landmark Peach Sunrise Improved ) constituted the third group, whose pollen stainability was between 20% and 30%. Overall larger seasonal variation was observed in Groups 2 and 3 (4.6 to 10.5%) than in Group 1 (0.1 to 4.9%). Tetraploids Three cultivars, Gold Dallas Red and Radiation had similar pollen stainability (26.2% to 32.3%), but overall, a large range of variation in pollen stainability was present among lantana tetr Pink Caprice had an average of 57.4% and 73.5% pollen stainability, respectively, statistically similar to that of the diploids. The average pollen stainability of Carlos was 49.4%, significantly different from diploids Cream and Lola but not significantly different from diploid Denholm White Morphological and molecular marker analysis has shown that Gold is likely an allotetraploid resulting from interspecific hybridization between L. camara and L. depressa while th e rest of the tetraploid cultivars included in this study are autotetraploids (Gong and Deng unpublished) In most plants, allotetraploids from two species with highly divergent genomes behave like diploids in chromosome paring and

PAGE 42

42 assortment during meiosi s, and thus they generally have higher pollen viability or fertility than autotetraploids. It is interesting that Gold a putative allotetraploid exhibited and especially Pink Caprice This may be due to some degree of genome homology that would lead to multivalents reducing successful meiotic products Pentaploids and hexaploids Two of the pentaploid cultivars had pollen stainabilities similar to tetraploids ( Cajun Pink and breeding line 629 2) Th e average pollen stainability of Spreading Sunset was 19.5%, similar to that of hexaploids and some triploids (Groups 2 and 3). Tangerine was the only hexaploid cultivar, and its average pollen stainability was 20.9%, similar to that of Groups 2 and 3 triploids. Two hexaploid breeding lines had similar pollen stainability (11.9% to 21.2%). Meiotic Abnormalities F lower clusters containing all unopen corollas ranging from 0.5 mm to 15 mm in size were examined in an attempt to identify the best stage for meiotic analysis. Fully developed pollen was found in all flower size s examined This indicates that meiosis must have occur red very early in flower developmen t but was variable in most of the sizes screened as most meiotic stages were found. Normal meiotic products were found in diploid and tetraploid cultivars (Figure 2 3) but abnormalities were also found in those same cultivars (Figure 2 4). Based on the cytogenetic analysis by Spies (1984a) there were no individuals with completely pairing sets of chromosomes. It seems likely the abnormalities found in this study are the result of the numerous univalents, trivalents, and quadravalents described in all ploidy levels assessed (Spies 1982b; Spies 1984ab). Chromosomal formations such as these would

PAGE 43

43 likely lead to lagging chromosomes, abnormal telophase, uneven tetrads, and numerous microspores. In addition other observed meiotic dysfunctions were fused microspores, which may be a resu lt of failed telophase II or additional aborted pollen. Meiotic analyses presented here would be the result of suggested chromosomal mispairings by Spies (1982 b ) Previous work has shown a large number of meiotic paring arrangements that could lead to pol len abortion and abnormalities (Spies 1984a). Th e current work demonstrates the results of abnormal meiosis suggested by Spies. It is likely that the cultivars assessed are an amalgam of diverse genotypes similar cytogenetically to the hybrid swarm studi ed by Spies. Summary Male fertility of L. camara as revealed by pollen stainability, varied greatly among cultivars, differing by 45 to 110 fold An important factor determining L. camara male sterility was the ploidy level. This was determined by grouping individuals by ploidy level. On average, diploids exhibited the highest male fertility, followed by tetraploids, pentaploids, and hexaploids. Triploids h ad the lowest pollen stainability or the highest sterility. Thus generating triploids could be an effective genetic approach to reduce L. camara s invasive potential. Significant pollen stainability variation was also found within certain ploidy levels This variation suggests that genetic background or pedigree may also play a significant role in determining L. camara Some of the triploids had pollen stainability approaching 20 30%. Thus new triploids require careful examination and screening to ensure that only highly sterile breeding lines or cultivars are selected. Overall, triploids, especially those with pollen stainability below 10%, showed little variation between growing seasons. The pollen stainability of a number of

PAGE 44

44 cultiva rs varied significantly between seasons, suggesting potential influence of environmental conditions on pollen stainability. This work provides a characterization of the maximum male fertility of commercial materials currently available in the U.S. horticu ltural trade Combined with our meiotic analyses which yielded similar results expected from previous research (Spies and Stirton, 1982b ) indicates that meiotic failures due to triploidy and abnormal pairing greatly reduce pollen viability.

PAGE 45

45 Table 2 1. An alysis of variance table for comparing tw o vital stains used to determine the pollen stainability of eight Lantana camara cultivars and a L. depressa accession Pollen stainability tests were conducted at the University of Florida Gulf Coast Research and Education Center, Balm, Florida in 2010. Source DF F value P value Cultivar Z 8 137.61 < 0 .0001 Stain Y 1 0.06 0.803 C ultivar Z *Stain 8 0.87 0.5508 Z Eight L. camara cultivars were tested and one accession of L. depressa Y The two vital stains tested to determine if a difference existed were Cottone Blue and Fluorescein Diacetate. Table 2 2 Analysis of variance table for pollen stainability of 32 Lantana camara and two Lantana montevidensis cultivars. Fresh pollen was stained with cotton blue at the U niversity of Florida, Gainesville, Florida and University of Florida Gulf Coast Research and Education Center, Balm, Florida in 2008 and 20 09 respectively. Source DF F value P value Ploidy Level Z 5 16 5.36 < 0 .0001 Lines Z 33 74.17 < 0 .0001 Z In total 32 lines of L. camara and two lines of L. montevidensis were included in this study.

PAGE 46

46 Table 2 3. Ploidy levels, p ollen counts, and stainability of 34 lantana lines detected with cotton blue. Season 1 experiment was conducted in 2008 in Gainesville FL and and Season 2 experiment was conducted in 2009 in Balm, FL. L antana spp. cultivars/ lines Ploidy X Season 1 Season 2 Average pollen stainability Y Pollen c ount Pollen stainability Pollen c ount Pollen stainability Cream 2 x 4441 87.1 1.1 649 70.3 0.3 78.7 8.4 a Denholm White 2 x 5116 68.3 1.0 1251 72.6 1.0 70.4 2.2 a c Lola 2 x 3492 85.6 1.8 276 76.7 3.3 81.1 4.4 a LAOP 9 2 x 4089 88.7 1.1 304 69.3 17.3 79.0 9.7 a LAOP 30 2 x 5691 54.4 12.3 768 74.9 9.6 64.6 10.3 a d Athens Rose 3 x 5348 20.8 3.1 694 20.3 3.5 20.5 0.3 g j Landmark Peach Sunrise Z 3 x 4055 21.8 2.8 402 32.3 0.9 27.1 5.2 f i Landmark Pink Dawn 3 x 4621 8.9 3.2 731 4.0 0.6 6.4 2.5 k m Lemon Drop 3 x 4318 5.7 0.9 800 1.7 0.6 3.7 2.0 l m Lucky Red Hot Z 3 x 4812 19.4 0.9 373 9.3 3.0 14.4 5.0 i k 3 x 6414 2.0 0.2 562 1.9 0.8 1.9 0.0 m 3 x 6074 0.8 0.1 260 2.7 2.0 1.8 0.9 m 3 x 5291 5.6 0.8 514 7.0 1.7 6.3 0.7 k m 3 x 4512 19.3 1.0 442 14.7 2.5 17.0 2.3 h j Red Butler 3 x 4911 7.5 1.8 236 4.0 1.7 5.8 1.8 k m Red Spread Lantana 3 x 4588 6.2 0.7 399 5.7 0.3 5.9 0.3 k m Samson Lantana 3 x 5208 6.4 0.4 483 5.2 0.7 5.8 0.6 k m Sunset Lantana 3 x 6465 5.2 0.7 1078 3.1 1.2 4.2 1.1 l m Carlos 4 x 4102 54.5 1.6 1416 44.2 2.4 49.4 5.2 b e Dallas Red 4 x 4234 34.5 2.3 374 29 1.0 31.7 2.8 e h 4 x 4727 31.0 9.6 559 21.4 8.2 26.2 4.8 f i 4 x 3994 55.1 1.2 1272 59.7 3.2 57.4 2.3 a d Pink Caprice 4 x 3227 75.9 2.3 966 71.1 1.9 73.5 2.4 a b 4 x 3961 40.7 11.2 576 23.9 2.7 32.3 8.4 e h 5 x 4241 41.2 2.5 229 23.4 2.0 32.3 8.9 e g 5 x 5423 41.3 4.3 938 42.5 3.1 41.9 0.6 d f 5 x 6609 13.6 2.0 2463 25.4 5.5 19.5 5.9 g j

PAGE 47

47 Table 2 3 Continued. L antana spp. cultivars/lines Ploidy Y Season 1 Season 2 Average pollen stainability Y Pollen c ount Pollen stainability Pollen c ount Pollen stainability 629 1 5 x 3946 33.0 3.1 700 29.5 3.8 31.2 1.8 e h 629 2 5 x 3646 52.9 3.0 1287 43.7 5.6 48.3 4.6 c e Tangerine 6 x 3881 24.5 5.2 285 17.3 3.0 20.9 3.6 g j 620 10 6 x 6668 9.9 0.9 1237 14.0 4.1 11.9 2 j l 621 4 6 x 3389 24.5 2.4 512 17.9 2.7 21.2 3.3 g j L. montevidensis (lavender) 3 x 3308 1.2 0.0 770 0.2 0.2 0.7 0.5 m L. montevidensis (white) 3 x 5966 2.4 0.3 248 0.6 0.6 1.5 0.9 lm Z Re moved Improved from cultivar name Y Lowercase letters indicate statistical groupings with Tukey s W procedure X Differences in ploidy level based on the average pollen stainability of all L antana spp. cultivars/lines. Table 2 4 Difference in pollen stain ability among ploidy levels of Lantana spp from two seasons. Species Ploidy level Lines sampled Pollen counted Lowest stainability Highest stainability Average p ollen stainability (%) Z Standard error L. camara 2 x a 6 26077 64.6 81.1 64.6 a 3.1 3 x d 13 73591 1.8 27.1 9.3 d 2.2 4 x b 6 29408 26.2 73.5 45.1 b 7.5 5 x b 5 29482 19.5 48.3 34.6 b 4.9 6 x c 3 15972 11.9 21.2 18.0 c 3.1 L. montevidensis 3 x e 2 10292 0.7 1.5 1.1 e 0.4 Z Lowercase letters indicate statistical groupings with Tukey s W procedure

PAGE 48

48 Figure 2 1. Example of stainable (S) and unstainable (U) Lantana pollen grains A) D etected with cotton blue under a bright field miscroscop e B ) Detected with f luorescein diacetate under a fluorescence microscope Pollen was collected from L. camara Tangerine Figure 2 2. Linear correlation of pollen stainability (%) of eight L. camara and one L. depressa lines detected by two vital stains, cotton blue and fluorescein diacetate. Pollen stainability of each line revealed by the two stains we re not significantly different (P= 0.803 )

PAGE 49

49 Figure 2 3. Stages of normal meiosis in Lantana camara A) Prophase I B) . D) Prophase II F) Late Anaphase II Lola . I) J Photo credit: Amanda Hershberger, Univeristy of Georgia.

PAGE 50

50 Figure 2 4 Abnormalities observed in Lantana camara me io sis A ) L agging chromosome in T elophase I Lola B ) Lagging chromosome in Telophase II . E ) Pentad formation ( 1 Miss Huff and 2 F ) G ) Uneven pollen size H ) Fused tetrad after tetrad sac rupture Miss Hu ff I ) Variable sized pollen (I 1 Athens Rose and 2 Miss Huff Photo credit: Amanda Hershberger, Univeristy of Georgia.

PAGE 51

51 CHAPTER 3 FEMALE FERTILITY OF LANTANA CAMARA Rationale In plants, the most important factors determining female fertility are seed (or fruit) production and seed germination. In lantana, a seed is borne inside a round, fleshy drupe (berr y ). The fruit is initially green, but turns purple then blue black as the fruit ripe ns L. camara can produce fruit all ye ar round if adequate temperature, moisture and light are available (reviewed by Sharma et al., 2005). Generally each drupe contains one seed that is 0.1 0.2 cm wide. Occasionally, the fruit may contain one additional seed (reviewed by Sharma et al. 2005 ). S everal studies have examined the fruit production of naturalized L. camara plants or seed densities in the soil seed bank under naturalized plants. Significant intraspecific variation seems to exist. An Australian study showed that each lantana infl orescence c ould bear about eight fruit (Barrows, 1976); whereas in India, as many as 25 28 fruits were observed on individual peduncles (inflorescences) (Sharma et al., 2005). An even greater variation has been observed in the density of lantana seed in t he soil seed bank. Reported seed densities ranged from less than 5 to 2,690 seed per square meter (Sharma et al., 2005). These seed have been shown to be able to germinate at any time of the year with sufficient soil moisture, light, and temperature (Gen tle and Duggin, 1997). Collectively little information is available in the literature regarding fruit production capacity and seed germination of commercial lantana cultivars or in U.S production. Several researchers have attempted to understand the relat ionships between ploidy levels and fruit or seed production in L. camara Natarajan and Ahuja ( 1957 ) suggested that ploidy level would be an influencing factor in fruit production as diploid

PAGE 52

52 plants had no seed to good seed production while no triploid s in their study produced any seed. Thirty percent of tetraploid plants were found to not produce seed while the rest ranged from none to good seed production. Two later studies by Raghavan and Arora (1960) and Khoshoo and Mahal (1967) indicated that triploid plants did produce a few seed. Spies (1984c) collected seed produced from all observed ploidy levels in South Africa and found a range in seed production across diploid to pentaploid plants of 0 2,485. These studies i ndicated that tetraploid and diploid plants were the highest seed producers at 856 (4 x ) and 565 (2 x ) seed per plant respectively. The triploid plants were expected to be sterile triploids but produced 342 seed per plant. Very few pentaploid and hexaploid plants were available and only one pentaploid survived the treatment (plants were cut back to the crown) to produce 638 seed on a single plant. Studies from both Australia and India indicated a range of seed germination : 12% in diploids, 28% in triploids, and 56% in tetraploids (Raghavan and Arora 1960; Spies 1984c). An older study (Heit 1946) investigating the best methods for seed germination of L. camara determined the highest average rate of seed germination to be 53% after 40 days with an individual accession reaching as high as 70% after 60 days. Only one individual was sampled from diploid, triploid, and tetraploid L. camara providing limited data (Raghavan and Arora 1960). These studies indicate that ploidy l evel likely ha s a significant effect on fruit production in lantana. The observed wide range of variation in fruit production within each ploidy level group also suggests that other factors such as genes and chromosomal constituents may play significant r

PAGE 53

53 production capacity. The current work outlined here will focus on assessing the variation in fruit production and seed germination of U.S. commercially produced L. camara in controlled field tri a ls and understanding the role of ploidy level variation in influencing L. camara fruit production and seed germination. Materials and Methods Plant Materials This study used the same 32 commercial L. camara cultivars and breeding lines used for pollen stainability work described in Chapter 2. All cultivars /breeding lines were propagated by cuttings on 17 19 September 2007. When plants were ~8 month old (a fter male fertility assessment) they were transplanted to ground beds on 29 May 2008. The beds were treated with Roundup (T he Scotts Company, Marsville, OH) and Image (BASF, Research Triangle Park, NC) and covered with white on black plastic mulch spaced at 1.8 m. The field was irrigated using drip tubes for one hour twice a week Each plant received five grams of a commerc ial control released fertilizer (Osmocote 15N 9P 12K, 5 6 months release at 21 C; The Scotts Company, Marsville, OH). Plants were grown in the ground beds until final plant harvest on 13 14 November 2008. Assessing Fruit Production Commonly L. camara ta ke s about 3 5 weeks from flower opening to produc e ripe fruit. Thus the first fruit collection was ma de 6 weeks after transplanting Fruit collection was then repeated about every 5 weeks until mid November when the air temperature became too low for lantana plants to produce flowers and set fruit regularly. A total of four collections were made during the growing season, on 17 18 July, 25 28 August, 28 30 September, and 4 11 November 2008 During each collection, 20

PAGE 54

54 flower peduncles were randomly ha rvested from an experimental unit (plant) and berries on each peduncle, regardless of maturity, w ere c ounted to calculate the percentage of flower peduncles setting fruit or percent fruiting peduncles (PFP) and the number of fruit per peduncle (FPP) ( Fig. 3 1). In addition, every plant in the study was inspected during each collection to determine if the plant set any fruit to calculate the percentage of plants setting fruit should the 20 flower peduncles collected not bear fruit. After each collection, ripe berries were stored in glycine bags in ambient lab oratory conditions at 22.2 C for subsequent seed extraction ( see below) and green/immature and visibly damaged berries were discarded. During the last collection, all ripe fruit on each plant, in addi tion to those from the 20 randomly harvested peduncles, were collected and stored for seed extraction and germination study. Insect D amage During the first fruit collection insect larvae were found burrowing through flower clusters and developing fruit i n the field Some l arvae were collected reared and sent to the Florida Department of Agriculture and Consumer Services, Division of Plant Industry in Gainesville, FL for species identification Samples were processed according to the guidelines of the Department of Plant Industry (Clemson University, Clemson, SC) Characteristic larvae damage consisted of spiraling grooves on peduncles, blackened depressed areas, and burrowed holes through leaves, flowers, and berries (Fig. 3 1 ) near the growing points of the plant All peduncles collected thereafter were scored for insect damage and the percent insect damaged peduncles (PID) were calculated

PAGE 55

55 Evaluating Seed Germination Seed was extracted on 7 9 January 2009 after fruit collection was completed in Nove mber 2008. Saved ripe berries were macerated manually using a fine metal mesh flour sieve and collected seeds were air dried under the ambient conditions in the laboratory for ~4 weeks. For most of the L. camara cultivars/breeding lines, the number of se ed from each experimental unit was limited, thus seed were combined by cultivar at each collection period for seed germination studies. The bulked seed was divided into three replicat ions Seed w as sown in plastic community trays on the surface of Fafard 2B (Anderson, SC) potting soil on 9 February 2009, and germinated in the greenhouse under intermittent mist. Germination rates were taken e very week for 16 weeks. Calculating Female Fertility Index As shown later in this chapter, L. camara cultivars/b reeding lines varied greatly in fruit production and seed germination. Some of them produced copious amounts of fruit but seed had low germination, while others set fewer fruit but their seed germinated readily. The female reproductive potential of a giv en L. camara would be expected to depend on both its fruit production capacity and seed germination. To take both into consideration, a female fertility index (FFI) was calculated by multiplying fruit per peduncle (FPP) and seed germination percentage. T his index was expected to give a Plant Dry Weight To determine if differences in vigor existed a fter the final fruit collection on 13 14 November 2008, the above ground parts of each plant w ere harvested by cut ting the stems off at the soil line and placing them in to a bag (116 x 95 cm trash bags). Bags

PAGE 56

56 were left open and t he harvested plant materials were dried at 60C in the drying room for 4 weeks before their weight was taken (kg). Experime ntal Design Thirty two L. camara cultivars and breeding lines were arranged in the field in a randomized complete block design. The experimental unit was a single asexually propagated plant. There were four blocks and one plant per cultivar in each block Statistical Analysis FPP and seed germination were analyzed using PROC GLM in SAS for Windows 9.2 (SAS Institute Cary, NC ) to determine the signific ance of differences among ploidy levels l antana lines, and collections Seed germination d ata were t r ansform ed using the a rcsine s quare root function. Seed collection intervals were combined to better determine differences among lines. When differences were significant, mean separation analysis was performed using the LSD Procedure in SAS A principal component analysis (PCA) was performed using NTSYS (NTSYSpc, version 2.2 [Rohlf 2005]) to obtain a graphic representation of relationships among cultivars/breeding lines (Kulakow 1999). Procedurally the data were standardized and then a correla tion was computed. The correlation allowed eigenvalues to be extracted (which determine the variance explained by a principal component). These values were then projected onto a two dimensional graph. Results and Discussion Fruit Production in L. camara Three variables percent fruiting plants, percent fruiting peduncles (PFP), and fruit per peduncle (FPP), were used to assess the fruit production capacity of each L. camara cultivar/breeding line. The percentage of plants setting fruit varied from 6.3% t o

PAGE 57

57 100.0% ; PFP rang ed from 0.3% to 98.8% ; and FPP ranged the highest from 0.003 to 7.173. Analysis of variance indicates that the differences among cultivars/breeding lines in FPP were highly significant (F value = 46.54, P < 0.0001). As described in Chapt er 2, the 26 cultivars and six breeding lines represented six ploidy levels, from diploid to hexaploid. As shown later in Chapters 4 and 5, 11 of the 13 triploids, three of the six tetraploids, all five pentaploids, and all three hexaploids included in thi s study are expected to produce unreduced female gametes (UFGs) (and apomictic seed). For the reasons to be discussed in Chapter 5 and simplicity, these polyploids are to be referred to as having the UFG producer ing trait ANOVA results indicate that bot h the ploidy level and the UFG producing trait played a role in fruit production. Statistical analysis also indicate d that the FPP values were significantly different among four collections (F value = 3.14, P = 0.0252). This was expected based on a preliminary study conducted in 2007 using 139 L. camara lines and their cyclic flowering and fruiting habit. Climatiologic al factors such as temperature and pollinator activity could greatly influence seed production at each collection period. Thus fruit collection was done four times over a period of five months. Seed Germination in L. camara Because of the large differen ces among cultivars/breeding lines in fruit production, the number of seeds available for the germination study varied considerably. No seed w as available for testing the seed germination of Athens Rose and 629 1. One seed was collected from Denholm W hite plants and it germinated. For five cultivars/breeding lines, 4 to 119 seed were processed for sowing, but none of them germinated. Excluding these cultivars/breeding lines, the seed germination percentage

PAGE 58

58 in the remaining 25 cultivars/breeding line s ranged from 9.4% to 57.1%. As expected, statistical analysis indicates that cultivars were significantly different in seed germination (F = 4.44, and P=<0.0001) (Table 3 2). Female Fertility Index (FFI) of L. camara As mentioned above, seeds of seven cul tivars/breeding lines did not germinate, resulting in a FFI of 0. Pink Caprice had the highest FFI, 2.998 (Table 3 3). The remaining 25 cultivars/breeding lines had a FFI between 0.003 and 0.599. In total there were 13 cultivars/lines (three diploids, three triploids, three hexaploids, and four pentaploids) whose FFI were 0.054. Fruit Production, Seed Germination and Female Fertility Index of Diploid L. camara Based on the FFI values, the five diploid cultivars/breeding lines could be separated into f percentage of plants setting fruit (100%) and a high FPP value (0.922) but a rather low germination per centage (16.2%), resulting in a FFI of 0.149. In the second group was LAOP 30, which had a lower FPP value (0.344) but a much higher germination percentage (60.0%), resulting in a similar FFI (0.261). The third group con sisted of 9. Their seeds did not germinate well (~10.0% germination), which led to a low FFI (0.020 or 0.034), although their FPP values were not very low (0.193 lowest FPP value (0.003) among all L. camara cultivars/breeding lines assessed in this study. Only one seed was collected from 303 flower peduncles surveyed. This seed germinated, and the cultivar had a FFI of 0.003.

PAGE 59

59 These results suggest that even with in diploids, L. camara cultivars/breeding lines can vary remarkably in fruit production (FPP from 0.003 to 0.922), seed germination (10.0% to 100%) and FFI (0.003 to 0.261) and certain diploids can be highly sterile. Understanding the genetic mechanism(s ) for the high level of female sterility in these could be very valuable for sterilizing L. camara To determine if a gene is controlling ould require cross pollination with fertil e diploids and assessment of the progeny to determine if the trait is heritable. Alternatively embryology could be described with cytological invesitigation to determine the mechanism causing sterility. Determination of the factors regulating this trait would indicate the ultimate stability of sterility in this line. Fruit Production, Seed Germination and Female Fertility Index of Triploid L. camara Of the thirteen triploid cultivars assessed for fruit production, the majority (11) were found to have the ability to produce UFGs and apomictic seeds (refer to Chapters 4 and 5 for further discussion). Only two triploid cultivars did not have this trait. The two groups of triploids had some differences in seed germination, but th eir most significant differences were in fruit production, and consequently in FFI (Tables 3 1 and 3 2). Non UFG producing t riploids Improved are the only cultivars that belong to this group. A total of two berries were Athens Rose the entire 5 Lucky Red Hot Improved i.e. a FPP value of 0.094. The se eds of this triploid had 11.1% germination. As a

PAGE 60

60 group, non UFG producing triploids had the lowest FPP value (0.05) and the lowest FFI (0.005) and thus were the least fertile in L. camara (Table 3 4). These results suggest that triploidy in combination w ith removal of the UFG trait could result in a high level of female sterility in L. camara U FG producing triploids These triploids were highly prolific in fruit production, with 100% of the plants or 16.0% to 60.7% of the flower s producing fruit for an FPP of 0.175 to 1.379 (Table 3 3). FPP value of 1.232 to 1.379, and produc ed prolific diploid (0.922). As a group, the ir average FPP was 0.236, higher than that of the non UFG producing triploids and 2.5 times higher than that of the diploids (Table 3 4). remaining 10 triploids had a seed g ermination percentage between 18.4% and 57.1%. The average seed germination of these triploids was 29.3%, approximately 264% that of of the non UFG producing triploids. The average FFI of these triploids reached 0.236, 0.231 higher than that of the non UF G producing triploids and 0.142 higher than that of the diploids. The UFG production trait was initially observed only in a number of L. camara tetraploids (Czarnecki and Deng, 2009). It was interesting that parents with this trait were widly used as pare nts and ha s caused UFG production to become wide spread in commercial triploid L. camara cultivars. In one experimental triploid (Czarnecki and Deng, 2009), this trait greatly increased the fruit or seed production capacity. As shown above, similar ferti lity restoring effects are also present in the commercial triploid cultivars. Thus in L. camara triploidy alone may not be able to provide adequate levels

PAGE 61

61 of female sterility. Rather, it would be critical and necess ary to eliminate the UFG production t rait in order to produce highly sterile L. camara cultivars. Fruit Production, Seed Germination and Female Fertility Index of Tetraploid L. camara Non UFG producing tetraploids d producing trait. Their FPP values were 1.870, 0.573, and 1.568, respectively, with the three, collectively, averag ing 1.340. Thus these tetraploids seemed to be more prolific in fruit production than L. camara diploids (group FPP average 0.225) (Table 3 4). The seed germination percentage of these tetraploids was 14.2, 39.1, and 11.8, respectively (Table 3 3). Their average seed germination was 21.7%, which is lower than that of the diploid L. camara (39.3%). As a group, their average FFI was 0.225, close to the average FFI of UFG producing triploids, but 239% higher than the average FFI of diploid L. camara Low fruit production and germination rates were unexpected in this group of plants. UFG producing t etraploids FPP values (1.401 and 1.594) but relatively low seed germination (9.3% and 12.4%), and consequently a moderate FFI (0.130 and 0.198), similar to the FFI of the n on UFG producing tetraploids and many of UFG producing triploids (Table 3 3). Conversely an FFI of 2.998 higher tha n all the other UFG producing triploids, the highest among all L. camara examined. This high FFI was due to its extreme ly high

PAGE 62

62 and ~2.4 fold higher respectively As a group, the average FPP of UFG producing tetraploids was 3.39, indicating that they were the most prolific fruit producers in L. camara (Table 3 4). The average seed germination of UFG producing tetraploids was 21.2%, which is similar to that of non UFG producing tetraploids and UFG producing triploids. The average FFI of UFG producing tetraploids was 1.109, which is at least four times the average FFI of non UFG producing tetraploids and UFG producing triploids and the highest in all groups of L. camara examined. The production of UFGs greatly increases the overall fecundit y of the species and b ecause of this high level of female fertility and their ability to pass the UFG production trait onto progeny (Czarnecki and Deng 2009) this group of plants should be avoided for most breeding purposes. Fruit Production, Seed Germi nation and Female Fertility Index of Pentaploid a nd Hexaploid L. camara 1, and 629 2) produced small amounts of fruit and had FPP values between 0.108 and 0.426 (Table 3 3). None of the seeds extracted from very close to that of non UFG producing triploids. None of its seeds germinated resulting in a FFI of FPP value of 0.906, 15.1% seed germination, and a FFI of 0.137, quite similar to the respective values of many of the diploids. The two hexaploid breeding lines (620 10 a nd 621 4) produced few fruit (0.013 or 0.053 FPP) and none of the few seeds extracted from the fruit germinated, thus their

PAGE 63

63 FFI was 0 (Table 3 that had a moderate FPP value (0.557) but a low seed germination percentage (7.1%), and a low FFI (0.040). The pentaploid and hexaploid plants were of moderate or low fertility and may not pose a significant threat in some environments. Correlation Analysis The strongest positive correlation was found between the F FI and FPP ( r = 0.93916, P <0.0001) (Table 3 5). The correlation between FFI and seed germination was not significant with r = 0.24989, P = 0.1678. This seems to indicate that FPP is of much greater influence to the overall female fertility of L. camara than seed germination. The highest correlation among all data collected was between FPP and the other data points. For this reason it was considered for further analysis. Interestingly insect damage was positively correlated to F P P (0.47797, P = 0.005 7) (T able 3 5). This was likely because insects feed on plants with more fruit rather than the insects caus ing higher seed set. The correlation analysis also indicates significant negative correlation between plant dry weight and pollen stainability (r 0.44461, P = 0.0108). Most likely this correlation was largely due to the fact that diploids had the smallest plant dry weight values and the highest pollen stainability This correlation was expected as diploid L. camara plants often have high pollen f ertility (Spies 1984c ) but are dwarf and small (Sander s 2001). The results of the correlation analysis could be largely influenced by several factors such as genes and ploidy levels. Studies among individuals with similar genetic backgrounds may incre ase the correlary rates among the fertility characteristics.

PAGE 64

64 Principal Component Analysis The process of PCA correlates data points then extracts the most influential data to plot on a graph allowing the relation of individuals or categories to be seen in a graphical representation. This analysis will provide the opportunity for numerous addition al categories of data and controls to be included in fertility assessments. In the PCA plot derived from both the FFI values mentioned above and the pollen sta inability data described in Chapter 2, the 32 L. camara commercial cultivars/breeding lines cluster ed into six groups (Figure 3 2) The largest group consisted of UFG producing triploids that had high FFIs but low pollen stainability. Diploid cultivars that had lower FFIs and high pollen stainability made up the second group. Located near the center of the graph were those with the lowest levels of male and female fertility av ailable in L. camara two non UFG triploids and some of the pentaploids and hexaploids T he most which ha d high levels of stained pollen, high FFP, and high germination rates leading to a high FF I. Two additional groups in the figure are seen among the pollen sterile high seed producing triploids and the pollen fertile low seed producing diploids. Summary The results of this study showed that L. camara cultivars differed considerably in fruit or seed production and seed germination. The difference was particularly obvious in the number of fru it produced per peduncle. P loidy level difference and the UFG producing trait played a very significant role in determining the fruit production capacity of L. camara Triploids without the UFG produc tion trait were most sterile. The UFG producing trait is wide spread in many L. camara cultivars. It is critical and necessary to eliminate this trait to achieve hi gh levels of female sterility in L. camara Results also

PAGE 65

65 showed that there are other genetic mechanisms causing female sterility in L. camara Principal component analysis based on both FFI and pollen stainability provides a useful way to visualize the r eproductive characteristics of different L. camara cultivars/breeding lines. These results indicate that low levels of fertility can be achieved through ploidy manipulation and have explained the lack of correlation between male and female fertility from previous authors.

PAGE 66

66 Table 3 1. Analysis of variance table for fruit production of 32 of L. camara cultivars/breeding lines. Fruit was collected from July through November at five week intervals in 2008 in Balm, FL. Source DF F value P value Cultivar /bree ding lines 31 46.54 <0.0001 Ploidy Level 4 40.97 <0.0001 UFG vs. Non UFG within 3x 1 44.46 <0.0001 UFG vs. Non UFG within 4 x 1 17.65 <0.0001 Table 3 2. Analysis of variance table for seed germination of 32 lines of L. camara cultivars/breeding lines. Seed was collected from July through November at five week intervals in 2008 in Balm, FL. Seed was sown on 9 February 2009. Source DF F value P value Cultivars 31 4.44 <0.0001 Ploidy Level 4 6.3 <0.0001 UFG vs. Non UFG within 3x 1 4.86 0.0292 UFG vs. Non UFG within 4 x 1 1.46 0.2316

PAGE 67

67 Table 3 3. Average seed production and female fertility characteristics of 32 L. camara lines. Cultivar Percent plants producing fruit Percent f lower peduncles producing fruit Flowering peduncles examined Fruit produced by the peduncles Fruit per peduncle (no.) Y Percent flower peduncles damaged by insects Seed g ermination (%) Y Female fertility i ndex W Plant dry w eight (kg) Cream (2x) 93.8 14.7 319 62 0.193 h j 77.3 10.3 i l 0.020 634 Denholm White (2x) 6.3 0.3 303 1 0.003 j 42.9 100 a 0.003 50 LAOP 9 (2x) 75 .0 21.3 191 107 0.435 f j 56.7 60 b 0.034 15 LAOP 30 (2x) 87.5 22 .0 223 93 0.344 g j 60.8 10 k l 0.261 26 Lola (2x) 100 .0 43.6 307 289 0.922 d f 68.4 16.2 f k 0.149 93 Athens Rose (3x) 18.8 0.6 305 2 0.006 j 63.4 0 l X 0.000 728 Landmark Peach Sunrise Z (3x) 100.0 16.0 319 56 0.175 h j 42.1 57.1 b c 0.100 218 Landmark Pink Dawn (3x) 100.0 53.1 318 392 1.232 c e 66.7 48.6 b d 0.599 996 Lemon Drop (3x) 100.0 60.7 315 401 1.270 c e 47.8 9.4 h l 0.119 506 Lucky Red Hot Z (3x) 81.3 9.1 316 30 0.094 h j 71.1 11.1 k l 0.010 167 M is s Huff (3x) 100.0 47.4 321 285 0.890 d f 77.5 18.4 e k 0.164 944 New Gold (3x) 100.0 43.9 319 243 0.763 e g 55.4 26.8 d j 0.205 622 New Red Lantana (3x) 100.0 46.0 318 264 0.832 e g 66.8 24.4 d k 0.203 615 Patriot Fire Wagon (3x) 100.0 44.7 315 255 0.809 e g 62.0 45.9 b e 0.372 517 Red Butler (3x) 100.0 39.0 318 184 0.579 f h 55.2 9.3 h l 0.054 672 Red Spread Lantana (3x) 100.0 33.5 317 164 0.518 f j 56.8 21.6 d k 0.112 687 Samson Lantana (3x) 100.0 58.6 318 436 1.379 b d 71.0 26.8 c i 0.370 647 Sunset Lantana (3x) 100.0 42.2 321 288 0.895 d f 57.9 33.5 c h 0.300 323 Carlos (4x) 100.0 72.3 317 592 1.870 b 73.3 14.2 f j 0.266 80 Dallas Red (4x) 81.3 26.0 312 179 0.573 f h 61.7 39.1 b g 0.224 161 Gold (4x) 100.0 60.0 317 446 1.401 b d 64.6 9.3 h l 0.130 663 Iene (4x) 93.8 68.3 298 478 1.568 b c 63.7 11.8 h k 0.185 101 Pink Caprice (4x) 100.0 98.8 317 2280 7.173 a 85.6 41.8 b f 2.998 967 Radiation (4x) 100.0 53.4 320 510 1.594 b c 71.9 12.4 g k 0.198 431

PAGE 68

68 Table 3 3 Continued. Cultivar Percent plants producing fruit Percent f lower peduncles producing fruit Flowering peduncles examined Fruit produced by the peduncles Fruit per peduncle (no.) Percent flower peduncles damaged by insects Seed g ermination (%) Female fertility i ndex W Plant dry w eight (kg) 629 1 (5x) 56.3 9.7 319 45 0.141 h j 63.3 0 l X 0.000 87 629 2 (5x) 87.5 10.1 317 34 0.108 h j 59.0 0 l 0.000 125 Cajun Pink (5x) 100.0 31.5 317 135 0.426 f j 54.5 0 l 0.000 383 Patriot Hallelujah (5x) 56.3 2.7 322 10 0.030 j 52.6 0 l 0.000 328 Spreading Sunset (5x) 100.0 50.5 317 287 0.906 d f 50.1 15.1 f k 0.137 404 620 10 (6x) 43.8 4.9 305 14 0.053 i j 71.3 0 l 0.000 282 621 4 (6x) 31.3 1.3 320 4 0.013 j 69.6 0 l 0.000 262 Tangerine (6x) 100.0 41.4 314 173 0.557 f i 50.0 7.1 i l 0.040 184 Z Re moved Improved from the cultivar name Y Letters differentiate based on LSD procedure 0.05 X No seed available to sow. T he highest seed germination rate was used from others months when seed was sown. W Index from the multiplicat ion of FPP and seed germination rate. Indicate s unreduced female gamete producer.

PAGE 69

69 Table 3 4. Average fruit production seed germination, and female fertility index of 32 L. camara by ploidy level and unreduced female gamete production. Ploidy Level UFG producers Average FPP Z Low FPP High FPP Low g ermination High g ermination Germination at 16 w eeks Z Female f ertility i ndex 2 x No 0.47 c d (0.38) 0.003 0.922 10 100 Y 24.1 (39.3) a b 0.094 3 x No 0.05 d 0.006 0.094 0 Y 21.6 11.1 b c 0.005 3 x Yes 0.85 c 0.175 1.379 9.3 57.1 29.3 a 0.236 4 x No 1.34 b 0.573 1.87 11.8 39.1 21.7 a b 0.225 4 x Yes 3.39 a 1.401 7.173 9.3 41.8 21.2 a c 1.109 5 x Yes 0.32 d 0.03 0.906 0 Y 15.1 3.8 c 0.027 6 x Yes 0.21 d 0.013 0.557 0 Y 7.1 2.4 c 0.013 Z Letters denote different statistical groupings from the LSD procedure. Y Only one seed was sown. Multiple observations. Number in parenthesis includes Denholm White which was excluded from statistical analysis. Table 3 5. Correlation of female and male fertility traits and statistical probability (p value). Dry weight Plants (%) setting seed Flowers (%) setting fruit Insect damage Fruit per peduncle Germination (November) Pollen staining Z Female fertility index 0.42025 (0.0166) 0.21442 (0.2386) 0.58338 (0.0005) 0.45796 (0.0084) 0.93916 (<.0001) 0.24989 (0.1678) 0.23396 (0.1975) Dry weight 0.28524 (0.1136) 0.41635 (0.0178) 0.35861 (0.0439) 0.4138 (0.0186) 0.01172 (0.9492) 0.44461 (0.0108) Plants setting seed (%) 0.70624 (<.0001) 0.12121 (0.5087) 0.35855 (0.0439) 0.05966 (0.7457) 0.17009 (0.352) Flowers setting fruit (%) 0.31844 (0.0757) 0.78601 (< .0001) 0.06826 (0.7105) 0.05698 (0.7568) Insect damage 0.47797 (0.0057) 0.2346 (0.1962) 0.13397 (0.4648) Fruit per peduncle 0.14744 (0.4206) 0.2129 (0.242) Germination (November) 0.17004 (0.3521) Z Pollen staining data from chapter two

PAGE 70

70 Figure 3 1 Effects of L. camara seed production. A) L. camara seed on peduncle. B) Crocidosema lantana found in seed peduncles of L. camara C ) D amage caused by Crocidosema lantana

PAGE 71

71 Figure 3 2 Principal component analysis of pollen stainability and female fertility index. The x axis and y axis represent D imensions one and two respectively of the principal component analysis.

PAGE 72

72 CHAPTER 4 OC C UR R ENCE OF UNREDUCED FE MALE GAMETES LEADS T O SEXUAL POLYPLOIDIZATION IN LANTANA CAMARA Rationale Lantana is a member of the Verbenaceae and has been widely grown as container plants, hanging basket plants, ground covers, and hedge or accent plants (Beaulie u 2008; Howard, 1969). It is popular in the nursery industry because of easy propagation short production cycle s, interest among gardeners because it attracts butterflies and has toleran ce to drought and poor soil conditions. Nursery production of lantana is widespread, especially in the Southern United States. For example, a survey of the Florida nursery industry, which consists of more than 5,000 nurseries, indicates that 19.0% of the responding nurseries grew lantana and that the annual sales value in Florida alone was over $40 million (Wirth et al., 2004). T he majority of the commercial lantana cultivars belong to Lantana camara L. This species can escape from cultivation through seed dispersal and invade agricultural and natural lands and can hybridize (as pollen donors) with native lantana species. Becaus e of these behaviors, L. camara has been listed as an invasive species in South and Central Florida (FLEPPC, 2007). In several other countries including Australia, India, and South Africa, lantana has been considered as a noxious weed or an invasive speci es (Sharma et al., 2005). Polyploid manipulation, particularly triploid production, has been proposed as a genetic approach to develop sterile, non invasive lantana cultivars (Czarnecki et al., 2008). Similar genetic approaches (polyploid production and selection) are being used to sterilize other ornamental plants for invasiveness control (Ranney, 2004). Polyploids are common in L. camara Triploids, tetraploids, pentaploids, and hexaploids have been reported in cultivated and naturalized L. camara (Cza rnecki et al.,

PAGE 73

73 2008; Natarajan and Ahjua, 1957; Spies and Stirton, 1982 a ). Several lines of evidence suggest that polyploidization may be associated with the species invasive behavior (Sanders, 2001): Tetraploids are rare in the native populations of th is species in tropical America, but very common in the naturalized populations in India, South Africa, and Australia, and overall, tetraploids have courser leaves, grow more vigorously, and set more seeds (Sanders, 2001), whereas diploids tend to be stunte d. Tetraploids have a much wider range of distribution than diploids, and pentaploids are frequently found at high altitudes (Sinha and Sharma, 1984). In plants, natural polyploidization can occur through somatic chromosome doubling or gametic non reduction (Bretagnolle and Thompson, 1995). The former results from mitotic abnormalities in somatic (zygotic and meristematic) cells, while the latter result s from meiotic abnormalities during gamete or gametophyte genesis and the formation of unreduced gametes (pollen and/or eggs). When an unreduced gamete unites with another unreduced gamete (bilateral) or with a normal haploid gamete (unilateral), the unio n leads to sexual polyploidization. To a certain extent, polyploidization via somatic chromosome doubling bears similarity to inbreeding, while polyploidization via unreduced gametes can retain heterozygosity. Thus, the two polyploidization processes can have significant differences in terms of genetic and evolutionary consequences to polyploidized species (Bretagnolle and Thompson, 1995; Hermsen, 1984). Little information is available regarding the origin of polyploids in L. camara except for a report by Khoshoo and Mahal (1967). The authors observed several tetraploids and a pentaploid in the open pollinated (OP) progeny of a triploid and two hexaploids in

PAGE 74

74 the OP progeny of a pentaploid. They inferred that these tetraploid and hexaploid progeny with chromosome numbers higher than their parents must have come from the union of unreduced female gametes (UFGs) with normally reduced pollen. The occurrence of UFGs has been reported in several plant species (Stelly and Peloquin, 1986; Ramsey and Schemske, 1998) On the contrary, unreduced pollen has been reported in numerous plants and seems to be a more common mode of sexual poloploidization in plants (Bretagnolle and Thompson, 1995). Several studies have examined the pollen size and morphology of Lantan a but none of them reported the occurrence of unreduced pollen in L. camara (Raghavan and Arora, 1960; Sanders, 1987 a ). In a recent pollen viability study, we examined tens of thousands of pollen grains and did not notice highly variable viable pollen gr ains within cultivars (D. Czarnecki and Z. Deng, unpublished). During the course of interploidy pollination and triploid generation, we observed pentaploid progeny from a tetraploid by diploid cross, which seems to indicate the occurrence of UFGs in the lantana cultivars used. Therefore, a study was undertaken to (1) confirm the occurrence of UFGs in lantana (2) determine its frequency and distribution in major commercial lantana cultivars, (3) test its transmissibility from generation to generation, an d (4) determine if this trait would affect seed set on lantana triploids Toward these objectives, progeny of commercial cultivars from self pollination (SP) and open pollination (OP) were first analyzed for ploidy levels, followed by controlled pollinati ons among cultivars and breeding lines and ploidy analysis of their progeny. This report presents the results from these pollinations and ploidy analyses. Based on the results, we further propose a model for the origin of the multiple levels of

PAGE 75

75 polyploids in cultivated (and naturalized ) L. camara and discuss the possible mechanisms for the formation of unreduced gametes in lantana and potential implications of this trait for lantana ploidy manipulation, particularly triploid generation and select ion for sterile lantana development. Materials and Methods Plant Materials Ten commercial cultivars and 6 breeding lines were used in this study (Table 4 1). Cuttings or rooted cuttings of these cultivars were generously provided by Dr. Bijan Dehgan, Un iversity of Florida Environmental Horticulture Department (Gainesville, FL) in 2004, and Robrick Nursery (Hawthorne, FL) in 2005 and 2006 respectively. Plants were grown in plastic containers filled with a commercial soilless mix, VerGro container mix A (Verlite Co., Tampa, FL) amended with controlled release fertilizer, Osmocote (15N 3.9P 10K, 8 9 months release at 21 C; The Scotts Company, Marysville, OH) at 7.12 kg m 3 Pollinations Three types of pollination were completed to generate as many seeds ( progeny) as possible for ploidy analysis: self, open, and controlled pollinations (SP, OP, and CP, respectively). In SP, lantana plants in 3 gallon containers were grown in a greenhouse with double doors and a thrips proof screen. Insecticides were appli ed at regular intervals in the greenhouse. As lantana is autogamous (Rambuda and Johnson 2004), hand pollination is not needed for SP. Therefore, to avoid damage to flowers and maximize seed set, flowers were allowed to self pollinate without manual emas culation or hand pollination. In OP, lantana plants in 7 gallon containers were arranged on a raised metal bench in open air. Different cultivars were interspersed to maximize cross

PAGE 76

76 pollination mediated by insect pollinators. Insecticides were not appli ed. Common insect pollinators observed were butterflies, moths, bees and wasps. Controlled pollinations were done on lantana plants grown in the greenhouse. Mature flowers were emasculated by pulling off the entire calyx to which the anthers were attach ed and pollinated immediately using a c amel brush with fresh pollen from flowers that just opened. Between pollinations, the brushes were soaked in 100 percent ethanol to kill residual pollen. Controlled pollinations were done in April June 2006, Augus t December 2006, February July 2008, June November 2008, and February 2009. When lantana fruit (berries) turned dark purple to black and became ripe, they were collected from self, open, or control pollinated flower heads. Seeds were extracted and cleaned within 1 to 2 weeks after harvest, with the exception of the SP seeds of Pink Caprice which were cleaned and stored for over 1 years before sowing. Progeny Growing Seeds were sown on the surface of a peat/vermiculite mix (VerGro container mix A, Verlite, Tampa, FL) and germinated under intermittent mist in a greenhouse. Temperatures in the greenhouse ranged from 16 C (night) to 30 C (day), and no artificial lighting was used. The majority of seeds germinated within four months after sowin g, but some took as long as a year to germinate. After young seedlings had developed true leaves, they were transferred to 12.7 cm plastic containers filled with VerGro container mix. Plants were fertilized by incorporating Osmocote (15N 3.9P 10K) at 7 .12 kg m 3 in the soilless mix.

PAGE 77

77 Ploidy Analysis Analysis was performed using fully expanded young leaves and the Partec PA I ploidy analyzer and the CyStain UV Ploidy Precise P dye (Partec, Germany ). The manufacturer recommended ploidy analysis procedure was followed with minor modifications. The ploidy level of a progeny was determined by comparing to one or more commercial cultivars (reference cultivars) with known ploidy levels that were includ ed in the same analysis. The ploidy levels of the reference cultivars had been confirmed by counting chromosomes in root tip cells. Growing root tips of Cream Gold Pink Caprice and Radiation were collected from rooted cuttings, chilled at 8 C overnight and pre treated with 0.05% colchicine at ambient temperature for 4 hrs, fixed in Carnoy s fluid for 2 d, and stored in 70% ethanol at 4 C. Fixed root tips were hydrolyzed in 1 N HCl at 60 C for 5 to 10 min, squashed in acetic carmine on glass s lides, and observed under a 100x objective on Olympus BH 2 microscope. Chromosome counting results showed that Cream is a diploid and Gold Pink Caprice and Radiation are tetraploids. Results Ploidy Analysis One diploid ( Lola ) and two tetraploi d cultivars ( Pink Caprice and Gold ) were allowed to self pollinate naturally in the greenhouse (without emasculation and hand pollination). More than 900 seed were collected and sown, and more than 500 progeny were analyzed for ploidy levels. All 103 progeny of Lola were diploids, indicating no UFG formation in this cultivar. The majority (88.7% and 93.4%) of the progeny of tetraploid cultivars Pink Caprice and Gold were tetraploids, as expected for normal n (2 x ) gamete formation and fertilization, but 7.6% of Gold SP progeny and 11.3% of

PAGE 78

78 Pink Caprice SP progeny were hexaploids (Table 4 1). The occurrence of these hexaploids indicates that 2n gametes (4 x ) had been formed and fertilized with n gametes (2 x ) f rom normal meiosis and sporegenesis, and the fertilized eggs had developed into embryos (6 x ). Additionally, two diploids were recovered from Gold SP seeds, but no diploids were identified in the SP progeny of Pink Caprice even though much more progen y (311 of Pink Caprice vs. 91 of Gold ) were examined. This is the first time that such ploidy level reduction from a tetraploid to a diploid has been observed in lantana. Understanding the origin of these diploids will require the use of cytological and molecular genetic analysis tools, which are not yet available in lantana. The above results clearly indicate the formation of 2n gametes in Gold and Pink Caprice but do not elucidate which side (maternal or paternal) produced the 2n gametes. Ploid y a nalysis of OP p rogeny To confirm the above findings and to determine if 2n gametes occur in other lantana cultivars, OP seeds (more than 2800 in total) were collected from Lola Pink Caprice and Gold as well as another two diploids ( Cream and Denholm White ) and three more tetraploids ( Carlos Dallas Red and Irene ) grown in a large screen house. If 2n gametes were not present, it would be expected that most OP progeny of a diploid cultivar would be diploid ( from self pollination or cros s pollinat ion by other diploid cultivars) and some progeny would be triploid ( from cross pollinat ion by tetraploid cultivars) (Table 4 2). Similarly, if 2n gametes did not occur, most OP progeny of a tetraploid cultivar would be tetraploids ( from self pol lination or cross pollinat ion by other tetraploid cultivars) and some progeny would be triploids ( from cross pollinat ion by diploid cultivars) (Table 4 2) On the other hand, should 2n gametes have been formed

PAGE 79

79 on the seed parent side, some OP progeny of a diploid cultivar would be expected to be tetraploids (from 2n gametes of the diploid seed parent and the normal gametes of tetraploid pollen parents) and some OP progeny of a tetraploid cultivar would be pentaploids and hexaploids (from 2n gametes of the tetraploid seed parent and the normal n gametes of diploid and tetraploid pollen parents) (Table 4 2) A limited number of seeds (and progeny) were obtained from Cream and Denholm White due to their very low female fertility, especially in Denholm White` where 2472 flowers were pollinated with five pollen sources and 0.0004% seed set was observed. All their OP progeny were diploids, indicating a possible lack of 2n gamete formation in these cultivars (more supporting data below). The ma jority (96.4%) of Lola OP progeny were diploids and 2 out of 55 progeny were triploids (Table 4 3) In an OP environment and with both diploids and tetraploids present, these two triploids could have resulted from normal gamete formation and fertilizati on, or 2n gamete formation and fertilization (Table 4 2) To determine the origin of the two triploids, we examined the leaf and flower morphology of these triploids as well as all the potential diploid and tetraploid parents present at the time when OP w as conducted. The most obvious difference was in flower color: all diploids had yellow, creamy yellow or white flowers, while all tetraploids had pink, magenta or red flowers. Our lantana flower color inheritance studies indicated that crosses among yell ow, creamy yellow or white flowers would result in progeny with shades of yellow to white, and crosses between yellow, creamy yellow or white flowers and pink, magenta or red flowers would result in progeny with shades of pink or red (D. Czarnecki and Z. D eng, unpublished). Flowers of the two OP triploids of Lola were in shades of magenta,

PAGE 80

80 indicating a probable origin from union of normal n ( x ) female gametes of Lola (yellow) with n (2 x ) male gametes of tetraploids (pink or magenta). Thus 2n gametes w ere not involved in the formation of the triploid progeny of Lola Four ploidy levels (3 x 4 x 5 x and 6 x ) were observed in 14 Pink Caprice OP progeny (Table 4 3) The presence of 5 x and 6 x progeny indicates the occurrence of 2n gametes in Pink Caprice (Table 4 2) Further, the presence of 5 x suggests that, more specifically, it was UFGs that led to the production of higher ploidy levels (Table 4 2). The percentage of 5 x and 6 x progeny was 35.7%, a much higher frequency of 2n gamete formation than observed when Pink Caprice was grown in the greenhouse and self pollinated. Formation of 2n gametes was observed again in Gold as indicated by the presence of 11 pentaploids and 4 hexaploids in the 81 Gold OP progeny (Table 4 3). Similarly, t he frequency of 2n gamete formation in Gold was much higher in the OP progeny than in the SP progeny (19.8% vs. 7.7% in the SP progeny). One diploid appeared in Gold OP progeny, and its causes remain to be understood. In Carlos and Dallas Red t he majority of the 77 or 50 OP progeny were tetraploids and a few were triploids. The absence of pentaploids or hexaploids indicates a lack of 2n gamete formation in these tetraploid cultivars. OP seeds were also collected from a commercial cultivar grown at a hotel site in Pittsburg PA (UPL) and its progeny were either tetraploid or hexaploid, suggesting a high frequency (57.7%) of 2n gamete formation in this cultivar (Table 4 3). These results from ploidy analysis of OP progeny confirm the resul ts from SP progeny analysis. They also suggest remarkable differences among lantana cultivars in 2n gamete formation and the non reduction is likely to occur on the maternal side.

PAGE 81

81 Ploidy a nalysis of CP p rogeny A total of over 2000 seed were sown from CP. Carlos (4 x ) produced tetraploid progeny when pollinated with Dallas Red (4 x ) and produced triploids (214) when pollinated with Denholm White (2 x ), Lola (2 x ), or a diploid breeding line LAOP 9 (progeny of Lola ) (Table 4 4). These results support the above mentioned observation in SP and OP progeny Carlos did not produce 2n gametes. It seems reasonable to make a similar conclusion for Dallas Red as it produced only tetraploid progeny when pollinated with Carlos and produced only triploid s when pollinated with Denholm White Lola or LAOP 9 (Table 4 4). When pollinated with three tetraploids ( Carlos Dallas Red and Irene ), Lola produced triploids, and when pollinated with diploid LAOP 9 or selfed, Lola produced diploids (Table 4 4). Cream produced triploids when pollinated with Carlos Dallas Red or Irene which indicates that Cream did not form 2n gametes and supports the observations from SP and OP progeny (Table 4 4). Although 1137 flower s of Gold were pollinated with Carlos or Dallas Red in 2007 and 2008, only 20 seeds were obtained and only 15 germinated for ploidy analysis. The high percentages of hexaploids (60% and 100%) appearing in these progeny indicate high percentages of 2 n gamete formation, more specifically, UFG formation, in Gold (Table 4 4). When Gold was pollinated onto Carlos all the progeny were tetraploids (Table 4 4). This contrasts with the ploidy level distribution in progeny of Gold Carlos and indicates a lack of 2n pollen formation in Gold and a lack of UFG formation in Carlos Pink Caprice produced pentaploids and hexaploids when pollinated with diploid Denholm White or tetraploid Radiation respectively, which supports the for mation of UFGs in Pink Caprice (25% to 100%) (Table 4 4).

PAGE 82

82 Transmission of UFG Formation T o gain understanding of the transmission of this 2n gamete formation trait, progeny of Gold and Pink Caprice were analyzed for ability to form UFGs. Two Gold progeny were selected because of the availability of flowers for pollination. Progeny 604 1 (P604 1, 4 x ) was from a cross pollination between Carlos (non 2n gamete producer) and Gold (UFG producer). P604 1 produced hexaploids at high frequencies in its progeny when self or open pollinated (Table 4 5). This indicates that P604 1 was able to produce 2n gametes and had possibly inherited the trait from Gold Transmission of the tra it from a paternal parent to progeny also suggests that the trait may be u nder nuclear gene control (Figure 4 2). Progeny GDGHOP 36 was a diploid progeny of self pollinated Gold When pollinated with Lola (2 x non 2n female gamete producer), GDGHOP 36 produced triploids only (100%; Table 5), indicating 2n gamete formation in GDGHOP 36, another line of evidence showing the transmission of 2n gamete formation from Gold to its progeny. The transmission of 2n gamete formation from Pink Caprice to its progeny was examined using PCOP 6 (4 x ), a progeny of Pink Caprice and PKGHOP 1 (2 x ), a second generation progeny of Pink Caprice PCOP 6 produced 2n female gametes, as shown by the occurrence of 5 x progeny when pollinated with LAOP 9 (2 x ) (Table 4 5) PKGHOP 1 produced 4 x progeny when pollinated with DROP 25, a 4 x OP progeny of Dallas Red (Table 4 5), indicating that PKGHOP 1 also carried the UFG formation trait from Pink Caprice Effect of 2 n Gamete Formation on Seed Production by Triploids Thre e 3 x breeding lines with or without 2n gamete formation background were pollinated with one common cultivar Lola (2 x ). Breeding lines 605 35 and 624 1 were

PAGE 83

83 from Carlos and Dallas Red respectively, and both did not carry the UFG trait. Their seed set ranged from 0 to 1%. Breeding line GDOP 4 carried the UFG formation trait from Gold and its seed set was 25%. Discussion In the 2 n gamete producing cultivars, the frequency of 2 n gam ete formation varied widely, from 7.7% to 100.0% in Gold and from 11.3% to 100.0% in Pink Caprice (Table 4 1). Similar variation seems common in other plants and has been noted among individuals, flowers of an individual, and even within different ant hers of a single flower. For example, the frequency of 2 n gamete (pollen) formation varied between 4% and 37% among flowers of an individual plant of Medicago sativa (McCoy, 1982) and ranged from 5.6% to 61.7% among anthers of a single flower bud in Solan um (Veilleux et al., 1982). It has been documented that the frequency of 2 n gamete formation in a plant may be influenced by genotype, environment, and their interactions. Temperatures, in particular, seem to have a strong impact on the production of 2 n gametes in Solanum (McHale, 1983). Additional important factors noted in previous studies are variable degrees of penetrance and expressivity in the genes responsible for 2 n gamete formation (Watanabe and Peloquin, 1989; Mok and Peloquin, 1975). In this study, except for the variable frequencies, a cultivar s 2 n gamete formation activity remained quite consistent under different pollination schemes (OP, SP, or CP) and under different growing conditions (shade house and greenhouse): 2 n gamete producers alw ays produced 2 n gametes, and non 2 n gamete producers did not. Mechanisms of UFG Formation in L. camara Two basic processes have been described as the mechanisms for 2n gamete (pollen and egg) formation in plants: first division restitution (FDR) and second division

PAGE 84

84 restitution (SDR) (Bretagnolle and Thompson, 1995). T he most critical step in gamete formation is meiosis. Normal meiosis consists of two successive divisi ons: the first leading to the separation of paired homologous chromosomes and the second leading to the separation of sister chromatids. In FDR, the first meiotic division occurs abnormally, with homologous chromosomes not pairing during Prophase I and/or not separating to opposite poles during Anaphase I. In SDR, the second meiotic division occurs abnormally, with sister chromatids not separating to opposite poles during Anaphase II. Two experimental approaches have been used to deduce the likely mode a nd the relative frequency of each mode in the formation of 2n gametes in plants: observing the cytological abnormalities in the meiotic mother cells and/or segregation analysis of morphological or molecular markers in the progeny (Bretagnolle and Thompson, 1995). Due to its fruit structure (drupe), clear identification of meiotic abnormalities in the ovules of lantana is expected to be very difficult. Morphological and molecular markers are yet to be identified or developed in lantana for segregation anal ysis. However, the ploidy analysis results from the present study and cytological observations from previous studies may be able to provide some indications as to the possible mechanisms underlying UFG formation in lantana. Possible explantaions could be non functional spindle apparatus, or dysfunctional gene(s) regulating meiosis. It has been shown that univalents and multivalents (during Prophase I) and laggards (during Anaphase I) are very common in lantana polyploids (Natarajan and Ahuja, 1957; Spie s and Stirton, 1982 b ). It is expected, therefore, that 2n gametes formed through SDR would be aneuploids and contain various numbers of chromosomes. The progeny of Gold and Pink Caprice resulting from UFGs were

PAGE 85

85 either pentaploids or hexaploids (Table s 4 3 and 4 4). Similarly, the progeny of Red Cap (3 x ) were either tetraploids or pentaploids, and the progeny of Purple Prince (5 x ) were hexaploids (Khoshoo and Mahal, 1967). These results suggest that the UFGs leading to the production of these hig her level polyploids should contain all the chromosome complements in the respective seed parents. This would be possible only through FDR, and the meiotic abnormalities occurred before or during the first division and all the unpaired (univalents) as wel l as paired homologous chromosomes (bivalents, trivalents and quadrivalents) did not separate during anaphase I. To confirm or refute the above inference, we have develop ed SSR (simple sequence repeat) markers that can be used for segregation analysis and for assessing the levels of heterozygosity in the progeny of 2n gamete producing cultivars. This will help determine the stage of non reduction. Occurrence of UFGs and Polyploidization in Lantana In the present study, diploids produced triploids and tetr aploids, and tetraploids produced pentaploids and hexaploids through UFG formation and fertilization with n male gametes (Tables 4 3 and 4 4). In a previous study, a triploid ( Red Cap ) produced tetraploid and pentaploid progeny, and a pentaploid ( Purpl e Prince ) produced hexaploid progeny through the same process (Khoshoo and Mahal, 1967). As summarized in Fig ure 4 1, these results suggests that (1) all the observed polyploid levels in L. camara (triploids to hexaploids) can be evolved through UFG form ation and fertilization with normal n male gametes, and (2) there exist two or more pathways for polyploidization at each of the ploidy levels in L. camara For example, tetraploids could be evolved from three pathways: the union of an UFG (3 x ) of a triploid with a n male gamete ( x ) of a diploid, an UFG (2 x ) of a diploid with the n male gamete (2 x ) of a

PAGE 86

86 tetraploid, or a n female gamete (2 x ) and a n male gamete (2 x ) of a tetraploid. Similarly, there are more than one pathway for the production and evolution of triploids, pentaploids, and hexaploids. It is conceivable that the existence of multiple polyploidization pathways could lead to an increased diversity (or complexity) in L. camara The significance of UFG formation in the evolution of polyploids in L. camara also lies in its potential to quickly generate a series of polyploids within a short period of time. A diploid with UFG forming capability could produce triploid progeny (in generation 1), which could produce tetraploid progeny (in generation 2) that could produce pentaploid and hexaploid progeny in generation 3. That is, all the polyploid levels observed in L. camara from triploids to hexaploids, could emerge in three generations. Considering the short juvenile periods and abilit y to flower year round and propagate vegetatively, once L. camara acquired the ability to produce UFGs, it could produce the observed polyploid series in three generations and in two or three years. With the ability to produce UFGs in the genetic backgrou nd, pentaploids and hexaploids could potentially produce heptaploids and octoploids in the presence of 2 x pollen from tetraploids. Should pentaploid or hexaploid eggs be formed in these pentaploid or hexaploid seed parents, the polar nuclei would be decap loids (10 x ) and dodecaploids (12 x ). If the UFGs and polar nuclei were fertilized by the n male gametes of a tetraploid parent, the ploidy level ratio between the embryo and the endosperm would be 7:12 (for the pentaploid seed parent) or 8:14 (for the hexa ploid parent). It will be interesting to find out if it is possible for fertilized eggs under these genetic conditions would develop into viable embryos, seeds, or individuals at these high ploidy levels.

PAGE 87

87 When this study was initiated, hexaploids were no t available for controlled pollination to address this question. However, previous studies of numerous naturalized populations from India, South Africa, and Australia ha ve not revealed any ploidy levels higher than hexaploids (Natarajan and Ahuja, 1957; R aghavan and Arora, 1960; Spies and Stirton, 1982), which may indicate negative or lethal effects associated with excessively high ploidy levels (heptaploidy or higher) in L. camara So far, the most common polyploids found in commercial cultivars as well as naturalized populations of L. camara have been tetraploids (Czarnecki et al., 2008; Natarajan and Ahuja, 1957; Raghavan and Arora, 1960; Sanders, 1987 a ; Spies and Stirton, 1982 b ). Compared to diploids or triploids, pentaploids and hexaploids, tetraploi ds occupy in the widest habits and show high levels of fertility, suggesting that tetraploidy might be the most adaptable (and aggressive) ploidy level in L. camara (Sanders, 1987 a ). Polyploids are common in other lantana species (such as Lantana involucra ta L.) and some other genera of Verbenaceae (such as Duranta L., Lippia L., Stachytarpheta Vahl, and Verbena L.) (Sanders, 2001). A good understanding of the mechanism and pathways for polyploidization in L. camara may shed light on the emergence and evol ution of polyploids and speciation in these highly complex species as well. UFGs and Seed Set in Polyploids One interesting phenomenon in L. camara has been the lack of correlation between meiotic irregularities during microsporegenesis and seed set in naturalized polyploids. In spite of high frequencies of meiotic irregularities in pollen mother cells, lantana polyploids produced abundant viable se eds (Natarajan and Ahuja, 1956; Raghavan and Arora, 1960). For example, the majority of tetraploids collected from different parts of India showed 2 to 3 quadrivalents, 2 to 6 trivalents, and 4 to 12

PAGE 88

88 univalents in pollen mother cells, but more than half o f these tetraploids produced the largest numbers of seeds, compared to diploids (Natarajan and Ahuja, 1957; Spies 1984 b ). Khoshoo and Mahal (1967) suggested that apoximis was responsible for the seed set in these highly male sterile polyploids. Spies and Stirton (1982 c ) examined embryo sac development in L. camara from South Africa, but did not find definitive evidence for apomixis. L antana produc ing UFGs provide s an explanation for such a lack of correlation in polyploids. Without undergoing the normal meiotic first division during megasporegenesis, polyploids, even though showing high levels of meiotic abnormalities during microsporegenesis in their pollen mother cells, could still produce megaspores and egg cells that contain all the chromosome comple ments, which would be viable and able to be fertilized and produce viable seeds. Effect of 2 n Gamete Formation on Seed Production by Triploids Sterile cultivars are needed to control the invasive potentials L. camara has shown in the Southern United States and other countries. Compared to diploids and tetraploids, triploid lantana plants show the lowest pollen stainability and seed set (Czarnecki et al., 2008). Because of this and the general high level of male and female sterility triploids have expresse d in banana, citrus and watermelon, production of triploids has been proposed as a main genetic approach to develop new sterile lantana cultivars. Results from the present study suggest the possibility of producing triploids through crosses between diploi ds and tetraploids, and through crosses between diploids as well (Fig ure 4 1). More importantly, the results indicate a strong need to screen and select breeding parents carefully and to avoid using UFG producing plants in crosses intended for high steril ity. If triploids have inherited the ability to form UFGs from either seed or pollen parents, these triploids will be likely to produce viable UFGs

PAGE 89

89 (3 x ), which can be fertilized with pollen from surrounding diploids or tetraploids and produce viable seeds which will defy the purpose of producing such triploids. None of the diploids ( Cream Denholm White and Lola ) in this study showed tendency to produce UFGs, but three out of the six tetraploids ( Gold Pink Caprice ) showed such tendency. To f ind more tetraploids that do not produce UFGs and can be used as breeding parents, screening of more commercial cultivars or germplasm and progeny through ploidy analysis will be required.

PAGE 90

90 Table 4 1. Ploidy level, ancestry, source of plant material, and formation of unreduced female gametes in lantana cultivars and breeding lines. Cultivars / breeding lines Ploidy level Ancestry Sources of plant material 2 n female gamete formation 2 n female gametes (%) Carlos 4x Unknown Commercial No Cream 2x Unknown Commercial No Dallas Red 4x Unknown Commercial No Denholm White 2x Unknown Commercial No Gold 4x Unknown Commercial Yes 7.7 to 100 Irene 4x Unknown Commercial No Lola 2x Unknown Commercial No Pink Caprice 4x Unknown Commercial Yes 11.3 to 100 Radiation 4x Unknown Commercial Yes 50.0 UPL Z 4x Unknown This study Yes 42.3 DROP 25 4x OP progeny of Dallas Red This study No GDGHOP 36 2x OP progeny of Gold This study Yes 100 GDOP 4 3x OP progeny of Gold This study Yes 100 LAOP 9 2x OP progeny of Lola This study No PCOP 6 4x OP progeny of Pink Caprice This study Yes 100 PKGHOP 1 2x Second generation OP progeny of Pink Caprice This study Yes 100 P604 1 4x Progeny of Carlos and Gold This study Yes 57.1 to 100 Z Unknown Pittsburg lantana

PAGE 91

91 Table 4 2. Expected distribution of ploidy levels in the progeny of diploid and tetraploid lantana with or without unreduced female gametes and/or unreduced pollen. Seed parents: ploidy level P ollen parents: ploidy level Pollination schemes E x pected ploidy levels in progeny of diploid and tetraploid parents with and/or without 2 n gamete formation during sporogenesis n gametes (female and male) 2 n female gametes (unilateral) 2 n male gametes (unilateral) 2 n female x 2 n male gametes (bilateral) 2 x 3 x 4 x 3 x 4 x 5 x 6 x 3 x 4 x 5 x 6 x 4 x 6 x 8 x 2 x 2 x SP, CP z yes yes yes yes 4 x CP yes yes yes yes 2 x and 4 x OP yes yes yes yes yes yes yes yes 4 x 2 x CP yes yes yes yes 4 x SP, CP Yes yes yes 2 x and 4 x OP yes yes yes Yes yes yes yes yes z CP, OP, SP: controlled, open, and self pollination, respectively.

PAGE 92

92 Table 4 3. Distribution of ploidy levels in the progeny from self and open pollination (SP and OP). Pollination was performed in 2006 and 2007, and ploidy analysis done in 2007 and 2008, Wimauma, FL. Seed parent (ploidy level) Pollination scheme Seeds collected & sown Progeny available for ploidy analysis Progeny in ploidy levels (no.) Polyploids indicating 2 n gamete formation Percentage of 2 n gametes 2 x 3 x 4 x 5 x 6 x Lola (2 x ) SP 263 103 103 3 x 4 x 0 Pink Caprice (4 x ) SP 403 311 276 35 6 x 11.3 Gold (4 x ) SP 256 91 2 84 5 6 x 7.7 Carlos (4 x ) SP 38 15 15 6 x 0 Cream (2 x ) OP 56 9 9 4 x 0 Lola (2 x ) OP 335 55 53 2 4 x 0 Pink Caprice (4 x ) OP 403 14 2 7 4 1 5 x 6 x 35.7 Gold (4 x ) OP 758 81 1 2 63 11 4 5 x 6 x 19.8 Carlos (4 x ) OP 639 77 9 68 5 x 6 x 0 Dallas Red (4 x ) OP 254 50 3 47 5 x 6 x 0 Irene (4 x ) OP 355 52 1 51 5 x 6 x 0 UPL (4 x ) OP 65 26 11 15 6 x 42.3

PAGE 93

93 Table 4 4. Distribution of ploidy levels in the progeny of controlled pollinations grouped by seed parent. Pollinations were made in 2005, 2006 and 2007, and ploidy analysis done in 2008, Wimauma, FL. Seed parent Pollen parent Flowers pollinated Progeny available for ploidy analysis Progeny in ploidy levels (no.) Polyploids indicating 2 n gamete formation Percentage of 2 n gametes 2 x 3 x 4 x 5 x 6 x Carlos (4 x ) Dallas Red (4 x ) 2202 31 31 6 x 8 x 0 Carlos (4 x ) Denholm White (2 x ) 731 23 23 4 x 5 x 6 x 0 Carlos (4 x ) LAOP 9 (2 x ) 1259 56 56 4 x 5 x 6 x 0 Carlos (4 x ) Lola (2 x ) 3656 135 135 4 x 5 x 6 x 0 Dallas Red (4 x ) Carlos (4 x ) 2051 19 19 6 x 8 x 0 Dallas Red (4 x ) Denholm White (2 x ) 468 5 5 4 x 5 x 6 x 0 Dallas Red (4 x ) LAOP 9 (2 x ) 1653 28 28 4 x 5 x 6 x 0 Lola (2 x ) LAOP 9 (2 x ) 1894 100 100 3 x 4 x 0 Lola (2 x ) Lola (2 x ) 567 18 18 3 x 4 x 0 Lola (2 x ) Carlos (4 x ) 2215 16 16 4 x 5 x 6 x 0 Lola (2 x ) Dallas Red (4 x ) 2230 5 5 4 x 5 x 6 x 0 Lola (2 x ) Irene (4 x ) 564 4 4 4 x 5 x 6 x 0 Cream (2 x ) Carlos (4 x ) 769 7 7 4 x 5 x 6 x 0 Cream (2 x ) Dallas Red (4 x ) 995 4 4 4 x 5 x 6 x 0 Cream (2 x ) Irene (4 x ) 472 3 3 4 x 5 x 6 x 0 Gold (4 x ) Carlos (4 x ) 469 10 4 6 6 x 8 x 60.0 Gold (4 x ) Dallas Red (4 x ) 668 5 5 6 x 8 x 100.0 Carlos (4 x ) Gold (4 x ) 450 5 5 6 x 8 x 0 (4 x ) Denholm White (2 x ) 213 2 2 5 x 6 x 100.0 Pink Caprice (4 x ) Radiation (4 x ) 273 4 3 1 6 x 8 x 25.0 LAOP 9 (2 x ) Carlos (4 x ) 1515 31 31 4 x 5 x 6 x 0 LAOP 9 (2 x ) Dallas Red (4 x ) 1326 44 44 4 x 5 x 6 x 0

PAGE 94

94 Table 4 5. Formation of 2 n female gametes in the first and/or second generation progeny of Gold and Pink Caprice Pollinations were made in 2006 to 2008, and ploidy analysis done in 2008, Wimauma, FL. Seed parents (ploidy level) Pollen parents (ploidy level) Flowers pollinated Progeny available for ploidy analysis Progeny in ploidy levels (no.) Polyploids indicating 2 n gamete formation Percentage 2 n gametes 2 x 3 x 4 x 5 x 6 x P604 1 (4 x ) Self pollinated 70 2 2 6 x 100.0 P604 1 (4 x ) Self pollinated 27 16 11 6 x 59.3 P604 1 (4 x ) Open pollinated 7 3 4 6 x 57.1 GDGHOP 36 (2 x ) Lola (2 x ) 265 21 21 3 x 100.0 PCOP 6 (4 x ) LAOP 9 (2 x ) 158 2 2 5 x 100.0 PKGHOP 1 (2 x ) DROP 25 (4 x ) ~80 2 2 4 x 100.0 PKGHOP 1 (2 x ) LAOP 9 (2 x ) ~80 1 1 3 x 100.0

PAGE 95

95 Table 4 6. Seed set of three lantana triploids pollinated with Lola Hand pollination was performed in February 2009 in Wimauma, FL. Triploid breeding lines UFG background Flowers pollinated Seed set standard error (%) Germination 12 weeks after sowing (%) GDOP 4 Yes 188 25 .0 5 .0 26 605 35 No 160 0 .0 0.0 0 624 1 No 222 1 .0 1 .0 0

PAGE 96

96 Figure 4 1. Summary of observed pathways for polyploid formation in Lantana camara Solid double lines represent polyploidization pathways involving 2 n female gamete and observed in this study; dashed double lines are polyploidization pathways involving 2 n female gamete formation and observed by Khooshoo and Mahal (1967); and solid single lines are observed pathways not involving 2 n female gamete formati on. Symbol indicates parents producing 2 n female gametes. Figure 4 2. Transmission of the unreduced female gamete formation trait from Gold to its progeny. Gold (4 x ) was the pollen parent of breeding line P604 1, which produced tetraploid (40 .7%) and hexaploid (59.3%) progeny when self pollinated. Carlos was the seed parent of P604 1. Symbol indicates unreduced female gamete production.

PAGE 97

97 CHAPTER 5 MULTIPLE MODES OF RE PRODUCTION AS REVEAL ED BY PLOIDY AND MICROSATELLITE MARKE R ANALYSIS OF LANTANA CAMARA Rationale In evaluating female fertility of L. camara produce abundant fruit and seed (Chapter 3 ). It was suspected initially that mos t of the collected seeds (from OP ) would have been developed from fertilized UFGs, because previous results had indicated that UFG production could result in high levels of female fertility in lantana triploids (Chapter 4). How ever, many of the progeny had a ploidy level identical to their maternal parents (triploids). The frequent occurrence of this type of triploid raised a question about the possibility of apomixis in L. camara A l iterature search revealed that attempts t o understand L. camara biology could trace back to the 1960s. Raghavan and Arora (1960) noted lantana progeny that looked identical to maternal plants in morpholog y K hoshoo and Mahal (1967) to o describe d a series of reproductive anomalies in eight accessions of L. camara They suspected that unreduced eggs and apomixis had led to the polyploid series found in naturalized populations. A t th at time, only morphological and cytological data could be analyzed to understand the relationships b etween parents and progeny. Nonetheless, facultative or obligate apomixis was first proposed to occur in diploid, triploid, tetraploid, and pentaploid lantana plants. Spies and Stirton (1982a) attempted to determine if apomixis was occurring in lantana. They found that the conditions for apomixis were present but they were not able to confirm this phenomenon. Their study suggested that apospory (somatic cell s form ing gametes) could be occurring. Recently the occurrence of UFGs was confirmed to exist in U.S. commercial cultivars of L. camara by conducting extensive ploidy analysis of progeny from OP and

PAGE 98

98 controlled pollinations (CP) (Czarnecki and Deng 2009). The significance of UFG production comes from the ability to increase the ploidy level rapidly from one generation to the next and to restore female fertility in triploid plants (Czarnecki and Deng 2009; Chapter 4 ). Ploidy analysis led to the discovery of the occurrence of triploids from the offspring of triploids and the possible occurrence of ap omixis. However, it was not possible to provide conclusive evidence. To facilitate understanding of the reproductive biology of L. camara simple sequence repeat (SSR) markers were developed (L. Gong and Z. Deng, unpublished). This chapter focuses on an alyzing the ploidy levels and SSR marker banding patterns of L. camara s progeny from CP or OP. The objectives were to determine 1) the likely stage for UFGs formation 2) the effect of UFGs on female reproduction, and 3) other possible modes of reproduction. Materials and Methods Plant Materials All OP materials for this study were produced from a previous experiment (Chapter 3) to determine the fruit production of L. camara The seedlings and parent s from this experiment were transplanted to 15 .4 cm pots on 12 May 2009 in Fafard 2B potting mix supplemented with slow release O smocote (15N 3.9P 10K, 8 9 months release at 21 C; The Scotts Company, Marysville, OH) at 6.51 kg m 3 Pollinations Pollinations were made from March 2006 through August 2009. Fresh pollen was collected and placed on the emasculated stigmas. Open pollinated seed were collected at five week intervals from field grown plants when the seed was ripe (Chapter 3)

PAGE 99

99 Seed g ermination Seed were collected when ripe and stored un til bulked, cleaned and sown between March 2006 and February 2010. Seed from c ontrolled crosses were sown in individual cell trays (Figure 5 1) and OP seed collected from the field were sown in community trays. Seed were sown on the surface of Fafard 2B potting soil and germinated in the greenhouse Flow c ytometry Parents and progeny from 10 controlled crosses and 15 commercially available cultivars were screened for flow cytometric analysis. Also, i ndividuals of i nterest from previous studies (Chapt ers 3 and 4) were kept for further molecular analysis. Analysis was performed using fully expanded young leaves and the Partec PA I ploidy analyzer and the CyStain UV Ploidy Precise P dye (Partec, Germany). The manufacturer recommended ploidy analysis pr ocedure was followed with minor modifications (supplemented with 2% w/v PVP and 0.01% mercaptoethanol) with dye mixture kept on ice The ploidy level of a progeny was determined by comparing to one or more commercial cultivars (reference cultivars) with known ploidy levels that were included in the same analysis. Base d on flow cytometry analysis materials were divided into six groups for analysis: apomeiosis, apomixis, double unreduced female gametes (DUFG), haploids, twins (Figure 5 1), and unreduced m ale gametes (UMG). DNA e xtraction Fresh tender leaves were collected and dessicated in a dark box for 3 7 days with silicon gel beads. The dried leaf tissue was used for DNA extraction by first desiccating the tissue The extraction protocol of Fulton et al. (1995) was used Extracted DNA was dissolved in TE buffer, and DNA concentrations were determined with a N D 1000

PAGE 100

100 (Thermo Scientific, Wilmington, DE USA ) DNA concentrations were adjusted to 8 ng/ L for PCR reactions. Microsatellite a nalysis Polym erase chain reactions (PCR) w ere done on an Eppendorf Vapo protect Mastercycler Pro 384. Each reaction (10 L ) contained 1 x PCR reaction buffer (New England Biolabs, Ipswich, MA USA), 1.5 mM of MgCl2, 2 mM dNTPs, 0.25 pmol of the forward primer with an 2.5 pmol of the reverse primer 0.25 unists of Taq DNA polymerase, and 8 g of genomic DNA Fluorescent labeling was done with 2.25 pmol of IRD700 and IRD800 infrared dye labeled M13 tail primer (MWG Biotech, Highpoint, NC USA). Primer sequences and PCR cycles used are listed in tables 5 1 and 5 2. PCR p roducts were separated on a Li cor 4300 DNA analyzer after denaturing at 95 C for 5 minutes. PCR products were diluted with formamide loading buffer at a 1:1 ratio and 0 .8 L of each sample was loaded into each well. Conditions for gel electrophoresis were 1500 V, 40mA at 45 C for 90 minutes after a 25 minute prerun at the same conditions Gels were composed 6.5% Li cor KB Plus Gel Matrix (20 m L ) with, 10% APS solution (150 L ), and TEMED (15 L ). B and ing results were scored as absent (0) or present (1). Each group of plants was compared to determine similarity to the parents. Results Ploidy Analysis of OP Progeny Seed germination experiments in C hapter 3 resulted in 1,517 progeny from 22 L. camara cultivars and two breeding lines. As plants of these cultivars were grown in the open field, their progeny were expected to have resulted from open pollinations. Due to the wide differences among cultivars/lines in fruit production and seed germination

PAGE 101

101 (refer to Chapter 3), the number of progeny available for ploidy analysis varied considerably, 1 to 8 for five cultivars/lines, 13 to 83 for 16 cultivars, 137 to 449 for three cultivars, and more than 1,000 for one cultivar Ploidy distribution in the OP progeny of diploid L. camara and the other a triploid (Table 5 3) In Chapter 4 and a previous report (Czarneck and Deng, 2009), progeny after controlled pollination with tetraploids. The controlled pollination involved more than 2,000 Cream flowers. It was concluded in Chapter 4 that Cream was not a UFG pro ducer. Taking this into consideration, it is likely that the diploid would have resulted from self pollination or cross pollination with another diploid, and that the triploid would have resulted from a cross pollination with a neighboring tetraploid grow n in the field. Ten of the 13 OP progeny of Lola were diploids one was triploid and one was (presumably from hexaploid pollen) tetraploid (Table 5 3) Previously Lola produced diploid progeny after controlled pollination with diploids and produced t riploid progeny after pollination with tetraploids, but did not produce tetraploid progeny (Chapter 4; Czarnecki and Deng, 2009). The previous analysis involved nearly 7,500 Lola flowers and nearly 200 progeny. Considering the diploidy of Lola and ot her possible modes of reproduction to be discussed later in this chapter, possible cause(s) for the observed tetraploid include 1) fertilization of a Lola UFG (2 x ) with a 2 x male gamete from a neighboring tetraploid, 2) the formation of a doubled UFG by Lola followed by apomictic seed development, and 3) fertilization of a normal reduced female gamete (1 x ) of Lola with a reduced male gamete (3 x ) of a neighboring hexaploid.

PAGE 102

102 One progeny was recovered from LAOP 9 OP seeds, and it was a diploid. Similarly the three OP progeny of LAOP 9, were diploids. Since LAOP 9 was a non UFG producer, as demonstrated by controlled pollinations and ploidy analysis, the diploid observed here would likely be a result of self pollination or cross pollination with a neighboring diploid. The same seems to be applicable to the diploid OP progeny of LAOP 30. OP progeny of triploid L. camara Only one progeny was recovered from Lucky Red Hot Improved OP seeds, and it was f ound to be aneuploid (3 4 x ) (Table 5 3) No progeny w ere recovered from the seeds of Athens Rose Except for these two triploid cultivars, appreciable numbers (14 to 154) of OP progeny were recovered for each of the remaining 11 triploid cultivars. Four ploidy levels (3 x 4 x 5 x and 6 x ) were observed in the 14 OP progeny of Improved x 4 x an x 4 x or 5 x and occasionally some aneuploid progeny (3 4 x 4 5 x or 5 6 x ). ere very diverse in ploidy level, from 2 x to 6 x plus aneuploids. For most triploid cultivars (9 out of 11), 20.5% to 38.7% of their OP progeny were triploids, 55.4% to 76.9% tetraploids, and 1.3% to 10.8% pent aploids. In two triploid cultivars, 54.2% to 57.1% of OP progeny were triploids, 21.4% to 37.5% were tetraploids, and 8.3% to14.3% were pentaploids. Hexaploid s appeared at 1.2 7.1% in three triploid cultivars.

PAGE 103

103 In s ummar y the majority of the OP progeny of the female fertile triploid cultivars were euploids, having the same ploidy level as their parents or having higher ploidy levels. Aneuploids were rare or did not occur in the OP progeny. When aneuploids were present, some of them had higher ploidy level s as well. Unreduced (2 n ) pollen grains are rare in L. camara as shown in a previous publication (Czarnecki and Deng, 2009). Thus the occurrence of OP progeny with x 5 x and occasionally 6 x ) suggests that these triploid parents had produced UFGs. As n pollen grains of 1 x 2 x or 3 x became available from 2 x 4 x or 6 x lantana plants grown in close proximity, these 3x UFGs would form 4 x 5 x and 6 x embryos (and OP progeny) after fertilization. Through the same process, these 3 x UFGs could produce some of the aneuploid progeny (4 5 x or 5 6 x ) if aneuploid pollen grains were produced by the polyploid paternal plants and participated in the fertilizati on. Conversely the occurrence of the numerous 3 x individuals in the progeny of 11 triploid cultivars suggest s that these triploid parents not only had formed UFGs, but also their UFGs had developed into embryos (and progeny) without double fertilization, i.e. apomixis. A similar phenomenon had been suspected by Spies and Stirton (1982c) As shown later in this chapter, apomixis did occur and was the source of the triploids in the OP progeny of the triploid cultivars. The suspected pathway for embryogene sis (Fig ure 5 3) also suggest s that when UFGs were formed during megaspor o genesis, they could follow two paths, sexual or asexual, to produce progeny. When UFGs took the sexual pathway, they produced 2 n + n progeny, and when UFGs took the asexual pathway, they produced 2 n + 0 progeny.

PAGE 104

104 Interestingly, when the frequencies of the two types of progeny were averaged across the 11 triploid cultivars, the ratio was close to 2:1 [65.8% (2 n + n ): 33.1% 2 n + 0] This ratio may indicate that on average, the UFGs mi ght be twice as likely to produce 2 n + n progeny through fertilization as to produce 2 n + 0 progeny through apomixis. One progeny of (Chapter 4; Czarnecki and Deng, 2009). The occurrence of these diploids in the progeny of tetraploids indicates that some reduced female gametes might also have the ability to develop into embryos (and progeny) through apom ixis, leading to low frequencies of haploidization in L. camara OP progeny of tetraploid L. camara The ploidy levels of the OP progeny of Carlos Dallas Red Gold Irene Pink Caprice and Radiation were described in chapter 4 and a previous publication (Czarnecki and Deng, 2009). Analysis of the ploidy level distribution in the OP progeny of these cultivars suggests that Gold Pink Caprice and Radiation produced UFGs while Carlos Dallas Red and Irene Carlo s Dallas Red and Irene produced triploids when pollinated with diploids and produced tetraploi d s when pollinated with tetraploids (Chapter 4) A similar ploidy level distribution was observed in the OP progeny (35 to 56 progeny ) of these cultivars: low frequencies of triploids (5.4% to 11.1%), high frequencies of tetraploids (77.1% to 91.1%), and low frequencies of aneuploids. This distribution was similar to the one shown in Chapter 4, except for the occurrence of low frequencies of aneuploids. An aneuploid (5 6 x ) appeared in the OP progeny This was likely the result of a viable pollen grain from a pentaploid o r hexaploid plant. The field planting map

PAGE 105

105 indicated that several pentaploids and hexaploids were wi thin 1.8 to 3.6 meters o f frequency (1.7%) of n + 0 progeny, a low frequency (3.4%) of 2 n + n progeny (5 x and 6 x ), and a high frequency (94.9%) of 4 x progeny (Table 5 3) This ploidy level distribution again indicates that UFGs had been formed and apomixis h a s occurred in Gold Considering these new possibilities, the 4 x progeny may be the result of normal fertilization of n female and male gametes ( n + n progeny) or apomictic seed development from UFGs without fertilization, 2 n + 0 progeny. A similar ploidy level distribution was found among Radiation OP progeny, confirming the ability of this cultivar to produce UFGs as shown in Chapter 4 and suggest s that apomixis le d to 2 n or n progeny in addition to n + n progeny. Compared to Gold the exception among Radiation OP progeny was the occurrence of a low frequency (11.9%) of triploids, which would have been the result of fertilization of n female and n male gametes. A large population of Pink Caprice OP progeny (449) was analyzed for ploidy levels (Table 5 3) One triploid and one aneuploid with a lower ploidy level (3 4 x ) were observed among the progeny. Haploids were not observed among this group of OP progeny, but one haploid did occur previously (Czarnecki and Deng, 2009), suggesting apomixis occurred in Pink Caprice Hexaploids (2 n + n ) occurred at a frequency of 14.3%. The majority of the OP progeny was 4 x (84.2%). The higher rates of 2 n + n than the other UFG tetraploids (Chapter 2). The observed

PAGE 106

106 tetraploid plants are from n + n fertilization of reduced female gametes with redu ced male gametes and/or 2 n + 0 progeny from UFGs undergoing apomixis. OP progeny of pentaploid and hexaploid L. camara There was a wide range of ploidy variation among the 49 OP progeny of x ) (Table 5 3). The ploidy level of euploids ranged from 4 x to 7 x and aneuploids had 3 4 x 4 5 x or 7 8 x This was the first time that 7 x (2) and 7 8 x progeny (3) were observed in this study. T his was also the first time that they we re observed in L camara as numerous reports have noted only 2 x to 6 x ploidy levels ( Khoshoo and Mahal, 1967; Natarajan and Ahuja 1957; Ojha and Dayalo, 1992; Raghavan and Arora 1960; Spies 198 4ab ; Spies and Stirton 1982a, b, c; Sanders 1987a b ). The occurrence of hexaploids and heptaploids in the OP progeny demonstr ates UFGs T he 5 x gametes had had been fertilized with 1 x or 2 x reduced male gametes from neighboring diploids or tetraploids and developed into 6 x or 7 x embryos and progeny. This answers a question rais ed in Chapter 4 if 7 x and 8 x plants could be formed and may suggest the possibility to continually increase ploidy levels beyond what has been observed. Of the eight progeny from Tangerine (6 x ) OP seeds were four triploids, one tetraploid, one hexaploi d, and one aneuploid (3 4 x ) (Table 5 3) The occurrence of a high frequency (50%) of triploids suggests that some of Tangerine s reduced female gametes (3x) might produce apomictic seeds and progeny. However, the observed ploidy level distribution could not discern whether or not Tangerine produced UFGs Ploidy Analysis of CP Progeny The p loidy analyses above and in Chapter 4 suggest several possible modes of reproduction in L. camara and the need to use molecular markers to elucidate some of

PAGE 107

107 the proposed hypotheses. Toward this objective, three diploids, one triploid and three tetraploids were selected as seed parents, and three diploids and two tetraploids as pollen parents to make seven controlled crosses (T able 5 4 ) Three of the seven se ed parents were known to produce UFGs (GDGHOP 36, GDOP 4, and PKGHOP 1) and 25). It was unknown would produce UFGs. Seeds from each cross were germinated individually. Progeny were first analyzed for ploidy levels and then analyzed with molecular markers ( see below). x x ) produced 77 progeny. All progeny were diploids, and many demonstrated flower color s simil ar to (mostly creamy yellow and white shades) pro geny emerged from two seeds ( Fig. 5 1 ). Of the 63 progeny from the cross between GDGHOP 36 (2 x ) and Denholm White (2 x ), 11 were diploids, 51 triploids, and one tetraploid (Table 5 4) In an earlier cross with diploid Lola 100% progeny of GDGHOP 36 we re triploids (Czarnecki and Deng, 2009). Thus, the high frequency (80.1%) of triploids was expected in th e current cross, but not the diploids and the tetraploid. Potentially the diploids could have resulted from fertilization of reduced female and reduc ed male gamete s or from UFGs developing into embryos through apomixis. Several possibilities might lead to the occurrence of a tetraploid in a diploid x diploid cross: fertilization of an UFG (2 x ) and a n unreduced male gamete (UMG) (2 x ) (2 n + 2 n progeny) the formation of an UFG followed by

PAGE 108

108 chromosome doubling (doubled UFG) and apomixis (4 n + 0 progeny) or fusion of UFGs in one embryo sac ( 4 n + 0 embryo ) followed by apomixis (4 n + 0 progeny) Spies and Stirton (1982a) noted that diploid L. camara will produce two sexual embryo sacs in one ovule at a frequency of 26.7% which could explain the twin seedlings The cross between PKGHOP 1 (2 x ) and PCOP 6 (4 x ) produced 10 progeny; half of them were diploids and the other half tetraploids. The tetrapl oids were expected, as PKGHOP 1 had been shown to produce UFGs (Chapter 4). A possible origin of the diploids could be PKGHOP n progeny). The majority (24 out of 25) of the progeny of GDOP 4 (3 x ) and Lola (2 x ) were tetraploi ds, indicating that GDOP 4 produced UFGs that fertilized with reduced male n + n progeny). This was anticipated based on the results from a previous controlled cross, in which GDOP 4 also produced a high freq uency of UFGs (Czarneck and Deng, 2009). However, one progeny was found, unexpectedly, to be a pentaploid. The occurrence of a progeny of this ploidy level seems to suggest that an UFG (3 x ) formed by GDOP 4 had fertilized with an UMG n + 2 n progeny). 25 x LAOP 9. A large number of progeny (135 or 92) were analyzed for each cross, and all of their progeny were triploids. The results support the conclusion in Chapter 4 and DROP 25 were not UFG producer s Twin seedlings were observed in 25 x LAOP 9.

PAGE 109

109 and they were all tetraploids, in agreement with the earlier conclusion ( Chapter 4 ) that d reduced female and male gametes. O ne pair of twin seedlings was also observed in this cross Summary of Ploidy A nalys is R esults P loidy analyses of OP and CP progeny above and in C hapter 4 describe six embryo and fertilization combinations for reproduction in L. camara (Figure 5 2) including 1) reduced female and male gamete s fertilized and developed into n + n progeny, 2) reduced female gametes underwent apomixis ( n + 0 progeny, or haploidization), 3) UFGs fertilized with reduced male gametes producing 2 n + n progeny, 4) UFGs took the apomixis pathway and directly developed into embryos and 2 n + 0 progeny, 5) d oubled UFGs fertilized with reduced male gametes, resulting in 4 n + n progeny, and 6) doubled UFGs developed into 4 n progeny through apomixis (4 n + 0 progeny) (Figure 5 3). In contrast the primary reproductive pathway during microsporogenesis was the for mation of reduced male gametes. UMGs were possibly observed, but they occurred at very low frequencies. SSR Marker Analysis Four SSR markers, M1, M51, M76, and M87, were used to analyze 195 progeny from 22 females crossed with five males (Table 5 4 ). T hey amplified 23 alleles in total. Twenty two of the 23 alleles were found to be polymorphic with one allele present in all lines used (M87 5) All allele banding patterns to be discussed can be seen in table 5 4.

PAGE 110

110 Putative 2 n + n progeny As described above, 2 n + n progeny were observed in the progeny of three diploid breeding lines, nine triploid cultivars, three tetraploid cultivars, one petaploid cultivar, and one hexaploid cultivar. Ploidy analyses had provided evidence for the occurre nce of UFGs and 2 n + n progeny. A key question about lantana UFG formation is when is the stage of nonreduction ? It was unknown whether it was first division restitution (FDR) or second division restitution (SDR). The high frequencies of 4 x and 5 x (eupl oids) individuals in the progeny of triploid cultivars lack of aneuploids and restoration of 3 x and 5 x fertility indicated a FDR mechanism A mechanism (such as diplospory) to restor e the fertility prior to a failure in prophase of meiosis I or apospory replacing the aborted gamete would both potentially lead to gamete s resulting in euploid progeny. The progeny of two crosses were analyzed using SSR markers to provide further evidence to explain how female gametes are formed The selected seed parents f or the two crosses were GDGHOP 36 (2 x ) and GDOP 4 (3 x ), known to produce UFGs and 2 n + n progeny known to be non UFG producers. UFG producing tetraploids, pentaploids or hexaploids were not used in the crosses in order to avoid complex marker banding patterns GDGHOP 36 and Denholm White and their 2 n + n progeny : T hree markers (M1, M51, and M76) detected 11 polymorphic alleles between GDGHOP 36 and Alleles M1 4, M1 7, M1 9, M51 4, M51 5, M76 2, and M76 3 were present in GDGHOP 36, while alleles M1 1, M51 1, M51 2 and M76 4 were present in x ) analyzed, 45 produced clear gel images for reliable marker scoring. All 45 progeny carried all the alleles from the maternal

PAGE 111

111 parent GDGHOP 36. The two alleles amplified by M51 (M51 1 and M51 1 present in 31 progeny and M51 2 present in another 16 progeny. These results show ed that all progeny had the complete sets of alleles (or chromosomes) of their maternal parent and so me of the marker alleles from the ir paternal parent. among the progeny (Table 5 4) GDOP 4 (3 x ) and Lola (2 x ) and their progeny: Controlled crosses produced 21 progeny (2 n + n 4 x ). F our SSR markers det ected seven alleles present only in the maternal parent (M1 1, M1 7, M51 1, M51 4, M51 5, M87 1, and M87 3) and three alleles present only in the paternal parent (M1 4, M76 1, and M87 2). There were five alleles shared by both parents (M1 9, M51 2, M76 2, M76 3, and M87 5). Complete marker data were obtained for 18 of the progeny. All progeny carried all the marker alleles unique to the maternal parent (7) M1 4 and M87 2, were present in all the progeny, ind icating that these alleles were alleles, M76 1, was segregating, present in 16 progeny and absent in 5 progeny. In summary, t he results from these marker analyses confir med that all the progeny analyzed were of the 2 n + n type with invariant female alleles present T he gamete s that are formed from the female are essentially clone s of the maternal plant that is then fertilized by a male gamete. Putative 2 n + 0 progeny from controlled pollinations PKGHOP 1 (2 x ) and PCOP 6 (4 x ) and their 2 n progeny: Four SSR markers revealed five alleles (M1 3, M1 5, M51 1, M76 3, and M87 4) unique to the maternal parent, four alleles (M1 4, M1 9, M51 3, and M87 1) unique to the paternal parent, and

PAGE 112

112 another five alleles (M51 2, M51 5, M76 1, M76 2, and M87 5) common to both parents. The five putative 2 n + 0 (2 x ) progeny identified above (refer to ploidy analysis of CP progeny) all carried the five alleles unique to the maternal pa rent. None of the unique alleles from the five paternal parent s were present in the progeny. These results, thus, confirm that all the suspected progeny were indeed 2 n + 0 progeny result ing from UFGs developing into seeds through apomixis. GDGHOP 36 and Denholm White and their 2 n + 0 progeny: As described above, the four SSR markers amplified seven alleles present only in the maternal parent, four alleles present only in the paternal parent, and four alleles present in both parents. All the 11 progeny suspected to be 2 n + 0 progeny contained all the alleles present in the maternal parent but none of the alleles unique to the paternal parent. Again, the results were as expected for 2 n + 0 progeny resulting from UFGs and apomixis. Putative 2 n + 0 progen y from open pollinated triploids : As described in the above ploidy analysis section, 11 triploid cultivars were suspected to have produced 2 n + 0 progeny (3 x ) through UFG formation and apomixis during open pollination. Three progeny were randomly selecte d out of the suspected individuals and subjected to SSR marker analysis. As the progeny were from OP with their paternal parents unknown their SSR marker profiles were compared only to that of their respective maternal parents. Marker M1 amplified a total of nine alleles among the triploid parents, and these parents each carried 2 to 4 alleles. All nine alleles were polymorphic. M51 revealed five alleles among the parents, and each parent contained 2 to 5 alleles. M76 detected 2 to

PAGE 113

113 4 alleles, all the mark er alleles were polymorphic. M87 amplified five alleles, four of which were polymorphic and one monomorphic. In total 20 polymorphic alleles were available for screening. M1, M51, M76, and M87 each produced three alleles in Landmark Pink Dawn : M1 1, M1 4, M1 9, M51 1, M51 2, M51 5, M76 1, M76 2, M76 4, M87 1, M87 3, and M87 5. Two of the three OP progeny had a marker profile identical to Landmark Pink Dawn and one progeny was missing the allele M76 4. Lemon Drop and its OP progeny: M1, M51, M76, and M87 amplified three (M1 4, M1 7, and M1 9), two (M51 3 and M51 5), three (M76 1, M76 3, and M76 4), and two (M87 3 and M87 respectively All three OP progeny shared the same marker Miss Huff and its OP progeny: 2, M1 4, M1 5, and M1 9), three alleles of M51 (M51 2, M51 3, and M51 5), two alleles of M76 (M76 1 and M76 4), and four alleles of M87 (M87 1, M87 2, M87 3, and M87 5). All M1 M51, M76, and M87 detected 10 alleles ( M1 1, M1 7, M1 9, M51 4, M51 5, M76 2, M76 3, M87 1, M87 3, and M87 5 ) in New Go ld : All OP progeny had an identical marker profile at all the 10 alleles amplified by the four SSR markers: M1 4, M1 8, M51 2, M51 3, M51 5, M76 2, M76 4, M87 2, M87 4, and M87 5.

PAGE 114

114 : M1, M51, M76, and M87 revealed 11 alleles in Patriot Fire Wagon : M1 4, M1 9, M51 2, M51 3, M51 5, M76 1, M76 2, M76 3, M76 4, M87 1, and M87 5. The marker profile of the three OP progeny wa s identical Red Butler Markers M1, M51, M76, and M87 amplified three (M1 4, M1 5 and M1 9), two (M51 2 and M51 3), three (M76 1, M76 2, and M76 3), and four (M87 2, M87 3, M87 4, and M87 the OP progeny. and their OP progeny: M1, M51, M76, and M87 each revealed a total of 12 alleles (M1 4, M1 7, M1 8, M51 2, M51 3, M76 2, M76 3, M76 4, M87 1, M87 2, M87 4, and M87 5). All three OP progeny carried these alleles from Sunset No additional alleles were observed in the progeny Putative 2 n + 0 progeny of open pollinated pentaploid and hexaploid at M76, and possessed the same ma rker profile as the maternal parent, as expected for offspring originating from UFGs and apomixis. providing a total of 11 alleles for the four SSR markers. Only one OP progeny was available for marker analysis, and it had the same marker profile as its maternal parent. Putative 2 n + 0 and n + n progeny of diploids and tetraploids The confirmed occurrence of U FGs and apomixis in triploids raises a question : what gametes form the diploid progeny of diploid cultivars and the tetraploid progeny of

PAGE 115

115 tetraploid cultivars ? Potentially the diploid progeny of a diploid cultivar could be from fusion of reduced female an d male gametes ( n + n progeny) or from UFGs and apoximis (asexual reproduction) (2 n + 0 progeny). A similar scenario applies to the tetraploid progeny of a tetraploid cultivar. They could be 2 n + 0 or n + n progeny. Cross between GDGHOP 36 (2 x ) and De nholm White (2 x ) and its diploid progeny: M1, M51, M76, and M87 revealed eight alleles that were present in GHGHOP diploid progeny had th e same marker profile and it was identical to the maternal parent GDGHOP 36, thus, they were 2 n + 0 not n + n progeny. Cross between LAOP 9 (2 x ) and Lola (2 x ) and its diploid progeny: The four SSR markers identified four alleles (M1 1, M5101, M76 4, and M87 1) that were present in LAOP analyzed in a ratio of 2:3, 5:1, 4:2, or 3:3 (present: absent), respectively. There were two additional alleles (M76 2 and M76 3) that were present in both parents. They segregated in the progeny in a ratio of 3:3 or 4:2 (present: absent). Each of the diploid progeny possessed a distinct marker profile. These results suggest that the progeny resulted from sexual reproduction or fertili zation of reduced female and male gametes. Thus these progeny were of n + n type, in contrast to the diploids from UFG producing GDGHOP 36 (2 n + 0) Cross between Dallas Red (4 x ) and Carlos (4 x ) and its tetraploid progeny: Allele M76 2 was present i but 3 and M76 1 but

PAGE 116

116 available for SSR marker analysis. M76 2 was not observed in progeny, but both M51 3 and M76 1 were present, ind icating that the progeny had resulted from fertilization of reduced female and male gametes, rather than UFGs and apomixis. Thus these progeny belong to the n + n type. Pink Caprice (4 x ) and its tetraploid OP progeny: Four SSR markers amplified 3, M1 4, M1 5, M1 6, and M1 9), four by M51 (M51 1, M51 2, M51 3, and M51 5), four by M76 (M76 1, M76 2, M76 3, and M76 4), and four by M87 (M87 1, M87 3, M87 4, and M87 5) The five tetraploid they were 2 n + 0 progeny resulting from UFGs and apomixis. Radiation (4 x ) and its tetraploid OP progeny: M1, M51, M76, and M87 detected two (M 1 4 and M1 9), three (M51 2, M51 3, and M51 4), three (M76 1, M76 2, and M76 3), and two (M87 1 and M87 three tetraploid OP progeny carried M87 1. One descendant had one allele (M87 3) not 1, M51 1, and M87 3) not type res ulting from fertilization of reduced female and male gametes. Putative 4 n + 0, 2 n + 2 n and 4 n + n progeny Tetraploids were recovered from the progeny of two diploid by diploid crosses, one (917 56) from the cross GDGHOP ne from the cross between Myst the fertilization of UFGs and UMGs. As described above, GDGHOP and four alleles

PAGE 117

117 present in the latter. The tetraploid descendant contained all five alleles from its maternal parent GDGHOP 36 but none of the four alleles from its paternal parent (Table 5 4) Thus it was a 4 n + 0 (4 x ) progeny. This i s likely the result of the fusion of two 2 n embryos or an additional mitosis like chromosome doubling process. T he tetraploid from Myst from a 4 n female gamete fused with an n male gamete Myst 107 carried five alleles (M1 3, M51 3, M76 1, M87 1, and M87 (M87 2) that was absent in Myst 107. The tetraploid progeny contained all the alleles from both parents. Thus the progeny belongs to a 2 n + 2 n type resulting fr om the fusion of a n UFG and an UMG (Table 5 4) To our knowledge this is the first observance of an unreduced pollen grain successfully fertilizing an embryo. One pentaploid was also identified among the progeny of Myst This pentaploid ca rried the same banding pattern as the tetraploid from the fusion of a UFG and UMG described above, but also was a ploidy level that can only be explained by the formation of a 4 n gamete from Myst 107 that was fertilized by a n gamete from Collect ively the formation of these three progeny from two controlled crosses indicate d two reproductive modes that formed these individuals. The first is the formation of a double UFG (DUFG). The tetraploid progeny was not likely to be the result of a n UMG due to the almost complete lack of pollen fertility of GDGHOP 36 (1.3% stainable pollen appendix A). In addition the formation of a pentaploid from Myst 107 could not be explained by self fertilization alone. The pentaploid progeny would have required the f ormation of a DUFG and pollination (Myst 107 also has largely

PAGE 118

118 non pollen did successfully fertilize this gamete. The other individual of interest was a tetraploid formed from the formation of a UFG and a UMG. Based on the ploidy level of the plant (4 x ) it was the result of a successful cross pollination or a DUFG. The banding pattern of the plant had alleles and a UMG to for m the tetraploid plant. As previously discussed UMG pollen appears to be extremely rare in L. camara and is likely an infrequent random occurrence. Putative n + 0 progeny From 2006 to 2010 11 haploids were identified in the OP progeny of three x (2 x x x ) from The continued occurrence of these individuals provided enough evidence to suspect haploidization and conduct marker analysis. Five of the alleles (M51 2 M76 1, M76 2, M76 3, and M87 3, M1 4, M1 5, M1 6, M1 9, M51 1, M51 3, M51 5, M87 1, M87 3, and M87 4) segregated among the four haploids, PCH1, PCH2, Myst 107, and PKGHOP 1. Each haploid possessed a distinct marker profile, containing 5 to 6 of the 11 segregating alleles in different combinations and the five non segregating alleles. Thus, these haploids lost 5 to 6 alleles of their maternal parent

PAGE 119

119 alleles for M51, three alleles for M76, and three alleles for M87. Seven alleles were not segregating, but the remaining four alleles (M51 3, M76 2, M76 4, and M87 1) segregated in carried two of the segregating alleles plus seven non segregating alleles, and they differed from each other at two alleles. h were segregating in 10, GDGHOP 36, and GDOP 31. These segregating alleles. carried eight alleles but missed two alleles (M51 3 and M87 1) Caprice Gold Radiation and e process of apomixis has been shown to occur with 2 n gametes (discussed above) it seems likely that at some frequency apomixis may be triggered with n gametes in the ovule causing haploidization to occur This confirms previous speculation from Chapter 4 and demonstrates that at a low frequency haploidization will occur. Potentially these haploids may have had served a purpose in evolution and could be manipulated for breeding. Twin progeny Carlos (4 x ) x Lola (2 x ): The six pairs of twins (3 x ) ident ified in the progeny of this cross were different among and within pairs at one to five alleles. Within the six pairs one differed at M87 1, two twins differed at all M76 alleles while at two different

PAGE 120

120 alleles in primer M 51 (M51 3 and M51 5). Another tw in only differed at M51 3 and M51 5. The other differed at M51 5, M76 2 and M76 3. The last twin differed only at the M76 primer (M76 1, M76 2, and M76 3). Within this group of twins differences were more common within the M76 locus as 13 of the 18 diff erences within the group were found with this primer. In total five of the alleles were missing from the maternal parent indicating different female gametes were fertilized to produce twins. LAOP 9 (2 x ) x Lola (2 x ): Two pairs of twin seedlings (2 x ) were identified from the progeny of this cross. The first pair of twins differed at three alleles (M1 1, M76 2, and M76 3), and the second pair differed at four alleles (M1 1, M76 2, M76 3, and M87 1), indicating that each pair of twins were developed fr om different zygotes within a seed. Alleles M1 1, M76 4, and M87 1 were present in the maternal parent LAOP 9 but 4 segregated among pairs of twins. M87 1 segregated within one pair and M1 1 segregated within bo th pairs of twins. These results suggest that the female gametes involved in the production of each pair of twin seedlings were different as well. Landmark White x Denholm White : Two pairs of twin seedlings were analyzed. The seedlings within the fi rst pair were different at three alleles, M51 1, M51 3, and M76 3, and the seedlings within the second pair were different at alleles M51 1, M51 2, and M87 1. M51 3 and M76 es were segregating within the first pair of twins, but present in both seedlings of the second pair. Nevertheless, M51 2, an allele present in both parents, segregated in the second pair of twins. Additionally, M87 1, an allele segregated in this pair of seedlings.

PAGE 121

121 DROP 25 (4 x ) x LAOP 9 (2 x ): A total of 10 pairs of twin seedlings (2 x ) were analyzed using SSR markers. They could be divided into two groups, based on their marker profile. The first group consisted of eight pairs o f seedlings from eight seeds. These individuals differed among and within pairs at 1 to 4 alleles (an average of 2.1). Thus each of the twins originated from separate zygotes. The second group consisted of two pairs of twins (3 x ). Their marker profile s were identical. Each progeny contained one (M76 1) of the three alleles that were present in DROP 25 and two (M51 1 and M87 2) of the three alleles that were present in LAOP 9. Thus these twins did carry some alleles from each parent, that is, these tw ins were from zygotes. Alternatively, the identical marker profile may suggest that each pair of twins might have originated from a single gamete GDOP 4 (3 x x ): Two pairs of twins were collected from the progeny of this cross, and they were tetraploids known to result from fertilization of UFGs and reduced male gametes (refer to above). The four progeny were identical in SSR marker profile, carrying all eight alleles (M1 1, M1 7, M51 1, M51 4, M51 5, M76 4, M84 1, and M84 3) that were presen t in GDOP 4, M76 2, and M87 4). M1 4 and M87 progeny of However, M76 segregated at 16:5 in all the progeny from this cross. Th e allele present of the more common band and may be from two pollen grains carrying the same allele. The identical marker banding pa ttern within each of the two pairs of twins does raise a

PAGE 122

122 question whether or not the twins originated from single zygotes similar to the previous cross discussed. GDGHOP 36 (2 x ) x Lola (2 x ): Two pairs of twins were identified in the progeny of this cross, and they were triploids known to originate from UFGs and reduced male gametes ( see above). These twins shared the same marker profile: containing all five alleles (M1 7, M51 4, M51 5, M87 1, and M87 3) of GDGHOP 36 and two alleles (M51 2 and M87 reproduction, but they were not able to discern whether or not each pair of twins originated from the same or separate zygotes. Discussion Multiple Modes of Reproduction in L. camara The above ploidy and microsatellite marker analyses have shown that L. camara can form three types of female gametes (reduced female gametes or RFGs, UFGs, and DUFGs) and two types of male gametes (red uced male gametes or RMGs and UMGs), and can develop seed through fertilization or apomixis. Taking into consideration these various types of gametes and apomixis, potentially there could be nine possible modes of reproduction. Seven of these (Modes 1 7 ) were observed in this current study (Figure 5 3). So far, the observed frequencies of DUFG and UMG seem to be quite low. The observed primary modes of reproduction have been Modes 1, 3, 4, and 5, which would result in n + n n + 0, 2 n + n and 2 n + 0 p rogeny (Figure 5 3). Differences in reproductive modes have been noted among L. camara cultivars/breeding lines. For primarily through Mode 1 ( n + n ), while three othe

PAGE 123

123 Formation of U FG s and DUFGs UFG formation is one of the two primary factors that cause such diverse types of reproduction in L. camara T wo basic proce sses have been described as the mechanisms for formation of unreduced gamete in plants: first division restitution (FDR) and second division restitution (SDR) (Bretagnolle and Thompson, 1995). In FDR, the first meiotic division occurs abnormally, with hom ologous chromosomes not pairing during Prophase I and/or not separating to opposite poles during Anaphase I. In SDR, the second meiotic division occurs abnormally, with sister chromatids not separating to opposite poles during Anaphase II. Univalents and multivalents (during Prophase I) and laggards (during Anaphase I) are common in lantana polyploids (Natarajan and Ahuja, 1957; Spies and Stirton, 1982b). Thus 2n gametes formed by SDR would be aneuploids and contain various numbers of chromosomes. Howev er, the progeny of from UFGs were euploids ( Chapter 4, and Tables 5 4 and 5 5). Microsatellite marker data showed that these euploids carried all the alleles of their respective seed parents. This would be possible only through FDR more specifically apomeiosis The occurrence of DUFGs was rare but how they were formed is very intriguing. Determining the mechanism by which DUFGs are formed may also be explained by the mechanism that forms UFGs or potentially a di fferent chromosome doubling mechanism may be involved. Conceivably, several possible processes could lead to the formation of this type of female gametes, including another chromosome doubling event after UFG formation, or merging of two embryo sacs. Rega rdless of the specific processes, DUFGs could result in rapid increase of ploidy levels in the progeny (Figure 5 3, Modes 6 and 7). This has similarly been seen in Fragaria (Bringhurst and Gill 1970).

PAGE 124

124 Apom i xis This type of asexual reproduction was suspect ed to occur in L. camara because of the appear ance of apparently matriclinical progeny ( Khoshoo and Mahal 1967; Raghavan and Arora 1960). Based on extensive ploidy and molecular marker analysis, this study confirmed that apomixis occur red frequently, espe cially in some cultivars/breeding lines. Apomixis can be either aposporous or diplosporous. Aposporous apomixis arises from somatic cells within the ovule, commonly from the nucellar or integumentary cells, while diplosporous apomixis initiates from mega spore mother cells, like the normal megasporogenesis. Three types of apomictic progeny have been observed in L. camara : those identical to their maternal parents in ploidy level and microsatellite marker banding pattern (2 n + 0 progeny), those identical to their maternal parents in microsatellite marker banding pattern but with an increased ploidy level (4 n + 0 progeny, Figure 5 3, Mode 7), and those with reduced ploidy levels (haploids) and containing a portion of the microsatellite marker alleles of the maternal parents ( n + 0 progeny) (Figure 5 3, Mode 3). The first two types of progeny could be either from somatic cells or doubled somatic cells undergoing aposporous apomixis or from UFGs or DUFGs going through diplosporous apomixis. However, the third type of apomictic progeny ( n + 0) could be produced only through the formation of reduced female gametes and diplosporous apomixis. Twin s eedlings Twin seedlings seem to be common i n some cultivars/breeding lines. Previously, Spies and Stirton (1982c) observed the occurrence of two sexual embryos in numerous embryo sacs in L. camara SSR marker analysis in this study showed that progeny in many pairs of twins had different marker banding patterns. This suggests that twins in

PAGE 125

125 these pairs have result ed from separate fertilization events in a single embryo sac rather than from the splitting of single embryos Evolutionarily this formation of two viable embryos within an embryo sac may seve two purposes. The first is the potential for two progeny to come from a single and the second is that two embryos may increase the succ essful rates of fertilization. Implications for Lantana Breeding The fact that L. camara can reproduce through multiple modes can greatly affect the manner in which lantana breeding is to be conducted. In order to achieve high levels of sterility, parental materials must have progeny screened prior to use to know their primary gamete types and modes of reproduction. Such screening can be timely especially when seed germination is consi dered In addition, a flow cytometer is needed to analyze the ploidy levels of progeny. In L. camara breeding, open pollinated (OP) seeds are often used to develop new germplasm. The use of this type of seed may save time, but this practice may introduce UFG production and apomictic traits into new germplasm. These traits would reduce our ablity to control the sexual reproduction of the species. Thus careful planning to preclude UFG producers is necessary before OP seed are collected and used for germp lasm development. The available diploid cultivars identified in this study (and previous studies) have creamy white to yellow flowers. Transferring other flower colors (red, pink, etc.) and other desirable ornamental traits into diploids has been an i mportant objective in many lantana breeding programs. For this reason, it was thought that the obtained haploids could be very valuable. However, these haploid lines carried the UFG trait and it was very difficult to transfer desirable traits from UFG pr oducing haploids to diploid

PAGE 126

126 germplasm. W hen used as seed parents, these haploids (haploids PKGHOP 1, GDGHOP 36, and MYST 107; Table 5 5) produced mostly 2 n + 0 or 2 n + n (occasionally 4 n + 0) progeny which are of little value for improving diploid germpl asm. These haploids potentially could be used as pollen sources, but presently all haploids evaluated had very low pollen viability (Appendix A). Evolutionary and Ecological Significance L. camara is a woody to semi woody perennial plant (Floyd 1998) wi th the ability to self and cross pollinate (Mathur and Ram 1986) and produce seed that can be dispersed by birds ( National Weeds Strategy Executive Committee, 2000) or other natural means such as gravity. Based on the analysis of its life history charact eristics, L. camara is expected to have moderate to high levels of genetic diversity (Hamrick and Godt 1996). High levels of genetic diversity could have contributed to L. camara success in colonizing and invading habitats. Results from this study have shown that L. camara has substantial reproductive versatility which may assist in genome evolution (Soltis and Soltis 1999) This versatility may be another important factor contributing to the success of this species in colonizing and invading new habit ats. Population dynamics of L. camara can be very complicated in nature as the individuals of a population may take different reproduction pathways ( n + n 2 n + n 2 n + 0, n + 0, etc.). Figure 5 4 presents a summary of likely interactions among individuals of a population and potential impacts on population dynamics. In this model normal n + n reproduction is shown by the red solid arrows. To simplify the model, it is assu med that only diploids, tetraploids, and hexaploids produce viable reduced male and female gametes. It is expected that in open pollination, these diploids, tetraploids, and hexaploids would produce some 3 x and 5 x progeny. Normally, these 3 x and 5 x

PAGE 127

127 proge ny would be highly sterile and not be able to cross pollinate with individuals of other ploidy levels. This would raise a question how does a species overcome such genetic isolation? L. camara is known t o produce UFGs, which have been shown to restore fe male fertility ( Czarnecki and Deng, 2009 ). This trait leads to the production of 2 n + n progeny in the population, represented by solid blue triangular pointed lines (Figure 5 4). L. camara may best be characterized as a successful polyploid metapopulati on as described by Rausch and Morgan ( 2005) Specifically, this trait would enable 3 x and 5 x plants to participate in seed production and population formation. Additional fitness to colonizing plants may come from the formation of apomeiotic gametes. Th is would prevent the formation of lethal allele combinations during establishment (Noyes 2006). Maintaining previously productive genotypes may provide a conservative yet successful colonization approach providing an explanation for the success of the s pecies Subsequent to colonization, how would this species begin to adapt? Ther e may be two answers to this question The first explanation may lie in genome or allele additions. Similar to Noyes (2005), L. camara is able to participate in 2 n + 0 and 2 n + n seed production (Figure 5 4 modes 2 and 4). The 2 n + n seed could incorporate relative genomes and locally adapted alleles safely. Evidence of this is seen in a cytological investigation of naturalized L. camara accessions. This work led to the co nclusion that L. camara was comprised of at least 3 genomes (Spies and Stirton 1982a). It could be expected that based on life history and the reproductive systems of L. camara genetic diversity would likely be maintained and further increased with local cross pollination. It is possible that L. camara may represent a compilospecies that assimilates local genes so successfully that it eventually invades

PAGE 128

128 and overtakes the nativ e or local species (Harlan and de Wet 1963) Another similar model is that of species homogenization causing the loss of relatives in the area and possibly endemic species (Lockwood and McKinney 2001) Both of these lead to the adaptation of one specie s at the expense of another. The second adaptive mechanism found in L. camara is an extensive plasticity in ploidy level. As previously discussed normal ploidal blocks (odd ploidy levels) may not exist on the female side of this species and is not always correlated to ploidy level ( Czarnecki and Deng, 2009; Khoshoo and Mahal, 1967; Raghavan and Arora 1960; Spies 1984b). Polyploidization has been thoroughly discussed in previous sections but little has been discussed about haploidization. A model simil ar to Harlan and de tetraploid diploid tetraploid cycling may be possible although more complex (Figure 5 4 modes 2, 3, 4, and 5). Al beit at different rates of production of n + 0, 2 n + 0, and 2 n + n similar reproductive means were also show n by Bicknell and Koltunow (2004) in Hieracium Spontaneous haploidization was shown to occur when reduced gametes go though apomixis. This process may force foreign alleles to recombine with invading genomes. Although fertility levels may vary amo ng ploidy levels (Ramsey and Schemske 2002) a ny lack of fertility is removed by the presence of the UFG trait thus allowing this process to continue. A previous study has shown that plants became larger at higher altitudes (Matthew 1971) possibly indicati ng adaptation from gene recombinations or an increase in ploidy level A similar phenomenon may be occurring with polyploidization, where increased ploidy level may encourage a rapid means for adaptation. The adaptation may have come from changing ploidy levels as Sanders (2001) suggested diploids are stunted, dwarfed, or generally less vigorous

PAGE 129

129 while polyploids are generally larger. The ability to increase ploidy levels from 2 x to 5 x in one generation (Figure 5 4 mode 3) suggests this to be highly plaus ible adaptive trait. It is our suspusion that both mechanisms are working together. The additional genomes would allow increased heterosis that could lead to improved fitness. Collectively this work provides an explanation of the reproductive capabilitie s leading to L. camara formation of UFGs leading to clonal apomixis that allows initial establishement, then UFGs provide a foundation for successful gametes to fuse with other local genotypes and then after colonization normal gametes and reproduction can occur to adapt more quickly to ecotype. For the purposes of varietal development UFG production appears to be a trait that can be selected against (Czarnecki and Deng 2009) thus removing a large component of the invasive potential of L. camara

PAGE 130

130 Table 5 1 Polymerase chain reaction cycle s to detect differences in L. camara parents and progeny produced from different modes of reproduction Lantana reproduction. 1345 Hold Cycle X 10 Cycle X 30 Hold Hold 95 C 95 C 55 C 72 C 95 C 45 C 72 C 72 C 8 C 2 min. 45 sec. 45 sec. 1 min. 45 sec. 45 sec. 1 min. 5 min. infinity 1350 95 C 95 C 68 C 72 C 95 C 50 C 72 C 72 C 8 C 2 min. 45 sec. 45 sec. 1 min. 45 sec. 45 sec. 30 sec. 5 min. infinity Table 5 2 Primer sequences for SSR markers used to detect differences between parents and progeny of L. camara produced from different modes of reproduction of Lantana spp Primer code Repeat motif Primer Primer Pairs PCR Cycle M 1 (AG) 12 Forward Z TGAGAACAGCTCAGTTGACCA 1350 Reverse CAACATGAATTAAAGGACTAAACTGC M 51 (GA) 9 Forward TGGAATGGAAAGCAAGCAG 13 45 Reverse TCCAGGGAAAAATCATCACC M 76 (TC) 10 Forward CCCGCATTTTAATTCAAGAC 13 45 Reverse GGAGGGTGTATGTCCATGAG M 87 (TC) 10 Forward TCACCTATTTTGCGTCTCTGTG 13 50 Reverse GGGGTGGAAAAAGGTTGTCT Z M13 tail code CCCAGTCACGACGTTG was added to the forward primer for fluorescence labeling

PAGE 131

131 Table 5 3 Distribution of the ploidy level of progeny from field open pollination experiments of L. camara commercial cultivars and breeding lines. Lines are followed with ploidy level in parenthesis. L. camara lines and progeny were used from seed collection in C hapter 3. Percentage of progeny in different ploidy levels L. camara cultivars/lines OP progeny analyzed (no.) 2 x 3 x 4 x 5 x 6 x 7 x 3 4 x 4 5 x 5 6 x 7 8 x Cream (2 x ) 2 50 .0 50 .0 LAOP 9 (2 x ) 1 100 LAOP 30 (2 x ) 3 100 Lola (2 x ) 13 84.6 7.7 7.7 Landmark Peach Sunrise Z (3 x ) 14 57.1 21.4 14.3 7.1 Landmark Pink Dawn (3 x ) 137 35.8 56.9 6.6 0.7 Lemon Drop (3 x ) 81 24.7 65.4 3.7 6.2 Lucky Red Hot (3 x ) 1 100 Miss Huff (3 x ) 78 23.1 65.4 9 2.6 New Gold (3 x ) 39 20.5 76.9 2.6 New Red Lantana (3 x ) 40 22.5 75 2.5 Patriot Fire Wagon (3 x ) 83 1.2 24.1 67.5 3.6 1.2 2.4 Red Butler (3 x ) 24 54.2 37.5 8.3 Red Spread Lantana (3 x ) 31 38.7 58.1 3.2 Samson (3 x ) 154 31.2 66.2 1.3 1.3 Sunset (3 x ) 65 32.3 55.4 10.8 1.5 Carlos (4 x ) 56 5.4 91.1 1.8 1.8 Dallas Red (4 x ) 36 11.1 83.3 5.6 Gold (4 x ) 59 1.7 94.9 1.7 1.7 Irene (4 x ) 35 5.7 77.1 17.1 Pink Caprice (4 x ) 449 0.2 84.2 14.3 0.2 0.2 0.9 Radiation (4 x ) 59 3.4 11.9 67.8 3.4 13.6 Spreading Sunset (5 x ) 49 4.1 51 2 4.1 8.2 20.4 4.1 6.1 Tangerine (6 x ) 8 50 .0 12.5 12.5 25 Z Removed Improved from cultivar name.

PAGE 132

132 Table 5 4 SSR marker analysis and ploidy level of parents and progeny ( in quotations ) from controlled and open pollinations of L. camara cultivars /breeding lines to determine reproductive modes The f ive modes of reproduction found were unreduced female gamete production (UFG) double UFG apomixis, haploidization, and twinning. When poss ible individuals with the same alleles are grouped and the number of individuals is indicated. SSR markers M1 M51 M76 M87 Allele 1 2 3 4 5 6 7 8 9 1 2 3 4 5 1 2 3 4 1 2 3 4 5 Mode Group Ploidy N o. 2 n + n GDOP 4 (F) Z 3 x 1 0 0 0 0 0 1 0 1 1 1 0 1 1 0 0 1 1 1 0 1 0 1 2 x 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 1 1 0 0 1 0 0 1 Progeny type 1 Y 4 x 16 1 0 0 1 0 0 1 0 1 1 1 0 1 1 0 1 1 1 1 1 1 0 1 Progeny type 2 4 x 5 1 0 0 1 0 0 1 0 1 1 1 0 1 1 0 0 1 1 1 1 1 0 1 G DGHOP 36 (F) Z 2 x 0 0 0 1 0 0 1 0 1 0 0 0 1 1 0 1 1 0 1 0 1 0 1 2 x 1 0 0 0 0 0 0 0 0 1 1 0 1 1 0 0 0 1 1 0 1 0 1 Progeny type 1 Y 3 x 31 1 0 0 1 0 0 1 0 1 0 1 0 1 1 0 1 1 1 1 0 1 0 1 Progeny type 2 3 x 16 1 0 0 1 0 0 1 0 1 1 0 0 1 1 0 1 1 1 1 0 1 0 1 2 n + 0 PKGHOP 1 (F) Z 2 x 0 0 1 0 1 0 0 0 0 1 1 0 0 1 0 1 1 0 0 0 0 1 1 PCOP 6 (M) 4 x 0 0 0 1 0 0 0 0 1 0 1 1 0 1 0 1 1 0 1 0 0 0 1 Progeny type 1 Y 2 x 5 0 0 1 0 1 0 0 0 0 1 1 0 0 1 0 1 1 0 0 0 0 1 1 G DGHOP 36 (F) Z 2 x 0 0 0 1 0 0 1 0 1 0 0 0 1 1 0 1 1 0 1 0 1 0 1 2 x 1 0 0 0 0 0 0 0 0 1 1 0 1 1 0 0 0 1 1 0 1 0 1 Progeny type 1 Y 2 x 12 0 0 0 1 0 0 1 0 1 0 0 0 1 1 0 1 1 0 1 0 1 0 1 Myst 107 (F) Z 2 x 1 0 0 1 1 0 0 0 0 1 0 1 1 0 1 1 1 1 1 1 0 1 1 1 2 x 0 0 0 1 0 0 0 0 1 0 1 0 0 0 1 1 1 0 1 0 1 0 1 Progeny type 1 Y 2 x 0 0 1 1 0 0 0 0 1 0 1 1 0 1 1 1 1 1 1 0 1 1 1 Z 3 x 0 0 0 1 0 0 1 1 0 0 1 1 0 0 0 1 1 1 1 1 0 1 1 Progeny type 1 Y 3 x 3 0 0 0 1 0 0 1 1 0 0 1 1 0 0 0 1 1 1 1 1 0 1 1 3 x 0 0 0 1 1 0 0 0 1 0 1 1 0 0 1 1 1 0 0 1 1 1 1 Progeny type 1 Y 3 x 3 0 0 0 1 1 0 0 0 1 0 1 1 0 0 1 1 1 0 0 1 1 1 1 3 x 0 0 0 1 1 0 0 0 1 0 1 1 0 0 1 1 1 0 0 1 1 1 1 Progeny type 1 Y 3 x 3 0 0 0 1 1 0 0 0 1 0 1 1 0 0 1 1 1 0 0 1 1 1 1 4 x 0 0 1 1 1 1 0 0 1 1 1 1 0 1 1 1 1 0 1 0 1 1 1 Progeny type 1 Y 4 x 5 0 0 1 1 1 1 0 0 1 1 1 1 0 1 1 1 1 0 1 0 1 1 1 3 x . . . . 0 0 1 0 1 1 1 1 0 1 0 1 1 1 Progeny type 1 Y 3 x 3 0 0 1 1 0 1 0 0 1 0 0 1 0 1 1 1 1 0 1 0 1 1 1

PAGE 133

133 T able 5 4 Continued. SSR markers M1 M51 M76 M87 Allele 1 2 3 4 5 6 7 8 9 1 2 3 4 5 1 2 3 4 1 2 3 4 5 Mode Group Ploidy N o. 2 n + 0 (cont) 3 x 1 0 0 0 0 0 1 0 1 0 0 0 1 1 0 1 1 0 1 0 1 0 1 Progeny type 1 Y 3 x 3 1 0 0 0 0 0 1 0 1 0 0 0 1 1 0 1 1 0 1 0 1 0 1 X (OP) 3 x 1 0 0 1 0 0 0 0 1 1 1 1 1 0 1 0 0 1 1 0 1 0 1 Progeny type 1 Y 3 x 3 1 0 0 1 0 0 0 0 1 1 1 1 0 0 0 0 0 1 1 0 1 0 1 3 x 0 1 0 1 1 0 0 0 1 0 1 1 0 1 1 0 0 1 1 1 1 0 1 Progeny type 1 Y 3 x 3 0 1 0 1 1 0 0 0 1 0 1 1 0 1 1 0 0 1 1 1 1 0 1 3 x 1 0 0 1 0 0 0 0 1 . . 1 1 0 1 1 0 1 1 1 Progeny type 1 Y 3 x 1 0 0 0 1 0 0 0 0 1 1 1 0 0 1 1 1 0 0 1 0 1 0 1 Progeny type 2 Y 3 x 2 0 0 0 1 0 0 0 0 1 1 1 0 0 1 1 1 0 1 1 0 1 0 1 3 x 0 0 0 1 0 0 0 1 0 0 1 1 0 1 0 1 0 1 0 1 0 1 1 Progeny type 1 Y 3 x 3 0 0 0 1 0 0 0 1 0 0 1 1 0 1 0 1 0 1 0 1 0 1 1 5 x 0 0 0 1 0 0 1 0 1 0 1 1 1 1 1 1 1 0 1 0 1 0 1 Progeny type 1 Y 5 x 3 0 0 0 1 0 0 1 0 1 0 1 1 1 1 1 1 1 0 1 0 1 0 1 3 x 0 0 0 1 0 0 1 0 1 0 0 1 0 1 1 0 1 1 0 0 1 0 1 Progeny type 1 Y 3 x 3 0 0 0 1 0 0 1 0 1 0 0 1 0 1 1 0 1 1 0 0 1 0 1 3 x 0 0 0 1 0 0 0 0 1 0 1 1 0 1 1 0 1 1 1 0 0 0 1 Progeny type 1 Y 3 x 3 0 0 0 1 0 0 0 0 1 0 1 1 0 1 1 0 1 1 1 0 0 0 1 6 x 0 0 0 1 0 0 1 0 1 0 0 0 1 1 1 1 1 0 1 0 1 0 1 Progeny type 1 Y 6 x 1 0 0 0 1 0 0 1 0 1 0 0 0 1 1 1 1 1 0 1 0 1 0 1 n + 0 4 x 0 0 1 1 1 1 0 0 1 1 1 1 0 1 1 1 1 0 1 0 1 1 1 Progeny type 1 Y (PCH1) 2 x 1 0 0 0 0 1 1 0 0 1 0 1 0 0 1 1 1 1 0 0 0 1 1 1 Progeny type 2 (PCH2) 2 x 1 0 0 1 1 0 0 0 0 1 1 1 0 0 0 1 1 1 0 0 0 0 1 1 Progeny type 3 (Myst 107) 2 x 1 0 0 1 1 0 0 0 0 1 0 1 1 0 0 1 1 1 0 1 0 1 0 1 Progeny type 4 (PKGHOP 1) 2 x 1 0 0 1 0 1 0 0 0 0 1 1 0 0 1 1 1 1 0 0 0 0 1 1 4 x 1 . . . . 1 1 0 1 1 0 1 1 1 1 0 1 0 1 Progeny type 1 Y (GDH1) 2 x 1 0 0 0 1 0 0 1 0 1 1 0 0 0 1 . . 1 0 1 0 1 Progeny type 2 (GDGHOP 36) 2 x 1 0 0 0 1 0 0 1 0 1 0 0 0 1 1 0 1 1 0 1 0 1 0 1 Progeny type 3 (GDGHOP 10) 2 x 1 1 0 0 0 0 0 1 0 1 1 0 0 0 1 0 1 0 0 1 0 1 0 1 Progeny type 4 (GDOP 31) 2 x 1 1 0 0 1 0 0 0 0 1 1 0 0 0 1 0 1 1 1 0 0 1 0 1

PAGE 134

134 Table 5 4 Continued. SSR markers M1 M51 M76 M87 Allele 1 2 3 4 5 6 7 8 9 1 2 3 4 5 1 2 3 4 1 2 3 4 5 Mode Group Ploidy N o. n + 0 (cont) 6 x 0 0 0 1 0 0 1 0 1 0 1 1 1 1 0 1 1 1 1 0 1 0 1 Progeny type 1 Y (TANG H1) 3 x 1 . . . . 0 1 0 1 1 0 1 1 1 0 0 1 0 1 4 x 0 0 0 1 0 0 0 0 1 0 1 1 0 1 0 1 1 1 1 1 0 0 1 Progeny type 1 Y (RAD H1) 2 x 1 0 0 0 1 0 0 0 0 1 0 1 1 0 1 0 1 1 0 0 1 0 0 1 Progeny type 1 Y (RAD H2) 2 x 1 0 0 0 1 0 0 0 0 1 0 1 0 0 1 0 0 1 1 1 1 0 0 1 4 n + 0 and 4 n + n GDGHOP 36 (F) 2 x 0 0 0 1 0 0 1 0 1 0 0 0 1 1 0 1 1 0 1 0 1 0 1 2 x 1 0 0 0 0 0 0 0 0 1 1 0 1 1 0 0 0 1 1 0 1 0 1 Progeny type 1 Y (917 56) 4 x 1 0 0 0 1 0 0 1 0 1 0 0 0 1 1 0 1 1 0 1 0 1 0 1 Myst 107 (F) 2 x 0 0 1 1 0 0 0 0 1 0 1 1 0 0 1 1 1 0 1 0 1 0 1 2 x 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 1 1 0 0 1 0 0 1 Progeny type 1 Y (815 7) 5 x 1 0 0 1 1 0 0 0 0 1 0 1 1 0 0 1 1 1 0 . . 2 n + 2 n Myst 107 (F) 2 x 0 0 1 1 0 0 0 0 1 0 1 1 0 0 1 1 1 0 1 0 1 0 1 2 x 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 1 1 0 0 1 0 0 1 Progeny type 1 Y (813 9) 4 x 1 0 0 1 1 0 0 0 0 1 0 1 1 0 0 1 1 1 0 1 1 1 0 1 2 n + n and n + n LAOP 9 (F) 2 x 1 0 0 1 0 0 0 0 1 1 1 0 0 0 0 1 1 1 1 1 0 0 1 2 x 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 1 1 0 0 1 0 0 1 9 1 W 2 x 1 0 0 0 1 0 0 0 0 1 1 1 0 0 0 0 1 0 1 1 1 0 0 1 9 2 2 x 1 1 0 0 1 0 0 0 0 1 1 1 0 0 0 0 0 1 1 1 1 0 0 1 13 1 2 x 1 . . . . . . . 0 0 1 0 0 1 0 0 1 13 2 2 x 1 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 1 1 0 0 1 0 0 1 15 1 2 x 1 0 0 0 1 0 0 0 0 1 1 1 0 0 0 0 0 1 1 1 1 0 0 1 15 2 2 x 1 1 0 0 1 0 0 0 0 1 1 1 0 0 0 0 1 0 1 0 1 0 0 1 GDOP 4 (F) 3 x 1 0 0 0 0 0 1 0 1 1 1 0 1 1 0 0 1 1 1 0 1 0 1 2 x 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 1 1 0 0 1 0 0 1 Twins (2 pairs) W 4 x 4 1 0 0 1 0 0 1 0 1 1 1 0 1 1 0 1 1 1 1 1 1 0 1 2 x 1 0 0 1 0 0 0 0 1 1 0 1 0 0 0 0 1 1 0 1 1 0 1 2 x 1 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 1 1 0 1 0 1 911 74 1 W 2 x 1 . . . . 1 1 0 0 0 0 0 0 1 1 1 0 0 1 911 74 2 2 x 1 1 0 0 1 0 0 0 0 1 0 1 1 0 0 0 0 1 1 1 1 0 0 1

PAGE 135

135 Table 5 4 Continued. SSR markers M1 M51 M76 M87 Allele 1 2 3 4 5 6 7 8 9 1 2 3 4 5 1 2 3 4 1 2 3 4 5 Mode Group Ploidy N o. 2 n + n and n + n (c ont) 911 75 1 2 x 1 1 0 0 1 0 0 0 0 1 0 1 1 0 0 0 0 1 1 1 0 1 0 1 911 75 2 2 x 1 1 0 0 1 0 0 0 0 1 1 0 1 0 0 0 0 1 1 0 0 1 0 1 DROP 25 (F) 4 x 0 0 0 1 0 0 0 0 1 0 1 1 0 1 1 0 0 1 1 0 0 0 1 LAOP 9 (M) 2 x 1 0 0 1 0 0 0 0 1 1 1 0 0 0 0 0 1 1 1 1 0 0 1 Twins (2 pairs) W 3 x 4 0 0 0 1 0 0 0 0 1 1 1 0 0 0 1 0 0 1 1 1 0 0 1 31 1 W 3 x 1 0 0 0 1 0 0 0 0 1 1 1 0 0 1 1 0 1 0 1 1 0 0 1 31 2 3 x 1 1 0 0 1 0 0 0 0 1 1 1 0 0 0 1 0 0 1 1 1 0 0 1 32 1 3 x 1 0 0 0 1 0 0 0 0 1 1 1 0 0 1 1 0 0 1 1 1 0 0 1 32 2 3 x 1 0 0 0 1 0 0 0 0 1 . . 1 0 0 1 1 0 0 0 1 33 1 3x 1 1 0 0 1 0 0 0 0 1 1 1 0 0 1 1 0 1 1 1 1 0 0 1 33 2 3x 1 0 0 0 1 0 0 0 0 1 1 1 0 0 1 1 0 0 1 1 1 0 0 1 34 1 3x 1 1 0 0 1 0 0 0 0 1 1 1 0 0 0 1 0 0 1 1 0 0 0 1 34 2 3x 1 0 0 0 1 0 0 0 0 0 1 1 0 0 0 1 0 0 1 1 0 0 0 1 35 1 3x 1 0 0 0 1 0 0 0 0 0 1 1 0 0 1 1 0 0 1 1 0 0 0 1 35 2 3x 1 0 0 0 1 0 0 0 0 0 1 1 0 0 0 1 0 0 1 1 0 0 0 1 36 1 3x 1 1 0 0 1 0 0 0 0 0 1 1 0 0 0 1 0 0 1 1 0 0 0 1 36 2 3x 1 1 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 1 1 1 0 0 1 37 1 3x 1 1 0 0 0 0 0 0 0 0 1 1 0 0 0 1 0 0 1 1 1 0 0 1 37 2 3x 1 1 0 0 0 0 0 0 0 0 1 1 0 0 1 1 0 0 1 1 0 0 0 1 38 1 3x 1 1 0 0 1 0 0 0 0 0 1 1 0 0 1 1 0 0 1 1 0 0 0 1 38 2 3x 1 0 0 0 1 0 0 0 0 0 1 1 0 0 0 1 0 0 1 1 0 0 0 1 4x 0 0 0 1 0 0 0 0 1 0 1 1 0 1 1 0 0 1 1 0 0 0 1 2x 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 1 1 0 0 1 0 0 1 19 1 W 3x 1 0 0 0 1 0 0 0 0 1 0 1 1 0 1 1 1 0 0 1 1 0 0 1 19 2 3x 1 0 0 0 1 0 0 0 0 1 0 1 1 0 1 1 1 0 0 0 1 0 0 1 20 1 3x 1 0 0 0 1 0 0 0 0 1 0 1 0 0 0 1 0 1 0 0 0 1 0 1 20 2 3x 1 0 0 0 1 0 0 0 0 1 0 1 1 0 0 0 1 0 1 0 0 1 0 1 21 1 3x 1 0 0 0 1 0 0 0 0 1 0 1 0 0 1 0 0 1 1 0 0 1 0 1 21 2 3x 1 0 0 0 1 0 0 0 0 1 0 1 0 0 0 1 1 0 0 0 0 1 0 1

PAGE 136

136 Table 5 4 Continued. SSR markers M1 M51 M76 M87 Allele 1 2 3 4 5 6 7 8 9 1 2 3 4 5 1 2 3 4 1 2 3 4 5 Mode Group Ploidy N o. 2 n + n and n + n (cont) 22 1 3x 1 0 0 0 1 0 0 0 0 1 0 1 0 0 1 1 0 0 1 . . 22 2 3x 1 0 0 0 1 0 0 0 0 1 0 1 1 0 0 1 0 0 1 0 0 1 0 1 23 1 3x 1 0 0 0 1 0 0 0 0 1 0 1 0 0 0 1 0 1 0 1 0 1 0 1 23 2 3x 1 0 0 0 1 0 0 0 0 1 0 1 0 0 1 1 1 0 0 1 0 1 0 1 26 1 3x 1 0 0 0 1 0 0 0 0 1 . . 1 0 1 1 1 0 1 0 1 26 2 3x 1 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 1 0 1 1 0 1 0 1 GDGHOP 36 (F) 2x 0 0 0 1 0 0 1 0 1 0 0 0 1 1 0 1 1 0 1 0 1 0 1 2x 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 1 1 0 0 1 0 0 1 Twins (2 pairs) W 3x 4 0 0 0 1 0 0 1 0 1 0 1 0 1 1 0 1 1 0 1 1 1 0 1 Z Begins new group of female and male controlled or open pollinations. Y Begins the progeny type for the female and male combination directly above. X Removed Improved in cultivar name. W Begins twin progeny. Twin numbers indicate the order in which they wer e discovered among crosses then as either the first of second twin within the pair. Bolded text indicate s differences between twins.

PAGE 137

137 Figure 5 1. Identification of twins from germination. A) Twin germination from a single tray cell B) Two plants germinating from a single seed. Figure 5 2 Microsatellite analysis of Lantana camara lines to confirm modes of reproduction Female and male lanes are represented with respectively. A) Unreduced female gametes. The lane with an indicates a double unreduced female gamete from GDGHOP B) Apomeiosis from GDOP C) Apomixis from GDGHOP 36 x D) Haploidization from progeny

PAGE 138

138 Figure 5 3. Multiple modes of reproduction demonstrated by L. camara

PAGE 139

139 Figure 5 4. Population interactions of L. camara f rom controlled and open pollination observations. Red lines indicate normally reduced gamete fusion events. Blue and green lines indicate polyploidization events with and without gamete fusion. Black dotted and dashed lines represent paths of clonal and haploidized seed formation. To reduce complexity only 2 x 4 x and 6 x ploidy levels were assumed to contribute normally reduced female gametes (RFG) and reduced male gametes. All ploidy levels are assumed to contribute unreduced female gametes (UFG). On ly observed double UFG (DUFG) events were included as this is expected to be uncommon. All polyploidization events greater than 7 x were not included but may be possible

PAGE 140

140 CHAPTER 6 HYBRIDIZATION POTENT IAL BETWEEN CULTIVATED LANTANA CAMARA AND LANTANA DEPRESSA Rationale Lantana depressa is a small native shrub endemic to Florida, growing in pine rockland of Miami Dade County in southern Florida. Plants of this species are small usually <0.5 m tall, with branches radiating, prostrate, reddish o r purplish, more or less hispid. Leaves of L. depressa are induplicate or longitudinally incurved ovate to elliptic, 1 3.5 cm long; the angle between the midrib and the basal margin of lamina is 50 or less. The flowers of L. depressa open yellow and tur n tawny orange with age, and they produce fleshy, shiny black fruits (drupes) (Sanders, 2001 ) All L. depressa accessions surveyed by Sanders (1987) were diploids (2 n = 2 x = 22) with normal bivalent pairing. In a subsequent publication, Sanders (2006) me ntioned autotetraploidy in L. depressa Sanders (1987) recognized the above described L. depressa as L. depressa var. depressa and two additional varieties that occupy ecologically and geographically distinct areas. L. depressa var. floridana occurs on s tabilized dunes of the Atlantic coast barrier islands and relictual dunes of central Florida; and var. sanibelensis occupies calcareous dunes along the Gulf of Mexico and wet limestone prairies in southwestern Florida (Sanders, 1987). These two varieties are nearly identical in vestiture, laminar shape, laminar curvature, laminar lustrousness, inflorescence bract shape and persistence, and fruit color to L. depressa var. depressa but they are large, bushy shrubs. Recently taxonomic authorities working on Florida taxa revised their classification and asserted that L. depressa var. floridana and var. sanibelensis were synonymous for L. camara while L. depressa var depressa remained a separate taxon (Wunderland and Hansen, 200 4 ).

PAGE 141

141 Lantana camara originated from South America through Mexico, and the West Indies ( Sanders, 2001; Sanders, 2006 ) It was introduced as early as 1687 to Europe as a landscape plant. Substantial selection and hybridization occurred in Europe, resulting in various combina tions of plant growth habits, flower colors, etc. By the late nineteenth century, there were over 600 named cultivars (Howard 1969). European colonists further introduced L. camara cultivars throughout both the Old and New World Tropics. L. camara h as naturalized in Florida (Sanders, 1987 a ) and is documented in nearly every Florida county and many habitats. Naturalized plants are vigorous, large, and weedy They can grow in thickets reaching 4 m in height and produce dense prickles on branches. T heir leaves are cordate to ovate elliptic, flat or slightly undulate, not induplicate or longitudinally incurved. Flowers are multi colored red, pink, orange, and yellow. Cytological studies revealed that Florida naturalized L. camara comprised only tet raploids and occurred almost exclusively in disturbed habits (Sanders, 1987). However, L. camara probably originated as a diploid, because some diploid cultivars are still available (Sanders, 2006). Post origin allopolyploidy has contributed to L. camara aggressive growth and success in the wild (Sanders, 2006). Numerous natural hybrids between L. depressa and naturalized L. camara have been observed sympatrically or in close proximity to one or both species (Sanders, 1987). Hybrids were triploid and m orphologically intermediate between L. depressa and L. camara These observations indicate that L. depressa and naturalized L. camara are hybridizing, which puts L. depressa in great danger. Over the last 40 50 years, the native habits for L. depressa

PAGE 142

142 ha ve been altered and/or destroyed substantially due to oceanside development, urban expansion, and agricultur e As the natural hab itat s are disturbed, naturalized L. camara invades and hybridize s with the native species and produce triploids. These triploids seem to have combined the vigor of L. camara and the adaptations to local habits of L. depressa and they can persist and spread (Sanders, 1987 a ). Sanders (1987 a ) confirmed by hand pollinati on in the greenhouse that naturalized L. camara tetraploids and L. depressa were fully crossable and interspecific pollinations resulted in 31% to 40% fruit set. As a popular nursery and landscape plant, L. camara receives a considerable amount of attentio n from corporate or independent breeders. Over the last several decades, numerous new cultivars have been developed in L. camara Many of these cultivars have much reduced pollen viability and/or female fertility. As shown in Chapter 2, pollen stainabil ity of current L. camara [ a 1 200 fold ] has been observed among these cultivars (Chapter 3). As shown in Chapters 4 and 5, some cultivars can adopt multiple modes of reproduction, either sexual or asexual, to produce seeds. These factors can have significant effects on the hybridization potentials of L. camara cultivars w ith L. depressa Little information on such aspects is available. Thus the objectives of this study are to 1) Assess the hybridization potential of L. camara as a male parent with L. depressa 2) Assess the hybridization potential of L. camara as a fema le parent with L. depressa and 3) analyze the ploidy level of the resultant interspecific hybrids.

PAGE 143

143 Materials and Methods Plant Materials Ten L. camara cultivars and 15 breeding lines were used includ ing two diploid s eight triploid s 10 tetraploid s two pentaploid s and three hexaploid s. They represented a range of ploidy levels and pollen stainability (from 1.5 to 79.1%) (Table 6 1). The L. depressa accession used was provided by Pro Native Consulting (Miami, FL). Flow cytometry analysis found the L. depressa material provided to be tetraploid. All plants were grown in a commercial soilless potting mix ( Fafard 2B Anderson, SC) supplemented with a controlled release fertilizer ( Osmocote 15N 3.9P 10K, 8 9 months release at 21 C; The Scotts Company, M arysville, OH) at the rate of 6.51 kg m 3. Hand Pollination P lants used for pollination were grown in the greenhouse and evenly spaced on metal benches with a spacing of 30.5 cm. Pollinations were conducted over two seasons The first was from 29 Septe mber to 11 November 2009 and the second from 6 April 2010 to 12 May 2010. Flowers were emasculated prior to anther dehiscence and pollinated immediat ely after emasculation with fresh pollen. Fruit set was recorded as ripe fruit was harvested ~3 4 weeks a fter hand pollination. Ten commercial cultivars and 15 lines of L. camara that represented five ploidy levels (2 x to 6 x ) and a wide range of pollen stainability (1.5% to 79.1%) were selected as the pollen source for pollinating L. depressa and assessing L. camara cause fruit (seed) set on the native species. Hand pollinations were conducted in early to late fall 2009 (the first season) and repeated in mid to late spring 2010 (the second

PAGE 144

144 season). More than 100 flowers were pollinated per cros s in a season with one exception when flowers were not available. Pollen Staining Pollen staining was done with cotton blue vital stain similar to chapter 2 on one plant of each line of L. camara upon completion of the first season of pollinations. L. de pressa stainability was assessed using three randomly selected plants from the crosses. The pollen stainability from Chapter 2 was averaged with those data collected during this study. Seed Germination Ripe fruit was collected and stored under ambient (22 .2 C ) conditions until seed extraction. All pulp was removed from the seed and soaked in water overnight (Heit 1946) cleaned and sown on 8 February and 16 July 2010 for the first and second seasons of pollinations respectively. Seed were sown in a green house on the surface of Fafard 2B potting soil under intermittent mist. After germination seedling phenotypes were examined to determine if their phenotypes were intermediate between the parents Progeny Ploidy Analysis Analysis was performed using fully expanded young leaves and the Partec PA I ploidy analyzer and the CyStain UV Ploidy Precise P dye (Partec, Germany). The manufacturer recommended ploidy analysis procedure was followed with minor modifications (supplemented with 2% w/v PVP and 0.01% me rcaptoethanol) with dye mixture kept on ice The ploidy level of progeny was determined by comparing to one or more commercial cultivars (reference cultivars) with known ploidy levels that were included in the same analysis.

PAGE 145

145 Results and Discussion L. camara a s t he Pollen Source The highest fruit set was 28.3% ) in the first season and 25.2% in the second season (Table 6 1). Seasonal effects seemed to be evident in a number of L. depressa the fruit set was 28.3% in the first season, source, the fruit set was 9.0% and 2.4% in the first and the second season, respectively. increased to 16.1% in the second season. Se asonal changes might be due to the difference in temperature between the two seasons, or plant response to temperature fluctuations, etc. Correlation analysis indicates that pollen stainability of L. camara is the most important factor determining the percentage of fruit set on L. depressa An R 2 value of 0.9405 was obtained by fitting the observ ed pollen stainability to the observed fruit set with a polynomial equation: y = 0.3832 0.096x + 0.0089x 2 0.0002x 3 + 0.000002x 4 (Figure 6 1), where y is fruit set (%) and x the pollen stainability (%) This relationship suggests that as the pollen st ainability decreases fruit set can drop more rapidly. Figure 6 1 suggests that when pollen stainability is below 10%, pollinated flowers would generally not set fruit. L. depressa seeds resulting from L. camara pollination had low germination (0 to 33.3 %), similar to the germination of L. depressa pollinated seeds (S. Wilson, personal communication) (Table 6 2).

PAGE 146

146 As described in Chapter 2, the main factor determining L. camara stainability is its ploidy level. Thus, the following discussio n of the various cause fruit set on L. depressa will be by ploidy level. Diploid L. camara c ultivars Th is group consisted of their pollen, 14.5% to 28.3% of L. depressa the most compatible with L. depressa Nevertheless, L. depressa seeds from crosses with the two cultivars had a low germination rate (5.5%) (Table 6 2 ). Triploid L. camara cultivars/bre eding lines Three commercial triploid cultivars and five triploid lines were used as pollen source s to pollinate nearly 1 300 L. depressa flowers. Four triploids with pollen iss 4, and 605 35) did not caus e any fruit set in either season. Of the four triploids with 11.5 to 14.9% pollen stainability, one (623 43, and 617 1 ) did cause fruit set on L. depressa ranging from 0.3% to 3.2 %. However, none of the seed extracted from the fruit germinated. These results show that triploid L. camara with low pollen stainability (<10%, even <15%) ha d little potential to cross pollinate L. depressa Tetraploid L. camara c ultivars/breeding l ines This group consiste d of four commercial cultivars and six breeding lines. When pollinated with these cultivars/lines, L. depressa had an average fruit set of 3.1%, only 14% of the pollination success rate of diploid L. camara but about 10 fold of that of the triploid L. ca mara As pollen stainability varied among tetraploids, so did the pollination success rate. In one pollination caused 16.1% of the flowers to set fruit on L.

PAGE 147

147 depressa close to the pollination success rate of diploid L. camara In most crosses, however, the observed fruit set was between 0.6% and 6.0% (Table 6 1). Two tetraploid lines, CAOP 73 and 60 4 5, did not cause any fruit set on L. depressa This result seems to be understandable for 604 5, as its pollen stainability was l ow (12.1%), but it was unexpected for CAOP 73, which had 40.7% pollen stainability. There might be other mechanism(s) in CAOP 73 that ma d e CAOP 73 and L. depressa less compatible. On the other hand, CAOP 88, even with only 3.2% pollen stainability, was a ble to cause some fruit set in one season. L. camara p entaploids and h exaploids 1) and three hexaploids (PIT 2, PIT 20, and 620 1) were available as the pollen source for pollinating L. depressa and two L camara lines affected 0.5% to 5.7% fruit set, and two lines did not, even though they had similar pollen stainability (24.6% to 39.1%) (Table 6 1). On average, pentaploids and hexaploids appeared to be less crossable with L. depressa than tetraploids. N one of the seeds (15) extracted from fruit of L. depressa pentaploid L. camara germinated, while the only seed from the L. depressa hexaploid L. camara cross (620 1) germinated. L. camara a s t he Female The same 10 commercial cultivars and 15 lines w ere used as female parents and pollinated with L. depressa which had an average pollen stainability of 63% (from two greenhouse tests) and an average of 67.3% from the plants used for pollinations. A total of 104 to 270 flower s were pollinated per cross in S eason one (September November 2009), and 140 to 304 flowers were pollinated per cross in S eason two (April May 2010 ). L. camara fruit set varied between 0 and 45.9% in S eason one and

PAGE 148

148 betw een 0 and 30.0% (Table 6 3) in S eason two among cultivars/lines. Seasonal effects 7 (plus or minus 28.0% to 28.8%). When po llinated with L. depressa thens Rose and 604 5 did not set any fruit in both seasons, suggesting that they are highly female sterile. It has been shown that ploidy level (Chapter 3) and reproducti ve mode (chapter 5) strongly affect L. camara et in open pollination (OP) and OP seed germination. Table 6 4 indicates that these two factors played a similar ly important role in L. camara s fruit (or seed) production in hand pollination with L. depressa as well as seed germination. Diploid L. camara seasons was 11.4% and 21.7%, respectively, being the fifth and second highest among the average fruit set data for all 25 L. camara cultivars/lines evaluated. As a ploidy group, the average fruit set was 16.5%, representing the highest among the five ploidy levels. Seeds resulting from L. depressa germinated, with 22.3% germination percentage. This germination rate was similar to that from chapter 3 (24.1%). All germinated plants we re found to be the result of successful pollinations based on intermediate phenotypes. Triploid L. camara Fruit set of three triploid cultivars and five triploid lines varied, with the two season average ranging from 0 to 31.4% (Table 6 3). Three of the triploids 4 are UFG producers, and their two season average fruit set was 31.4%, 1.4%, and 7.8%, respectively. Among the five non UFG producing triploids, 605

PAGE 149

149 35, 605 43, 617 1 and 623 any fruit in both seasons. The average fruit set of these non UFG producing triploids was 0.4% (from 1,396 flowers pollinated), in contrast with the 13.5% average fruit set of the UFG producing triploids (33.8 fold higher). Because of the low fruit set on non UFG producing triploid cultivars/lines, only 11 seeds were obtained (Table 6 4). The germination percentage of these seeds was 27.2% compared to 11.1% germination in C hapter 3. A total of 145 seeds were obtained from the three UFG producing L. camara triploids, and the seed germination was 47.5% (29.3% chapter 3), approximately 75% higher than that of non UFG producing triploids. These results show that certain triploid L. camara lines were highly crossable as a female with L. depressa and c ould produce a large number of fruit S eeds from these fruit c ould germinate at high percentages. UFG production played a very important role in the high fruit set and seed germination of these triploids. In contrast non UFG producing tri ploid L. camara were the least crossable as a female with with L. depressa among L. camara polyploids and these triploids set only a very low percentage of fruit. Tetraploid L. camara Fruit set varied considerably among L. camara tetraploid cultivars and lines. Line 604 88, and 611 4.1%, respectively (Table 6 3). Three of the 10 tetraploids were known to produce UFGs and seven were not. When data were pooled and averaged, the UFG producing tetraploids had 7.9% fruit set, and the non UFG producing tetraploids had 9.1% fruit set. It seems that the UFG production trait did not make any difference in changing fruit set in tetraploid L. camara

PAGE 150

150 L. depressa crosses. T he role of UFG production seems to be very obvious in seed germination (Table 6 4). Seeds from non UFG producing L. cama ra tetraploids had 18.5% germination, similar to those from non UFG producing diploid or triploid L. camara but seeds from UFG producing L. camara tetraploids had 54.4% germination, nearly 2 fold increase compared to the non UFG producing tetraploids. P entaploid and h exaploid L. camara 1) over the two seasons was between 2.2% and 4.0%, which was relatively low. A total of 22 seeds were obtained from the 690 flowers pollinated, of which only two germinated, resulting 9.1% germination, lower than that of diploids, triploids and tetraploids (Table 6 4). The average fruit set of three hexaploids (PI T 2, PIT 20 and 620 1) over the two seasons was 0.3 to 1.2% (Table 6 4). As a ploidy level, the average fruit set was 0.8%, only slightly higher than that of non UFG producing triploids (0.3%), suggesting that hexaploids were overall highly female sterile and rarely crossable with L. depressa None of the six seeds obtained from hexaploid L. camara L. depressa crosses germinated. Ploidy a nalysis of p rogeny from L. camara and L. depressa c rosses Ten of the 25 crosses between L. camara and L. depressa resu lted in a total of 150 progeny. The number of progeny per cross varied from one to 63. The most successful crosses involved two L. camara diploids, one triploid, and three tetraploids. The cross between diploid L. camara L. depressa produced 12 progeny, known not to produce UFGs and the L. depressa accession was expected to be a diploid from previous reports (Sanders 1987a). The L. depressa accession provide d for

PAGE 151

151 this experiment was non distinguishable morphologically from typical L. depressa but further analysis showed that the L. depressa accession was a tetraploid. The discovery of this tetraploid suggests that natural polyploidization had occurred in L. depressa Similarly L. depressa produced triploids. When L. depressa (4 x x ), the progeny were not triploids as expected; instead they were pentaploids (Table 6 5). This type of 2 n + n progeny were also observed in the cross between L. depressa x ), between L. depressa and breeding line 611 7. These results suggest that the L. depressa accession used in this study carried the UFG production trait like some L. camara cultivars. Apomixis seems to have occurred in this L. depressa accession as well. The cross between L. depressa 5). It could be either a n + n or 2 n + 0 progeny. Morphologically it was nearly ident ical to L. depressa Thus th e descendant is more likely to be apomictic than zygotic. L. depressa ) produced 26 progeny, 20 of which were tetraploids and six hexaploids. These hexaploids would be 2 n + n type, but the tetraploids could be 2 n + 0 or n + n produce UFGs and apomictic seed. Thus their origin remains to be determined. Morphological and /or molecular marker characterization could help address this question. The crosses between DROP 25 or 611 7 and L. depressa also produced tetraploid progeny (Table 6 5). As DROP 25 and 611 7 were known not to produce UFGs or

PAGE 152

152 apomictic seeds, these tetr aploid progeny would be more likely to be n + n type than 2 n + 0 type. L. depressa yielded 63 progeny. A great morphologically, indicating tha and apomixis (Table 6 L. depressa (2 n + n ). Summary Hand pollinations were made between 25 L. camara cultivars /breeding lines and one L. depressa accession to assess hybridization potential between the two species. The L. camara cultivars/breeding lines used represented five ploidy levels (diploid to pentaploid) a wide range of pollen stainability (1.5% to 79.1% ) and fruit production capacity (0.006 to 7.137 fruit per peduncle from Chapter 3 ). Results indicate that pollen stainability of L. camara was the most important factor determining the potential of L. camara as a male parent to hybridize L. depressa and c ause fruit production. As L. camara L. camara could cause on L. depressa dropped rapidly. As L. camara pollen stainability dropped to below 10%, it caused very little fruit set on L. depressa Among the five ploidy levels, diploid L. camara cultivars most compatible with L. depressa followed by tetraploids T riploid L. camara with low pollen stainability (<10%, even <15%) ha d little potential to cross pollina te L. depressa Pentaploids and hexaploids appeared to be less crossable with L. depressa than tetraploids. The same L. camara cultivars /breeding lines were used as female parents and pollinated with L. depressa Results reveal that L. camara ploidy level and mode of

PAGE 153

153 reproducti on play ed important role s in fruit (or seed) production after hand pollination with L. depressa D iploid L. camara produced the highest number of fruit among the five ploidy levels. UFG producing triploid L. camara lines were highly crossable as a female with L. depressa and produced large numbers of fruit and seeds, whereas non UFG producing triploid L. camara were the least crossable as female with L. depressa The U FG production trait made no difference in fruit set in tetraploid L. camara L. depressa crosses but caused large differences in seed germination. Ploidy analysis results revealed that natural polyploidization had occurred in L. depressa Ploidy level distribution among the progeny of crosses between L. camara and L. depressa suggest that the L. depressa accession used in this study carried the UFG production and apomixis traits like some L. camara cultivars did. This is the first report of such traits in L. depressa Further analys e s of morphologica l and molecular markers are needed to confirm the presence of UFGs and apomixis in L. depressa Collectively results from this chapter suggest that hybridization potential was more prevalent when L. depressa was the pollen donor and L. camara the pollen recipient. This may be due to the fact that the male fertility of many L. camara cultivars/breeding because of the UFG formation and apomixis traits. Thus eliminating U FG formation and apomixis is critical for developing highly sterile L. camara cultivars.

PAGE 154

154 Table 6 1. Fruit set on L. depressa flowers pollinated with L. camara Hand pollination was performed in greenhouse condition s in fall 2009 (Season 1) and repeated spring 2010 (Season 2). L. camara cultivar/line Ploidy Stainable p ollen (%) Z L. depressa f lowers p ollinated (no.) Fruit set (%) Average fruit set (%) Season 1 Season 2 Season 1 Season 2 Cream 2 x 79.1 106 129 20.6 25.2 22.9 Lola 2 x 74.5 107 49 28.3 14.5 21.4 Athens Rose 3 x 14.9 117 80 3.2 0.0 1.6 M is s Huff 3 x 1.8 125 119 0.0 0.0 0.0 New Gold 3 x 1.5 18 100 0.0 0.0 0.0 GDOP 4 3 x 6 .0 99 111 0.0 0.0 0.0 605 35 3 x 7.4 104 105 0.0 0.0 0.0 605 43 3 x 14.8 105 109 1.0 0.0 0.5 617 1 3 x 11.5 110 0.3 0.3 623 5 3 x 14.6 111 125 0.0 0.0 0.0 Bandana Cherry 4 x 37.4 95 93 4.4 2.3 3.3 Carlos 4 x 51.2 109 114 2.7 6.0 4.3 Dallas Red 4 x 29.4 102 105 0.9 0.0 0.5 Pink Caprice 4 x 68.7 305 93 1.6 16.1 8.9 CAOP 73 4 x 40.7 126 0.0 0.0 CAOP 88 4 x 3.2 109 71 1.0 0.0 0.5 DROP 25 4 x 37.4 338 0.6 0.6 PCOP 6 4 x 39 .0 113 1.6 1.6 604 5 4 x 12.1 118 0.0 0.0 611 7 4 x 47.6 111 94 6.3 1.1 3.7 Cajun Pink 5 x 39.1 123 122 9.0 2.4 5.7 629 1 5 x 32.8 102 112 0.0 0.0 0.0 PIT 2 6 x 26.1 106 79 1.0 0.0 0.5 PIT 20 6 x 24.6 101 129 2.0 0.0 1.0 620 1 6 x 31.5 78 0.0 0.0 Z Pollen staining rates from Chapter 2 were combined with another pollen sample during the first season of pollinations.

PAGE 155

155 Table 6 2. Differences among L. camara ploidy levels in causing fruit set on L. depressa flowers and seed germination. Fruit set (%) Seeds s own (no.) Seeds g erminated (no.) Seed germination (%) Ploidy level of L. camara as Season 1 Season 2 Average 2 x 24.4 19.9 22.2 91 5 5.5 3 x 0.6 0.0 0.3 3 0 0 .0 4 x 1.9 4.3 3.1 47 4 8.5 5 x 4.5 1.2 2.8 15 0 0 .0 6 x 1.0 0.0 0.5 3 1 33.3

PAGE 156

156 Table 6 3. Fruit set of L. camara flowers pollinated with L. depressa Hand pollination in greenhouse conditions was performed fall 2009 (Season 1) and repeated spring 2010 (Season 2). L. camara cultivar/line Ploidy UFG p roducer L. camara f lowers p ollinated Fruit set (%) Average fruit set (%) Season 1 Season 2 Season 1 Season 2 Cream 2 x No 270 249 14.8 7.9 11.4 Lola 2 x No 224 304 18.5 24.9 21.7 Athens Rose 3 x No 165 169 0.0 0.0 0.0 M is s Huff 3 x Yes 191 158 45.9 17.0 31.4 New Gold 3 x Yes 124 140 1.5 1.3 1.4 GDOP 4 3 x Yes 149 204 7.1 8.4 7.8 605 35 3 x No 184 188 0.5 0.4 0.5 605 43 3 x No 191 280 0.5 0.3 0.4 617 1 3 x No 219 1.7 1.7 623 5 3 x No 200 221 0.0 0.5 0.3 Bandana Cherry 4 x No 104 162 9.6 8.8 9.2 Carlos 4 x No 182 197 0.6 7.6 4.1 Dallas Red 4 x No 162 151 1.5 1.8 1.7 Pink Caprice 4 x Yes 235 239 14.4 8.2 11.3 CAOP 73 4 x No 128 8.7 8.7 CAOP 88 4 x No 182 197 14.5 10.9 12.7 DROP 25 4 x No 148 8.1 8.1 PCOP 6 4 x Yes 178 8.2 8.2 604 5 4 x Yes 176 0.0 0.0 611 7 4 x No 149 189 2.0 30.0 16.0 Cajun Pink 5 x Yes 141 191 5.4 2.6 4.0 629 1 5 x Yes 202 156 4.4 0.0 2.2 PIT 2 6 x Yes 161 147 0.6 0.0 0.3 PIT 20 6 x Yes 140 166 0.0 1.9 1.0 620 1 6 x Yes 153 1.2 1.2

PAGE 157

157 Table 6 4. Total seed set and germination rates when L. camara was used as a female for L. depressa pollen over two seasons. Fruit set (%) L. camara ploidy level Gamete type Season 1 Season 2 Average Seed sown (no) Seed germ Germ. % 2 x Non UFG 16.7 16.4 16.5 179 40 22.3 3 x Non UFG 0.6 0.3 0.4 11 3 27.2 3 x UFG 18.2 8.9 13.5 145 69 47.5 4 x Non UFG 6.4 11.8 9.1 168 31 18.5 4 x UFG 7.5 8.2 7.9 68 37 54.4 5 x UFG 4.9 1.3 3.1 22 2 9.1 6 x UFG 0.6 1.0 0.8 6 0 0 .0 Table 6 5. Ploidy level distribution of crosses between L. camara and L. depressa Seed parent ( p loidy) Pollen parent ( p loidy) Progeny available (no.) Ploidy level distribution in progeny Possible modes of reproduction 3 x 4 x 5 x 6 x Cream (2 x ) L. depressa (4 x ) 12 12 n + n L. depressa (4 x ) Cream (2 x ) 3 3 2 n + n Lola (2 x ) L. depressa (4 x ) 24 24 n + n L. depressa Lola (2 x ) 1 1 2 n + n Miss Huff (3 x ) L. depressa (4 x ) 63 61 2 2 n + 0, 2 n + n Pink Caprice (4 x ) L. depressa (4 x ) 26 20 6 2 n + 0, 2 n + n L. depressa (4 x ) Pink Caprice (4 x ) 1 1 n + n or 2 n + 0 DROP 25 (4 x ) L. depressa (4 x ) 6 6 n + n 611 7 (4 x ) L. depressa (4 x ) 13 13 n + n L. depressa (4 x ) 611 7 (4 x ) 1 1 2 n + n

PAGE 158

158 Figure 6 1. Polyno mial relationship between L. camara pollen stainability and fruit set of L. depressa flowers pollinated with L. camara

PAGE 159

159 CHA PTER 7 DEVELOPING STERILE TRIPLOIDS IN LANTANA CAMARA Rationale The ornamental plant industry is a major contributor to Florida with an annual wholesale value of nearly $ 700 million (USDA report 2010) ( Hodges and Haydu 2006 ; www.usda.gov ). L. camara is a very important crop of this industry, contributing to more than 1% (~$40 million) of the total annual wholesale value. According to a survey of the Florida nurseries, 19% of the responding n urseries produce d L. camara (Wirth et al. 2004). Yet L. camara can hybridize with L. depressa a native lantana species endemic to southern Florida (Sanders 1987a). Because of this, L. camara is listed as a C ategory I invasive species in Florida (FLEPPC 2009) Sterile cultivars are n production. Research and Education Center to develop new sterile L. camara cultivars The poly ploid nature of the species (Sanders 2006) was though t to provide numerous opportunities and avenues for breeding (Eigsti 1957). Initial efforts produced a number of highly sterile triploids, but they had poor growth habits for ornamental use. To overcome this difficulty, additional tetraploids were used as breeding parents in interploid crosses. Triploids produced from these cros ses had mounding or semi spreading growth habits but produced more fruit (seed) than desired. To improve the effectiveness and efficiency in genetic sterilization of L. camara the male and female fertility of 26 L. camara cultivars was investigated (Chap ters 2 and 3). The majority of the triploid cultivars had pollen stainability below 10%, even as low

PAGE 160

160 as 1.8%. Thus, high levels of male sterility can be achieved in L. camara through ploidy manipulation, particularly triploid production. However, most t riploid cultivars produced as much fruit (or seed) as diploids or tetraploids did. Very few of the existing commercial L. camara cultivars have the desirable levels of male and female sterility. Onl y a weak correlation was found between male and female fe rtility in L. camara polyploids (Chapter 3; Khoshoo and Mahal 1967; and Spies 1984c). To understand this phenomenon, ploidy and molecular marker analysis were performed on L. camara progeny from controlled as well as open pollinations. Results have sho wn that L. camara has evolved the ability to produce unreduced female gametes (UFG) and to develop seed through apomixis (Czarneck i and Deng, 2009; Chapter 5). These traits have enabled L. camara triploids to produce fruit (seed). Many L. camara cultivar s can reproduce through multiple pathways. Therefore, it is necessary to identify diploids and tetraploids that do no t have these traits to develop highly sterile cultivars in L. camara ,. The objectives of this study we re to 1) produce new triploids thr ough interploid crosses, 2) evaluate the effects of parental combination, direction of pollination, and environmental factors on pollination success, 3) select promising sterile triploids, 4) assess their male and female sterility, 5) determine their hybri dization potential with L. depressa 6) evaluate their plant performance, and 7) select the best triploids as candidates for releas e Materials and Methods Parental Plant Materials breeding line (LAOP (Table 7

PAGE 161

161 1) Parental plants were propagated by cuttings and grown in 30.5 cm plastic containers filled with a commercial soilless mix, VerGro container mix A (Verlite Co., Tampa, FL) amended with controlled release fertilizer, Osmocote (15N 3.9P 10K, 8 9 months release at 21 C; The Scotts Company, Marysville, OH) at 7.12 kg m 3 Containerized parental plants were grown in the greenhouse set to 29 / 21 C day/night temperature under natural light and ranged from 32 t o 15 C Parental plants used for growth chamber experiments were grown in 15.2 cm plastic containers in the so il mix previously described from 23 April 2007 to 7 August 2007. Three Percival Scientific growth chambers (model E 30B, Boone, IA) set to 16 hour days and held at 21.1 26.7 and 32.2 C were used. After pollinations were completed and seed were collecte d, the temperatures were changed in rotation across the three chambers. Plants were given two weeks to acclimate at each temperature before pollinations were started again. Hand Pollination F lowers were emasculated by removing the corolla prior to opening Emasculated flowers were pollinated by transferring fresh pollen from a newly opened flower with a small paint bush. The brush was cleaned in 100% ethanol and dried before it was used to pollinate another flower. When l antana fruit turned dark purple to black they were collected for seed extraction Crosses were made in 20 April 2006 28 June 2006, 20 September 2006 30 January 2007, and 15 February 2007 31 July 2007. Progeny Growing and Evaluation Seeds were extracted and cleaned within 1 to 2 weeks aft er fruit was harvested. Seeds were sown on the surface of a peat/vermiculite mix (VerGro container mix A,

PAGE 162

162 Verlite, Tampa, FL) and germinated under intermittent mist in a greenhouse. Temperatures in the greenhouse ranged from 16 C (night) to 30 C (day), and no artificial lighting was used. The majority of seeds germinated within four months after sowing, but some took as long as a year to germinate. After young seedlings had developed true leaves, they were transferred to 1 0.2 cm plastic contai ners filled with VerGro container mix. Plants were fertilized by incorporating Osmo cote (15N 3.9P 10K) at 7.12 kg m 3 in the soilless mix. When seedlings were 2 to 4 month old, ploidy a nalysis was performed using fully expanded young leaves and the Part ec PA I ploidy analyzer and the CyStain UV Ploidy Precise P dye (Partec, Germany). The manufacturer recommended ploidy analysis procedure was followed with minor modifications (supplemented with 2% w/v PVP and 0.01% mercaptoethanol) with dye mixture kep t on ice Identified triploids were first evaluated in the greenhouse for their growth habit, leaf characteristics, and flower characteristics. Plants were grown in 10.2 cm pots with the potting soil as described above. Those that had weak vigor, errat ic growth habits, few leaves, and/or undesirable flowers were eliminated. The selected triploids were then planted in the ground beds mulched with white plastic. Plants in the ground beds were irrigated biweekly and fertilized with 29.7 cm 3 O smocote (15 N 3.9P 10K, 8 9 months release The Scotts Company). Again, triploids that had weak vigor, erratic growth habits, few leaves, and/or undesira ble flowers were eliminated. Statewide t rials of p romising t riploids Mature cuttings were taken on 14 16 March 2009 Grow and rooted in 128 cell

PAGE 163

163 speedling trays in Fafard 3B (Anderson, SC) potting soil. Rooted cuttings were grown in 10.2 cm pots with the potting soil as describe d above. When plants were 2 month old, they were distributed to three locations. The two northern sites were pure plantings of only L. camara lines. Northern sites were located at the North Florida Research and Education Center (NFREC) Quincy, Florida and at the University of Florida Plant Science Research and Education Unit located near Citra, Florida The two locations in central Florida we re mixed plantings of L. camara and L. depressa (the Florida native lantana) at Balm, Florida at the Gulf Coast Research and Education Center (GCREC) and Ft. Pierce, Florida at the Indian River Research and Education Center (IRREC). Planting was completed the week of 5 May 2009. All the trial sites included the GCREC. To prevent fertile pollen drift location planted in the main plot with the other trialed triploid lines. The plants were grown in raised beds under artificial white on black plastic mulch on 1.8 m centers and fertili zed with two 29.7 cm 3 of 15 9 12 slow release 12 14 month O smocote (The Scotts company, Marysville Ohio). To provide adequate spacing for growth p lants were planted in 2 plant plots separated by the native lantana for the mixed planting sites across thre e blocks at each site providing a 3.6 m separation from the experimental lines in each plot. Watering durations and rates were similar across locations ( two hours twice a week from drip irrigation ).

PAGE 164

164 Evaluating m ale s terility Ten UFBLs and two commercial c ultivars ( Pink Caprice and ) and the Florida native L. depressa planted in raised bed s at the GCREC and IRREC were screened. Collection of anthers was done on three intervals, 24 September, 6 October, and 16 November of 2009. Pollen wa s stain ed with cotton blue vital stain ( Eng. Scientific, Inc. Product No. 6730, Clifton, NJ). Anthers were collected from predehiscent flower clusters. Four a nthers from three flower clusters were collected from each plant sample into 1.5 m L eppendorf tubes with approximately 100 L of stain. A total of 8 12 anthers were placed in stain were then allowed to soak in the stain overnight at 65C. Anthers were triple rinsed with distilled water placed onto microscope slides and fixed with a 4: 1 glycerol to water ratio. Slides were observed at a 40 0 x magnification and pictures were taken for manual scoring later. Evaluating f emale s terility Determining the female fertility of the triploid and commercial cultivars was done by randomly collecti ng 20 fresh peduncles with entirely spent flowers from each plant in a plot (similar to Chapter 3). Peduncles were bulked by plots then stripped of any seed present and an average seed per peduncle was calculated. Evaluating p lant p erformance Plants were evaluated at four week intervals starting at planting (5 May 2009) until the conclusion of the study on 14 December 2009. The two characteristics considered at all four sites were plant quality and flowering intensity. Plant quality (PQ) was rated on a s cale of 1 to 5. A rating of 1 indicated that the plant was growing poorly with few shoots, sparse foliage, and/or severe diseases, while a rating of 5 indicated that the plant grew well with dense foliage. A rating of 5 would be a very desirable

PAGE 165

165 plant. Flowering intensity (FI) was a visual rating of the percentage of maximum bloom. A categorical scale from 1 5 representing 0 20%, 21 40%, 41 60%, 61 80% and 81 100 of the maximum flower coverage of the plant. Statistical Analysis D ata were analyzed using PROC GLM in SAS for Windows 9.2 (SAS Institute Cary, NC) to determine the signific ance of differences among ploidy levels l antana lines, and seasons Percentage d ata w ere t ransform ed using the a rcsine s quare root function when necessary. D ifferences were determined among evaluation weeks, mean separation in SAS Results Triploid Generation Selecting p arent s for interploid crosses Pollination success rates ranged from 0 18.3%, with an average of 2 .5% (data not later habits and light flower colors ( personal observations the best parent among the three diploid cultivars in terms of pollination success rate. When hand pollinated, 0.8 s flowers set fruit were tetraploids or hexapl oids, rather than triploids, because these two tetraploid cultivars formed unreduced female gametes and apomictic seeds (Chapters 4 and 5).

PAGE 166

166 Thus they were not desirable parents for producing sterile triploids. When crossed with mor e desirable tetraploid parent. The majority of Dallas Red erratic growth habit, but this tetraploid cultivar did not produce unreduced female gametes or apomictic seeds and was one of only a few cultivars available at the time with a red flower color, thus it was selected as a tetraploid parent. pollination progeny in early 2007, one line with a more compact growth habit and darker yellow flower tha n as LAOP 9 and use as a compact mounding type Effects of parental combination on pollination success LAOP 9) were paired in eight interploid crosses to evaluate the effects of parental combinations on pollination success. As controls for the interploid crosses, the two tetraploid parents were paired in two crosses, and the two diploid parents were paired in another two crosses. Thus a total of 12 crosses were designed from the four parents. Without considering the ploidy level differences, these crosses resembled a full diallel mating design. Hand pollinations were performed in February through July 2007. Fruit set percentages of pollinated flowers were reco rded as pollination success rates. The number of flowers pollinated over the 6 month period for each interploid cross ranged from 1204 to 1894 (Table 7 2). The observed pollination success rates varied considerably among the interploid crosses, from 0.1% to 9.1% during the Feb. May pollination period and 1.2% to 38.2%

PAGE 167

167 during the June July pollination period (Table 7 2). Overall, pollination success rates were higher when the diploid pa rent was LAOP A strong seasonal effect on pollination success seemed to exist (to be discussed separately later). The direction of crossing and the level of compatibility between parents seemed to affect pollination success to a la rge extent. Taking the June July f pollinated flowers set fruit the maternal parent. The parents were reciprocally crossed and 20.5% of the pollinated flowers set fruit, which is 3.8 fold higher (Table 7 3). A similar 9 and in the cross 9. In these crosses, the pollination success rate was 3 or 1.5 fold higher when the maternal paren t was a tetraploid than when it was a diploid (38.2% vs. 9.5%, or 25.3% vs. 10.1%). A similar effect was also evident in the Feb. 9. Using tetraploids as f emale parents when crossed with diploids is likely to be more productive than the reciprocal. had the lowest pollination success rates (0.3% and 0.1% for the Feb .May pollination period and 1.2% and 4.1% for the June July pollination period), regardless of the direction of crossing (Table 7 1), in contrast to the results when LAOP 9 was the diploid parent. As described above, when LAOP 1% for the Feb. May pollination period and 10.1% to 25.3% for the June July pollination period (Table 7 2). As described above, LAOP

PAGE 168

168 cause of the lower level of cross s to be elucidated. Seasonal effects on pollination success rates The monthly pollination success rate of each of the eight interploid crosses was similar among February, March, and April (thereafter referred to as the first pollination period) and similar between June and July (referred to as the second pollination period) (data not shown). Thus the pollination data were combined and averaged within each period. The average pollination success rates were very different between the two periods. The diffe rence in pollination success rates ranged from 2.3 fold fold 9). Similar differences were observed in two 2 x x 2 x crosses (3.0% to 7.2% in the first period vs. 41.0% to 28.3% in the second p eriod) and two 4 x x 4 x crosses (1.1% to 1.7% in the first period vs. 15.8% to 14.7% in the second period). These differences indicated a strong seasonal effect on pollination success, as the fertilization and irrigation regime was held consistent in the g reenhouse from February to July. Several environmental factors were different between the two pollination periods, including temperature, light level, photoperiod, humidity, etc. As shown below, experimental results indicate d that temperature was a facto r causing this seasonal effect on pollination success. Effects of growing temperatures Significant differences (F = 31.08, P=<0.0001) were observed in pollination success between different pollination periods. It was hypothesized that one of the main environmental factors for such differences was growing temperatures. To test this parental plants were subjected to three treatments: growing at 21.1 C, 26.7 C, or

PAGE 169

169 32.2 C. The average pollination success rate was the highest (25.0%) when parental plants were grown at 26.7 C (Figure 7 2). The average success rate was 9.1% for plants subjected 21.1 C. T hese success rates were similar to the results from the first and secon d pollination periods (Table 7 1). For the 32.2 C treatment, more than 230 flowers were pollinated, but none of them set any fruit. Additionally, the parental plants grown at 32.2 C seemed to produce ~60% fewer flowers compared to those grown at 21.1 C o r 26.7 C (~230 vs. ~540). Selection of n ew t riploids Over the course of hand pollinations in 2006 and 2007, 393 triploids were generated. They were first evaluated in the greenhouse and then in the field for growth habit and leaf and flower characteristic s. Fourteen of them received high scores and were selected for propagation tests. Four of them did not root well or grow well in containers. The remaining 10 triploids rooted well and grew well in containers, and they were selected for further evaluatio n of male sterility, female sterility, and plant perf ormance. Male Sterility of Promising Triploid Selections Pollen stain ing Three experiments were conducted to assess the pollen stainability of the 10 promising triploids. In each experiment, more than 2000, up to 5141 (Table 7 4), pollen grains were examined per triploid, except for T8, which produced less pollen. Similar pol len stainabilitiy was observed for each triploid among the experiments. Thus, the average pollen stainability will be used for discussion. The commercial triploid cultivar included in the experiments had a pollen stainability of 0.3% The male

PAGE 170

170 4). Two triploids had pollen stainability slightly above 10% (12.2% for T6 and 10.4% for T7). The pollen stainability of the remaining eight tri ploids was below 10.0%, ranging from 2.8% (T8) to 9.7% (T2). Hand p ollination Two hand pollination experiments were performed in fall 2009 and spring 2010 to cause fruit set on L. depressa to confirm that low pollen stai ning lines will not cause seed set at seen in Chapter 7 In each experiment, fresh pollen was collected from unopen flowers and immediately applied onto emasculated L. depressa not cause any fruit set in both experiments (Table 7 5 caused 3.2% fruit set in the first caused fruit set in both experiments: 1.6% in the first one and 16.1% in the second one, resulting in an average of 8.9% fruit set. These results were similar to the results described in Chapter 6. Seven triploids, T1, T2, T4, T5, T6, T8, and T9, did not cause fruit set in both caused 2.8% fruit set in the first experiment, but none in the second one. Similarly, hand fruit set on L. depressa Female s terility of p romising t riploid s elections month period of evaluation, producing 3 to as many as 18 fruit per peduncle (Table 7 5). Interestingly it set more fruit when grown in Quincy and Citra without L. depressa planted in the plots (pure planting) than when grown in Balm and Ft. Pierce with L. depressa planted in the plots (mixed planting) (14.118 vs. 6.783 on average). The female sterile control uit per

PAGE 171

171 peduncle in the six evaluations from 30 July to 14 December 2009 in Quincy and Citra when L. depressa planted with L. depressa and grown in Balm and Ft. Pierce, this triploid cultivar set, on ave rage across two sites and over six evaluations, 10.5 fold more fruit (0.540 per peduncle). The average fruit set of the new triploids across four sites over 6 months was below 0.100 fruit per peduncle, except for T1 grown in Quincy and Citra. Thus these triploids were highly fertility of these triploids was reduced by ~137 fold FPP), the new triploids maintained their low level of fruit set (40% of high level of female sterility even when L. depressa was inter planted and viable pollen was available from L. depressa plants. Plant p erformance of p romising t riploid s elections Controls. Within 8 weeks post planting (WPP) quality rating of 3.0 or above at all four sites. Its plant quality ratings remained at or plants grown in Quincy, Citra, and Balm received a quality rating between 3.0 and 5.0 and had an average PQ r ating per site over the six evaluations of 4.1 (Citra), 4.3 (Balm), and 4.5 (Quincy) (Table 7 7). When and received an average PQ rating of 3.4. With all four sites and all six evaluations entries in the experiments. at 8 WPP of 3.3 and 4.8 and maintained its rating between 4.0 and 5.0 (Table 7 FI rating of 2.7 at 8 WPP and 3.2 to 5.0 thereafter until

PAGE 172

17 2 the end of the evaluation. The average FI rating per site over the 6 month period was 4.7 for Quincy, 4.8 for Citra, FI ratings (2.0 to 3.5 vs. 3.2 to 5.0) in five out of the six evaluations and a lower average sites over six evaluations was 4.2, the highest among all entries. Plants gr ew better and open ed more flowers in Quincy, Citra, and Balm than in Ft. t quality rating in Quincy, Citra, and Balm was 3.3, average FI rating in Quincy, Citra, and Balm was 3.7, 3.3, and 3.4, respectively, but it PQ and FI ratings were 2.9 and 2.8, respectively (Tables 7 7 and 7 8). To determine if the time period and site influenced PQ evaluations were analyzed by evaluation week and site. PQ evaluations were found to be influenced by both evaluation week and site. Evaluations of plant quality revealed that weeks were found to be significantly different at P <0.0001 (F value 10.41, DF 5) and sites were significantly different at P <0.0001 (F value 80.14, DF 3). The evaluation periods when separated indicated WPP 12 through 24 to be the highest two statistical group ings WPP eight and 28 were the lowest evaluations of PQ When separating sites Quincy was rated the highest rated followed by Balm and Citra, and Ft. Pierce was found to be the lowest rated site statistical ly Analysis of FI data were done similar to PQ for evaluation weeks and sites. F lowering intensity was also found to be influenced by the evaluation WPP and site.

PAGE 173

173 Statistical groups did not follow PQ groups completely. Eva luations of FI revealed that WPP were significantly different at P <0.0001 (F value 21.17, DF 5) and sites were significantly different at P <0.0001 (F value 233.51, DF 3). The WPP evaluations when separated indicated WPP 12 through 28 to be the highest t wo statistical group ing s. Only the initial evaluation eight WPP was in the lowest statistical group. Site separation indicated that Citra and Quincy were in the highest evaluation group followed by Balm and lastly Ft. Pierce in the lowest statistically r ated site. T1 selection. When grown in Quincy, Citra, and Balm, plants of this triploid received a quality rating between 3.0 and 5.0 in 17 out of 18 evaluations and a FI rating between 3.5 and 5.0 in 15 out of 18 evaluations (Table 7 7 and 7 8). The average PQ in these locations was 4.4, 3.9, and 3.4, respectively, and the average FI rating in t hese locations was 4.5, 4.3, an PQ rating in Ft. Pi erce was 2.5 to 3.5, with an average of 2.8, and its FI rating was 1.2 to 2.5, with an average of 2.0. The overall average PQ rating and FI rating of T1 was 3.3 and 3.5, respectively. T2 selection This line is a sibling of the T1 selection. They were similar in 16 out of 18 and 13 out of 18 evaluations, with ratings between 3.0 and 5.0 for PQ and FI (Table 7 7 and 7 8). The highest PQ rating s were from Balm ( 3.8 ) and Quincy ( 4.2 ) The other two site s had PQ ratings of 3.1 (Citra) and 3.6 (Quincy). The FI ratings were 3.4 and 4.0 for Balm and Citra, respectively. Evaluations from Ft. Pierce only exceeded 3.0 for week 24 (PQ) and week 28 (FI), with averages of 2.7 and 2.2 for PQ and FI respectively. T3 selection T3 was rated the highest in Quincy (4.1) for PQ and the same in Citra and Quincy for FI (4.2) (Table 7 7 and 7 8). From Balm and Citra 15 observations

PAGE 174

174 were above a 3.0 rating with averages of 3.2 (Balm) and 3.8 (Citra). PQ of T3 rated well with 16 observations over 3.0. Balm had an avera ge FI of 3.1. Ft. Pierce was again lower than the other sites with PQ and FI averages of 2.6 and 2.3. The collective averages of PQ and FI across all sites were 3.4 and 3.5 respectively. T4 selection This selection had the highest PQ and the lowest FI of all the triploid selections (Table 7 7 and 7 8). Looking more closely at this line 16 PQ and 13 FI observations were at or above 3.0 for the sites excluding Ft. Pierce. For this particular line the averages were a ffected largely by Ft. Pierce as it was the lowest rated FI of all triploids at that site. The other sites averaged 2.5 (Balm), 3.4 (Citra) and 3.8 (Quincy). The best attribute of this plant is its PQ as it was the highest rated of the triploid selections at 3.7. At Ft. Pierce this plant was tied for the highest rated PQ of all the triploids with an average of 2.8. The other sites average 3.9, 3.8, and 4.4 for Balm, Citra, and Quincy respectively. T5 selection This selection was rated a little higher on average and had 17 observations for PQ and FI were above the 3.0 rating (Table 7 7 and 7 8). The three site averages for PQ were 3.5, 3.6, and 4.2 for Balm, Citra and Quincy respectively. The FI rating for the previously mentioned sites was 3.3, 4.2, and 4.4 respectively. Ft. Pierce a veraged 2.7 (PQ) and 1.9 (FI). The overall averages were 3.5 for PQ and 3.4 for FI. T6 selection Plants of this triploid did not grow well at the four test sites, with a quality rating of below 3.0 in 15 out of 24 evaluations (Table 7 7). Its FI rat ing ranged from 1.7 to 5.0 when grown in Quincy, Citra, and Balm, and from 1.2 to 2.8 in Ft. Pierce (Table 7 8). Its overall average PQ and FI ratings were 2.4 and 3.3, respectively.

PAGE 175

175 T7 selection. Plants of this triploid seemed to be weak, not growing or flowering well at all test sites. Its overall PQ and FI ratings were 2.4 and 2.9, respectively, below the acceptable level for commercial production (Table 7 7 and 7 8). T8 selection This selection was similar to T6: Plants did not grow well, but they flowered acceptably. Its average PQ rating per site was below 3.0 for three out of the four sites (Table 7 7 and 7 8). Its average FI per site was 4.1 for Quincy 4.3 for Citra, 3.6 for Balm, and 1.9 for Ft. Pierce. T9 selection Plants of this selection received a quality rating of 3.0 or above (3.0 to 4.7) and a FI rating of 3.0 or above (3.3 to 5.0) in 15 out of the 18 evaluations in Quincy, Cita, and Balm (Table 7 7 and 7 8). The average PQ and FI ratings per site over 6 months were 4.1 and 4.1 for Quincy, 3.8 and 4.3 for Citra, and 3.0 and 3.1 for Balm. Its PQ rating in Ft. Pierce was between 2.0 and 2.8, averaged to 2.6. Its FI was 1.3 to 2.7, averaged to 1.9. T10 selection Plants of this triploid did not grow well at all four sites, and its PQ rating was below 3.0 in 13 out of 24 evaluations (Table 7 7). Its FI rating was 3.7 or 3.8 in Quincy and Citra, but was 2.7 and 1.8 in Balm and Ft. Pierce (T able 7 8). Thus the ornamental value of this selection was lower than other triploids. Discussion Breeding L. camara For S terile C ultivar R elease Optimizing environment Developing sterile L. camara cultivars is a multi step, lengthy process. Increas ing the efficiency in each step is very important. Each step should be optimized as information becomes available. This study has revealed that environmental conditions, especially growing season and temperature, had significant effects on pollination

PAGE 176

176 su ccess. The genetic by environemental interactions are extensive and important when breeding. These discoveries could lead to much improved rates of seed set. In the case of difficult crosses, it may be possible to manipulate the growth conditions to enh ance seed production. Optimizing crosses As discussed in Chapters 4 and 5, the selection of tetraploids that do not produce unreduced female gametes (UFGs) and apomictic seeds is critical for producing highly sterile triploids. However, this critical piece of information was not available when the first 30 crosses were undertaken to produce triploids. Serendipitously three non UFG producing cultivars were included in those crosses. As the information became available pollination efforts were focused on these non UFG producing tetraploids. Results from this study showed that triploid selections resulting from non UFG producing tetraploids were both highly male and female sterile. Currently the number of non UFG producing tetraploids is limited. Mo re non UFG producing tetraploids need to be identified or developed. The results from this study (Tables 7 3) also indicate that the direction of crossing played a critical role in successful triploid production. Diploid plant s were much more successful at causing seed production on tetraploid plants than the reverse. This may be due to the violation of the endosperm balance number (EBN) theory. This theory states that when the maternal contribution to the endosperm to paternal contribution exceeds a 2 :1, embryo abortion may occur ( Burton and Husband, 2000; Carputo et al ., 2003 ; Poehlman and Sleper 1995 ). In the case of interploid crosses, some tolerance may exist if the higher ploidy level plant is used as the maternal source so as not to exceed this ratio. The other explanation for the lack of seed set is that some parents

PAGE 177

177 could have very low pollen fertility levels. As seen in table 7 3 the diploid plants have on average nearly double the pollen staining rates of the tetraploids. Given that the diploids have higher pollen staining and crosses of the 4 x x 2 x nature are expected to be more effective these factors are likely not mutually exclusive. Another factor that needs to be considered in lantana interploid crosses is cultivar x x 2 x ) had the lowest fruit setting cross. This cross normally would be expected to have high seed set rates based on the high pollen stainability of 9, a direct descendant indicate d strong incompatibilit y between the two cultivars. Selecting t riploids as c andidates for r ele asing When male and female sterility, leaf and flower characteristics, and plant performance were considered, four triploids, T2, T3, T4, and T9, seemed to have the best potential to be released as new cultivars. Their pollen stainability was below 10% a safe level from controlled pollinations in Chapter 6. These triploids all had extremely low fruit production (below one seed for rile produced much more fruit (over 1 fruit per flower cluster), while the four new tri ploids remained highly sterile.

PAGE 178

178 Summary A productive breeding program working with L. camara will include a number of factors that have been explained from multiple experiments over several years of study. For the purpose of releasing sterile cultivars of L. camara it was necessary to determine the maximum levels of male sterility possible to achieve (Chapter 2) Similarly information about the the level of seed production and seed viability (Chapter 3) present in commercial cultivars was needed This pr ovided a basis for comparison for these new selections. After the discovery of UFG production and other reproductive modes (Chapters 4 and 5) it became necessary to ensure the parents and progeny did not contain any of the genes related to those traits. For a plant that is valued primarily for its ornamental attributes the last stage of selection pressure was exerted on PQ and FI. Collection of clear fertility data and using it in conjunction with performance analyses greatly assisted plant selection. It was clear that the site a plant is grown can drastically affect its performance characteristics. However, these plants were generally consistent across sites. This made the selection process straightforward in most cases. In one particular incident wit h T1 and T2 sibling plants, T2 was selected over T1 even though the plant performance was just a little lower. The reason for this was that T1 had slightly higher seed production rates than T2. The statewide evaluation of these triploids successfully dem onstrated the selection of ornamentally desirable, high ly male and female sterile lines across several months indicating a high level of landscape performace and durable supression of sexual reproduct ion.

PAGE 179

179 Table 7 1. Parental lines used in interploi d cross pollination s Line Ploidy Form Foliage density Opening flower color Finishing flower color 4 x Mounded High Yellow Magenta 4 x Er ratic Low Yellow Red 2 x Mounded High Cream White with yellow center 4 x Mounded High Yellow Magenta LAOP 9 2 x Mounded High Dark Yellow Dark Yellow 2 x Er ratic Medium Yellow Yellow Table 7 2. Pollination success rates for eight interploid crosses between two tetraploids and two diploids for triploid generation. Two tetraploid by tetraploid and two diploid by diploid crosses were included as controls. Seed parent Pollen parent Nature of Cro ss Flowers pollinated Pollination success rate (%) Feb. May June July Feb. May June July Carlos Lola 4 x x 2 x 715 489 9.1 20.5 Lola Carlos 2 x x 4 x 875 541 1.4 4.3 Carlos LAOP 9 4 x x 2 x 783 476 6.6 38.2 LAOP 9 Carlos 2 x x 4 x 936 579 1.0 9.5 Dallas Red Lola 4 x x 2 x 1298 541 0.3 1.2 Lola Dallas Red 2 x x 4 x 975 460 0.1 4.1 Dallas Red LAOP 9 4 x x 2 x 1149 504 1.8 25.3 LAOP 9 Dallas Red 2 x x 4 x 702 624 2.1 10.1 Carlos Dallas Red 4 x x 4 x 837 398 1.1 15.8 Dallas Red Carlos 4 x x 4 x 931 491 1.7 14.7 Lola LAOP 9 2 x x 2 x 1433 461 7.2 41.0 LAOP 9 Lola 2 x x 2 x 860 668 3.0 28.3

PAGE 180

180 Table 7 3. Results of full diallel crosses of four lines of Lantana camara indicating the rates of seed production compared to the rates of pollen stainability. Parent Carlos (49.4 Z ) Dallas Red (31.7) Lola (81.1) LAOP 9 (79) Carlos (4x, 1.9 Y ) 15.8 20.5 38.2 Dallas Red (4x, 0.6) 14.7 1.2 25.3 Lola (2x, 0.9) 4.3 4.1 41.0 LAOP 9 (2x, 0.4) 9.5 10.1 28.3 Z Pollen stainability (%) with cotton blue stain along the top of the table from chapter 2 Y Ploidy level followed by the average seed production per seed cluster in open pollination from chapter 3

PAGE 181

181 Table 7 4. Pollen stainability of 10 new triploid lines, three commercial cultivars and L. depressa Gulf Coast (REC) 1 Gulf Coast (REC) 2 Ft. Pierce (REC) 1 P ollen counted total Z Grand a vg X Line P ollen counted Avg. Z P ollen counted Avg. Z Pollen counted Avg. Z T1 624 1 4171 9.6 3598 7.1 2770 6.6 10539 7.8 def T2 624 4 3417 12.3 2919 8.7 2412 8.3 8748 9.7 cde T3 705b 3 5141 6.5 3752 6.4 4025 2.4 12918 5.1 gh T4 702a 3 3992 3.1 2983 4.6 3808 1.9 10783 3.2 ih T5 605 16 3500 9.7 2099 10.3 4202 6.1 9801 8.7 def T6 625 2 2429 17.8 2517 11.8 3693 6.9 8639 12.2 c T7 603 8 2826 13.7 2417 9.3 2828 8.1 8071 10.4 cd T8 605 7 471 3.3 1813 2.5 933 2.6 3217 2.8 i T9 613 3 3679 6.8 2237 8.5 4846 2.8 10762 6.1 gf T10 603 17 3381 11.0 2936 8.7 3835 3.7 10152 7.8 efg Athens Rose 6042 20.6* New Gold 2245 0.9 1816 0.3 2550 0.1 6611 0.4 j Pink Caprice 2211 62.0 2030 65.1 1752 69.9 5993 65.6 a L. depressa Y 722 42.3 780 45.7 1006 33.4 2508 40.5 b Z Indicates average pollen staining (%) with cotton blue from plants in field conditions at the University of Florida Gulf Coast Research and Education Center, Balm, Florida (GC) across two time periods and plants at the Indian River Research and Education Center, Ft Pierce, Florida (FP). Y Three L. depressa plants were sampled from each field. X Pollen staining from two replicated garden trials in 2009.

PAGE 182

182 Table 7 5. Fruit set of Lantana depressa when nine Lantana camara triploids and three commercial cultivars were used as a pollen source. L. camara line used L. depressa flowers pollinated Fruit set (%) Average fruit set (%) Z Fall 2010 Spring 2011 Fall 2010 Spring 2011 T1 624 1 107 116 0.0 0.0 0.0 a T2 624 4 114 119 0.0 0.0 0.0 a T3 705b 3 64 114 2.8 0.0 1.4 ab T4 702a 3 133 107 0.0 0.0 0.0 a T5 605 16 209 85 0.0 0.0 0.0 a T6 625 2 79 114 0.0 0.0 0.0 a T7 603 8 317 1.4 1.4 ab T8 605 7 112 0.0 0.0 a T9 613 3 114 97 0.0 0.0 0.0 a Athens Rose 117 80 3.2 0.0 1.6 ab New Gold 18 100 0.0 0.0 0.0 a Pink Caprice 305 93 1.6 16.1 8.9 b Z

PAGE 183

183 Table 7 6. Fruit production per peduncle of 10 Lantana camara triploids and two commercial cultivars with Lantana depressa planted at Quincy, Citra, Balm, and Ft. Pierce in Florida. Lines were found to be significantly different at P <0.0001 (F value 19.41, DF 10). Line Site Type of p lanting W PP 8 W PP 12 W PP 16 W PP 20 W PP 24 W PP 28 Average Line a verage X T1 Quincy Pure Z 0.008 0.000 0.092 0.217 0.225 0.592 0.189 0.117b Citra Pure 0.017 0.033 0.025 0.000 0.700 0.208 0.164 Balm Mixed Y 0.025 0.059 0.149 0.017 0.017 0.016 0.047 Ft. Pierce Mixed 0.067 0.133 0.017 0.058 0.100 0.033 0.068 T2 Quincy Pure 0.025 0.000 0.017 0.017 0.008 0.000 0.011 0.062ab Citra Pure 0.008 0.042 0.025 0.000 0.167 0.033 0.046 Balm Mixed 0.000 0.127 0.308 0.103 0.075 0.025 0.106 Ft. Pierce Mixed 0.117 0.167 0.075 0.017 0.075 0.025 0.079 T3 Quincy Pure 0.008 0.000 0.008 0.000 0.000 0.033 0.008 0.019a Citra Pure 0.025 0.008 0.000 0.000 0.000 0.017 0.008 Balm Mixed 0.000 0.017 0.074 0.033 0.017 0.016 0.026 Ft. Pierce Mixed 0.025 0.033 0.025 0.025 0.042 0.033 0.031 T4 Quincy Pure 0.000 0.000 0.000 0.125 0.000 0.358 0.081 0.023a Citra Pure 0.000 0.017 0.000 0.000 0.000 0.000 0.003 Balm Mixed 0.000 0.017 0.016 0.008 0.000 0.000 0.007 Ft. Pierce Mixed 0.008 0.008 0.000 0.000 0.000 0.000 0.003 T5 Quincy Pure 0.025 0.000 0.000 0.008 0.000 0.008 0.007 0.023a Citra Pure 0.000 0.008 0.000 0.000 0.083 0.033 0.021 Balm Mixed 0.000 0.017 0.058 0.000 0.016 0.016 0.018 Ft. Pierce Mixed 0.017 0.108 0.058 0.017 0.042 0.008 0.042 T6 Quincy Pure 0.008 0.000 0.008 0.008 0.000 0.008 0.006 0.007a Citra Pure 0.017 0.008 0.000 0.000 0.017 0.000 0.007 Balm Mixed 0.000 0.017 0.008 0.000 0.000 0.009 0.006 Ft. Pierce Mixed 0.008 0.000 0.000 0.008 0.008 0.017 0.007

PAGE 184

184 Table 7 6 Continued. Line Site Type of Planting WPP 8 WPP 12 WPP 16 WPP 20 WPP 24 WPP 28 Average Line Average X T7 Quincy Pure Z 0.025 0.000 0.108 0.167 0.133 0.483 0.153 0.062ab Citra Pure 0.017 0.008 0.000 0.000 0.017 0.000 0.007 Balm Mixed Y 0.008 0.009 0.008 0.025 0.016 0.049 0.019 Ft. Pierce Mixed 0.050 0.075 0.058 0.017 0.033 0.017 0.042 T8 Quincy Pure 0.000 0.000 0.033 0.100 0.025 0.008 0.028 0.036ab Citra Pure 0.000 0.017 0.008 0.000 0.083 0.033 0.024 Balm Mixed 0.000 0.025 0.093 0.066 0.042 0.007 0.039 Ft. Pierce Mixed 0.033 0.067 0.067 0.017 0.050 0.083 0.053 T9 Quincy Pure 0.008 0.000 0.017 0.100 0.008 0.150 0.047 0.025a Citra Pure 0.017 0.000 0.025 0.000 0.000 0.117 0.026 Balm Mixed 0.000 0.020 0.058 0.000 0.000 0.026 0.017 Ft. Pierce Mixed 0.008 0.033 0.000 0.000 0.008 0.008 0.010 T10 Quincy Pure 0.008 0.000 0.075 0.008 0.058 0.150 0.050 0.021a Citra Pure 0.008 0.008 0.008 0.000 0.017 0.000 0.007 Balm Mixed 0.017 0.000 0.067 0.008 0.000 0.000 0.015 Ft. Pierce Mixed 0.000 0.025 0.000 0.008 0.033 0.008 0.013 Quincy Pure 0.017 0.017 0.125 0.017 0.008 0.075 0.043 0.294c Citra Pure 0.175 0.025 0.017 0.000 0.033 0.058 0.051 Balm Mixed 0.430 0.581 0.833 0.746 0.673 0.227 0.582 Ft. Pierce Mixed 0.208 0.250 1.142 0.408 0.675 0.308 0.499 Quincy Pure 7.150 22.838 20.825 17.000 11.138 11.275 15.038 10.451 Citra Pure 15.808 10.867 16.092 9.175 12.783 14.467 13.199 Balm Mixed 1.143 10.683 12.416 4.226 8.883 7.532 7.481 Ft. Pierce Mixed 5.067 6.608 9.525 8.000 4.583 2.733 6.086 L. depressa Quincy N/A N/A N/A N/A N/A N/A N/A N/A 1.918 Citra N/A N/A N/A N/A N/A N/A N/A N/A Balm Mixed 3.805 2.349 1.159 1.810 2.281 Ft. Pierce Mixed 1.367 3.517 2.092 1.105 0.800 0.444 1.554 Z Only Lantana camara lines were planted at this location. Y Lantana camara and Lantana depressa lines were planted at this location. X Superscript letters indicate mean se 0.05).

PAGE 185

185 Table 7 7. Plant quality ratings of new triploid line s and commercial cultivars with Lantana depressa planted at four sites: Quincy, Citra, Balm, and Ft. Pierce in Florida. Plants were planted on 5 May 2009. Lines were found to be significantly different at P <0.0001 (F value 29.64, DF 11). Plants were eva luated in a scale of 1 5. Increasing evaluations indicated more compact and healthy plants. Line Site WPP 8 WPP 12 WPP 16 WPP 20 WPP 24 WPP 28 Site a vg. Grand a vg. Z T1 Balm 2.5 3.2 4.2 4.2 3.2 3.2 3.4 3.6ab Citra 3.2 4.5 4.8 4.0 4.0 3.0 3.9 Ft. Pierce 2.2 2.5 2.5 2.8 3.5 3.3 2.8 Quincy 3.3 4.8 5.0 4.2 4.3 4.8 4.4 T2 Balm 3.5 3.5 3.8 4.0 3.8 3.8 3.8 3.3bcd Citra 2.2 3.3 4.2 3.3 3.0 2.7 3.1 Ft. Pierce 2.2 2.7 2.5 2.8 3.2 2.8 2.7 Quincy 3.3 4.0 4.0 3.0 3.8 3.7 3.6 T 3 Balm 2.7 2.8 3.5 3.7 3.3 3.3 3.2 3.4bc Citra 3.3 4.0 4.5 4.2 3.5 3.5 3.8 Ft. Pierce 2.0 2.3 2.7 2.8 3.2 2.7 2.6 Quincy 2.7 4.8 5.0 3.8 4.2 4.2 4.1 T4 Balm 3.2 4.0 4.2 4.3 4.0 3.7 3.9 3.7ab Citra 4.7 5.0 3.5 4.0 3.5 2.3 3.8 Ft. Pierce 2.7 3.2 2.7 2.8 2.7 2.5 2.8 Quincy 2.2 5.0 5.0 4.5 4.8 5.0 4.4 T5 Balm 3.8 3.7 4.0 3.5 3.2 3.0 3.5 3.5bc Citra 3.7 4.0 4.3 3.8 3.2 2.7 3.6 Ft. Pierce 2.8 3.0 2.8 2.8 2.5 2.0 2.7 Quincy 3.5 4.8 4.3 4.3 3.8 4.5 4.2 T6 Balm 3.7 3.3 4.0 3.5 2.7 2.8 3.3 3.1cd Citra 4.3 3.7 3.5 3.2 2.7 2.2 3.3 Ft. Pierce 2.8 2.8 2.0 2.2 2.3 2.2 2.4 Quincy 3.7 3.5 3.5 3.0 2.7 3.2 3.3

PAGE 186

186 Table 7 7 Continued. Line Site WPP 8 WPP 12 WPP 16 WPP 20 WPP 24 WPP 28 Site a vg. Grand a vg. Z T 7 Balm 2.2 2.8 3.3 3.2 3.0 2.8 2.9 2.4f Citra 1.0 2.5 2.7 2.3 2.7 2.3 2.3 Ft. Pierce 1.7 1.8 1.7 2.0 2.2 1.8 1.9 Quincy 3.5 2.7 2.5 2.3 1.8 2.8 2.6 T8 Balm 2.5 2.5 3.0 3.0 2.8 1.8 2.6 2.6ef Citra 2.5 3.0 4.0 3.2 2.8 1.8 2.9 Ft. Pierce 1.3 1.8 2.0 1.5 1.3 1.2 1.5 Quincy 2.5 3.2 3.5 3.0 3.3 3.7 3.2 T9 Balm 2.5 2.7 3.7 3.3 3.0 3.0 3.0 3.4bc Citra 4.2 4.3 4.0 4.2 3.5 2.7 3.8 Ft. Pierce 2.8 2.7 2.5 2.3 2.5 2.5 2.6 Quincy 3.0 4.7 4.5 3.8 4.3 4.2 4.1 T10 Balm 2.0 2.2 2.2 3.0 2.2 2.0 2.3 2.4f Citra 1.0 3.0 3.0 3.0 2.3 2.0 2.4 Ft. Pierce 1.2 2.0 2.2 2.2 2.2 2.3 2.0 Quincy 3.5 3.5 3.0 3.0 2.3 2.7 3.0 Balm 4.0 4.7 4.7 5.0 3.8 3.5 4.3 4.1a Citra 4.5 5.0 4.0 4.0 4.0 3.0 4.1 Ft. Pierce 3.5 3.2 3.3 3.0 3.8 3.8 3.4 Quincy 3.2 5.0 5.0 5.0 4.5 4.3 4.5 Balm 3.3 3.7 4.2 3.7 3.0 2.7 3.4 2.9de Citra 4.2 3.0 3.2 2.0 2.0 2.0 2.7 Ft. Pierce 1.7 2.5 3.2 2.7 2.2 1.5 2.3 Quincy 2.5 3.5 3.0 3.0 3.3 4.8 3.3 Z

PAGE 187

187 Table 7 8 Flower intensity ratings of new triploid lined and commercial cultivars with Lantana depressa planted at four sites: Quincy, Citra, Balm, and Ft. Pierce in Florida. Plants were planted on 5 May 2009. Lines were found to be significantly different at P <0.0001 (F value 6.73, DF 11). Plants were evaluated in a scale of 1 5. Increasing evaluations indicated more open flowers Line Site WPP 8 WPP 12 WPP 16 WPP 20 WPP 24 WPP 28 Site a vg. Grand a vg. Z T1 Balm 2.0 1.8 4.0 4.0 3.5 4.2 3.3 3.5 ab Citra 3.5 4.5 4.8 5.0 4.0 4.0 4.3 Ft. Pierce 1.2 1.5 2.3 2.0 2.5 2.5 2.0 Quincy 2.8 5.0 5.0 4.3 4.8 5.0 4.5 T2 Balm 3.3 1.3 3.0 3.8 4.0 4.7 3.4 3.4 bc Citra 2.2 4.3 5.0 5.0 4.0 3.3 4.0 Ft. Pierce 1.2 2.0 2.2 1.8 2.5 3.3 2.2 Quincy 4.3 4.2 4.0 3.0 4.7 4.8 4.2 T3 Balm 1.8 2.0 3.8 3.7 3.3 4.2 3.1 3.5 b Citra 3.3 4.2 4.7 5.0 3.3 4.5 4.2 Ft. Pierce 1.3 2.2 3.2 1.7 2.2 3.2 2.3 Quincy 2.7 4.7 4.5 3.7 4.8 5.0 4.2 T4 Balm 1.3 1.0 3.3 3.0 3.2 3.0 2.5 2.7 c Citra 2.5 4.0 3.8 5.0 3.0 2.3 3.4 Ft. Pierce 1.0 1.3 1.7 1.2 1.2 1.0 1.2 Quincy 2.2 4.0 4.7 3.2 4.3 4.7 3.8 T5 Balm 3.3 2.2 3.0 3.7 3.3 4.3 3.3 3.4 b Citra 3.0 4.3 5.0 5.0 4.0 3.7 4.2 Ft. Pierce 1.7 1.8 3.0 1.2 1.8 1.8 1.9 Quincy 3.5 4.8 4.5 4.3 4.2 5.0 4.4 T6 Balm 2.7 1.7 3.5 4.0 3.8 4.2 3.3 3.3 bc Citra 3.7 4.0 4.0 5.0 3.8 3.5 4.0 Ft. Pierce 1.5 1.2 2.0 1.5 1.5 2.8 1.8 Quincy 3.8 5.0 4.2 3.7 3.8 4.5 4.2

PAGE 188

188 Table 7 8 Continued. Line Site WPP 8 WPP 12 WPP 16 WPP 20 WPP 24 WPP 28 Site a vg. Grand a vg. Z T 7 Balm 1.7 1.8 3.2 4.0 3.2 3.8 2.9 2.9 bc Citra 1.0 3.7 3.5 4.3 3.8 3.2 3.3 Ft. Pierce 1.7 1.8 1.8 1.5 1.8 1.7 1.7 Quincy 3.0 3.2 3.5 3.5 4.0 4.8 3.7 T8 Balm 2.5 2.3 4.3 4.2 4.0 4.0 3.6 3.5 ab Citra 2.8 4.3 5.0 5.0 4.7 4.0 4.3 Ft. Pierce 1.3 1.7 2.7 1.5 2.0 2.5 1.9 Quincy 2.3 4.2 4.2 4.2 4.7 5.0 4.1 T9 Balm 2.5 1.2 3.8 3.3 3.8 4.0 3.1 3.4 bc Citra 4.2 4.7 4.7 5.0 3.7 3.7 4.3 Ft. Pierce 1.3 2.0 2.2 1.2 2.7 2.2 1.9 Quincy 2.7 5.0 4.3 3.7 4.5 4.7 4.1 T10 Balm 1.3 1.8 2.7 4.0 3.2 3.2 2.7 3.0 bc Citra 1.2 4.0 4.7 5.0 3.3 3.8 3.7 Ft. Pierce 1.0 1.7 2.2 1.7 1.8 2.3 1.8 Quincy 3.2 4.3 4.2 3.0 3.7 4.5 3.8 Balm 2.7 3.2 4.7 5.0 5.0 5.0 4.3 4.2 a Citra 4.8 5.0 5.0 5.0 5.0 4.0 4.8 Ft. Pierce 3.5 3.2 3.5 2.0 2.7 2.5 2.9 Quincy 3.3 5.0 5.0 5.0 5.0 4.7 4.7 Balm 2.8 1.7 4.5 4.3 3.3 3.5 3.4 2.9 bc Citra 3.3 3.0 3.7 4.8 2.0 2.8 3.3 Ft. Pierce 1.0 1.8 2.0 1.0 1.0 1.0 1.3 Quincy 2.7 3.5 3.0 4.0 4.0 5.0 3.7 Z ( 0.05).

PAGE 189

189 Figure. 7 1 Effect of three growth chamber temperatures on the percent f ruit set of Lantana camara pollinated with Temperature effect w as significantly different at P=0.0232 (F value 8.76, DF 2). Mean sepa ration was ; letters of different groups represent different statistical groups.

PAGE 190

190 CHA PTER 8 CONCLUSIONS Rationale Lantana camara is a important ornamental and landscape plant in Florida. Yet, it is a Category I invasive species that can hybridize with the Florida native species Lantana depressa Genetic sterilization has potential as an economical, preventive measure to control the in vasiveness of L. camara This study sought to identify the primary biological factors that affect L. camara L. camara L. depressa and to develop new sterile L. camara cultivars Male a nd Female Fertility of L. camara Male fertility was assessed based on pollen stainability. Results revealed that pollen stainability varied from 1.8% to 81.1% among cultivars. Ploidy level was found to be the most important factor affecting male fertilit y. On average, diploids exhibited the highest male fertility, followed by tetraploids, pentaploids, hexaploids, and triploids. There was significant pollen stainability variation within certain ploidy levels and genetic background P edigree may also pl ay a significant role in determining L. camara well, ranging from 1.8% to 27.1%. Pollen stainability rates this high in triploids is unusual but not as high as some previousl y reported. Thus newly produced triploids need to be screened to identify triploids that can meet the required male sterility level. The most important factors determining female fertility in plants are seed (or fruit) production and seed germination. Lit tle information is available in the literature regarding fruit production capacity and seed germination of lantana cultivars. Results

PAGE 191

191 showed that L. camara cultivars differed considerably in fruit production, ranging from fruit per peduncle. Ploidy level, unreduced female gamete (UFG) production, and apomixis played significant roles in determining the fruit production capacity of L. camara Results indicate that triploids without the UFG production an d apomixis traits were most sterile. However other genetic mechanisms might cause female sterility in L. camara Principal component analysis based on female fertility index (product of fruit per peduncle and seed germination) and pollen stainability pr ovide d a useful way to visualize the reproductive characteristics of different L. camara cultivars/breeding lines. Multiple Modes of Reproduction in L. camara Ploidy analysis was performed on progeny from open and controlled pollinations and complimented with simple sequence repeat (SSR) based molecular marker analysis. These analyses showed that L. camara could form three types of female gametes [reduced female gametes (RFGs), UFGs, and double UFGs (DUFGs)] and two types of male gametes [reduced male gam etes (RMGs) and unreduced male gametes (UMGs)] and could develop seed through fertilization or apomixis, leading to six primary modes of reproduction in L. camara These modes include: 1) RFGs and RMGs fertilized and developed into n + n progeny, 2) RFGs under going apomixis ( n + 0 progeny, or haploidization), 3) UFGs fertilized with RMGs producing 2 n + n progeny, 4) UFGs developed via apomoxis directly into embryos and 2 n + 0 progeny, 5) DUFGs fertilized with RMGs, resulting in 4 n + n pro geny, and 6) D UFGs developed into 4 n + 0 progeny through apomixis. The UFG production trait was found to be controlled by nuclear gene(s). The primary reproductive pathway during microsporogenesis was the formation of RMGs. UMGs were observed, but they occurred at very low frequencies.

PAGE 192

192 T he se results indicate a strong need to screen breeding parents carefully and to avoid using those materials that can form UFGs or apomictic seed in crosses intended for developing sterile triploids None of the diploids evaluated in this study showed tendency to produce UFGs or apomictic seeds but three out of the six tetraploids studied showed such tendency. There is a dire need t o identify or develop more tetraploids that do not carry the UFG production or apomixis tr ait for sterile triploid development L. camara Lantana depressa Pollen stainability of L. camara was the most important factor determining the potential of L. camara as a male parent to hybridize with L. depressa and cause f ruit production. A s L. camara pollen stainability was reduced its ability to cause fruit set on L. depressa dropped rapidly. When L. camara s pollen stainability dropped to below 10%, it could hardly cause fruit set on L. depressa Among the five ploidy levels, diploid L. camara cultivars seemed to be most compatible with tetraploid L. depressa followed by tetraploids. Triploid L. camara with low pollen stainability (<10%) ha d very little potential to cross pollinate L. depressa When L. camara was pollinated with L. depressa p loidy level and mode of reproducti on of L. camara were the primary factors determining fruit (or seed) production. D iploid L. camara produced the largest number of fruit among the five ploidy levels U FG producing triploid L. camara lines were highly crossable as a female with L. depressa and produced large numbers of fruit and seed, whereas non UFG producing triploid L. camara lines were the least crossable as a female with L. depressa

PAGE 193

193 Ploidy analysis results revealed that natural polyploidization had occurred in L. depressa Ploidy level distribution among the progeny of crosses between L. camara and L. depressa suggest that the L. depressa accession used in this study carried the UFG production and apomixis traits like some L. camara cultivars did. This is the first report of such traits in L. depressa Developing Sterile Triploid Selections Interploidy crosses were made among two diploid cultivars, one diploid breeding line, and th ree tetraploid cultivars that did not carry the UFG production and apomixis traits. These crosses resulted in 393 triploids, from which 10 triploids were selected. These selections showed high levels of male and female sterility when tested at four sites in southern, central and northern Florida. Four of the sterile triploid selections also performed and flowered well in these tests and have shown potential to be released as new cultivars. Future Opportunities for Lantana Breeding This study has demonstr ated that high levels of male and female sterility can be achieved through selecting appropriate tetraploids and generating and selecting triploids. In the process of investigating male and female sterility of L. camara one diploid cultivar and one tetra ploid breeding line were found to have unusually low levels However, it set very little seed despite thousands of flowers that were hand or open pollinated (fruit per peduncle of 0.003). CAOP 88, a tetraploid individual, showed good female fertility but exceptionally low pollen stainability (3.2%). Although the exact remain to be elucidated, it is expected that the sterility should be

PAGE 194

194 caused by genes rather t han meiotic failure due to imbalanced c hromoso mes since they are a diploid and tetraploid. These materials may offer additional genetic means to reduce the fert ility of L. camara and to develop new sterile cultivars in this species.

PAGE 195

195 APPENDIX A FULL DATASET OF POLL EN STAINING OF ALL L ANTAN A LINES STAINED Table A 1 Pollen stainab ility of all lines screened in S easons 1 and 2. Line Ploidy Season 1 Season 2 Season 1 Y Season 2 Y Cream 2 x 87.1 1.1% 70.3 0.3% d e a f Denholm White 2 x 68.3 1 .0 % 72.6 1% a f GDGHOP 10 2 x N/A 0 .0 0% 8 GDGHOP 36 2 x N/A 1.3 0.7% n o 5 8 GDOP 31 2 x N/A 0 .0 0% 8 Landmark Flame Z 2 x N/A 47.5 1.8% d n Landmark White 2 x N/A 45 4.6% e o LAOP 30 2 x 54.4 12.3% 70.2 14.5% e f a f LAOP 9 2 x 88.7 1.1% 69.3 17.3% a g Lola 2 x 85.6 1.8% 76.7 3.3% a e Lucky Pot of Gold 2 x N/A 87.6 2% a c Myst 107 2 x N/A 1.4 0.8% 5 8 PKGHOP 1 2 x N/A 2.4 1.5% l n 5 8 Samantha 2 x N/A 5.2 2.9% i h 1 8 624 1 3 x N/A 17.2 2.2% i q 1 712b 7 3 x N/A 10.1 1.8% k l t 5 713 1 3 x N/A 4.7 0.5% y 8 Athens Rose 3 x 20.8 3.1% 20.3 3.5% i h o y GDOP 4 3 x N/A 5.1 0.9% o y 8 Landmark Peach Sunrise Z 3 x 21.8 2.8% 32.3 0.9% h r Landmark Pink Dawn 3 x 8.9 3.2% 4 .0 0.6% z 8 Lemon Drop 3 x 5.7 0.9% 1.7 0.6% c d 5 8 X 3 x 19.4 0.9% 9.3 3% a v 5 Luscious Lemonade 3 x N/A 4.2 3% a b 3 8 Miss Huff 3 x 2 .0 0.2% 1.9 0.8% 4 8 New Gold 3 x 0.8 0.1% 2.7 2% 5 8 New Red Lantana 3 x 5.6 0.8% 7 .0 1.7% w 7

PAGE 196

196 Table A 1 Continued. Line Ploidy Season 1 Season 2 Season 1 Y Season 2 Y Patriot Fire Wagon 3 x 19.3 1 .0 % 14.7 2.5% r 3 PCOP 12 3 x N/A 10.7 0% s 5 PCOP 2 3 x N/A 10.7 1.8% s 5 Red Butler 3 x 7.5 1.8% 4 1.7% z 8 Red Spread Lantana 3 x 6.2 0.7% 5.7 0.3% x 7 Samson Lantana 3 x 6.4 0.4% 5.2 0.7% x 8 Silver Mound 3 x N/A 0.5 0.5% n o 6 8 Sunset Lantana 3 x 5.2 0.7% 3.1 1.2% 3 8 604 1 4 x N/A 49.7 4.4% c m 604 5 4 x N/A 9.7 1.3% u 5 Anne Marie 4 x N/A 43.2 2.4% d e e o Bandana Cherry Sunrise 4 x N/A 35.1 6% a g q Bandana Red 4 x N/A 24.3 2.2% l m l w CAOP 88 4 x N/A 0.4 0.4% a 6 8 Carlos 4 x 54.5 1.6% 44.2 2.4% f h e o Confetti 4 x N/A 47.6 3.1% d m Dallas Red 4 x 34.5 2.3% 29 1% j u DROP 25 4 x N/A 30.2 2% i j h t Gold 4 x 31 .0 9.6% 21.4 8.2% o y Irene 4 x 55.1 1.2% 59.7 3.2% i j a i Landmark Gold 4 x N/A 53.5 4.7% b k Lucky Peach 4 x N/A 40 6.3% f q Patriot Cherry 4 x N/A 16.3 2.4% l q 2 Patriot Desert Sunset 4 x N/A 60.2 1.8% l m a h Patriot Parasol 4 x N/A 43.3 4% e o Patriot Passion 4 x N/A 43.3 2.2% e o Patriot Rainbow 4 x N/A 31.2 3.2% l h s PCOP 6 4 x N/A 35.8 2.1% j k g q Pink Caprice 4 x 75.9 2.3% 71.1 1.9% g i a f Radiation 4 x 40.7 11.2% 23.9 2.7% c d l w

PAGE 197

197 Table A 1 Continued. Line Ploidy Season 1 Season 2 Season 1 Y Season 2 Y 629 1 5 x 33 .0 3.1% 29.5 3.8% i u 629 2 5 x 52.9 3% 43.7 5.6% k l e o Cajun Pink 5 x 41.2 2.5% 23.4 2% m w Patriot Hallelujah 5 x 41.3 4.3% 42.5 3.1% e p Sonshine Lantana 5 x 13.6 2% 21.7 5% n x Spreading Sunset 5 x N/A 25.4 5.5% m o k v 620 1 6 x N/A 34.3 1.8% r 4 620 10* 6 x 9.9 0.9% 14 4.1% g q 621 4 6 x 24.5 2.4% 17.9 2.7% p z Tangerine 6 x 24.5 5.2% 17.3 3% q 1 Pale Blue 3x N/A 0.9 0.9% f 6 8 L. montevidensis ( Lavender ) 3x 1.2 0% 0.2 0.2% f g 7 8 L. montevidensis (W hite ) 3x 2.4 0.3% 0.6 0.6% b c 6 8 L. canescens N/A N/A 87.9 0.4% l m a b L. involucrata N/A N/A 52 3% b l L. depressa var. depressa N/A N/A 58.6 2.2% a j L. depressa var. floridana N/A N/A 84.8 2.3% i j a d L. depressa N/A N/A 95.1 3.1% e f a L. depressa (ft meyers) 3 x N/A 7.3 2.2% w 6 L. depressa (zugar) 3 x N/A 3.3 1.4% 2 8 Z Removed improved from the cultivar name. Y Letters A Z followed by 1 8 indicates statistical groupings by Tukey s W procedure. *Indicates significant difference between season 1 and 2.

PAGE 198

198 APPENDIX B BRANCH CUTTINGS FROM BALM, FLORIDA TRIAL Table B 1. Quantitative data for potential analysis of ornamental characteristics of L. camara for evaluation and sel e ction. Data was taken from 30 cm branch cuttings on 17 November 2009. Sample Flower clusters per node St.er. Nodes per dry weight index St.er. Flowers per dry weight index St.er. L ateral branches per dry weight St.er. T1 1.46 0.12 3.93 0.30 5.53 0.23 0.91 0.07 T2 1.35 0.05 4.56 0.18 6.00 0.02 1.19 0.05 T3 1.48 0.27 4.04 0.64 5.59 0.32 1.04 0.12 T4 0.93 0.17 4.89 0.77 4.27 0.28 0.97 0.12 T5 1.38 0.10 4.35 0.28 5.84 0.27 0.98 0.04 T6 1.32 0.14 5.87 0.31 7.34 0.60 1.12 0.15 T7 1.41 0.24 5.43 0.70 7.15 0.72 1.21 0.16 T8 1.73 0.39 4.72 1.13 7.12 0.37 1.26 0.09 T9 1.49 0.31 3.98 0.96 5.27 0.20 0.89 0.07 T10 1.54 0.26 4.54 0.74 6.32 0.33 0.77 0.06 1.08 0.06 10.86 0.50 11.49 0.21 2.70 0.18 1.18 0.14 4.96 1.15 5.21 0.35 0.74 0.11 Lantana depressa var. depressa 0.83 0.05 15.04 0.50 12.32 0.88 2.05 0.32

PAGE 199

199 LIST OF REFERENCES Anderson, N.O. 2001. New o rnamental c rops: A prima ry source of invasive species? Chicago Botanic Garden, New Ornamental Crops Res Symp Program 13. (abstr.). Anderson, N.O. 2004a. Invasive horticultural crops: Part I. Why be concerned? MNLA News. 28(3): 36 40. Anderson, N.O. 2004b Invasive horticultural crops: Part II. Where do they come from? MNLA News. 28(4):28 33. Anderson, N.O. 2006. Issues, c hallenges and o pportunities for the 21 st Century p. 177 214 In: N.O. Anderson (ed.) Flower b reeding and g enetics. Springer Dordrecht The Netherlands. Anderson, N.O. and P.D. Ascher. 1994. Erosion of style/anther length integrity in introgressive Lythrum hybrids p. 269 272 In: A.G. Stephenson and T h Kao (eds.). Pollen pistil interactions and pollen tube growth. Am. Soc. Plant Physiologists Rockville, MD. Arnold, MA. 2002. Landscape plants for Texas and e nvirons. Stipes publishing Champaign, Illinois. Arora, R.K. and Kohli, R.K. 1993. Autotoxic effect of decomposed leaf and inflorescence of Lantana camara var camara on its se ed germination parameters. Indian J of Ecol 20(2):109 112. Barrows E.M. 1976. Nectar r obbing and p ollination of Lantana camara (Verbenaceae). Biotropica. 8(2 ):132 135. Beaulie u D. 2008. Lantana plants for Florida landscaping Anywhere. About, Inc. 15 Ap ril 2008. < http://landscaping.about.com/od/flowerseed/p/lantana_plant.htm >. Bicknell, R.A. and A.M. Koltunow. 2004. Understanding a pomixis: r ecent a dvances and r emaining c onundrums. The Plant Cell. 16:228 245. regeneration in four invasive Reynoutria taxa (Polygonaceae). Plant Ecol 166:1 11. Brando, A.D., L.F. Viccini, S.M. Recco Pimentel. 2009. C ytogenetic c haracterization of Aloysia virgata Ruiz and Pavan (Verbenaceae). Brazilian Arch of Biol and Technol 52(4):893 899. Bretagnolle, F. and J.D. Thompson. 1995. Gametes with the somatic chromosome number: mechanisms of their formation and role in the evolution of autopolyploid plants. New Phytologist. 129:1 22.

PAGE 200

200 Brewbaker, J.L. and B.H. Kwack. 1963. The essential rold of calcium ion in pollen germination and pollen tube growth. Amer. J. Bot. 50(9):859 865. Brewbaker, J.L. 1967. The distribution and phylogenetic significance of binucleate and trinucleate pollen grains in the angiosperms. J. Bot. 54(9):1069 1083. Bringhurst, R.S. and T. Gill. 1970. Origin of Fragaria p olyploids. II. Unreduced and double unreduced gametes. Amer. J. Bot. 57(8):969 976. Broughton, S. 2000. Review and e valuation of Lantana b iocontrol programs. Biological Control. 17:272 286. Brown, B.J. and R.J. Mitchell. 2001. Copetition for pollination: effects of pollen of an invasive plant on seed set of a native congener. Oecologia. 1 29:43 49. Burton, T.L. and B.C. Husband. 2000. Fitness differences among diploids, tetraploids, and their triploid progeny in Chamerion angustifolium : mechanisms of inviability and implications for polyploidy evolution. Evolution. 54(4):1182 1191. Carputo, D., L. Frusciante, and S.J. Peloquin. 2003. The role of 2 n g ametes and e ndosperm b alance n umber in the o rigin and e volution of p olyploids in the Tuber Bearing Solanums. Genet ics 163:287 294. Castillo, J.M., P. Leira Doce, J. Carrin Tacuri, E. Mu oz Guacho, A. Arroyo Sol s, G. Curado, D. Doblas, A.E. Rubio Casal, A.A. Alvarez Lpez, S. Redondo Gmez, R. Berjano, G. Guerrero, A. De Cires, E. Figueroa, A. Tye. 2007. Contrasting strategies to cope with drought by invasive and endemic species of Lantan a in Galapagos. Biodiversity Conservation. 16:2123 2136. Cilliers, C.J. and S. Neser. 1991. Biological control of Lantana camara (Verbenaceae) in South Afri ca. Agr., Ecosystems, and Environ. 37:57 75. Cronk, Q.C.B. and J.C. Fuller. 1995. Plant Invaders. Th e Threat to Natural Ecosystems. Chapman and Hall, London. Culley, T.M. and N.A. Hardiman. 2007. The beginning of a new invasive plant: a history of the ornamental call ery pear in the United States. BioScience 57(11):956 964. Czarnecki II, Dav id M., Z. Deng, and D.G. Clark. 2008. Assessment of ploidy levels, pollen viability, and seed production of Lantana camara cultivars and breeding lines. HortSci ence 43(4): 1195 1196 (Abstr.). Czarnecki, D.M. and Z. Deng. 2009. Occurrence of u nreduced f emale g ametes l eads to s exual p olyploidization in Lantana J. Amer. Soc. Hort. Sci. 134(5):560 566. Day M. C.J. Wiley J. Playford and M.P. Zalucki. 2003. Lantana: Current m anagement s tatus and f uture p rospects. ACIAR, Canberra, ACT, Australia.

PAGE 201

201 Dotson, S.B., M.B. Lanahan A.G. Smith, and G.M. Kishore. 1996. A p hosphonate monoester hydrolase from Burkholderia caryophilli PG2982 is useful as a cond itional lethal gene in plants. The Plant J 10(2):383 392. Dozier,H. 1999. Plant introductions to invasion: History, public awareness, and the case of Ardisia crenata Univ. of Florida, Gainesville, PhD. Diss Duggin, J.A. and C.B. Gentle. 1998. Experimental evidence on the importance of disturbance intensity for invasion of Lant ana camara L. in dry rainforest open forest ecotones in north eastern NSW, Australia Forest Ecol and M gt. 109:279 292. Duvick, D.N. 1959. The use of male steril ity in hybrid seed production. Economic Bot 13(3):167 195. Eastham, K. and J. Sweet. 2002. Ge netically modified organisms (GMOs): The significance of gene flow through pollen transfer. Environmental Issue Report 28:1 75. Eigsti, O.J. 1957. Induced polyploidy. Amer J of Bot 44:272 278. El konin, L.A. and M.I. Tsvetova. 2008. Genetic and c ytological a nalyses of the m ale s terility m utation i nduced in a Sorghum t issue c ulture with streptomycin. Plant Genet 44(5):663 673. Florida Exotic Pest Plant Council. 2007. 2007 List of invasive plant species. 16 Apr. 2008. < http://fleppc.org/07list.htm >. Florida Exotic Pest Plant Council FLEPPC. 2009. 2009 L ist of invasive plant species. Florida Exotic Pest Plant Council. 16 April 2011. < http://fleppc.org/09list.htm > Florida Exotic Pest Plant Council FLEPPC. 2011. Florida Ex otic Pest Plant Council. 16 April 2011. < http://fleppc.org > Floyd Jr., J.A. 1998. Southern Living Garden Book Oxmoor House Inc., Birmingham, AL Foxcroft, L.C., D.M. Richardson, and J.R.U. Wilson. 2008. Ornamental p lants as i nvasive a liens: Problems and s olutions in Kruger National Park, South Africa. Environmental M gt. 41:32 51 Freeman, J.E., R. Albritton, S. Jose, and J.R.R. Alava lapati. 2009. The e conom ics, l aw, and p olicy of i nvasive s pecies m anagement in the United States: Responding to a g rowing c risis CRC Press, Taylor and Francis group, Boca Raton, FL. Fulton, T.M., J. Chunwongse, and S.D. Tanksley. 1995. Microprep protocol for extraction of DNA fr om Tomato and other herbaceous plants. Plant Mol. Biol. Rep. 13(3):207 209.

PAGE 202

202 Gentle, C.B., and J.A. Duggin. 1997. Lantana camara L. invasions in dry rainforest open ecotones: The role of disturbances associated with fire and cattle grazing. Aust ral. J. of Ecol. 22: 298 306. Ghisalberti, E.L. 2000. Lantana camara L. (Verbenaceae). Fitoterapia. 71:467 486. Gle ditsch, J. M. and T.A. Carlo. Fruit quantity of invasive shrubs predicts the abundance of common native avian frugivores in central Pennsylvania. Diversity and Distrib 17:244 253. Goulson, D. and L.C. Derwent 2004. Synergistic interactions between an exotic honeybee and an exotic weed: pollination of Lantana camara in Australia. European W eed R es S oc W eed R es 44:195 202. Graff, J.L. 1986. Lantana camara the plant and some methods for its control. The South African For J 13: 26 30. Graff, J.L. 1987. The seedfly Ophiomyia lantanae and other factors responsible for reducing germination in Lantana camara forms found in Natal. South African Pl ant Mag 53(1): 104 106. Hammer R.L. 2004. The Lantana mess : A c ritical l ook at the genus in Florida. The Palmetto. 23(1):21 23. Hamrick, J.L. and M.J.W. Godt. 1996. Effects of l ife h istory t raits on g enetic d ive r sity in p lant species. Philosophical transac tions: Biological Sci 351(1345):1291 1298. Harlan J.R. and J.M.J. de Wet. 1963. The c ompilospecies c oncept. Evol ut ion. 17(4):497 501. Harlan J.R. and J.M.J. de Wet. 1970. Polyploidy and s peciation in Dicanthium Evolution. 24(2):270 277. Heit, R.J. 1946. Laboratory germination resu lts with certain flower seeds. Proc of the Ass n. of Official Seed Analysts. 36:141 149. Helsop Harrison, J. and Y. Helsop Harrison. 1970. Evaluation of pollen viability by enzymatically induced fluorescence; intracellular hydrolysis of fluorescein diacetate. Stain Technology. 45(3) 115 120. Henderson, R J .F. 1969. A cytological study of Lantana montevidensis (Spreng) Briq. In Queensland. Co ntrib from the Queensland Herbarium. No. 3. Hermsen, J.G. 1984. Mechanisms and genetic implications of 2 n gamete formation. Iowa State J. Res. 58:421 434. Heywood, V.H. 1993. Flowering p lants of the world. Updated Edition. Oxford University Press, New York.

PAGE 203

203 Hodges, A.W. and J.J. Haydu. 2006. Economic impacts of the Florida e nvironmental h orticulture i ndustry in 2005. Food and Resource Economics Report, FE675, University of Florida, Gainesville. 24 April 2011 < http://economicimpact.ifas.ufl.edu > Howard, R. 1969. A check list of cultivar names used in the genus Lantana Arnoldia 29:73 109. Kandori, I., T. Hirao, S. Matsunaga, and T. Kurosaki. 2009. An invasive dandelion unilaterally reduced the reprod uction of a native congener throu gh competition for pollination. Oceologia. 159:559 569. Kao, K.N. 1975. A chromosomal stain ing method for cultured cells. Plant T issue C ult M ethods. Keller, R.P., D.M. Lodge, and D.C. Finnoff. 2007. Risk assessment for invasive species pro duces net bioeconomic benefits. Proc of the Nat l. Acad of Sci 104(1):203 207. Khaleel, T.F. 1972. Embryology of Lantana aculeate Linn. Var. nivea bailey. Current Sci 41(13):494. Khoshoo, T.N. and C. Mahal. 1967. Versitile reproducti on in Lantana camara Current Sci Bangalore 36:201 203. Kihara, H. Breeding for seedless fruits. Seiken Zih 9:1 7. Klingaman, G. 2000. Plant of the w eek ew Gold Lantana University of Arkansas Coop Ext Serv Ext N ews May, 12, 2000. Kulakow P A. 1999. Variation in Illinois bundleflower ( Desmanthus illinoensis (Michaux) MacMillan): A pot ential perennial grain legume. Euphytica 110:7 20. Knowler, D., E. Barbier. 2005 Importing exotic plants and the risk of invasion: A re market ba sed instrument s adequate? Ecol Economics 52:341 354. Knox, G.W. and S.B. Wilson. Evaluating n orth and s outh Florida l andscape p erformance and f ruiting of t en c ultivars and a w ild type s election of Nandina domestica a p otentially i nvasive shrub. J of Environ Hort 24(3):137 142. Li, Y., Z. Cheng, W. Smith, D. Ellis, Y. Chen, L. Lu, R.J. McAvoy, Y. Pei, W. Deng, C. Thammina, X. Zheng, H. Duan, K. Luo, and D. Zhao. 2006. Problems and challenges of invasive ornamental plants and molecular tools to c ontrol their spread. J of C rop I mprovement 17(1): 279 300. Loc kwood, J.L. and M.L. McKinney. 2001. Biotic Homogenization. Kluwer Academic/ Pl enum Publishers, New York, NY.

PAGE 204

204 Loomis, E.S. and L. Fishman. 2 009. A c ontinent w ide c lone: Population g enetic v ariation of the i nvasive p lant Hieracium aurantiacum (orange hawkweed; Asteraceae) in North America. Int. J. Plant Sci. 170(6):759 756. Lowe S., Browne M., Boudjelas S., De Poorter M. 2000 100 of the w orld s w orst i nvasive a lien s pecies : A selection from the g lobal i nvasive s pecies d atabase The Invasive Species Specialist Group (ISSG ) a specialist group of the Species Survival Commission (SSC) of the World Conservation Union (IUCN ) Mack, R.N. and M. Erneberg. 2002. The United States naturalized flora: Largely the product of deliberate introductions. Ann. of the Missouri Botanical Garden. 89(2)176 189. Mack, R.N. and W.M. Lonsdale. 2001. Humans as g lobal p lant d ispersers: Getting m ore t han we bargained for. Amer Inst of Biological Sci 51(2):95 102. Mathur, G. and H.Y. Ram. 1986. Floral b iology and p ollination of Lantana camara Phytomorphology. 36(12 ):79 100. McCoy, T.J. 1982. Inheritance of 2 n pollen formation in diploid alfalfa ( Medicago sativa L.). Can. J. Genet. Cytol. 24:315 323. McHale, N.A. 1983. Environmental induction of high frequency 2 n pollen formation i n diploid Solanum. Can. J. Genet. Cytol. 25:609 615. Miller, J.H. 2003. Nonnative invasi ve plants of southern forests. A field guide f or identification and control. USDA forest Service. Moody, M.L. and D.H. Les. 2007. Geographic distribution and genotypic compositon of invasive hybrid watermilfoil. ( Myriophyllum spicatum x M. sibiricum ) populations in North America. Biological Invasions. 9:559 570. Mok, D.W.S. and S.J. Peloquin. 1975 a The inheritance of thre e mechanisms of diplandroid (2 n pollen) formation in diploid potatoes. Hered. 35:295 302. Morton, J.F. 1994. Lantana or Red Sage ( Lantana camara L., [Verbenaceae]), Notorious w eed and p opular g arden flower: Some c ases of p oisoning in Florida Economic Bot 48(3): 259 270. Mugnai, S., F. Tognoni, and G. Serra. 1999. Water consumption and growth in nine cont ainer grown ornamental species. Agr icoltura Mediterranea. 129(2): 143 147. Myers, R.L. 1983. Site s usceptibility to invasion by the exotic tree Melaleuc a quinquenervia in s outhern F lorida. The J of Appl Ecol 20(2):645 658. N atarajan, A.T. and M.R. Ahuja. 1957. Cytotaxonomical studies in the genus Lantana J. Ind. Bot. Soc. 36:35 45.

PAGE 205

205 National Weeds Strategy Executive Committee. 2000 Weeds of National Significance: Lantana ( Lantana camara ) Strategic Plan. National Weeds Strategy Executive Committee, Launceston, Tasmania, Australia. Noyes, R.D. 2005. Inheritance of apomeiosis (diploispory) in fleabanes (Erigeron, Asteraceae ). Hered 94: 193 198. Noyes, R.D. 2006. Apomixis via recombination of genome regions for apomeiosis (diplospory) and parthenogenesis in Erigeron (daisy fleabane, Asteraceae). Sexual Plant Reproduction. 19:7 18. Olsen R.T. and T.G. Ranney. 2006 Fertility and i nheritance of v ariegated and p urple f oliage a cross a p olyploid s eries in Hypericum androsaemum L. J. Amer. Soc. Hort. Sci. 131(6):725 730. Ojha, B.M. and N. Dayal. 1992. Cytological i nvestigations in the g enus Lantana in India. Cytologia. 57:9 13. Park, J. R. Halitschke, H.B. Kim, I.T. Baldwin, K. A. Feldman, and R. Feyereisen. 2002. A knock out mutation in allene oxide synthase results in male sterility and defective wound signal transduction in Arabidopsis due to a block in jasmonic aci d biosynthesis. The Plant J 31(1):1 12. Pasha, C., M. Nag avalli, and L. Venkateswar Rao. 2007. Lantana camara for fuel ethanol product ion using thermotolerant yeast. Letters in Appl Microbilogy. 44:666 672. Perrings, C., M. Williamson, E. Barbier, D. Delfino, S. Dalmazzone, J. Shogren, P. Simmons, and A. Watkinson. 2002. Biological invasion risks and the public good: an economic perspecitive. Conservation Ecol 6(1):1. Pimentel, D., L. Lach, R. Zuniga, and D Morrison. 2000. Environmental and e conomic costs of noni ndigenous species in the United States. Bioscience. 50(1):53 65. Pimentel, D. R. Zuniga, D. Morrison. 2005. Update on the environmental and economic costs associated with alien invasive species in the United States Ecol Economics. 52:273 288. Poehlman, J.M. and D.A. Sleper. 1995. Breeding field crops. 4 th editon. Iowa State Press. Raghavan, R.S. and C.M. Arora. 1960. Morphological and c ytological studies in the genus Lantana L. Bull. Bot. Surv. India. 2:299 303. Rambuda, T.D and S.D. Johnson. 2004. Breeding systems of invasive alien plants in South Africa: D oes Baker s rule apply? Diversity and Distrib 10:409 416.

PAGE 206

206 Ramey, V. 1999. Invasive Plants. Lantana, shrub verbena, Lantana camara L. Aquaphyte Online. 19: 1 3. Ramsey, J. and D.W. Schemske. 1998. Pathways, mechanisms, and rates of polyploidy formation in flowering plants. Annu. Rev. Ecol. Systematics. 29:467 501. Ramsey, J. and D.W. Schemske. 2002. Neopolyploidy in flowering plants. Annu. Rev. Ecol. Syst. 33:589 639. Ranney, T.G. 2004. Population control: Developi ng non invasive nursery crops. Combined Proc. Intl. Plant Propagators Soc. 54:604 607. Rausch, J.H. and M.T. Morgan. 2005. The effect of self fertilization, inbreeding depression, and population size on autopolyploid establishment. Evolution. 59(9):1867 1875. Rayamajhi, M.B. T.K. Van, T.D. Center, J.A. Goolsby, P.D. Pratt, and A.R. Racelis. 2002. Biological a ttributes of t he c anopy held Melaleuca s eeds in Australia and Florida, U.S. J. Aquatic Plant M gt. 40:87 91. Reichard, S.E. 199 4. Assessing the potential of invasiveness in woody plant s introduced in North America. University of Washington, Seattle WA Reichard, S.H. and C.W. Hamilton 1997. Predicting i nvasions of w oody p lants introduced into North America. Conservation Biol 11 (1):193 203. Reichard, S.H. and P. White. 2001. Horticulture as a p athway of i nvasive p lant i ntro ductions in the United States. BioScience. 51(2):103 113. Rohlf, F.J. 2005. NTSYSpc numerical taxonomy and multivariate analysis system, version 2.2. Exeter Software, Setauket, NY. Roush, M .L. and S.R. Radosevich. 1985. Relationships b etween g rowth and c ompetitiveness of f our a nnual weeds. J of Appl Ecol 22(3):895 905. Sahu, A.K. and S. Panda. 1998. Population dynamics of a few dominant plant species around industrial complexes in west Bengal, India J Bombay Natural History Soc 95:15 18. Sanders, R.W. 1987a. Identity of Lantana depressa and L. ovatifolia (Verbenaceae) of Florida and the Bahamas Syst. Bot. 12:44 60. Sanders, R.W. 1987 b Taxonimic significance of chromosome observations in Caribbean spe cies of lantana (verbenaceae). Amer J l of Bot 74(6):914 920. Sanders, R.W. 2001. The genera of Verbenaceae in the s outheastern United States Harvard Papers in Bot ( 5 ) :303 358.

PAGE 207

207 Sanders, R.W. 2006. Taxonomy of Lantana sect. Lantana (Verbenaceae): I. correct application of Lantana camara and associated names. SIDA 22(1):381 421. Schemske, D.W. 1976. Pollinator s pecificity in Lantana camara and L. trifolia (Verbenaceae). Biotropica. 8(4):260 264. Singh, RJ. 2003. Plant Cytogenetics 2 nd edition. CRC Press New York, NY Sharma, G.P., A.S. Raghunbanshi, and J.S. Singh. 2005. Lantana invasion: An overview. Weed. Biol. Mg t. 5:157 165. Simberloff, D. 1996. Impacts of i ntroduced species in the United Sta tes. Consequences. 2(2). Sinha, S. and A. Sharma. 1984. Lantana camara L. A review. Feddes Rep o rt. 95:621 633. Spie s, J.J. and C.H. Stirton 1982a. C hromosome numbers of southern African plants: J. S outh Afr ican Bot. 48( 1 ):21 22. Spies, J.J. and C.H. Stirton. 1982b. Meiotic studies of some south African cultivars of Lantana camara (verbenaceae). Bothalia. 14: 101 111. Spies, J.J. and C.H. Stirton. 1982c. Embryo sac development in some South African cultivars of Lantana camara Bothalia. 14:113 117. Spie s, J.J. 1984 a A cytotaxonomic study of Lantana camara (verbenaceae) from South Africa. South African J of Bot 3:231 250. Spi es, J.J. 1984b Embryo sac development in some South African cultivars of Lantana camara Bothalia. 1 5 :1 61 1 66 Spies, J.J. 1984 c Hybridization potential of Lantana camara (Verbenaceae). Garcia de Orta, Ser. Bot., Lisboa. 6:145 150. Soltis, D.E. and P.S. Soltis. 1999. Polyploidy: recurrent formation and genome evolution. Trends in E col and E volution. 14(9):348 352. Starman, T. and L. Lombardini. 2006. Growth, g as e xchange, and c hlorophyll f luorescense of f our o rnamental h erbaceous p erennials during w ater d eficit conditions. J of Amer Soc for Hort Sci 131(4):469 475. Steere, W.C. 1931. Chromosome behavior in triploid Petunia hy brids. American J of Bot 19:340 357. Stelly, D.M. and S.J. Peloquin. 1986. Formation of 2 n megagametophytes in diploid tuber bearing solanums. Amer J. of B ot 73(9):1351 1363.

PAGE 208

208 Suehs, C.M., S. Charpentier, L. Affre, and F. Mdail. 2006. The Evolutionary potential of invasive Carpobrotus (Aizoaceae) taxa: A re pollen mediated gene flow potential and hybrid vigor levels connected? Evolution Ecol 20:447 463 Thomas, S.E., C.A. Ellison, and A.J. Tomley. 2006. Studies on the rust Prospo dium tuberculatum a new classical biological control agent released against the invasive alien weed Lantana camara in Australia. 2. Host range. Austral Plant Pathol 35:321 328. Trueblood, C.E., T.G. Ranney, N.P. Lynch, J.C. Neal, and R.T. Olsen. 2010. Evaluating f ertility in t riploid c lones of Hypericum androsaemum L. for u se as non invasi ve l andscape plants. Hortsci ence 45(7):1026 1028. USDA 201 0 N ASS. Floriculture Crops 2009 summary. < http://plants.usda.gov > USDA 2011 NRCS Lantana L. Lantana 24 Jan. 2011. < http://plants.usda.gov/java/profile?symbol=LANTA > Veilleux, R.E., N.A. McHale, and F.I. Lauer. 1982. Unreduced gametes in diploid Solanum phureja Juz. and Buk. Theor etical and Appl. Genet. 59:95 100. Veracion, V. P. 1983. Live fence in forest research. Canopy International (Philippines): March, p. 7. Virtue, J.G., S .J. Bennett, and R.P. Randall. 2004. Plant introductions in Australia: H ow c an we resolve Fourteenth Austral Weeds Conf 42 48. Wagstaff, D .J. 2008. International Poisonous Plants Checklist : An Evidence based Reference. CRC Press, New York, NY. Wardle, D.A. K.S. Nicholson, M. Ahmed, and A. Rahman. 1994. Interference effects of the invasive plant Carduus nutans L. against the nitrogen fixation ability of Trifolium repens L. Plant and Soil. 163:287 297. Watanabe, K. and S.J. Peloquin. 1989. Occurrence of 2 n pollen and ps gene frequencies in cultivated groups and their related wild species in tuber bearing Solanums. Theoretical and Appl. Genet. 78:329 336. Weihe, P.E. and R.K. Neely. 1997. The effects of shading on competition between purple loosest rife and broad leaved cattail. Aquatic Bot 59:127 138. Wirth, F.F ., K.J. Davis, and S.B. Wilson. 2004. Florida nursery sales and economic impacts of 14 potentially inv asive ornamental plant species. J. of Env. Hort. 22:12 16.

PAGE 209

209 Wilson, S.B. and L.K. Mecca. 2003. Seed p roduction and g ermination of e ight c ultivars and w ild t ype of Ruellia tweediana : A p otentially i nvasive ornamental. J of Env Hort 21(3):137 143. Wolf son, S.L. and T.W.G. Solomons. 1964. Poisoning by fruit of Lantana camara Amer J of Diseases of Children. 107:173 176. Wright, S. 1931. Evol ution in Mendelian populations. Genet ics 16:97 159. Zalucki, M.P. M.D. Day, and J. Playford. 2007. Will biological control of Lantana camara ever succeed? Patterns, processes & prospects. Biological Control. 42:251 261.

PAGE 210

210 BIOGRAPHICAL SKETCH David Mark Czarnecki II was born in PA in 1982. He grew up mostly in Bedford, TX. In 2004 he received h is B.S. in horticultural sciences from Texas A&M University in 2004. In May of 2006 he co mpleted his M.S. degree in the e nvironmental h orticulture dep artment from the University of Florida studying the diversity of Coreopsis leavenworthii Upon finishing his M.S. degree he started a Ph.D. program studying the genetics of sterilizing Lantana camara He will graduate with his Ph.D. in the Environmental Horticulture department in August of 201 1