<%BANNER%>

Hybridization of Various Races of Vaccinium darrowi with Cultivated Highbush Blueberry, V. arboreum, and V. fuscatum

Permanent Link: http://ufdc.ufl.edu/UFE0022827/00001

Material Information

Title: Hybridization of Various Races of Vaccinium darrowi with Cultivated Highbush Blueberry, V. arboreum, and V. fuscatum
Physical Description: 1 online resource (179 p.)
Language: english
Creator: Chavez Velasquez, Dario
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: blueberry, breeding, colchicine, colchiploid, corymbosum, cytogenetics, darrow, evergreen, gametes, heteroploid, homoploid, meiosis, pmc, polyploidy, unreduced, vaccinium
Horticultural Science -- Dissertations, Academic -- UF
Genre: Horticultural Science thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Plants from 3 Florida V. darrowi races were self-pollinated and crossed with southern highbush blueberry cultivars and with other Florida Vaccinium species. Partial to complete self-incompatibility was present in V. darrowi. V. arboreum was self-sterile. Some crosses between diploid V. darrowi and tetraploid highbush produced tetraploid hybrids because of unreduced gametes from V. darrowi. Variation in frequency of unreduced gamete production in diploid V. darrowi was present within plants (megaspores vs. microspores) and among plants within races. Overall, diploid V. darrowi, diploid V. fuscatum, and diploid F-1 (V. darrowi x V. fuscatum, natural hybrids), when crossed with southern highbush cultivars, were equally productive of hybrids whether used as males or females. Crosses of V. arboreum with V. fuscatum and V. darrowi were easy to make, especially when V. arboreum was the pollen parent. F-1 hybrids from these crosses and their reciprocals were highly vigorous. Pollen stainability of F-1 V. darrowi x V. arboreum hybrids was low. Numerous meiotic abnormalities were observed in these F-1 hybrids. F-1 V. darrowi x tetraploid V. corymbosum hybrid pollen stainability was highly variable among clones. F-1 hybrids with high pollen fertility were easy to backcross to tetraploid highbush cultivars. Several tetraploid V. darrowi plants were produced by colchicine treatment of seed. Stomata and pollen size measurements were efficient and accurate indicators of chromosome-doubled V. darrowi plants and periclinal chimeras. Crosses of colchicine-derived tetraploid V. darrowi plants with southern highbush cultivars and colchicine-derived tetraploid V. arboreum plants were easy to make.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Dario Chavez Velasquez.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Lyrene, Paul M.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022827:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022827/00001

Material Information

Title: Hybridization of Various Races of Vaccinium darrowi with Cultivated Highbush Blueberry, V. arboreum, and V. fuscatum
Physical Description: 1 online resource (179 p.)
Language: english
Creator: Chavez Velasquez, Dario
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: blueberry, breeding, colchicine, colchiploid, corymbosum, cytogenetics, darrow, evergreen, gametes, heteroploid, homoploid, meiosis, pmc, polyploidy, unreduced, vaccinium
Horticultural Science -- Dissertations, Academic -- UF
Genre: Horticultural Science thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Plants from 3 Florida V. darrowi races were self-pollinated and crossed with southern highbush blueberry cultivars and with other Florida Vaccinium species. Partial to complete self-incompatibility was present in V. darrowi. V. arboreum was self-sterile. Some crosses between diploid V. darrowi and tetraploid highbush produced tetraploid hybrids because of unreduced gametes from V. darrowi. Variation in frequency of unreduced gamete production in diploid V. darrowi was present within plants (megaspores vs. microspores) and among plants within races. Overall, diploid V. darrowi, diploid V. fuscatum, and diploid F-1 (V. darrowi x V. fuscatum, natural hybrids), when crossed with southern highbush cultivars, were equally productive of hybrids whether used as males or females. Crosses of V. arboreum with V. fuscatum and V. darrowi were easy to make, especially when V. arboreum was the pollen parent. F-1 hybrids from these crosses and their reciprocals were highly vigorous. Pollen stainability of F-1 V. darrowi x V. arboreum hybrids was low. Numerous meiotic abnormalities were observed in these F-1 hybrids. F-1 V. darrowi x tetraploid V. corymbosum hybrid pollen stainability was highly variable among clones. F-1 hybrids with high pollen fertility were easy to backcross to tetraploid highbush cultivars. Several tetraploid V. darrowi plants were produced by colchicine treatment of seed. Stomata and pollen size measurements were efficient and accurate indicators of chromosome-doubled V. darrowi plants and periclinal chimeras. Crosses of colchicine-derived tetraploid V. darrowi plants with southern highbush cultivars and colchicine-derived tetraploid V. arboreum plants were easy to make.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Dario Chavez Velasquez.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Lyrene, Paul M.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022827:00001


This item has the following downloads:


Full Text

PAGE 1

HYBRIDIZATION OF VARIOUS RACES OF Vaccinium darrowi WITH CULTIVATED HIGHBUSH BLUEBERRY, V. arboreum AND V. fuscatum By DARIO JAVIER CHAVEZ-VELASQUEZ A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Daro Javier Chvez-Velsquez 2

PAGE 3

To my family and to my love 3

PAGE 4

ACKNOWLEDGMENTS I thank my parents for their love and support in my life. I thank my mentor, Dr. Paul Lyrene, for his support on my rese arch project and for his voice of experience to keep me going. I thank him most for teaching me to love plan t breeding. I express gratitude to my committee members, Dr. Eileen Kabelka and Dr. Jose Chaparro. I thank David Norden, Vicky Ackroyd, Roger Haring, Santiago Torres, Lorena Luna and Laura Patio. I thank my love, Rachel. Finally, I am grateful to My Lord for giving me strength and companionship during this stage of my life. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................8 LIST OF FIGURES .......................................................................................................................13 ABSTRACT ...................................................................................................................................15 CHAPTER 1 INTRODUCTION................................................................................................................. .17 2 INTER-SPECIFIC CROSSING BEHAVIOR OF Vaccinium darrowi Camp WITH SOUTHERN HIGHBUSH CULTIVARS..............................................................................20 Introduction .............................................................................................................................20 Materials and Methods ...........................................................................................................23 Studies with V. darrowi Races ........................................................................................23 Inter-Specific Hybridization Experiments .......................................................................24 Control crosses .........................................................................................................24 4x 2x crosses .........................................................................................................26 2x 4x crosses .........................................................................................................28 Statistical analysis ....................................................................................................31 Morphological Studies of F-1 Hybrids (V. darrowi Southern Highbush Cultivars) ....31 Fertility Studies of F-1 ( V. darrowi Southern Highbush Cultivars) Hybrids ...............33 Pollen stainability .....................................................................................................33 Backcrossing experiments ........................................................................................34 Results and Discussion ...........................................................................................................35 Studies with V. darrowi Races ........................................................................................35 Inter-Specific Hybridization Data ...................................................................................36 Morphological Studies of F-1 Hybrids (V. darrowi Southern Highbush Cultivars) ....40 Leaf characteristics ...................................................................................................40 Flower characteristics ...............................................................................................41 Berry characteristics .................................................................................................42 Plant architecture ......................................................................................................42 Fertility Studies of F-1 ( V. darrowi V. corymbosum ) Hybrids .....................................43 Pollen stainability data .............................................................................................43 Backcross data ..........................................................................................................44 Conclusions .............................................................................................................................44 3 INTER-SPECIFIC CROSSES OF Vaccinium darrowi Camp. V. arboreum Marsh..........62 Introduction .............................................................................................................................62 Materials and Methods ...........................................................................................................66 5

PAGE 6

Inter-Specific Hybridization Experiments .......................................................................66 Diploid section Cyanococcus diploid section Batodendron .................................66 Control crosses .........................................................................................................68 Statistical analysis ....................................................................................................69 Fertility Studies of F-1 ( V. darrowi V. arboreum ) Hybrids .........................................70 Cytogenetics of F-1 ( V. darrowi V. arboreum ) Hybrids ..............................................71 Results and Discussion ...........................................................................................................72 Inter-Specific Hybridization Experiments .......................................................................72 Diploid section Cyanococcus diploid section Batodendron data ..........................72 Overall comparisons .................................................................................................74 Fertility Studies of F-1 ( V. darrowi V. arboreum ) Hybrids .........................................75 Cytogenetics of F-1 ( V. darrowi V. arboreum ) Hybrids ..............................................76 Hybrid FL98-146 (2n=2x=24).................................................................................76 Hybrid FL98-132 (2n=2x=24).................................................................................77 Hybrid FL98-129 (2n=2x=24).................................................................................78 Hybrid FL98-115 (2n=2x=24).................................................................................79 Hybrid FL98-95 (2n=2x=24)...................................................................................79 Hybrid FL98-93 (2n=2x=24)...................................................................................80 Hybrid FL98-53 (2n=2x=24)...................................................................................81 Hybrid FL98-61 (2n=2x=24)...................................................................................81 Hybrid FL98-64 (2n=2x=24)...................................................................................82 Hybrid FL98-165 (2n=2x=24).................................................................................83 Overall results ..........................................................................................................83 Conclusions .............................................................................................................................85 4 SELF-POLLINATION OF V. darrowi AND OTHER LOW CHILL BLUEBERRY TAXA ..................................................................................................................................104 Introduction ...........................................................................................................................104 Materials and Methods .........................................................................................................107 Hybridization Experiments ............................................................................................107 Vaccinium darrowi selfand cross-pollination ......................................................107 Vaccinium arboreum selfand cross-pollination ...................................................109 Southern highbush cultivars (V. corymbosum ) selfand cross-pollination ............111 Vaccinium fuscatum self-pollination experiments .................................................112 Hybrids ( V. darrowi V. corymbosum ) selfand cross-pollination ......................113 Results and Discussion .........................................................................................................114 Hybridization Experiments ............................................................................................114 Vaccinium darrowi selfand cross-pollination ......................................................114 Vaccinium arboreum selfand cross-pollination ...................................................116 Southern highbush cultivars (V. corymbosum ) selfand cross-pollination ............117 Vaccinium fuscatum self-pollination experiments .................................................118 Hybrids ( V. darrowi V. corymbosum ) selfand cross-pollinated .......................118 Conclusions ...........................................................................................................................118 6

PAGE 7

5 FERTILITY AND PLOIDY OF V. darrowi SOUTHERN HIGHBUSH F-1 HYBRIDS AND THEIR ABILITY TO BACKCROSS TO TETRAPLOID SOUTHERN HIGHBUSH BLUEBERRY...........................................................................125 Introduction ...........................................................................................................................125 Materials and Methods .........................................................................................................128 Fertility of F-1 ( V. darrowi V. corymbosum and Reciprocals) Hybrids .....................128 Backcrossing Experiments ............................................................................................129 Hybrids ( V. darrowi V. corymbosum and reciprocals) V. corymbosum ..........129 Hybrids ( V. darrowi V. corymbosum ) V. ashei and reciprocals ......................131 Hybrids ( V. darrowi V. corymbosum ) V. darrowi and reciprocals ..................132 Results and Discussion .........................................................................................................133 Fertility of F-1 ( V. darrowi V. corymbosum and Reciprocal) Hybrids ......................133 Backcrossing Experiments ............................................................................................134 Hybrids ( V. darrowi V. corymbosum ) V. corymbosum ....................................134 Hybrids ( V. darrowi V. corymbosum ) V. ashei and reciprocals ......................135 Hybrids ( V. darrowi V. corymbosum ) V. darrowi and reciprocals ..................136 Conclusions ...........................................................................................................................137 6 CROSSING VALUE OF V. darrowi COLCHICINE-DERIVED TETRAPLOIDS IN CROSSES WITH SOUTHERN HIGHBUSH. .....................................................................144 Introduction ...........................................................................................................................144 Materials and Methods .........................................................................................................147 Colchicine Treatment ....................................................................................................147 Stomata and Pollen Screening .......................................................................................149 Morphological Studies of Colchiploid Derived Plants ..................................................150 Inter-Specific Hybridization Experiments .....................................................................151 Vaccinium corymbosum (4x) colchicine-derived V. darrowi (4x) ......................151 Colchicine-derived V. darrowi (4x) colchicine-derived V. arboreum (4x) ........151 Vaccinium darrowi (2x) colchicine-derived V. darrowi (4x) .............................152 Control Crosses .............................................................................................................153 Vaccinium darrowi (2x) V. corymbosum (4x) and reciprocals ...........................153 Vaccinium darrowi (2x) V. arboreum (2x) and reciprocals ................................153 Statistical Analysis ........................................................................................................154 Cytogenetics of Colchici ne-Derived Tetraploids ..........................................................154 Results and Discussion .........................................................................................................155 Results of Colchicine Treatment ...................................................................................155 Stomata and Pollen Screening Data ..............................................................................156 Analysis of Morphological Characteri stics of Colchiploid-Derived Plants ..................157 Inter-Specific Hybridization ..........................................................................................158 Cytogenetics of Colchicine-Derived Tetraploid ............................................................160 Conclusions ...........................................................................................................................160 LIST OF REFERENCES .............................................................................................................175 BIOGRAPHICAL SKETCH .......................................................................................................179 7

PAGE 8

LIST OF TABLES Table page 1-1. Mean number of flowers that opened per plant per day on V. darrowi plants of two Florida races as measured in a green house over 52 weeks, in Gainesville, Fl. .................47 1-2. Result of intraspecific hybridization of southern highbush cultivars in 2007.......................47 1-3. Result of intra-spec ific hybridization of V. darrowi clones in 2007. .....................................47 1-4. Result of inter-specific hybridization be tween southern highbush cultivars and V. darrowi in 2007. Highbush was used as the female parent. ...............................................48 1-5. Result of reciprocal inter-spe cific hybridization between diploid V. darrowi V. fuscatum natural hybrid and southe rn highbush cultivars in 2007. ...................................48 1-6. Result of inter-specific hybridization between southern highbush cultivars and V. darrowi in 2008. Highbush was used as the female parent. ..............................................49 1-7. Result of reciprocal inter-specific hybridiza tions between diploid Vaccinium fuscatum and southern highbush cultivars in 2008. ..........................................................................49 1-8. Result of inter-speci fic hybridization between V. darrowi from the Ocala Forest race and southern highbush cultivars in 2001. In all crosses, V. darrowi was the female parent. .................................................................................................................................50 1-9. Result of inter-speci fic hybridization between V. darrowi from the Istokpoga race and southern highbush cultivars in 2005. V. darrowi was the female parent. ..........................50 1-10. Result of inter-specific hybridization between V. darrowi and southern highbush cultivars in 2006. V. darrowi was the female parent. ........................................................51 1-11. Result of inter-specific hybridization between V. darrowi and southern highbush cultivars in 2007. V. darrowi was the female parent. ........................................................51 1-12. Result of inter-specific hybridization between V. darrowi and southern highbush cultivars in 2008. V. darrowi was the female parent. ........................................................52 1-13. Result of inter-specific hybridization between V. darrowi and southern highbush cultivars with their reciprocalsz in 2007. ...........................................................................53 1-14. Crossing behavior of dipl oid and tetraploid Florida Vaccinium species in section Cyanococcus in 2007. ........................................................................................................54 1-15. Crossing behavior between di ploid and tetraploid Florida Vaccinium species in 2008. ......54 1-16. Leaf, flower and berry charac teristics of three taxa: F-1 ( V. darrowi southern highbush hybrids), southern highbush (HB), and V. darrowi (VD). .................................55 8

PAGE 9

1-17. Mean pollen stainability from Vaccinium darrowi, V. corymbosum V. arboreum and V. darrowi V. corymbosum hybrids. ...............................................................................56 1-18. Crossing behavior between F-1 ( V. darrowi southern highbush cultivars) and southern highbush cultivars. ..............................................................................................56 2-1. Hybridization experi ments between diploid V. darrowi in section Cyanococcus and diploid V. arboreum in section Batodendron in 2007. V. darrowi was female parent. ....87 2-2. Hybridization experi ments between diploid V. arboreum in section Cyanococcus and diploid V. darrowi in section Batodendron in 2007. V. arboreum was the female parent. .................................................................................................................................87 2-3. Result of inter-specific hybridiz ation experiments between diploid Vaccinium fuscatum in section Cyanococcus and V. arboreum in section Batodendron in 2007. V. fuscatum was the female parent .........................................................................................87 2-4. Result of inter-specific hybridization between diploid V. arboreum in section Batodendron and diploid V. fuscatum in section Cyanococcus in 2007. V. arboreum was the female parent. ........................................................................................................88 2-5. Hybridization experi ments between diploid V. darrowi in section Cyanococcus and diploid V. arboreum in section Batodendron in 2008. V. darrowi was the female parent. .................................................................................................................................88 2-6. Hybridization experi ments between diploid V. arboreum in section Batodendron and diploid V. darrowi in section Cyanococcus in 2008. V. arboreum was the female parent. .................................................................................................................................88 2-7. Result of intr a-specific hybridiz ation of diploid V. arboreum clones in 2007. ......................89 2-8. Summary of crossing results between diploid V. arboreum diploid V. darrowi and diploid V. fuscatum in 2007. ..............................................................................................89 2-9. Weight and seed count per be rry in crosses between diploid V. arboreum diploid V. darrowi and diploid V. fuscatum in 2007. ..........................................................................89 2-10. Mean pollen stainability of ten-year-old F-1 ( V. darrowi V. arboreum ) hybrids of the Horticultural Plant Science Unit at Gainesville, Florida. ..................................................90 2-11. Mean pollen stainability of ten-year-old F-1 ( V. darrowi V. arboreum ) hybrids and V. darrowi and V. arboreum clones. ..................................................................................90 2-12. Meiotic analysis of PMCs in seedling FL98-146 a V. darrowi V. arboreum hybrid. .......90 2-13. Meiotic analysis of PMCs in seedling FL98-132 a V. darrowi V. arboreum hybrid. .......91 2-14. Meiotic analysis of PMCs in seedling FL98-129 a V. darrowi V. arboreum hybrid. .......91 9

PAGE 10

2-15. Meiotic analysis of PMCs in seedling FL98-115 a V. darrowi V. arboreum hybrid. .......92 2-16. Meiotic analysis of PMCs in seedling FL98-95 a V. darrowi V. arboreum hybrid. .........92 2-17. Meiotic analysis of PMCs in seedling FL98-93 a V. darrowi V. arboreum hybrid. .........93 2-18. Meiotic analysis of PMCs in seedling FL98-53 a V. darrowi V. arboreum hybrid. .........94 2-19. Meiotic analysis of PMCs in seedling FL98-61 a V. darrowi V. arboreum hybrid. .........94 2-20. Meiotic analysis of PMCs in seedling FL98-64 a V. darrowi V. arboreum hybrid. .........95 2-21. Meiotic analysis of PMCs in seedling FL98-165 a V. darrowi V. arboreum hybrid. .......96 3-1. Results of self-pollinating six V. darrowi clones in 2007. ...................................................120 3-2. Results of self-pollinating four V. darrowi clones in 2008. .................................................120 3-3. Results of self-pollinating three V. arboreum clones in 2008. .............................................120 3-4. Results of self-pollinating five southern highbush cultivars in 2008...................................120 3-5. Results of self-pollinating three V. fuscatum clones in 2008. ..............................................121 3-6. Result of cross-pollination experiments between F-1 (V. darrowi southern highbush cultivars) in 2007. ............................................................................................................121 3-7. Results of selfpollinating 17 F-1 hybrid ( V. darrowi V. corymbosum ) plants in 2008. ..121 3-8. Selfand cross-pollination results in Florida V. darrowi races measured by fruit set (%), pollination-to-ripening interval (pol-ripe interval), average pol-ripe-interval and number of seedlings per pollinated flower .......................................................................122 3.9. Results of selfand cro ss-pollination in Florida V. darrowi races measured by berry weight (g), seeds per berry, plump seeds per pollinated flower and seedlings per pollinated flower. .............................................................................................................122 3-10. Results of six crosses and three self-pol linations in V. arboreum Each cross involved two unrelated V. arboreum parents. Three different V. arboreum seedlings were selfpollinated. .........................................................................................................................122 3-11. Results of selfand cross-pollinati on in Florida southern highbush clones and seedlings. ..........................................................................................................................123 3-12. Results of selfand cross-pollination fo r southern highbush cultivars measured by berry weight (g), number of large seeds, number of small seeds, number of seeds per berry and number of plump seeds per pollinated flower.. ...............................................123 10

PAGE 11

3-13. Results of two crosses and 17 self-pollinations involving F-1 ( V. darrowi V. corymbosum ) hybrids. Each of the 17 self-pollin ations used as di fferent F-1 hybrid. ....123 3-14. Results of selfand cro ss-pollination involving F-1 ( V. darrowi V. corymbosum ) hybrids measured by berry weight (g), seeds per berry, plump seeds per pollinated flower and seedlings per pollinated flower. .....................................................................124 4-1. Result of hybridization of F-1 ( V. darrowi southern highbush cultivars) hybrids with southern highbush cultivars in 2007. ...............................................................................139 4-2. Result of reciprocal hybridizations of F-1 (V. darrowi southern highbush cultivars) hybrids with V. ashei Florida Rose in 2007. .................................................................139 4-3. Result of hybridization of F-1 ( V. darrowi southern highbush cultivars) hybrids with southern highbush cultivars in 2008. F1 hybrids were the female parents. ....................140 4-4. Result of reciprocal hybridizations of F-1 (V. darrowi southern highbush cultivars) hybrids with diploid Vaccinium darrowi high 2n egg producersin 2008. ...................140 4-5. Mean pollen stainability of F-1 ( V. darrowi V. corymbosum ) hybrids selected from crosses made in 2006. The population was planted in Citra, Florida. .............................140 4-6. Pollen stainability of five F-1 ( V. darrowi V. corymbosum ) hybrids from crosses made in 2004 and 2005. .............................................................................................................141 4-7. Crossing behavior of F-1 hybrids (V. darrowi southern highbush), V. darrowi and southern highbush cultivars in 2007. ...............................................................................141 4-8. Weight and seed count per be rry in crosses of F-1 hybrids ( V. darrowi southern highbush), V. darrowi and southern highbush cultivars in 2007. ....................................141 4-9. Crossing behavior of F-1 hybrids (V. darrowi southern highbush), V. darrowi and southern highbush cultivars in 2008. ...............................................................................142 4-10. Weight and seed count per be rry in crosses of F-1 hybrids ( V. darrowi southern highbush), V. darrowi and southern highbush cultivars in 2008. ....................................142 5-1. Leaf and corolla characte ristics of ten diploid, four periclinal chimeras, and one tetraploid V. darrowi seedlings grown from colchicine-treated seed. Seedlings were approximately two-year-old when measured. ..................................................................162 5-2. Result of inter-specific hybridization between southern highbush cultivars and colchicine-derived V. darrowi clones in 2008. Highbush was used as the female parent. ...............................................................................................................................163 5-3. Result of inter-specific hybridiz ation between colc hicine-derived V. darrowi and colchicine-derived V. arboreum clone in 2008. Colchicine-derived V. darrowi were used as the female parent. ................................................................................................163 11

PAGE 12

5-4. Result of hybridiza tion between diploid V. darrowi clones and colchicine-derived V. darrowi clones in 2008. Diploid V. darrowi was used as the female parent. ..................163 5-5. Screening measurements of stom ata length and pollen diameter for V. darrowi seedlings grown from colchicine-treated seed. ................................................................................164 5-6. Colchicine doubling rate af ter aqueous treatments in V. darrowi seed planted. ..................165 5-7. Result of crossing colchicine-derived V. darrowi plants with other Vaccinium species in 2008..................................................................................................................................165 5-8. Results of crossing V. darrowi with V. arboreum at the tetraploid and diploid levels in 2008..................................................................................................................................166 5-9. Chromosome associations at me taphase I for colchiploid-derived V. darrowi plant FL08-403..........................................................................................................................166 12

PAGE 13

LIST OF FIGURES Figure page 1-1. Flower production of V. darrowi per week by region from October 2006 to August 2007....................................................................................................................................57 1-2. Leaf morphology for diploid V. darrowi (left), F-1 hybrids (center) and tetraploid V. corymbosum (right). ...........................................................................................................58 1-3. Scatter plot for leaf charac teristics by taxa. Each symbol represents the mean of five leaves from one plant. ........................................................................................................58 1-4. Flower and fruit morphology for diploid V. darrowi (left), F-1 hybrids (center) and tetraploid V. corymbosum (right). ......................................................................................59 1-5. Scatter plot for coro lla characteristics by taxa. Each sym bol represents the mean of five leaves from one plant. ........................................................................................................59 1-6. Plant architecture of diploid V. darrowi (left), F-1 hybrids (cen ter) and tetraploid V. corymbosum (right). The pot on the left is 30 cm tall. .......................................................60 1-7. Histogram of the F-1 (V. darrowi V. corymbosum ) hybrid population pollen stainability percentage. ......................................................................................................61 1-8. Pollen from one plant with poor pollen st aining (left panel) a nd one plant with good pollen staining (right panel) from a population of F-1 ( V. darrowi V. corymbosum ) hybrids; 250x. ....................................................................................................................61 2-1. Microphotographs of pollen from two F-1 ( V. darrowi V. arboreum ) hybrids; FL98165 (left) and FL98-61 (right), 250x. .................................................................................97 2-2. Ten-year-old F-1 ( V. darrowi V. arboreum ) hybrid population bushes at University of Florida Horticultural Unit in Gainesvi lle, Florida. Plant height approx. 3 m. ...................97 2-3. Ten-year-old F-1 ( V. darrowi V. arboreum ) hybrid plant at University of Florida Horticultural Unit in Gainesville, Florida. Heavy flowering in November 2007. .............98 2-4. One plant from a ten-year-old F-1 ( V. darrowi V. arboreum ) hybrid at the University of Florida Horticultural Unit in Gainesvill e, Florida. Large number of shoots per plant. The photo shows the lower 1 m of the plant. ...........................................................99 2-5.Meiotic analysis at prophase I and metaphase I of F-1 ( V. darrowi V. arboreum ) hybrids PMCs...................................................................................................................100 2-6. Meiotic analysis at metaphase I, anaphase I and tel ophase I of F-1 ( V. darrowi V. arboreum ) hybrids PMCs.. ..............................................................................................101 13

PAGE 14

2-7.Meiotic analysis at anaphase II and telophase II of F-1 ( V. darrowi V. arboreum ) hybrids PMCs...................................................................................................................102 2-8.Meiotic analysis at telophase II of F-1 ( V. darrowi V. arboreum ) hybrids PMCs. ............103 4-1. Microphotographs of shrunken and shri veled pollen (arrow), and plump and well stained 2n sporads (arrow head) of FL07-110 F-1 ( V. darrowi V. corymbosum ) hybrid.. .............................................................................................................................143 5-1. Tetraploid V. darrowi seedling (Clone 43) after treatmen t with colchicine. The stake is 20 mm wide. .....................................................................................................................167 5-2. Tetraploid V. darrowi seedling (Clone 40) after treatmen t with colchicine. The stake is 20 mm wide. .....................................................................................................................168 5-3. Tetraploid V. darrowi seedlings (Clone 17) after treatment with colchicine. The stake is 20 mm wide. .....................................................................................................................169 5-4. Stomata microphotographs. Normal LI laye r (top) and doubled LI layer (bottom). 250 ......................................................................................................................................170 5-5. Pollen microphotographs. Normal LII layer (top) and doubled LII layer(bottom) 250. ....171 5-6. Box plots, a) and b), of stomata length measurements for V. darrowi colchicine treated plants. The height of the box tells 75th percentile and the bottom 25th percentile. Line within the box tells median. External lines, upper level maximum and lower level minimum. .........................................................................................................................172 5-7. Box plots a) and b), of pollen diameter measurements for V. darrowi colchicine treated plants. ...............................................................................................................................173 5-8. Meiotic metaphase I in colchicine-deriv ed FL08-403 (2n=4x=48) (Clone 40). Left: 8 I, 8 II, 4 III and 3 IV. Right: 3 I, 5 II, 5 II and 5 IV. 400. .................................................174 14

PAGE 15

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science HYBRIDIZATION OF VARIOUS RACES OF Vaccinium darrowi WITH CULTIVATED HIGHBUSH BLUEBERRY, V. arboreum AND V. fuscatum By Daro Javier Chvez-Velsquez December 2008 Chair: Paul Lyrene Major: Horticultural Science Plants from 3 Florida V. darrowi races were self-pollinated and crossed with southern highbush blueberry cultivars and with other Florida Vaccinium species. Partial to complete selfincompatibility was present in V. darrowi V. arboreum was self-sterile. Some crosses between diploid V. darrowi and tetraploid highbush produced tetraploid hybrids because of unreduced gametes from V. darrowi. Variation in frequency of unre duced gamete production in diploid V. darrowi was present within plants (megaspores vs. microspores) and among plants within races. Overall, diploid V. darrowi diploid V. fuscatum and diploid F-1 ( V. darrowi x V. fuscatum natural hybrids), when crossed with southern highbush cultivars, were equally productive of hybrids whether used as males or females. Crosses of V. arboreum with V. fuscatum and V. darrowi were easy to make, especially when V. arboreum was the pollen parent. F-1 hybrids from these crosses and their reciprocals were highly vigorous. Pollen stainability of F-1 V. darrowi x V. arboreum hybrids was low. Numerous meiotic abnormalities were observed in these F-1 hybrids. F-1 V. darrowi x tetraploid V. corymbosum hybrid pollen stainability was highly variable among clones. F-1 hybrids with high pollen fertility were easy to backcross to tetraploid highbush cultivars. Several tetraploid V. darrowi plants were produced by colchicine treatment of seed. Stomata and pollen size measur ements were efficient and accurate indicators 15

PAGE 16

of chromosome-doubled V. darrowi plants and periclinal chimeras Crosses of colchicine-derived tetraploid V. darrowi plants with southern highbush cultiv ars and colchicine-derived tetraploid V. arboreum plants were easy to make. 16

PAGE 17

CHAPTER 1 INTRODUCTION The start of the cultivated highbush blueberry industry is attributed to F. V. Coville of the United States Department of Agriculture. Covill es first hybridization stud ies, in 1911, used wild blueberries as parents (Coville, 1937). These cr osses produced several cultivars (Moore, 1965). The blueberry industry expanded slowly but steadily until 1950 and more rapidly thereafter, with continuous introduction of new va rieties. New breeding programs were developed in other regions based on the earl y work of Coville. Florida native species were first used in breeding in 1940 when the USDA made crosses using two rabbiteye ( Vaccium ashei Reade) selections (Sharpe, 1953). In 1949, Sharpe, at the University of Florida, began breeding southern blueberries, using advanced selections and cultivars from breeding programs in the northern Un ited States as parents (Lyrene, 1997). Sharpe saw the need to develop highbush cultivars that were adapted to Florida conditions because the northern cultivars, when planted in north Florid a, had problems with high chilling requirements, poor adaptation, and susceptibility to diseases. Three native Florida Vaccinium species rabbiteye blueberry ( Vaccinium ashei Reade, 6x), shiny blueberry ( V. myrsinites Lam., 4x), and Darrows evergreen blueberry ( V. darrowi Camp, 2x) were used in crosses to reduce chilling requirement of the highbush blueberries (Sharpe, 1953). Rabbiteye blueberry was the principal cultiv ated blueberry species in north and west Florida during the 1920s. Low chilling requirement, resistance to stem canker ( Botryosphaeria corticis ), drought resistance and heat re sistance were traits that ma de possible its cultivation in Florida. Rabbiteye blueberries are hexaploid and northern highbush cult ivars are tetraploid. Crosses between rabbiteye and highbush cultivar s resulted in pentaploids, which had reduced fertility, although they were more fertile than or iginally expected (Darrow, et al., 1949; Moore, 17

PAGE 18

1965). In addition, crosses of rabbiteye blueberry with the Florida diploid, V. darrowi were made, with the goal of producing te traploid plants that could be crossed easily with tetraploid northern highbush. From over 7500 pollinations only 5 hybrids were obtained using V. ashei and V. darrowi (Sharpe and Darrow, 1959). Furthermore, Goldy and Lyrene (1983) found that these hybrids from V. ashei V. darrowi crosses were pentaploid instead of tetraploid as expected, a result of unreduced gametes produced by V. darrowi Sharpe and Darrow (1959) found that V. myrsinites readily produces fertile hybrids with highbush cultivars. In Gainesville, Florida, in 1952, forty hybrids were grown. These had several undesirable characteristics, incl uding low-growing bushes, low berry number, and unattractive dark berry color. V. darrowi which is diploid, was considered to have low potential for crossing with highbush cultivars. The triploid block is strong in Vaccinium and any hybrids were expected to be triploid and sterile. Success was greater than expected, an d from about 1600 pollinations, 31 tetraploid hybrids were selecte d. Unreduced gamete production in V. darrowi and the strong triploid block permitted this hybridization (Sharpe and Darrow, 1959). The V. darrowi clones used in these crosses, Fla. 4A and Fla. 4B, had been collected near Winter Haven in Polk County, Florida. Florida 4B is in the pedigree of nearly all southe rn highbush cultivars. Other distantly related groups have potentia l importance in blueberry breeding. These include Vaccinium species found in western North America, tropical America, Africa and other areas of the world. Potentially useful Vaccinium relatives that are native in Florida include Vaccinium species in sections Polycodium Raf. (the deer berry) and Batodendron Nutt (the sparkleberry), and species in the genus Gaylussacia H.B.K. (the huckleberry) (Sharpe and Sherman, 1971). 18

PAGE 19

The fact that there are large populations of wild Vaccinium plants in Florida and the narrow foundation of V. darrowi genes in the cultivated southe rn highbush gene pool led to the following study of the crossa bility and combining abil ity of various Florida Vaccinium species and the value of these sp ecies in the southern highbush breeding program. 19

PAGE 20

CHAPTER 2 INTER-SPECIFIC CROSSING BEHAVIOR OF VACCINIUM DARROWI CAMP WITH SOUTHERN HIGHBUSH CULTIVARS Introduction Ericaceae is an ancient group, and even the more primitive members have undergone evolutionary modifications (Camp, 1942). The cultivated blueberries belong to the genus Vaccinium L, section Cyanococcus Species in Vaccinium section Cyanococcus are predominantly distributed in eas tern North America. Related Vaccinium species in other sections are widespread in western North America, tropical America, Africa, Asia and other areas of the world. Vaccinium section Cyanococcus includes diploid species (2 n=2x=24), tetraploid species (2n=4x=48) and hexaploid specie s (2n=6x=72) native to North America (Camp, 1945). The main cultivated groups are lowbush (V. angustifolium ), highbush ( V. corymbosum and their hybrids), and rabbiteye ( V. ashei ). Improved cultivars within these species constitute the primary gene pool for breeding cultivars. The secondary gene pool consists of uncultivated species in section Cyanococcus, and the tertiary gene pool is made up of species in other sections of Vaccinium (Lyrene and Ballington, 1986). Since the establishment of blueberry breedi ng programs, numerous attempts have been made to introgress genes from wild relatives in to cultivated forms. Cr osses between homoploid species within section Cyanococcus are usually successful, with no major sterility barriers, and even some heteroploid crosses have been made (Meader and Darrow, 1944; Darrow, et al., 1949; Sharpe, 1953; Sharpe and Darrow, 1959; Rous i, 1966). In 1945, Camp described the native Vaccinium species present in Florida: five diploids, five tetraploids and two hexaploids. Of these, two species, V. myrsinites Lam. (4x), and V. darrowi Camp (2x), are found almost as far south as acid soils are present in Flor ida. These species, along with V. ashei which is common in north 20

PAGE 21

and west Florida, were used as parents to reduce the chilling requirement of northern highbush blueberries (Sharpe, 1953). Northern highbush blueberry cultivars were bred by the USDA using wild plants ( V. corymbosum and V. angustifolium ) collected in New Hampshire and New Jersey. By 1950, several thousand hectares of improved highbush bl ueberries were being cultivated, mostly in Michigan and New Jersey. Sharpe and Darrow (1959) produced the foundation crosses which brought together genes from V. darrowi V. myrsinites V. ashei and northern highbush cult ivars. Northern highbush cultivars hybridized with V. myrsinites produced forty seedlings. These showed little promise when grown in Gainesville, Florida (Sharpe, 1953; Moore, 1965). V. ashei (hexaploid), was expected to give tetraploid plants when crossed with V. darrowi (diploid). Sharpe and Darrow planned to cross tetraploid hybr ids produced in this way with tetraploid northern highbush cultivars. From about 7500 flowers pollinated in V. darrowi V. ashei crosses, 5 hybrids were obtained. Goldy and Lyrene (1983) later made si milar crosses, and the hybrids produced were pentaploid, a result of unreduced gametes from V. darrowi Hybridization between northern highbush and V. darrowi was successful. From 1600 pollinations, 31 fertile selections were made (Sharpe and Darrow, 1959). The production of unreduced gametes in V. darrowi and a strong triploid block in Vaccinium allowed fertile hybrids to be obtained. The V. darrowi clones used in the foundation cr osses that led to southern highbush cultivars, Florida 4A and Florida 4B, were selected from the wild from Winter Haven, Florida in about 1950 (Sharpe and Sherman, 1971). Qu and Hancock (1995) studied US 75, a hybrid generated by crossing V. darrowi Fla. 4B with highbush cultivar Bluecr op. US 75 preserved approximat ely 70% of the heterozygosity 21

PAGE 22

present in Fla. 4B. US 75 constituted one of the most successful tetraploid hybrids used in the development of the southern highbush blueberry cultivars (Qu, et al., 1998) It was the maternal parent of the cultivars Georgiagem, Cape Fear, Cooper, Gulf Coast and Legacy (Multiple authors, 1997). Lyrene (1986) found a morphological discre pancy between Camps description of V. darrowi and plants collected in Highlands County, Flor ida, and in the Ocala National Forest east of Ocala, Florida. Camp (1942) stated that V. darrowi occurred in extensive colonies, 0.15-0.40 m high, from Louisiana to Florid a. He noted that the leaves were coriaceous, evergreen, and somewhat glaucous (especially when young), an d the abaxial leaf su rface was non-glandular. In 1986, Lyrene concluded that V. darrowi plants from the Ocala Forest and from the Istokpoga region of south-central Florida were different from t hose of the Florida panhandle. Most plants from the panhandle had few or no st alked glands on the abaxial leaf surface, and produced colonies 0.34-0.70 m high. Plants from the Ocala Forest region ranged from nonglandular to heavily glandular, with colonies 1.10-1.61 m high. Pl ants from the Istokpoga region were highly variable, ranging fr om short to tall, with varyi ng levels of introgression from V. fuscatum (2x), a tall species resembling V. corymbosum (4x). The Florida short-stat ured panhandle race of V. darrowi inhabits sandy, well-drained, upland soils associated with pine forest that are frequently burned. This form is drought tolerant, and has a large underground rhizome system and sm all, shiny, evergreen leaves. This form of V. darrowi has not previously been used in breeding bl ueberry cultivars. In the Ocala Forest region, V. darrowi is taller, larger-leafed, and is found on moister sites, around the margins of sinkhole lakes or on poorly-drained flatwoods soils, associated with fre quent fires. Two ornamental V. darrowi cultivars, Johnblue and Everblue, whic h were released by North Carolina State 22

PAGE 23

University, belonged to the Istokpog a race, as did Fla. 4A and Fla. 4B, which were extensively used in southern blueberry breeding (Lyrene, 1997). V. darrowi was used in crosses to obtain low ch illing requirement, to improve drought tolerance in highbush blueberry, and to increase to lerance to upland soils low in organic matter. Breeders are working to develop la rge-fruited, evergreen highbush blueberry varieties that can be grown in areas with little or no chilling. In the University of Florida blueberry br eeding program, thousands of flowers were pollinated in an attempt to cross tetraplo id southern highbush cultivars with diploid V. darrowi clones from the Ocala Forest re gion. Relatively few hybrids were obtained due to the triploid block. Some of the F-1 hybrids that were obtaine d had low male and female fertility and were probably triploid. Other F-1 hybrid s were completely fertile, a nd were probably tetraploid. The triploid F-1 hybrids were crossed with hexaploid V. ashei cultivars to produce hexaploid hybrids through unreduced gametes from the triplo id plants. The tetraploid highbush V. darrowi F-1 hybrids had an intermediate phenoty pe between the parents, and were backcrossed to tetraploid southern highbush cultivars. B ackcross-1 seedlings were high ly variable in morphology, but were mostly quite fertile (Lyrene, 1997). The following study was conducted to determine how V. darrowi clones representing three provenances in Florida perform in inter-specifi c crosses with southern blueberry cultivars. Materials and Methods Studies with V. darrowi Races Flowering studies V. darrowi clones were propagated by softwood cutti ngs taken from wild plants selected during April and May 2006 by P. Lyrene and K. Hu mmer. The selected plants were vigorous and healthy and had highly glaucous foliage. Flor ida panhandle clones were selected from Cotton 23

PAGE 24

Landing, Sumatra, Wilma, and Chattahoochee, all in or near the Apalachicola National Forest. The Istokpoga clones were selected from Daven port, near Haines City, and from Highlands County. Softwood cuttings from the selected plants were rooted in a 1:1 mixture of sphagnum peat and perlite. The cuttings were maintained in a greenhouse with retr actable shade cloth and intermittent mist for 2-3 months. Rooted cuttings were transplanted to 4-liter pots containing sphagnum peat and perlite (1:1). In October 2006, plants of eightee n clones were grown in a beeproof greenhouse with five pots per clone, 2 plants per pot, and each plant being an experimental unit. Plants were allowed to flower in the greenhouse. Illumination was sunlight. Photoperiod was not modified and varied with the season. Greenhouse temperatures also varied seasonally, but were maintained between 5C and 30C. Open flowers were counted and removed each seven days from October 2006 to October 2007 (Table 1-1). Date of maximu m flowering and total flower count per plant were measured for clone s from each region. Means for number of flowers per plant per day, total number of flowers per pl ant per region and date of maximum flowering were separated using least squares means by Tukeys test, with significance level of 5%. Inter-Specific Hybridization Experiments Control crosses In January 2007, two-year-old potted V. darrowi clones, two from the Florida panhandle region and four from the Istokpoga region, and twoyear-old plants of fi ve tetraploid highbush clones were selected as parents for intra-specific crosses. The tetraploid highbush plants were cultivars or advanced selections from the Univ ersity of Florida blueberry breeding program and they are here considered V. corymbosum even when other species are present in their pedigree (Muoz and Lyrene, 1984b). In February 2007, the highbush plants were brought into a beeproof greenhouse after having been chilled for more than 1000 hours in a walk-in cooler at 4C with no light. The V. 24

PAGE 25

darrowi clones are evergreen, have no chilling require ment and were not chilled. Open flowers from the plants selected as females were re moved to avoid unplanned pollination. Various types of cross-pollinations were made using thes e plants (Table 1-2 and Table 1-3). For V. darrowi female plant-to-plant variation was eliminat ed as a confounding factor by dividing individual plants into two sections to compare selfand cross-pollination (Morrow, 1943). Self-pollination experiments of V. darrowi clones will be described in chapte r III. Emasculation was performed in all experiments. Pollination was done by touching the stigma with pollen that had been previously collected on the thumbnail. These first crosses were designed to determine the fertility of crosses among different clones within each species. The pollen donors were different clones but the same species as the seed parents. A pproximately 250 flowers were pollinated per plant for V. darrowi and 150 flowers were pollinated per pl ant for southern highbush cultivars. Berries were harvested when fully ripe. The first 20 berries that ripened from each cross were individually opened to examine the seed content. Additional berries were harvested, and seed was removed by shredding the berries in a food blender and removing the pulp and skins from the fruit using water separation (Moore, 1965) Seeds were washed, and dried for storage in coin envelopes at 5C. In November 2007, prior to planting, a sub-sa mple of 0.100 g of seed per cross or the complete seed package per cross (in cases where the seed quantity was low) were classified and counted in two classes: plump seeds or shrivele d seeds. Number of plump seeds per pollinated flower was calculated for each cross. Seeds were treated with fungicide (0.100 g Captan) before planting. The seed was spread on the top layer of 4-liter pots of sphagnu m peat. The pots were maintained in a greenhouse with intermittent mist for 2-3 months to allow seed germination. In 25

PAGE 26

February 2008, the number of seedlings per cr oss were counted. Numb er of seedlings per pollinated flower was calculated for each of the control crosses. 4x 2x crosses In January 2007, twelve tetraploid hi ghbush blueberry cultivars, ten diploid V. darrowi clones, and two diploid natu ral inter-specific hybrids ( V. darrowi V. fuscatum ) were selected as parents for tetraploid diploi d crosses (Table 1-4 and Table 1-5). The highbush cultivars were two-year-old plants potted in sphagnum-perlite (1:1) mixture. They were all cultivars or advanced selections from the southern hi ghbush blueberry breeding program. The diploid V. darrowi clones were selected from the wild during April and May 2006 by P. Lyrene and K. Hummer. Five were from the Florida panhandle race collected from Chattahoochee, Sumatra, Apalachicola, and Wilma. The other five were from the Istokpoga race collected from the Archbold Biological Station, from the area near Lake Istokpoga ( both in Highlands County), and from Davenport near Haines City (Polk C ounty). The two natural inter-specific hybrids ( V. darrowi V. fuscatum ) were collected from Davenport. In 2006, softwood cuttings from the diploid plants selected from the wild were rooted in a 1:1 mixt ure of sphagnum peat and perlite. The cuttings were kept for 2-3 months in a greenhouse with intermittent mist and shade cloth cover. The rooted cuttings were transferred to 4-liter pots of sphagnum peat and perlite (1:1), and grown until the plants flowered. In February 2007, all the plan ts were brought into a beeproof greenhouse after receiving adequate chilling. Southern highbus h cultivars were previously maintained in a walk-in cooler with no light at 4C for two months. V. darrowi clones are evergreen and were not chilled. Open flowers from the plants selected as females (tetraploid southern highbush cultivars) were removed to avoid unintended pollinations. Female plants of similar age, plant structure, size and growing conditions were used in the experiment (Morrow, 1943). Pollination was done by 26

PAGE 27

touching the stigmas of previously emasculated flowers with pollen previously collected on the thumbnail. Female and male plant cross-combinat ions were randomly assigned. Approximately 500 flowers were pollinated per plant. Berries were harvested when fully ripe. The first 20 berries that ripened were individually opened to determine seed content. Seeds from additional berries were removed using a food blender, after which they were dried at room temperature on a desk top and stored in coin envelopes at 5C. In November 2007, seed packages were tr eated with fungicide (Captan, 0.100 g per package) prior to planting. Subsamples or comple te seed packages, depending on the quantity of seeds produced from the cross, were counted and classified into two categories: plump seeds or shriveled seeds. The number of plump seeds per pollinated flower was cal culated for each cross. Seeds were planted on the top la yer of 4-liter pots of sphagnum peat. The pots were maintained in a greenhouse with intermittent mist for 2-3 months, until most seeds had germinated. The number of seedlings per pollinated flower for each cross was calculated. In February 2008, when the seedlings were 1-2 cm high, they were tr ansferred to plastic trays, 100 seedlings each, filled with sphagnum p eat. All the seedlings from each cross were transferred to trays. In May 2008, when the seedli ngs were 4-8 cm high, they were planted in a high density nursery in Citra, Florida, with plant-to-plant spacing of 10 cm and row-to-row spacing of 40 cm. The high density nursery was maintained with overhead irrigation and fertilization when required. After 60 days in the high density nurse ry, the seedlings were counted and classified into two categories: hybrids or selfs. The plants were separated visually by noting whether they had hybrid morphology or resembled the female parent. Because highbush blueberry and V. darrowi differ so greatly in morphology, the hybrids could easily be 27

PAGE 28

distinguished from the selfs. The number of hyb rids per pollinated flower for each cross was obtained. In February 2008, two-year-old plants of fift een tetraploid southern highbush cultivars were used as female parents in crosses with twelve diploid V. darrowi clones and three diploid V. fuscatum clones selected by P. Lyrene and K. Hummer in 2006 (Table 1-6 and Table 1-7, respectively). Of the V. darrowi clones, three belonged to the Istokpoga race and nine to the Florida panhandle race. Pollination, harvesting and seed extraction were the same as previously stated. Fruit percentage and nu mber of plump seeds per pollinat ed flower were determined. 2x 4x crosses Several experiments had previously been ma de by P. Lyrene in the southern highbush blueberry breeding program at Univer sity of Florida involving diploid V. darrowi as female parent, and tetraploid southern highbush cultivars as male parent In one experiment made in 2001, two-year-old plants of fifteen diploid V. darrowi clones were crossed as females with tetraploid southern highbush cultivars (Table 1-8). These V. darrowi clones had been collected in 1999 and belonged to the Ocala National Forest race. They were selected because they had high vigor, stout canes, and tall shoots (u p to 2.5 m), but fully expressed the V. darrowi characteristics of small, shiny, highly-evergreen leaves (Lyr ene personal communication) Fifteen tetraploid southern highbush cultivars from the southe rn highbush blueberry breeding program at University of Florida were used as the male parents. Both V. darrowi and the highbush cultivars were grown in 4-liter pots of sphagnum peat. In February 2001, the plants were brought in to a bee-proof greenhouse after chilling was satisfied. Southern highbush plants were previously kept for 2 months in a walk-in cooler with no light at 5C. The V. darrowi plants are evergreen and were not chilled. Pollination was done by touching the stigmas of V. darrowi flowers with pollen from the highbush parent that had 28

PAGE 29

been previously collected on the thumbnail. Flow ers were emasculated to avoid contamination. Approximately 300 flowers per plan t were pollinated. Berries were harvested when fully ripe, and seed was extracted using a food blender. Seed s were washed free of pulp and skins, dried on a table top, and stored in coin envelopes at 5C. The seed was treated with fungicide (Captan, 0.100 g per package) prior planting. The seeds were planted in November 2001 in a cool greenhouse under conditions known be favorable for germination of Vaccinium seed. Seedlings from the V. darrowi V. corymbosum crosses were carefully examined when they were two months old in order to eliminate pl ants that were not inter-specific hybrids. The hybrids had morphological characters in termediate between the parents. In 2005, twelve diploid V. darrowi clones were selected as female parents for crosses with twelve tetraploid southern hi ghbush cultivars (Table 1-9). The V. darrowi clones were collected in 2003 around Lake Istokpoga and at the Arc hbold Biological Station. They were chosen because of the light-blue color (high glaucescence) of their leaves and fruit. Two-year-old plants of twelve southern highbush cultivars were sel ected as male parents. They were advanced selections from the University of Florida southern highbush bluebe rry breeding program. Southern highbush cultivars were maintained in a walk-in cooler with no light for approximately 1000 hours at 5C to fulfill their chilling requirement. V. darrowi clones are evergreen and were not chilled. Plants were brought into the greenhouse to bloom in February 2005. Pollination and harvest procedures were similar to those descri bed before. Seeds were planted in November 2005 on the top layer of 4-liter pots of sphagnum peat. Pots were maintained with intermittent mist for 2-3 months in a greenhouse in Gainesville, Florid a. Seedlings were transplanted to plastic trays of peat, in February 2006, when they were 1-2 cm high. Seedlings were cat egorized as self or 29

PAGE 30

hybrids before being transplanted to a high-density nursery at Univ ersity of Florida Plant Science Unit in Citra, Florida, in June 2006. In February 2006, nine V. darrowi clones were selected as female parents for crosses with southern highbush cultivars (Table 1-10). Two we re from the Florida panhandle race and seven from the Istokpoga race. They had been select ed from the wild in 2003, and some had been previously used in crosses in 2005. The male pa rents were nine tetrap loid southern highbush cultivars and advanced selec tions selected from the blue berry breeding program. Crosscombinations between parents were randomly a ssigned. Pollination, harvesting, seed extraction and seed planting were as desc ribed before. Seedlings were cl assified by plant morphology as hybrids or selfs. Seedlings were transplanted to plastic trays in February 2007 and grown in a greenhouse in Gainesville, Florida. These plants were re-transplan ted to a field nursery at the University of Florida Plant Science Unit in Citra, Florida, in June 2007. In January 2007, ten diploid V. darrowi and two diploid natural inter-specific hybrids ( V. darrowi V. fuscatum ) were selected as female parents for crosses with twelve tetraploid southern highbush cultivars (Table 1-11 and Table 1-5, respectively). The V. darrowi clones and the F-1 natural hybrids were thos e collected in 2006 by P. Lyrene and K. Hummer as described earlier. A few of the V. darrowi clones had been previously used for crosses in 2005. Of the V. darrowi clones, three were from the Florida panha ndle race and seven from the Istokpoga race. The plants were grown in a greenhouse through the summer and fall, and were not defoliated or chilled before the crosses were made. Pollina tion, berry harvest, seed extraction and seed planting were as described above. In February 2008, fifteen diploid V. darrowi clones and three diploid V. fuscatum clones selected by P. Lyrene and K. Hummer in 2006, were used as fe male parents in crosses with 30

PAGE 31

eighteen tetraploid southern highbush cultiv ars (Table 1-12 and Table 1-7). Of the V. darrowi clones, six were from the Istokpoga race and nine from the Florida panhandle race. Pollination, berry harvesting and seed extraction were as prev iously stated. Fruit percentage and number of plump seeds per pollinated flower were determined. Statistical analysis Fruit set percentage, number of plump seeds per pollinated flow er, number of seedlings per pollinated flower and number of hybrids per pollina ted flower were used to measure the results of the pollination treatments. Means for number of plump seeds per pollinated flower, number of seedlings per pollinated flower and number of hybrids per pol linated flower for different treatments were separated using least squares m eans by Tukeys test, with significance level 5%. Means for fruit set percentage were separated using Chi-square test of independence, with significance level 5%. Data analysis was ANOVA by the PROC GLM and PROC FREQ procedures of SAS (Statistical Analysis Sy stem Version 9.1, SAS Institute, Cary, NC). Morphological Studies of F-1 Hybrids ( V. darrowi Southern Highbush Cultivars) In April 2008, two-year-old plants of eleven tetraploid southern highbush cultivars, threeyear-old plants of twelve diploid V. darrowi and two-year-old plants of ten F-1 hybrids ( V. darrowi southern highbush blueberry) were characterized for va rious morphological characteristics of the leaves flowers and berries. The V. darrowi plants were those collected in 2006 by P. Lyrene and K. Hummer and were rando mly selected from the Florida panhandle and Istokpoga clones available. The southern highbush pl ants were cultivars or advanced selections from the southern highbush blueberry breeding program, and were those previously used as parents. The F-1 hybrids were selected from the population produced in 2006. The plants were dug in December 2007 from a field nursery at the University of Florida Plant Science Unit in Citra, Florida, where they had been growing since April 2007. They were used as parents for 31

PAGE 32

crosses in 2008. The V. darrowi clones, southern highbush pl ants, and F-1 hybrids were maintained in a greenhouse in 8-liter pots of a 1: 1 mixture of sphagnum peat and perlite. Leaves and flowers were collected from the selected plan ts from the greenhouse. Fi ve leaves and flowers per clone were measured. Leaf and flower characteristics were measured using a Traceable Carbon Fiber Caliper from Fish er Scientific, Pittsburg, PA. Leaf characteristics measured in the three taxa were length, width, and pubescence on the lower leaf surface. Pubescence was scored as non e, low, medium or high, and marginal glands on the leaves were scored as none, sunken, exsert ed or both. Flower characteristics measured were corolla length, corolla wi dth, corolla aperture, pedicel length, peduncle length, bracteole length, bracteole width, presence of anther awns and presence of persistent bracteoles. Berries were harvested for weighing in June 2008 when fully ripe from V. darrowi plants, southern highbush advanced selections and F-1 hybrids (V. darrowi V. corymbosum ). The berries that were weighed came from open-pollinated plants in a fi eld nursery at the University of Florida Plant Science Unit in Citra, Florida. Berries were harvested at random. Berries from seventy plants, two berries per plan t, were weighed for F-1 hybrids ( V. darrowi southern highbush cultivars). Berries from sixty-seven plants two berries per plant, were weighed for the tetraploid southern highbush cultiv ars, and berries from eight plan ts, two berries per plant, were weighed for V. darrowi Means for leaf, flower and berry ch aracters were separated using least squares means by Tukeys test, with significan ce level 5%. Data were subjected to ANOVA by the PROC GLM procedure of SAS (Statistical Analysis System Version 9.1, SAS Institute, Cary, NC). 32

PAGE 33

Fertility Studies of F-1 ( V. darrowi Southern Highbush Cultivars) Hybrids Pollen stainability An estimate of potential male sterility in plants can be obtained by looking for morphologically abnormal pollen gr ains. Aborted pollen grains, wh en moistened with water or aceto-carmine stain, will be shriveled, show a bnormal shapes, fail to stain, or have varying degrees of pollen inflation (Dermen, 1940). Pollen morphological aberrations can result from chromosome pairing abnormalities during meiosis. Goldy and Lyrene (1983) described how several meiotic abnormalities in interspecific hybrids between V. ashei (6x) and V. darrowi (2x) produced a high percentage of shrunken pollen. Lyrene and Sherman (1983) studied inter-specific hybrids between V. corymbosum (4x) and V. elliottii (2x), and stated that both mitotic and meiotic instability in wide hybrids would reduce their fertility. The plants obtained from V. darrowi V. corymbosum crosses in 2006 were classified as hybrids or selfs based on morphologi cal characters in comparison w ith their parents. The plants were grown in a field nursery at the University of Florida Plant Science Unit in Citra, Florida. The classification was done in January 2008 when the plants were one year-old by comparing their morphological char acteristics with the V. darrowi clones and southern highbush selections growing next to them. Pollen fertility of 109 F-1 hybrids was studied to predict whether they would cross with tetraploid southern highbush cult ivars. In addition, eleven tetr aploid highbush cultivars, eight V. darrowi clones from the Florida panhandle race, fifteen V. darrowi clones from the Istokpoga race, and three V. arboreum clones collected from the woods in northeast Florida were studied to determine their pollen fertilit y. For each plant, pollen from two flowers was spread on a microscope slide. Pollen was stained using 1% a ceto-carmine. Slides were analyzed using a light 33

PAGE 34

phase-contrast Leitz microscope, 250 magnification. Pollen fertility has been used to separate triploids from tetraploids following inter-specific crosses between diploid and tetraploid species. Triploid plants produce pollen that is mostly a borted. Pollen from tetraploids is usually well developed. Pollen stainability was measured as percentage of well-stai ned pollen grains in normal tetrads, triads, dyads and monads compar ed with the numbers of sporads that had only one, two, or three stained pollen grains (each spor ad normally has four pollen grains). This result was averaged by the total of counted sporads. Backcrossing experiments In February 2007, two F-1 ( V. darrowi V. corymbosum ) hybrids from crosses made in 2005, and three F-1 hybrids from crosses made in 2006, were backcrossed with southern highbush cultivars. In addition, in February 2008, one-year-old plants of seventeen F-1 ( V. darrowi V. corymbosum ) hybrids, were backcrossed to sout hern highbush cultivars. The F-1 hybrid parents were selected from larger populati ons of F-1 hybrids growing in a field nursery at the University of Florida Plant Science Unit in Citra, Florida. The clones selected were those whose leaves were most glaucous (as opposed to green), had highest vigor, and had berries with the lightest-blue color (when pr esent). Pollination, berry harvesting, seed extraction, and seed planting were as previously described. Fruit set percentage, number of plump seeds per pollinated flow er and number of seedlings per pollinated flower were calculated for crosse s made in 2007. For crosses made in 2008, fruit set and number of plump seeds per pollinated flower were determined. Means for fruit set percentage were separated using Chi-square test of independenc e, with significance level 5%. Means for number of plump seeds per pollinated flower and number of seedlings per pollinated flower for the crossing experiments were analyzed using Tukeys test, w ith significance level 34

PAGE 35

5%. Data were subjected to ANOVA by the PROC GLM and PROC FREQ procedures of SAS (Statistical Analysis System Ve rsion 9.1, SAS Institute, Cary, NC). Results and Discussion Studies with V. darrowi Races Flowering data This study was done in a greenhouse under natu ral day-length and temperatures that averaged 10C night/25C. The temperatures ch anged with the seasons (Gainesville, FL). The flowering season for the Florida panhandle race of V. darrowi extended from November 2006 to July 2007. The flowering season for the Istokpoga race extended from December 2006 to May 2007 (Figure 1-1). The two races had similar flow ering patterns, with ma ximum flowering from February 21 st to March 24 th 2007. During this period, the Florida panhandle race and the Istokpoga race opened an average of 12.8 and 20.9 fl owers per day per plant, respectively. The Istokpoga clones produced almost twice as many flowers per plant as the Florida panhandle clones (Table 1-1). V. darrowi and V. corymbosum flower bud initiation is pho toperiod sensitive and is promoted by short days, while flower bud break is promoted by long days. In addition, temperature can have an important ro le in flower induction (Spann, et al ., 2003). Several factors influence flower induction and pr oduction: genetic factors, envi ronmental factors (temperature, photoperiod, and their interaction), plant size, species, and plant structure (Spann, et al 2003; Arora, et al., 2003; Ba ptista et al., 2006; Ba ados and Strik, 2006). The average date of 50% open flowers for field grown plants of southern highbush cultivars Emerald, Star, Jewel, and Windsor was February 3 rd in 2007 at Windsor, FL. In 2008, V. darrowi V. corymbosum inter-specific F1 hybrids had an average date of 50% open flowers on January 11 th in the field at Citra, FL., which is near (30 km) Windsor. Thus, hybridization 35

PAGE 36

between V. darrowi and southern highbush cultivars is one way of producing clones that flower very early. Inter-Specific Hybridization Data In V. corymbosum V. corymbosum and in V. darrowi V. darrowi crosses made in 2007, fruit set averaged 78.06% for V. corymbosum and 74.28% for V. darrowi (Tables 1-2 and 1-3). Variation for fruit set percentage in these crosse s was attributed to female-male interactions and plant fertility. In V. corymbosum (4x) V. darrowi (2x) crosses made in 2007, fruit set ranged from 0.90% to 44.01%, (Table 1-4). The reduced fruit set in these crosses was attributed to the difference in ploidy levels between the pare nts. Three crosses, 06-446-419-I, 07-92-418I, and 95-50-423-P, gave su rprisingly high numbers of plum p seeds per pollinated flower (PPF), numbers of seedlings per pollinated flow er (SPF) and numbers of hybrids per pollinated flower (HPF) compared with the mean of the seven other V. corymbosum (4x) V. darrowi (2x) crosses made in 2007 (P <0.0001). Unusually high levels of 2n gamete production in the diploid V. darrowi parents are the most likely reason for th e high success rate in these crosses. Fruit set in 2008 V. corymbosum (4x) V. darrowi (2x) crosses ranged from, 0.27% to 46.04% (Table 1-6). PPF varied greatly among crosses. Seeds from the 2008 crosses will be planted in November 2008. Unlike the situation in 2007, no V. darrowi clones produced a PPF>1.000 when crossed with highbush. PPF is not a very reliable pr edictor of hybrid production and does not accurately identify V. darrowi unreduced gamete producers, because some of the seeds that are classified as plump do not germinate and some seeds that are classified as shriveled can germinate. Results of V. darrowi (2x) V. corymbosum (4x) crosses made in 2001, 2005, 2006, 2007 and 2008 are described in Table 1-8, Table 1-9, Table 1-10, Table 1-11 and Table 1-12, 36

PAGE 37

respectively. SPF and HPF were variable among crosses. Some crosses in 2005 produced numerous seedless berries (Table 1-9). Most of the seedlings obtained from that years crossing efforts were not hybrids. From 4739 flowers of the Istokpoga race of V. darrowi that were pollinated in 2005, only 3 inter-spe cific hybrids were obtained (HPF 0.001). This compares with 133 hybrids obtained by pollinating 7340 V. darrowi flowers of the Ocala race in 2001 (HPF 0.021) (Table 1-8). In neither 2001 nor 2005 did any cross produce more than 0.100 hybrids per pollinated flower (Table 1-8 and Table 1-9). In 2006 crosses, fruit set varied from 0% to 93.4%. One V. darrowi clone from the Florida panhandle race (FL 03-421-P) crossed with a composite of pollen from highbush cultivars, gave unusually high values for PPF, SPF and HPF (140 F-1 hybrids were obtained by pollinating 1406 flowers), compared to the eight other crosses made in 2006 (3 F-1 hybrids were obtained from 2498 po llinated flowers from all other crosses combined) (Table 1-10). The higher production of hybrids (higher HPF) was attributed to high 2n egg production by FL03-421-P. In 2007, three V. darrowi clones: FL03-421-P (previously described), FL03-422-P and FL06-660-I, had on average a higher PPF, SPF and HPF than the additional seven V. darrowi clones used in V. darrowi V. corymbosum crosses in 2007 (P <0.05) (Table 1-11). No associ ation between higher fruit set and higher HPF was found in 2006 and 2007. For crosses made in 2006, PPF increased wh en fruit set increased. Some of the plump seeds did not germinate. Crosses made in 2007 confirmed the resu lts of 2006. FL03-421-P produced a high 2n egg frequency when crossed with southern highbush cultivars. Crossing experiments were a laborious method of screening for 2n gamete fre quency, but gave positive and repeatable results. Bretagnolle and Thompson (1995) counted the 4x seeds or seedlings from 2x-4x or 4x-2x crosses to detect 2n pollen and/or 2n egg production in potato and other autopolyploid plants and to 37

PAGE 38

estimate 2n gamete frequency in individual plan ts. Megalos and Ballington (1988) stated that test-crossing diploids and tetraploids may not be an accurate way to determine the frequency of 2n egg production in Vaccinium Stelly and Peloquin (1983) found diploid tetraploid crosses were a labor-intensive way to test for 2n gamete production in Solanum and did not always give reliable conclusions. PPF of 2008 crosses of V. darrowi V. corymbosum (Table 1-12) and their reciprocals (Table 1-6) was not a reliable pr edictor of the hybrid production. Eight V. darrowi clones had PPF values higher than 1.000 in crosses made in 2008, but only one of them, FL03-422-P, produced high HPF in crosses made in 2007. For crosses made in 2007, HPF for 4x-2x and 2x4x crosses were not significantly di fferent (P<0.05) (Table 1-14). Overall comparisons In 2007, six V. darrowi clones were used both as pollen a nd seed parents in crosses with tetraploid highbush cultivars (Table 1-13). One V. darrowi clone (FL03-421-P) gave far more hybrid seedlings when used as the seed parent, bu t three others gave more hybrids when used as pollen parents. V. darrowi FL03-421-P gave 40.8 hybrids per 100 pollinated flowers when used as seed parent but only 5.9 when used as pollen parent (P<0.0001). FL03-419-I gave only 4.5 hybrids per 100 pollinated flower when used as a seed parent compared to 34.2 when used as pollen parent. Two other V. darrowi clones, FL03-418-I and FL03-423-P, also produced far hybrids if used as pollen parent s (P<0.01 in each case). When V. darrowi clones were compared, a clones success as a pollen parent was not correlated with its su ccess as a seed parent. Because most of the hybrids from our crossing experiments ha ve been fertile (and presumably tetraploid), hybrid production rate from these crosses probabl y indicated the rate of 2n gamete production in the V. darrowi parent. 38

PAGE 39

Megalos and Ballington (1988) found that frequency of 2n gametes varied among Vaccinium species and among clones within species. They also found that the frequency of 2n gamete production differed for megaspores and microspores in the same plant. Two diploid natural hybrids of V. darrowi V. fuscatum found in moist woods near Davenport, Florida, were propagated by cuttings. In 2007, they were used in reciprocal crosses with tetraploid highbush cultivars (Table 1-5). Fruit set percentage was 15x higher when the natural hybrids were used as the se ed parent compared to the recipr ocals, but the berries had very few seeds (P<0.05) (Table 1-14). Lyrene (personal communication of unpublished data) obtained no hybrid seedlings after pollinating approximately 1000 flowers of tetraploid cultivars with pollen from two diploid V. fuscatum plants from eastern Alac hua County, Florida. In 2008 V. corymbosum (4x) V. fuscatum (2x) and their recipr ocals were not statistically different from each other for fruit set percentage or PPF (prelim inary results). The seeds will be planted in November 2008 (Table 1-15). As expected, the intra-specific homoploid crosses made in 2007 ( V. darrowi V. darrowi and V. corymbosum V. corymbosum ) had a higher fruit set, 74.28% and 78.06 respectively, than the tetraploid diploid crosses (P < 0.05, Table 1-14). The homoploid crosses had higher PPF and SPF than either the 4x-2x or the 2x-4x crosses. Although the means differed substantially, they were not signi ficantly different at P=0.05, due to the low number of replicates and high coefficient of variation among crosses. Brooks and Lyrene (1998a) reported a similar result for crosses between V. darrowi and V. arboreum Crosses between homoploid species and even some heteroploid crosses within section Cyanococcus are, in general, successful, with no majo r sterility barriers (Meader and Darrow, 1944; Darrow, et al., 1949; Sharpe, 1953; Sharpe and Darrow, 1959; Rousi, 1966). 39

PAGE 40

Screening for diploid V. darrowi clones with higher unreduced gamete production disclosed several clones that were high 2n gamete producers. These clones allowed us to produce a relatively large number of F-1 hybrids from 4x2x and 2x-4x crosses. Selection pressure for desired traits and plant morphology can be in creased when the size of the population is increased. Morphological Studies of F-1 Hybrids ( V. darrowi Southern Highbush Cultivars) Leaf characteristics The F-1 hybrids studied in th is report were from the V. darrowi (2x) V. corymbosum (4x) crosses made in 2006. For leaf length and leaf width, the F-1 hybrids had averages that were between those of their parents, V. darrowi (small) and V. corymbosum (large) (P<0.05) (Table 116, Figure 1-2). Leaf, flower a nd berry morphological differences permitted Brooks and Lyrene (1998a) to identify hybrid plants in V. darrowi V. arboreum crosses. Expressions of parental and intermediate character levels in hybrids depend on the type of genetic control of the character and inte ractions with the envir onment (Rieseberg, 1995). For the three taxa, V. corymbosum V. darrowi and F-1 hybrids, pubescence was not usually present on the lower leaf surface. No southern highbush clones and only 8% of the V. darrowi clones had leaf pubescence. Leaf pubescen ce was present in 62% of the F-1 hybrids (Table 1-16). Brooks and Lyrene (1998a) descri bed the presence of leaf pubescence on 75% of the V. darrowi clones they studied 100% of the V. arboreum clones, 100% of the F-1 ( V. darrowi V. arboreum ) hybrids, and 95% of the seedli ngs from open-pollinating F-1 ( V. arboreum V. darrowi ) plants. However, none of the sout hern highbush cultivars they studied had pubescence. Presence of leaf pubescence is highly variable between and within Vaccinium species. 40

PAGE 41

Marginal glands were present for all three ta xa (Table 1-16). Nine ty-six percent of the southern highbush cultivars had sunken marginal glands. For the V. darrowi clones, 42% of the plants had sunken and 50% had exerted marginal glands. For the F-1 hybrids, 54% had sunken and 14% exserted marginal glands. This increas e in sunken marginal glands in the F-1 hybrids was attributed to southern highbush introgression. Brooks and Lyrene (1998a) reported the presence of marginal glands in V. darrowi and southern highbush. They found that 95% of the V. darrowi plants and 20% of the southern highbush cultivars had sunken marginal glands. Leaf pubescence and presence of marginal glands were not distinctiv e for the three taxa. Hybrid field identification through these two char acteristics was not possible. The identification of the F-1 hybrids by leaf length and leaf wi dth was possible and e fficient (Figure 1-3). Flower characteristics Corolla length, corolla widt h, corolla aperture peduncle length, bracteole length and bracteole width of the F-1 (V. darrowi V. corymbosum ) hybrids were intermediate between V. darrowi clones (small) and southern highbush cultivar s (large) (P<0.0001) (Table 1-16, Figure 14). Mean pedicel length of the F-1 hybrids was not significantly different from the means for V. darrowi and southern highbush cultivars. The m ean values of pedicel length for V. darrowi and V. corymbosum differed significantly (P<0.05). Brooks and Lyrene (1998a) found that corolla length and corolla width of the F-1 V. darrowi V. arboreum hybrids they studied were intermediate between the values of their parents. In addition, the F-1 hybrids had a longer corollas and bracteoles than either V. darrowi or V. arboreum All flowers of V. darrowi southern highbush cultivars and F-1 hybrids lacked anthers awns. Persistent bracteoles were present in 100% of the flowers for the three taxa. Brooks and Lyrene (1998a), found that none of the V. darrowi plants and southern highbush cultivars had awns and that 93% of the F-1 ( V. darrowi V. arboreum ) hybrids had awns. Presence of anther 41

PAGE 42

awns and persistent bracteoles was not distinctive for the three ta xa in our study. Field identification using these two flower characterist ics was time consuming and not distinctive. Field identification of F-1 ( V. darrowi V. corymbosum ) hybrids using corolla length, corolla width, corolla aperture, peduncle le ngth, bracteole length a nd bracteole width was distinctive, clear and efficient. Clear separati on for corolla width and corolla length of the F-1 hybrids, V. darrowi clones and southern highbush cultivars can be seen from the individual plant means for the three taxa (Figure 1-5). Berry characteristics For berry weight, the F-1 ( V. darrowi V. corymbosum ) hybrids were between their parents (P<0.0001) (Table 1-16, Figure 1-4). Th e F-1 hybrids had a mean weight of 0.81 g, less than half that of the southern highbush cultiv ars (mean weight 2.25 g), and much closer to the V. darrowi parent than to the highbush parent. Succe ssive backcrosses to southern highbush cultivars and strong selec tion will be required to return to th e large berries of the cultivars. Following the first hybridizations between V. darrowi and northern highbush cultivars, elimination of undesired characteristics from V. darrowi (i.e. short, twiggy plant structure, small fruit, long fruit-development period) in the sout hern highbush genetic pool was accomplished by multiple generations of inter-crosses, backcrosses, and selection (Sharpe and Sherman, 1971). Plant architecture The plant architecture of the hybrids was intermediate betwee n their parents (Figure 1-6). Florida V. darrowi was classified by morphological characte rs by Lyrene (1986) into three races: (1) Florida panhandle race, with colonies 0.34-0.70 m high, (2) Ocala Forest race, with colonies 1.10-1.61 m high, and (3) Istokpoga race, in which pl ants were highly variable, ranging from short to tall, presumably due to va rying levels of in trogression from V. fuscatum (2x), a tall species resembling V. corymbosum (4x). V. darrowi plants used in our study were from the 42

PAGE 43

Florida panhandle race. To obtain height m easurements, southern highbush cultivars (V. corymbosum ) plants were grown for two years in a highdensity nursery at University of Florida Plant Science Unit in Citra, Florida. After two years, the plants were 1.5-2.0 m tall. Southern highbush plants without pruning can eventually reach a height up to 4 m. The F-1 ( V. darrowi V. corymbosum ) hybrids after two years in the field were 0.9-1.3 m high. The F-1 hybrids were somewhat branchy and twiggy like their V. darrowi parents. This phenotype was also identified in F-1 V. darrowi V. arboreum hybrids (Brooks and Lyrene, 1998a). In conclusion, field identification of F-1 ( V. darrowi V. corymbosum ) hybrids using plant architecture and the morphological characters pr eviously described is possible and efficient. Fertility Studies of F-1 ( V. darrowi V. corymbosum ) Hybrids Pollen stainability data The F-1 ( V. darrowi V. corymbosum ) hybrids in this study were produced in crosses made in 2006. Pollen staining studies of the F1 hybrid population showed plants with low and high stainability (Figure 1-7). Pollen stainability 0-50% was found in six plants of the F-1 hybrids (5.5%). These plants had shriveled pollen, pollen with abnormal shapes, and pollen that failed to stain with 1% aceto-carmine and had di fferent degrees of pollen inflation as described by Dermen (1940) (Figure 1-8a). Pollen staina bility 50.1-100% was found in 103 plants (94.9%) of the F-1 hybrid population (Figure 1-8b). For F1 hybrids produced in 4x-2x and 2x-4x crosses, low pollen stainability can indicate chromosome pairing abnormalities during meiosis, which can be the result of triploidy or aneuploidy. High pollen stainabili ty implies normal chromosome pairing during meiosis and a te traploid plant (Goldy and Lyre ne, 1983; Lyrene and Sherman, 1983). In our populations, F-1 hybrids with high polle n stainability are thought to be tetraploid, and those with lower pollen stainability may be triploid or aneuploid. Th e tetraploid F-1 hybrids should be easy to backcross to the tetraploid southern highbu sh gene pool. The triploid F-1 43

PAGE 44

hybrids can be backcrossed to hexaploid V. ashei through unreduced gamete production of the triploid plants. In this way, genes from V. darrowi and V. corymbosum can be moved to the V. ashei gene pool (Ehlenfeldt and Vorsa, 1993). Th e number of hexaploid hybrids from such crosses is expected to be low, and limited by th e rate of 2n gamete production in the triploid parent. Mean pollen stainability (%) for panhandle race V. darrowi clones, Istokpoga race, southern highbush cultivars, V. arboreum clones, and F-1 ( V. darrowi V. corymbosum ) hybrids, was not significantly di fferent (Table 1-17). Male fert ility for 109 F-1 hybrid plants averaged high, with mean pollen staining of 88.6%. Mean pollen staining for the five least fertile of these 109 F-1 hybrid plants was 26.48%. Backcross data When F-1 hybrids ( V. darrowi V. corymbosum ) were pollinated with highbush pollen (Table 1-18), fruit set percentage was not signifi cantly different from that measured in highbush highbush crosses (Table 1-2). Number of plump seeds per polli nated flower and number of hybrids per pollinated flower were not signi ficantly different in 2007 or 2008 for F-1 ( V. darrowi V. corymbosum ) hybrids pollinated with highbush pollen co mpared to intra-specific crosses of southern highbush. Even though the F-1 hybrids were hard to produce, once obtained, they were easy to backcross to southern highbush. This agrees with the observations of Sharpe and Sherman (1971), who found their first F-1 hybrids between V. darrowi and northern highbush cultivars easy to b ackcross to highbush. Conclusions The Florida panhandle race a nd the Istokpoga race of V. darrowi have a similar period of maximum flowering. The Istokpoga race produced twice as many flowers as the Florida 44

PAGE 45

panhandle race. Phenotypic variation is present in V. darrowi races and needs to be studied in more detail at the genetic level. Intra-specific crosses (tetraploid V. corymbosum tetraploid V. corymbosum and diploid V. darrowi diploid V. darrowi ) had high fruit set and produced numerous plump seeds and seedlings per pollinated flow er. Crosses between diploid V. darrowi tetraploid V. corymbosum and their reciprocals had low fruit set compared to the intra-specific, homoploid crosses. High number of hybrids per pollinated flower in cert ain 4x-2x crosses was attributed to high 2n pollen production of the V. darrowi clone used in those crosses. V. darrowi clones, FL03-419-I, FL03418I, and FL03-423-P were high 2n pollen producers when crossed with southern highbush cultivars (HPF >0.100). High number of hybrids per pollinated flow er in certain 2x-4x crosses was attributed to high 2n egg production of the V. darrowi clone used in those crosses. V. darrowi clones, FL03-421-P, FL03-422-P, and FL06660-I were high 2n egg producers when crossed with southern highbush cultivars (HPF>0.100). No association was found between 2n pollen and 2n egg production for V. darrowi clones. Variation in frequency of unreduced gamete production in diploid V. darrowi was present within plants (megaspores vs. microspores) and among plants within races. Number of plump seeds per pollinated flow er was not a useful method for selecting V. darrowi unreduced gamete producers be cause of variable rates of germination and because not al l the seedlings were hybrids. FL03-421-P, when used as female parent in cr osses with southern highbush cultivars, gave similar results in 2006 and 2007, confir ming the high 2n egg production of this V. darrowi clone. Test-crossing was a laborious technique for finding V. darrowi 2n gamete producers, but they gave consistent and positive results. HPF was significantly different between 4x-2x and 2x-4x 45

PAGE 46

crosses for V. darrowi 2n gamete producers, FL03-421-P, FL03-419-I, FL03-418-I, and FL03423-P (P<0.01). Overall, diploid V. darrowi diploid V. fuscatum and diploid F-1 ( V. darrowi V. fuscatum natural hybrids), when crossed with sout hern highbush cultivars, were equally productive of hybrids whether used as males or females. Identification of F-1 ( V. darrowi V. corymbosum and reciprocals) hybrids in the field through leaf, flower and berry characteristics wa s accurate and efficient. Southern highbush and V. darrowi structures (leaf, flower a nd berry) were larger and sma ller, respectively, than F-1 hybrids structures. Mean pollen stainability (%) of the F-1 hybrid s was similar to that of their parents. Variation was found for pollen stainability in the F-1 population. Lower pollen stainability may indicate that a few plants were triploid or aneuploid. Highly fe rtile hybrids indicate tetraploid plants with normal chromosome pairing. Fruit set, PPF, and SPF for the F-1 hybrids, when pollinated with pollen from southern highbush cultivars, were not significantly different from southern highbush cultivars pollinated with southern highbush pollen (P >0.05). This indicated that most of the F-1 hybrids were tetraploid. It should be possible to obtain plants of cultivar quality in a few generations of backcrosses starting with F-1 hybrids V. darrowi southern highbush cultivar crosses. 46

PAGE 47

Table 1-1. Mean number of flowers that opened per plant per day on V. darrowi plants of two Florida races as measured in a green house over 52 weeks, in Gainesville, Fl. Race Flowers per day per plant Total flowers per plant Date of maximum flowering z Florida panhandle 12.8 a* 338.1 b 83(24-Mar-07) a Istokpoga 20.9 a 700.5 a 52(21-Feb-07) a zDay maximum flowering in Julian days (date).*Similar letters within a column indicates means not significantly different, Tukeys test, =0.05. Table 1-2. Result of intra-speci fic hybridization of southe rn highbush cultivars in 2007. Female Male Flowers (No.) Berries (No.) Fruit set (%) PPF y SPFx 07-92-Hz 01-15-H 144 129 89.58 8.555 43.493 06-446-H 00-202-H 130 10278.468.70917.105 07-108-H 05-08-H 1 24 46 37.10 3.079 06-468-H Emerald-H 1 22 115 94.26 11.500 02-16-H 00-75-H 132 12090.917.72711.864 6.232 8.908 zH = Tetraploid V.corymbosum southern highbush cultivars. The first column gives the seed parent; the second column the pollen parent.yPPF = number of plump seeds per pollinated flower. xSPF = number of seedlings per pollinated flower. Table 1-3. Result of intra-specific hybridization of V. darrowi clones in 2007. Female Male Flowers (No.) Berries (No.) Fruit set % PPF y SPFx 03-421-Pz 03-418-I 291223 76.63 3.7501.363 03-423-P 06-660-I 2 40 212 88.33 37.9597.008 03-412-I 03-405-I 250254 101.60w 12.5135.061 03-404-I 03-415-I 2 79 188 67.38 20.5669.828 03-418-I 03-412-I 246126 51.22 3.3931.232 03-405-I 03-417-I 1 90 115 60.53 13.4692.724 zDiploid V. darrowi Darrows evergreen blueberry. P Florida panhandle race. I Istokpoga race. The first column gives the seed parent; the second column the pollen parent. yPPF = number of plump seeds per pollinated flower. xSPF = number of seedling s per pollinated flower. wFruit set over 100 percent, due to sel f-pollination or parthenocarpic fruit 47

PAGE 48

Table 1-4. Result of inter-specifi c hybridization between southe rn highbush cultivars and V. darrowi in 2007. Highbush was used as the female parent. Highbushz V. darrowiy Flowers (No.) Berries (No.) Fruit set (%) PPFx SPFw HPFv 06-446 03-419-I 569 13 8 24.25 2.333 0.342 0.342 02-16 03-421-P 596 29 4.87 0.153 0.059 0.059 07-92 03-418-I 400 133 33.25 1.219 0.460 0.460 06-468 06-712-P 549 11 2.00 0.007 0.004 0.004 07-108 03-404-I 557 5 0.90 0.034 0.011 0.011 06-25 06-716-P 528 46 8.71 0.061 0.002 0.002 95-50 03-423-P 549 155 28.23 0.350 0.093 0.093 01-277 06-702-P 505 27 5.35 0.103 0.081 0.079 Southern Belle 03-414-I 53 3 28 5.25 0.139 0.043 0.043 00-45 06-719-I 359 158 44.01 0.248 0.042 0.042 zTetraploid V. corymbosum southern highbush. yDiploid V. darrowi Darrows evergreen blueberry. P Florida panhandle race. I Istokpoga race. xPPF = number of plump seeds per pollinated flower. wSPF = number of seedlings per pollinated flower. vHPF = number of hybrids per pollinated flower. Table 1-5. Result of recipr ocal inter-specific hybri dization between diploid V. darrowi V. fuscatum natural hybrid and southe rn highbush cultivars in 2007. Femalez Male Flowers (No.) Berries (No.) Fruit set (%) PPFy SPFx HPFw 06-721-N 07-48-H 259 229 88.42 0.402 0.012 0.012 06-720-N 02-16-H 240 113 47.08 0.452 0.033 0.033 02-16-H 06-721-N 250 12 4.80 0.040 0.004 0.004 02-16-H 06-720-N 256 10 3.91 0.203 0.023 0.023 zN Diploid Natural inter-specific hybrid, diploid V. darrowi diploid V. fuscatum H Tetraploid V. corymbosum southern highbush. yPPF = number of plump seeds per pollinated flower. xSPF = number of seedlings per pollinated flower. wHPF = number of hybrids per pollinated flower. Here reciprocal crosses are defined in terms of species contrary to individual clones or cultivars. 48

PAGE 49

Table 1-6. Result of inter-speci fic hybridization between sout hern highbush cultivars and V. darrowi in 2008. Highbush was used as the female parent. Highbushz V. darrowiy Flowers (No.) Berries (No.) Fruit set (%) Plump seeds per pollinated flower 01-151 03-421-P 202 93 46.04 0.050 Sweet Crisp 03-422-P 288 5 1.74 0.045 00-34 06-719-I 504 2 0.40 0.018 95-138 06-716-P 516 6 1.16 0.056 Southern Belle 06-715-P 566 7 1.24 0.019 01-245 06-718-I 371 1 0.27 0.003 03-293 06-704-P 273 35 12.82 0.335 01-243 06-708-P 338 57 16.86 0.348 01-25 06-702-P 493 29 5.88 0.069 02-106 06-701-P 514 11 2.14 0.012 00-34 06-710-P 473 9 1.90 0.015 97-46 03-404-I 335 3 0.90 0.015 zTetraploid V. corymbosum southern highbush. yDiploid V. darrowi, Darrows evergreen blueberry. P Florida panhandle race. I Istokpoga race. Table 1-7. Result of recipr ocal inter-specific hybridi zations between diploid Vaccinium fuscatum and southern highbush cultivars in 2008. Femalez Male Flowers (No.) Berries (No.) Fruit set (%) PPF y 06-722-F 02-22-H 510 7 1.37 0.061 06-723-F 01-170-H 478 54 11.30 0.517 06-724-F 00-211-H 398 9 2.26 0.369 02-22-H 06-722-F 338 36 10.65 0.192 01-170-H 06-723-F 407 17 4.18 0.032 00-211-H 06-724-F 590 15 2.54 0.046 zF Diploid V. fuscatum H Tetraploid V. corymbosum southern highbush. yPPF = number of plump seeds per pollinated flower. Reciprocal crosses are defined in terms of species instead of individuals clones or cultivars. 49

PAGE 50

Table 1-8. Result of inter-sp ecific hybridization between V. darrowi from the Ocala Forest race and southern highbush cultivars in 2001. In all crosses, V. darrowi was the female parent. V. darrowiz Highbushy Flowers (No.) Berries (No.) Fruit set (%) SPFx HPFw 99-156 98-405 1093 62 1.74 0.017 0.009 99-156 98-409 257 7 5.06 0.051 0.000 99-156 97-75 862 22 2.55 0.026 0.009 99-156 Sunshine 448 27 1.12 0.011 0.009 99-156 98-18 347 11 1.15 0.012 0.012 99-158 Emerald 326 14 1.53 0.015 0.015 99-158 98-18 653 73 8.27 0.083 0.061 99-161 95-67 230 79 7.83 0.078 0.043 99-161 98-18 310 93 9.03 0.090 0.045 99-162 98-18 463 58 0.65 0.006 0.000 99-163 98-405 639 216 0.63 0.006 0.002 99-163 98-409 681 217 0.88 0.009 0.003 99-164 97-75 358 84 3.63 0.036 0.011 99-164 98-18 449 192 9.13 0.091 0.040 99-167 Emerald 224 113 5.80 0.058 0.058 zDiploid V. darrowi, Darrows evergreen blueberry. yTetraploid V. corymbosum southern highbush. xSPF = number of seedlings per pollinated flower. wHPF = number of hybrids per pollinated flower. Table 1-9. Result of inter-sp ecific hybridization between V. darrowi from the Istokpoga race and southern highbush cultivars in 2005. V. darrowi was the female parent. V. darrowiz Highbush y Flowers (No.) Berries (No.) Fruit set (%) SPFx HPFw 03-418 00-270 300 32 10.67 0.050 0.000 03-417 04-44 237 278 117.30v0.093 0.000 03-412 05-04 387 0 0.00 0.010 0.000 03-418 Emerald 457 32 7.00 0.011 0.000 03-418 01-06 545 21 3.85 0.006 0.000 03-405 Emerald 507 39 7.69 0.004 0.002 03-405 01-06 605 174 28.76 0.036 0.000 03-419 Emerald 210 88 41.90 0.010 0.005 03-418 05-65 600 29 4.83 0.017 0.000 03-404 01-06 -u 0.014 0.002 03-404 98-406 491 0 0.00 0.000 0.000 03-407 Millenia 400 0 0.00 0.000 0.000 zDiploid V. darrowi, Darrows evergreen blueberry. yTetraploid V. corymbosum southern highbush. xSPF = number of seedlings per pollinated flower. wHPF = number of hybrids per pollinated flower. vOver 100% fruit set due to parthenocarpic fruit.u = missing data. 50

PAGE 51

Table 1-10. Result of inter-spe cific hybridization between V. darrowi and southern highbush cultivars in 2006. V. darrowi was the female parent. V. darrowiz Highbushy Flowers (No.) Berries (No.) Fruit set (%) PPFx SPFw HPFv 03-423-P HCC 842 615 73.0 7.506 0.017 0.002 03-421-P HCC 1406 600 42.7 2.308 0.104 0.102 03-419-I Abundance 15011677.32.407 0.0330.013 03-416-I HCC 120 10 8.3 0.300 0.050 0.000 03-416-I 00-75 413 224 54.2 4.041 0.191 0.000 03-411-I Emerald 267 43 16.1 1.387 0.022 0.000 03-411-I HCC 423 395 93.4 2.949 0.045 0.002 03-402-I 96-138 28300.00.000 0.0000.000 03-405-I Emerald -u0.194 0.016 0.00 0 zDiploid V. darrowi Darrows evergreen blueberry. P Florida panhandle race. I Istokpoga race. yTetraploid V. corymbosum, southern highbush. HCC highbush cultivars compose. xPPF = number of plump seeds per pollinated flower. wSPF = number of seedlings per pollinated flower. vHPF = number of hybrids per pollinated flower. u = missing data. Table 1-11. Result of inter-spe cific hybridization between V. darrowi and southern highbush cultivars in 2007. V. darrowi was the female parent. V. darrowiz Highbushy Flowers (No.) Berries (No.) Fruit set (%) PPFx SPFw HPFv 03-414-I 03-228 540 94 17.41 0.449 0.078 0.022 03-404-I 03-286 667 112 16.79 0.354 0.024 0.000 03-421-P 00-59 544 220 40.44 1.242 0.426 0.408 03-419-I 92-84 644 237 36.80 1.967 0.087 0.045 03-418-I 01-297 554 38 6.86 0.106 0.023 0.000 03-422-P 88-53 541 225 41.59 1.798 0.153 0.125 03-416-I Jewel 573 268 46.77 0.792 0.058 0.003 03-423-P 01-63 701 229 32.67 1.088 0.023 0.003 06-660-I Southern Belle 551 302 54.81 6.229 0.161 0.152 03-402-I 07-49 551 7 1.27 0.192 0.087 0.000 zDiploid V. darrowi Darrows evergreen blueberry. P Florida panhandle race. I Istokpoga race. yTetraploid V. corymbosum southern highbush. xPPF = number of plump seeds per pollinated flower. wSPF = number of seedlings per pollinated flower. vHPF = number of hybrids per pollinated flower. 51

PAGE 52

Table 1-12. Result of inter-spe cific hybridization between V. darrowi and southern highbush cultivars in 2008. V. darrowi was the female parent. V. darrowiz Highbushy Flowers (No.) Berries (No.) Fruit set (%) Plump seeds per pollinated flower 03-421-P 01-151 515 74 14.37 0.533 03-422-P Sweetcrisp 505 221 43.76 3.331 06-719-I 00-34 517 238 46.03 3.146 06-716-P 95-183 522 266 50.96 2.692 06-715-P Southern Belle 283 36 12.72 0.339 06-718-I 01-245 507 209 41.22 0.649 06-704-P 03-293 526 327 62.17 0.633 06-708-P 01-243 448 197 43.97 0.232 06-702-P 01-25 468 46 9.83 0.304 06-701-P 02-106 161 14 8.70 0.584 06-710-P 00-34 482 159 32.99 0.284 03-404-I 97-46 507 14 2.76 0.120 03-419-I 00-59 506 175 34.58 1.212 03-405-I 96-22 203 130 64.04 4.538 06-660-I Emerald 509 198 38.90 2.469 zDiploid V. darrowi, Darrows evergreen blueberry. P Florida panhandle race. I Istokpoga race. yTetraploid V. corymbosum southern highbush. 52

PAGE 53

Table 1-13. Result of inter-spe cific hybridization between V. darrowi and southern highbush cultivars with their reciprocals z in 2007. Female Male Flowers (No.) Berries (No.) Fruit set (%) P:100Fw S:100Fv H:100Fu 03-414-Iy 03-228-HB 540 94 17.41 44.9 7.8 2.2 Southern Belle-HBx 03-414-I 533 28 5.25 13.9 4.3 4.3 p-value0.0106 r <0.0001 0.3143 0.4101 03-404-I 03-286-HB 667 112 16.79 35.4 2.4 0.0 07-108-HB 03-404-I 557 5 0.90 3.4 1.1 1.1 p-value 0.0002 <0.0001 0.4871 0.2943 03-421-P 00-59-HB 544 220 40.44 124.2 42.6 40.8 02-16-HB 03-421-P 596 29 4.87 15.3 5.9 5.9 p-value<0.0001 <0.0001 <0.0001 <0.0001 03-419-I 92-84-HB 644 237 36.80 196.7 8.7 4.5 06-446-HB 03-419-I 569 138 24.25 233.3 34.2 34.2 p-value 0.1082 0.0776 <0.0001 <0.0001 03-418-I 01-297-HB 554 38 6.86 10.6 2.3 0.0 07-92-HB 03-418-I 400 133 33.25 121.9 46.0 46.0 p-value<0.0001 <0.0001 <0.0001 <0.0001 03-423-P 01-63-HB 701 229 32.67 108.8 2.3 0.3 95-50-HB 03-423-P 549 155 28.23 35.0 9.3 9.3 p-value 0.5694 <0.0001 0.0399 0.0037 VD HB 3650u 930 25.16s 86.7 11.0 7.9 HB VD 3204 488 16.13 70.5 16.8 16.8 p-value 0.159 0.1963 0.2713 0.0733 zNote that the V. darrowi clone in each case is identical in the reciprocal pairs, but the southern highbush clones change. All the southern highbus h clones used in these crosses are known to be highly fertile both as males and females. yDiploid V. darrowi (VD), Darrows evergreen blueberry. P Florida panhandle race. I Istokpoga race. xTetraploid V. corymbosum (HB), southern highbush represented by HB. wP:100F = number of plum p seeds per 100 pollinated flower. vS:100F = number of seedlings per 100 pollinated flower. uH:100F = number of hybrids per 100 pollinated flower. tTotal values per cross type. sMean averages per cross type. rChisquare, test of independence, p-value<0.01. Comparison within a column between reciprocal crosses. 53

PAGE 54

Table 1-14. Crossing behavior of diploid and tetraploid Florida Vaccinium species in section Cyanococcus in 2007. Crossz Ploidy of parents Flowers (No.) Fruit set (%) PPFy SPFx HPFw HB HB 4x 4x 652 78.06a 7.916 ab 17.518 a -v HB NH 4x 2x 506 4.36 d 0.120 b 0.010 b 0.010a HB VD-P 4x 2x 2727 9.83 d 0.134 b 0.046 b 0.046a HB VD-I 4x 2x 2418 21.53c 0.794 b 0.178 b 0.178a NH HB 2x 4x 499 67.75a 0.425 b 0.020 b 0.020a VD-P HB 2x 4x 1786 38.23b 1.377 ab 0.200 b 0.180a VD-I HB 2x 4x 4080 25.82bc 1.441 ab 0.074 b 0.031a VD VD 2x 2x 1496 74.28a 15.275 a 4.535 ab zHB = V. corymbosum southern highbush. NH = natural hybrid, V. darrowi V. fuscatum VDP = V. darrowi Florida panhandle race. VD-I = V. darrowi Istokpoga race. yPPF = number of plump seeds per pollinated flower. xSPF = number of seedlings per pollinated flower. wHPF = number of hybrids per pollinated flower. v = no data.*Similar letters within a column indicates means not significantly different. T ukeys test for PPF, SPF and HPF, =0.05. Chi-square, test of independence for fruit set (%), =0.05. Table 1-15. Crossing behavior between diploid and tetraploid Florida Vaccinium species in 2008. Crossz Ploidy of parents Flowers (No.) Berries (No.) Fruit set (%) Plump seeds per pollinated flower HB VF 4x 2x 1335 52 5.79 b 0.090 NS HB VD 4x 2x 4873 125 7.61 b 0.082 VF HB 2x 4x 1386 36 4.98 b 0.316 VD HB 2x 4x 5122 277 33.80 a 1.404 zHB = V. corymbosum southern highbush. VF = V. fuscatum VD = V. darrowi. NS= indicated means not significantly different within a column, Tukeys test, =0.05. *Chi-square, test of independence, =0.05. 54

PAGE 55

Table 1-16. Leaf, flower and berry ch aracteristics of three taxa: F-1 ( V. darrowi southern highbush hybrids), southern highbush (HB), and V. darrowi (VD). Taxa F-1(VDHB)z HB VD No. clones 10 11 12 No. reps per clone 5 5 5 Leaf characteristics Length (mm) 27.98y(0.85)49.41 (0.81) 13.55 (0.78) b* a c Width (mm) 14.24 (0.47)27.15 (0.45) 5.30 (0.43) b a c Pubescensex None 38% 100% 92% Low 62% 0% 8% Medium 0% 0% 0% High 0% 0% 0% Margin glands None 32% 4% 8% Sunken 54% 96% 42% Exserted 14% 0% 50% Both 0% 0% 0% Flower characteristics Corolla length (mm) 7.72 (0.16) 9.01 (0.13) 5.41 (0.12) b a c Corolla width (mm) 5.19 (0.12) 7.28 (0.09) 3.97 (0.09) b a c Corolla aperture (mm) 2.07 (0.09) 3.44 (0.08) 1.70 (0.07) b a c Pedicel length (mm) 6.88 (0.29) 7.18 (0.23) 6.15 (0.22) ab a b Peduncle length (mm) 8.91 (0.66) 14.74 (0.53) 6.91 (0.51) b a c Bracteole length (mm) 3.69 (0.14) 4.93 (0.11) 2.35 (0.11) b a c Bracteole width (mm) 2.06 (0.12) 3.15 (0.09) 1.40 (0.08) b a c Anther awns 0% 0% 0% Persistent bracteoles 100% 100% 100% Berry characteristics Weight (g) 0.81 (0.05) 2.25 (0.05) 0.29 (0.14) bw av cu zF-1 ( V. darrowi southern highbush hybrids). HB = highbush ( Vaccinium corymbosum). VD = V. darrowi yMeans and (standard errors). xPubescence on lower surface. w70 plants, 2 berries per clone. v67 plants, 2 berries per plant. u8 plants, 2 berries per plant. *Similar letters within a row indicates means not significan tly different, Tukeys test, =0.05. 55

PAGE 56

Table 1-17. Mean pollen stainability from Vaccinium darrowi, V. corymbosum V. arboreum and V. darrowi V. corymbosum hybrids. Taxa Plants (No.) Pollen stainability (%)z V. darrowiFlorida panhandle race 8 97.42 N S V. darrowi-Istokpoga race 15 94.26 Southern highbush 11 93.48 V. arboreum 3 92.23 F-1 hybrids ( V. darrowi V. corymbosum ) 97 88.60 zPercent stainability was calculated as the number of stained spores per pollen grain (tetrad, triad, dyad and monad), averaged by the to tal number of pollen grains counted. NS= not significantly different within a column, Chi-s quare, test of independence, =0.05. Table 1-18. Crossing behavior between F-1 ( V. darrowi southern highbush cultivars) and southern highbush cultivars. Crossesz Number of crosses Flowers (No.) Fruit set (%) PPFy SPFx HB HB-07 5 652 78.06 NS 7.914 NS 17.52 NS F1 HB-08 5 4489 64.867.995 -wF1 HB-07 17 1351 57.499.658 2.46 zF1= F-1 hybrid ( V. darrowi southern highbush cultivars). HB = southern highbush. -07 = crosses made in 2007. -08 = crosses made in 2008. yPPF = number of plump seeds per pollinated flower. xSPF = number of seedlings per pollinated flower. w = no data. NS= indicated means not significantly different within a column, Tukeys test, =0.05 for PPF and SPF. Chi-square, test of independence for fruit set (%) =0.05. 56

PAGE 57

Figure 1-1. Flower production of V. darrowi per week by region from October 2006 to August 2007. 57

PAGE 58

Figure 1-2. Leaf morphology for diploid V. darrowi (left), F-1 hybrids (center) and tetraploid V. corymbosum (right). 0 10 20 30 40 50 60 70 01 02 03 04 0Length (mm)Width (mm) F1(VDHB) VD HB Figure 1-3. Scatter plot for leaf ch aracteristics by taxa. Each symbol represents the mean of five leaves from one plant. 58

PAGE 59

Figure 1-4. Flower and fr uit morphology for diploid V. darrowi (left), F-1 hybrids (center) and tetraploid V. corymbosum (right). 0 2 4 6 8 10 12 024681 0Length (mm)Width (mm) F1(VDHB) VD HB Figure 1-5. Scatter plot for coroll a characteristics by taxa. Each sy mbol represents the mean of five leaves from one plant. 59

PAGE 60

Figure 1-6. Plant archit ecture of diploid V. darrowi (left), F-1 hybrids (center) and tetraploid V. corymbosum (right). The pot on the left is 30 cm tall. 60

PAGE 61

0 10 20 30 40 50 60 70 80 900-10 10.1-20 20.1-30 30.1-40 40.1-50 50.1-60 60.1-70 70.1-80 80.1-90 90.1-100Plants (No.)Pollen stainability (%) Figure 1-7. Histogram of the F-1 ( V. darrowi V. corymbosum ) hybrid population pollen stainability percentage. a) b) Figure 1-8. Pollen from one plant with poor polle n staining (left panel) and one plant with good pollen staining (right panel) from a population of F-1 ( V. darrowi V. corymbosum ) hybrids; 250x. 61

PAGE 62

CHAPTER 3 INTER-SPECIFIC CROSSES OF VACCINIUM DARROWI CAMP. V. ARBOREUM MARSH. Introduction A potential obstacle with cultivation of temperat e fruits in Florida is slow growth and poor fruiting due to insufficient chilling during wi nter to overcome dormancy. Breeding low-chill southern highbush blueberry cultivars began with crosses between high-chill cultivars from New Jersey and Michigan which had high fruit qualit y, large berries, and earl y ripening and wild blueberries from Florida, mo st important of which was V. darrowi clone Fla. 4B. Several generations of crossing and selec tion were carried out to recombine desired traits from superior parents to create an improve d southern highbush blueberry. In Vaccinium section Cyanococcus crosses between homoploid species in general have been successful and without sterility barriers, an d even some heteroploid crosses have given at least some progeny (Meader and Darrow, 1944; Da rrow, et al., 1949; Sharpe, 1953; Sharpe and Darrow, 1959; Rousi, 1966). The cultivated southe rn highbush blueberry is a genetic complex in which different species have been brought together The first attempts to cross low-chill Florida native blueberry species with high-chill cultivars that had been bred using V. corymbosum and V. angustifolium were made in 1949 by Sharpe at Univers ity of Florida and Darrow at the USDA in Washington D.C. Sharpe and Sherman (1971) described some mo re-distantly related species with potential value in blueberry breeding that could be found in western North America, in tropical America, Africa and other areas of the world. In the southeas tern U.S., in addition to the close relatives of cultivated blueberry, which are in section Cyanococcus, other Vaccinium relatives include Vaccinium species in sections Polycodium Raf. ( V. stamineum the deerberry) and Batodendron 62

PAGE 63

Nutt, ( V. arboreum the sparkleberry), and species in the genus Gaylussacia H.B.K. (the huckleberry). Sharpe and Shoemaker (1958) described the first crosses which brought together adaptation to southern climates from Floridas native species and commercial berry quality from northern cultivated blueberry vari eties. The Florida species used in the first hybridizations were Florida evergreen blueberry ( Vaccinium myrsinities Lam.), which is widespread in Florida, Darrows evergreen blueberry ( Vaccinium darrowi Camp.), native in central and northern Florida, and rabbiteye blueberry ( V. ashei Reade.), widespread in northwestern and northeast Florida. V. myrsinites (4x) was crossed with northern highbush cu ltivars (4x) (Sharpe and Darrow, 1959). Forty seedlings were grown, but the hybrids were short and twiggy and the fruit small and dark V. ashei (6x) was crossed with V. darrowi (2x). The goal was to produce tetraploid hybrids that could be crossed to tetraploid highbush blueberry cultivars. From 7500 pollinations, 5 hybrids were obtained (Sharpe and Darrow, 1959). Thes e crosses were later re-examined, and the F-1 hybrids were found to be pentaploid due to unreduced gamete production by V. darrowi (Goldy and Lyrene, 1983). Finally, northern highbush cultivar s (4x) were crossed with V. darrowi (2x). Because of the strong triploid block that was known to exist in Vacccinium these crosses had been expected to be more difficult than the V. ashei V. darrowi crosses, but from 1600 pollinations, 31 selections of tetraploid hybrids were made. Fertile progeny were obtained by inter-crosses and backcrosses to northe rn highbush. Unreduced gamete production by V. darrowi and the existence of a strong triploid block in Vaccinium allowed tetraploid hybrids to be obtained from tetraploid highbush V. darrowi crosses (Sharpe and Darrow, 1959). V. 63

PAGE 64

darrowi clones used in these crosses were Fla. 4A and Fla. 4B, which had been selected from Winter Haven, Florida, because they had large fruit with light-blue color. In 1986, Lyrene studied V. darrowi plants from south-centr al Florida (Lake Istokpoga area), the Ocala National Forest, and the Florida panhandle region. He concluded that the populations differed from each other. The V. darrowi clones previously most used in breeding, Fla. 4A and Fla. 4B, belong to the Istokpoga race. The Florida panhandle race has never been used in breeding. V. darrowi has also been used in southern highbu sh breeding as a bridging species to overcome sterility barriers between te traploid highbush and several other Vaccinium species. When tetraploid highbush was crossed with diploid V. elliottii the yield of tetraploid hybrids per pollinated flower was less than 0.002, less than 10% of what is typically obtained from highbush V. darrowi crosses (Lyrene and Sherman, 1983). V. darrowi was crossed with V. elliottii and the F-1 hybrids were crossed with V. corymbosum (4x). V. darrowi had a 62.2% fruit set when pollinated by V. elliottii (both are diploid) (M eader and Darrow, 1944). The diploid F-1 hybrids were easily crossed with tetraploid southern highbush blueberry (Lyrene and Ballington, 1986). V. darrowi has been crossed with species in other Vaccinium sections. V. darrowi V. stamineum ( Cyanococcus Polycodium ) crosses produced F-1 hybrids fertile enough to produce F-2 and BC-1 progenies (Lyrene and Ba llington, 1986). In 1991, Lyrene crossed V. darrowi (section Cyanococcus) with V. arboreum (section Batodendron ). Over 40 seedlings were selected with characteristics intermediate betw een the parents. The F-1 hybrids were not difficult to obtain. These crosses were made using V. darrowi as seed parent. V. darrowi has not previously been used as the pollen parent in crosses with V. arboreum Brooks and Lyrene (1995) studied the characteristic s of sparkleberry blueberry derivatives. The introgression of 64

PAGE 65

genes from V. arboreum into highbush blueberry was promis ing, producing some upright plants with very open flower clusters. V. arboreum Marsh, known as sparkleberry or farkle berry, is a small tree, 2-10 m. high. The berry is small, dark, somewhat astringent, an d with gritty flesh due to sclerids and large seeds. The range of sparkleberry, as given by Camp (1945) is from North Carolina south to Florida and westward to Kent ucky, Indiana, and through the O zarks to the Edwards Plateau, Texas. Over several years various attemp ts have been made to directly cross V. arboreum with highbush cultivars, but all have failed (Lyrene, 1997). Several characteristics from sparkleberry could be useful if transferred to highbush cu ltivars. When grown on deep, well-drained soil, sparkleberry has a deep r oot system that allows it to tolerate dry soils in which highbush cultivars will not survive. It can grow well on soils with low organic matter, and on soils with pH up to 6.2 (Stockton, 1976). The V. arboreum flower has a shorter coroll a tube and a wider corolla aperture than the highbush flower. This flower architecture could improve insect pollination and fruit set in highbush cultivars. Late flowering from V. arboreum could help avoid crop loss due to freezes (Lyrene, 1991; Brooks and Lyrene, 1995; Lyrene, 1997; Brooks and Lyrene, 1998a; Brooks and Lyrene, 1998b). Seedlings obtained by open-pollination of V. darrowi V. arboreum F-1 hybrids in the presence of other Vaccinium species, including southern highbush cultivars, were studied by Brooks and Lyrene (1998a). The V. darrowi V. arboreum hybrids flowered heavily but produced very few seed in a field containing ma ny highbush and rabbiteye plants. Superior openpollinated seedlings were backcrossed to highbush cultivars, and over 1000 seedlings were fruited. Selected plants were backcrossed again to highbush cultivars, and the best plants were again selected. It was possible to retain high vi gor and fertility through multiple generations, and 65

PAGE 66

berry size was increased easily by selection. Se veral traits were improved by this method, drought resistance, increase of coro lla size and others (Lyrene, 1997). Another diploid Vaccinium species, V. elliottii (section Cyanococcus) was crossed with V. arboreum (section Batodendron). Like V. darrowi it was easily crossed with sparkleberry. Thousands of seedlings were obtained when V. elliottii was used as seed parent (Lyrene, 1997). Wenslaff and Lyrene (2003a and 2003b) made the cross, V. elliotii ( Cyanococcus) V. arboreum ( Batodendron ) and the reciprocal. With V. elliottii as seed parent, from approx. 3000 flowers, more than 4000 hybr ids were obtained. When V. arboreum was used as female, no hybrids were produced. None of the F-1 hybrids produced viable pollen. Here we report the results of reciprocal crosses between V. arboreum and various races of V. darrowi and discuss the possible value of the F-1 hybrids in producing three-way hybrids useful in breeding southern highbush cultivars. Materials and Methods Inter-Specific Hybridization Experiments Diploid section Cyanococcus diploid section Batodendron In January 2007, four diploid V. darrowi clones and three diploid V. fuscatum clones -in section Cyanococcuswere selected as parents for crosses with V. arboreum in section Batodendron The V. darrowi and V. fuscatum clones were chosen from the wild during April and May 2006 by P. Lyrene and K. Hummer. Of the V. darrowi clones, three were from the Istokpoga race, collected in Dave nport, Florida near Haines C ity and from Highlands County, Florida, and one V. darrowi clone was from the Florida panhandle race collected in Chattahoochee, near the Apalach icola National Forest. Three V. fuscatum clones were collected from Davenport, Polk County, Florida. In 2006, softwood cuttings from the selected V. darrowi and V. fuscatum plants were rooted in a 1:1 mixture of sphagnum peat and perlite. Rooted 66

PAGE 67

cuttings were transplanted to 4-liter pots cont aining sphagnum peat and perlite (1:1). They were grown in a greenhouse at Gainesvi lle, Florida. Three-year-old V. arboreum clones were selected as parents. They were grown from seed coll ected in Jennings State Forest, in Clay County, Florida, in 2003. The seedlings were transpla nted to a field nurser y in April 2005 at the University of Florida Plant Science Unit in Citra, Florida, and were late r dug and potted for use in crosses. In February 2007, V. darrowi V. fuscatum and V. arboreum clones were brought into a bee-proof greenhouse to avoid unintended pollination. Flowers were emasculated before pollination. V. darrowi V. arboreum crosses and reciprocals (Table 2-1 and Table 2-2, respectively), and V. fuscatum V. arboreum crosses and reciprocals (Table 2-3 and Table 2-4, respectively), were made. V. arboreum plants (used as females) were divided into three sections to avoid female plant-to-plant variation as a confounding factor. The results of cross-pollination with V. arboreum V. darrowi and V. fuscatum were compared (Morrow, 1943). Pollen from the pollen parents was collected on the thumbnail. The stigma was touched by the thumbnail to transfer pollen. For V. darrowi V. arboreum crosses and reciprocals, 500 flowers were pollinated per plant. For V. fuscatum V. arboreum crosses and reciprocals, 250 flowers were pollinated per plant. Berries were harvested when fully ripe. The first twenty berries were weighed and individually opened by hand. The seeds were counted and classified as plump or shriveled. Seeds from the remaining berries were extracted using a food blender. They were separated and washed in wate r, then dried and stored in coin envelopes at 5C. Number of plump seeds per pollinated flower per cross, number of large seeds per berry, number of small seeds per berry and total number of seeds per berry were calculated. Seeds were treated with a fungicide (0.100g Ca ptan per seed lot) prior to planting. In November 2007, seeds 67

PAGE 68

were planted on the top layer of 4-liter pots of sphagnum peat. Pots were kept in a greenhouse with intermittent mist for 2-3 months until germination was completed. Seedlings were kept in pots until they were 2-3 cm high. Seedlings were transplanted to plastic trays of sphagnum peat, where they were grown until 4-8 cm high. In Ma y 2008, seedlings were transplanted to a highdensity field nursery at University of Florida Plant Science Unit in Citra, Florida. Plant-to-plant spacing in rows was 10 cm and row-to-row spaci ng was 40 cm. The seedlings were irrigated and fertilized when required. In the field, seedlings were classified as hybrids or selfs based on vegetative morphology. Number of seedlings per pollinated flower and nu mber of hybrids per pollinated flower for each cross were calculated. In January 2008, six V. darrowi clones collected in 2006 by P. Lyrene and K. Hummer, (as described before), were selected as parents for crosses with V. arboreum. Three V. darrowi clones were from the Florida panhandle ra ce and three from the Istokpoga race. These V. darrowi clones were chosen from the germplasm collected in 2006 and were grown in pots during 2007. For V. darrowi V. arboreum crosses and reciprocals, approximately 500 flowers of each plant were pollinated (Table 2-5 and Table 2-6, respectively). Harvesting and seed extraction were as described before. Control crosses In January 2007, six V. arboreum plants and six V. darrowi plants were selected as parents for intra-specific crossing experiments. Two-year-old potted V. darrowi clones, two from the Florida panhandle region and four from the Ist okpoga region, were selected from the germplasm collected in 2006 by P. Lyrene and K. Hummer. In February 2007, the plants were brought into a bee-proof greenhouse. The growing conditi ons were as previously described. V. arboreum plants (used as females) were divided in to three sections. Each section received pollen from one of the 68

PAGE 69

three pollen sources used: V. arboreum (Table 2-7), V. darrowi (Table 2-2), or V. fuscatum (Table 2-4). The V. arboreum pollen was from a distinct V. arboreum clone, not self-pollination. Open flowers were removed to avoid contam ination. Flowers were emasculated before pollination. Pollen was collected on the thumbnail and rubbed onto the stigma of the flower previously emasculated. Berries were harveste d and weighed when fully ripe. Seeds from the first twenty berries were counted and classified as plump or sh riveled. Seeds from the remaining berries were extracted using a food blender. Nu mber of large seeds per berry, number of small seeds per berry and total number of seeds per be rry were calculated. Se ed planting and seedling transplanting were carried out as described above. Statistical analysis Fruit set, number of plump seeds per pollinate d flower, number of seedlings per pollinated flower, number of hybrids per pollinated flower, berry weight, number of large seeds per berry, number of small seeds per berry and total number of seeds per berry were used to measure the results of the pollination treatments. Means for berry weight, number of large seeds per berry, number of small seeds per berry, total number of seed per berry, number of plump seeds per pollinated flower, number of seedlings per pollinated flower and number of hybrids per pollinated flower for different treatments were separated using least squares means by Tukeys test, with significance level 5%. Means for fruit set percentage were separated using Chi-square test of independence, with significance level 5%. Data an alysis was ANOVA by the PROC GLM and PROC FREQ procedures of SAS (Statistical Analys is System Version 9.1, SAS Institute, Cary, NC). 69

PAGE 70

Fertility Studies of F-1 ( V. darrowi V. arboreum ) Hybrids Pollen stainability Pollen of ten F-1 ( V. darrowi V. arboreum ) inter-specific hybrids was collected from tenyear-old plants growing in a field nursery. Th ese hybrids were obtained from crosses made in 1997. The seeds that produced the hybrids were pl anted in November 1997 in 4-liter pots of sphagnum peat. The seedlings were transplanted to a high density nursery in April-May 1998, at the University of Florida Horticultural Unit in Gainesville, Florida. The following year, 100 vigorous plants which showed di stinctive hybrid morphology were tr ansplanted to 1m 4m field spacing. Flowers from the hybrids were collecte d in November 2007. Pollen fertility was determined as stainability of dry pollen moistened with 1% aceto-carmine. Slides were prepared from two to three flowers. Pollen stainability wa s measured as percentage of well-stained pollen grains in normal tetrads, triads, dyads and mona ds compared with the numbers of sporads that had only one, two, or three stained pollen grains (each sporad normally has four pollen grains). This result was averaged by the total of counted sporads. Leit z phase-contrast microscope, 250 magnification, was used. Abnormal pollen grains were shriveled, had abnormal shapes, failed to stain, or had varying degrees of pollen inflation (Dermen, 1940 ). Morphological abnormality of pollen grains in these hybrids was expected to result from chromosome pairi ng abnormalities during meiosis. Several meiotic abnormalities in pent aploid inter-specific hybrids between V. ashei (6x) and V. darrowi (2x) resulted in a high percentage of sh runken pollen (Goldy and Lyrene, 1983). Lyrene and Sherman (1983) studied in ter-specific hybrids between V. corymbosum (4x) and V. elliottii (2x), and stated that both mitotic and meiotic instability in wide hybrids could reduce their fertility. 70

PAGE 71

Cytogenetics of F-1 ( V. darrowi V. arboreum ) Hybrids Flower buds at various developmental stages were collected from the ten-year-old F-1 hybrids growing at University of Florida Horticultu ral Unit in Gainesville, Florida, in November 2007. Flower buds were fixed in a 3:1 solution of glacial acetic acid absolute ethanol. Samples were kept in fixative until the pigments were removed and then stored in fixative at 5C. The flower buds were then assessed for stage of development by cutting open a random floret from the several floret buds in one axillary bud, larg est or smallest, and squashing two anthers on a microscope slide in 45% acetic acid. The slide was analyzed with a phase-contrast Leitz microscope (250 and 400) to determine wh ether the pollen mother cells were undergoing meiosis. When a meiotic bud was found, the additiona l eight anthers were digested for 3 hours at room temperature in a 5% solution (diluted in citrate buffer at pH 6.0) of cell wall degrading enzyme complex from Aspergillus sp. (Viscozyme from Novoz ymes Corp.). After digestion, the anthers were stored in 70% ethanol solution at 5C. For ch romosome counting, anthers were imbibed for 20 min at room temperature in 45% a cetic acid. Four anthers were used to prepare each slide. Each anther was placed in one corn er on the microscope slide area that could be covered by one cover slip. Each anther was divi ded into two or three portions to produce an optimal squashing. Anthers were macerated in 45% acetic acid on the microscope slide. The cover slip was then placed on the slide. A medium -light tapping with the head of a pen was done over the cover slip at the points where the anth ers were covered. Then the slide was heated lightly over a flame for one second and allowed to cool. Excess acetic acid was removed from the slide. The edges of the microscope cover glass were sealed with nail polish. Cells were observed in a phase-contrast Leitz micr oscope at 250 and 400 magnifications. Microphotographs were taken us ing Moticam 1000 1.3MPixel micros cope digital camera with the Motic Images Plus Version 2.0ML software. 71

PAGE 72

For each hybrid that was studied, chromoso me pairing behavior during meiosis was studied in 10 or more cells during metaphase I. The number of univalents, bivalents, trivalents, and other chromosome associati ons at metaphase I was recorded. Pollen mother cells were studied at anaphase I and anaphase II to de tect any meiotic abnormalities that might be occurring. Results and Discussion Inter-Specific Hybridization Experiments Diploid section Cyanococcus diploid section Batodendron data In 2007, fruit set of V. darrowi (2x) V. arboreum (2x) crosses ranged from 30.83% to 76.00% (Table 2-1), higher than their reciprocal s, 0.19% to 27.45% (Table 2-2). Mean fruit set for V. darrowi V. arboreum crosses, 63.27%, was significan tly higher than for their reciprocals, 7.36%. (Table 2-8). Number of plump seeds per pollinated flower (PPF), number of seedlings per pollinated flower (SPF), and number of hybrids per pollinated flower (HPF) were 30x, 13x, and 9x higher when the V. darrowi clones were used as the seed parents compared to the reciprocals. The reciprocal means differed s ubstantially, but were not significantly different due to low replication and high coefficient of va riation (Table 2-8). Brooks and Lyrene (1998b) reported a similar result for crosses between V. darrowi and V. arboreum V. darrowi V. arboreum crosses were higher compared to their reciprocals for number of large seeds per berry, number of small seeds per berry, and total number of seeds per berry (P<0.05). Mean berry weight of V. darrowi V. arboreum crosses was significantly higher than for their reciprocal (P<0.05) (Table 2-9). Seed from V. arboreum V. arboreum germinated faster than seed of V. darrowi V. darrowi crosses. Seed from V. darrowi V. arboreum crosses made in 2007 started to germinate in the second week of March 2008, almost one month and two weeks later than V. arboreum V. 72

PAGE 73

darrowi crosses. V. arboreum have a larger seeds than V. darrowi and southern highbush cultivars (Brooks and Lyrene, 1998b). Later germination of V. darrowi V. arboreum crosses might be related to the seed size difference between the seed parents, or to differences in the way the endosperm develops in the reciprocal crosses. Poor endosperm development has been associated with embryo abortion. In V. elliottii V. corymbosum crosses, endosperm malfunction and embryo abortion were delayed when V. corymbosum was the female parent (Muoz and Lyrene, 1984b). In 2008, two of the six V. arboreum V. darrowi crosses produced ripe berries (Table 26). Fruit set of these two crosse s was 3.70% and 7.13%. Fruit set of V. darrowi V. arboreum ranged from 22.02% to 78.98% (Table 2-5). PPF of V. darrowi when used as seed parent with V. arboreum was higher than when used as male parent. These results were similar to those in 2007. Incompatibility with V. arboreum as female parents depended upon which V. darrowi clone was used as the pollen parent. Fruit set of V. fuscatum (2x) V. arboreum (2x) crosses in 2007 ranged from 19.63% to 52.41%, and was higher than for the reciprocal V. arboreum V. fuscatum crosses, 0% to 17.24% (Tables 2-3 and 2-4). Mean fruit set of V. fuscatum V. arboreum crosses, 40.16%, was significantly higher than mean fruit set of V. arboreum V. fuscatum crosses, 10.65% (P<0.05). PPF was 8x, and SPF and HPF were 80x higher when V. fuscatum was used as the seed parent compared to the reciprocals. Means of the differe nt reciprocal crosses di ffered substantially, but not significantly (Table 2-8). Number of large se eds, small seeds and total number of seeds were significantly higher for V. fuscatum V. arboreum than for the reciprocal crosses (Table 2-9). 73

PAGE 74

In 2008, seeds of V. fuscatum V. arboreum crosses started to germinate on the second week of March, almost one month an d two weeks later than seeds from V. arboreum V. fuscatum crosses. Similar results were found for V. darrowi V. arboreum crosses. Intra-specific crosses, V. darrowi V. darrowi and V. arboreum V. arboreum are described in Table 1-3 and Ta ble 2-7. Fruit set variation wa s attributed to female-male interactions and to variability in plan t fertility. In intra-specific crosses, V. arboreum had an average fruit set of 33.85% and V. darrowi averaged 74.28%, (P<0.05) (Table 2-8). PPF and SPF for V. darrowi V. darrowi crosses were four and six times higher, respectively, than th e corresponding means for V. arboreum V. arboreum crosses. Berry weight, number of large seeds per berry, and total number of seeds per berry for V. darrowi V. darrowi crosses were also significantly higher than the means for V. arboreum V. arboreum crosses (P<0.05) (Table 2-9) It is not known whether V. arboreum is naturally less fecund than V. darrowi or whether it is less tolerant of artif icial emasculation and pollination or of the greenhouse environment in our study. Overall comparisons Fruit set of V. arboreum V. arboreum crosses was higher than fruit set from V. arboreum cross-pollinated with V. darrowi or V. fuscatum (P<0.05) (Table 2-8). Fruit set from V. arboreum crosses with V. darrowi and V. fuscatum was significantly lower when V. arboreum was the seed parent (P<0.05) (Table 2-8). In addition, PPF, SPF and HPF of V. arboreum when used as seed parent with V. darrowi or V. fuscatum were lower than their recipr ocals (Table 2-8). In crosses of V. arboreum with V. darrowi and V. fuscatum mean values for number of large seeds per berry, number of small seeds per berry and total number of seeds per berry were significantly lower when V. arboreum was the seed parent compared to the reciprocals (P<0.05) (Table 2-9). 74

PAGE 75

Wenslaff and Lyrene (2003a) studied V. elliottii V. arboreum and their reciprocal crosses and found unilateral incomp atibility. Numerous hybrids were obtained when V. elliottii was the seed parent, but none when V. arboreum was the seed parent. In our studies of V. darrowi V. arboreum and their reciprocal crosses, and V. fuscatum V. arboreum and their reciprocal crosses, more hybrids were obtained when V. arboreum was the pollen parent, but at least some were obtained with V. arboreum as the seed parent. Since im portant cytoplasmic effects are sometimes seen in wide hybrids, the abil ity to produce at least some hybrids with V. arboreum cytoplasm could be important. F-1 hybrids from V. darrowi V. arboreum crosses and their reciprocals, and from V. fuscatum V. arboreum and their reciprocals, were not difficult to obtain. The number of seedlings per pollinated flower obtained by th ese crosses was low compared to homoploid crosses in Vaccinium section Cyanococcus and compared to V. arboreum V. arboreum crosses. Lyrene (1991) obtained similar results for V. darrowi V. arboreum crosses. Fertility Studies of F-1 ( V. darrowi V. arboreum ) Hybrids Pollen stainability data Pollen staining was studied in 10 ten-year-old F-1 ( V. darrowi V. arboreum ) hybrids. The plants were randomly selected from a population of about 100 V. darrowi V. arboreum F1 hybrids that were growing at University of Fl orida Horticultural Unit in Gainesville, Florida (Figure 2-1, Table 2-10). Only one of the ten hybrids had over 10% polle n stainability. The low fertility of the hybrids was probably due to poor pairing between the ch romosomes of the two parents (Brooks and Lyrene, 1995; Brooks and Lyre ne, 1998b). In the field, the F-1 hybrids were highly vigorous (approximately 3m high), with numerous shoots. The plants flowered heavily each year (Figure 2-2, Figure 2-3 and Figure 2-4). The plants once formed part of a large collection of Vaccinium germplasm that containe d hundreds of rabbiteye ( V. ashei ) and highbush 75

PAGE 76

selections ( V. corymbosum hybrids). The field was abandoned 7 years ago when the University moved its plant resear ch location. At present, 90 to 95% of the V. darrowi V. arboreum seedlings originally planted are still alive a nd growing vigorously, but essentially every other blueberry plant in the germplasm collection has died, probably due in large part to occasional periods of prolonged drought. The f act that the surviving plants are sterile relieved them of the stress of carrying a crop load, and that many have contributed to their survival. However, both V. arboreum and V. darrowi are noted for their drought toleran ce, so drought tolerance in their hybrids would be expected (L yrene personal communication). Meiotic abnormalities are expected during gamete formation of the F-1 ( V. darrowi V. arboreum ) hybrids. Pollen stainability of the hybrid s was much lower than that of the parent clones (P<0.05) (Table 2-11). Brooks and Lyrene (1998b) found that pollen stainability of F-1 ( V. darrowi V. arboreum ) hybrids averaged only 0.9%. Cytogenetics of F-1 ( V. darrowi V. arboreum ) Hybrids Hybrid FL98-146 (2n=2x=24) Data on the meiotic behavior is described in Table 2-12. The majority of PMCs (pollen mother cells), 89.6%, were observed to have multiv alent pairing different from the chromosome configuration (12 bivalents) for diploid V. darrowi and diploid V. arboreum Pairing relationship varied among PMCs. At least some bivalent pairing was observed in all the PMCs at diakenesis and metaphase I. Of the 48 PMCs, 21%, 6.3 % a nd 87.5% had at least on e univalent, trivalent and quadrivalent, respectively. Th e most frequent configurations 4IV+4II, and 5IV+2II, were observed in 20.8% and 16.7% of the PMCs. Of the 18 PMCs studied at anaphase I and telophase I, normal chromosome assortment, 12-12, was observed in 22.2%. Unbalanced a ssortment (12-11, 12-10, 11-11, 11-10) was observed in 77.8% of the PMCs. The most freque nt unbalanced chromosome assortment was 1276

PAGE 77

11. Lagging chromosomes were observed in 5.6 % of the cells. Frequently, unbalanced assortments and lagging chromosome s at anaphase I were associated with chromosomes that did not paired with other chromosomes at metaphase I. Goldy and Lyrene (1983) reported that whether or not lagging chromosomes are include d in the end products of meiosis would depend on whether they continue to lag and (or) where pollen walls form. Chromatid disjunction at anaphase II wa s normal. One PMC had asynchronized disjunction. Misaligned spindles at anaphase II were observed in seven PMCs. This arrangement may produce unreduced gametes depending on where the pollen wall is formed at the end of meiosis II. Two PMCs were observed at telophase II with unbalanced ch romatid assortment, 1010-11-12 and 9-10-11-11. These results were proba bly produced by unbalanced assortments and lagging chromosomes at anaphase I and telophase I. Pollen stainability averaged 7.87% for 108 sporads scored (Table 2-10). Hybrid FL98-132 (2n=2x=24) Data on the meiotic behavior is describe d in Table 2-13. Of the 55 PMCs, 10.9% were observed to have 12 bivalents at diakenesis and metaphase I. Pair ing configuration were variable among PMCs. Univalents were observed in 34.5% of the cells. The most frequent multivalent pairing association was bivalent s in 98.2% of the cells, followe d by quadrivalents in 89.1% and trivalents in 10.8%. The most fre quent configurations, 3IV+6II, 4IV+4II and 12II, were observed in 41.8% of the PMCs. At anaphase I and telophase I, the most frequent chromosome assortment, 12-10, was observed in 37.5% of the 16 PMCs. Normal assortment (12-12) was observed in 6.3% of the cells. The most frequent unbalan ced assortments, 12-10 and 12-11, were observed in 37.5% and 25.0% of the cells. Assortments with more than 12 chromosomes in either one or both poles were 77

PAGE 78

observed in 18.9% of the cells. These results ma y be produced by equational division or early sister chromatid separation of univalents during anaphase I. Of the 10 PMCs at anaphase II and telophase II, chromatid separation was normal. The most frequent chromatid assortment, 12-11-11-10 and 11-10-10-10, were observed in two PMCs each. Three PMCs had more than 12 chromatid s in one of their poles. Probably lagging chromatids were included in the pole during anaphase II. One PMCs had 23-11-7 assortment. This assortment was probably produced by misali gned spindles at anaphase II. No PMCs were observed with misaligned spindles at anaphase II. Pollen stainability averaged 2.17% for 150 sporads scored (Table 2-10). Hybrid FL98-129 (2n=2x=24) Data on the meiotic behavior is described in Table 2-14. Of the 55 PMCs at diakenesis and metaphase I, 96.4%, were observed to have at least one univalent. Partial desynapsis was observed in the majority of the PMCs. One PM C had 24 univalents. Just one cell was observed to have 12 bivalents. Pairing relationship varied among PMCs. Multivalent associations were present in large number of PMCs. Bivalent, triv alent and quadrivalent pairing was observed in 92.8%, 30.8% and 45.5% of the cells, respectively.. The most frequent configurations, 6II+12I, 4II+16I, were observed in 7.3% of the cells for each one. Of the 5 PMCs observed at anaphase I a nd telophase I, two had 11-10 chromosome assortment, one had 16-15, one had 11-11 and one ha d 10-9 + 2 laggards. Assortments with more than 12 chromosomes, 16-15, may be produced by equational division or early sister chromatid separation of univalents during anaphase I. No PM Cs were observed at anaphase II and telophase II. Pollen stainability averaged 9.06% for 138 sporads scored (Table 2-10). 78

PAGE 79

Hybrid FL98-115 (2n=2x=24) Data on the meiotic behavior is described in Table 2-15. At diakenesis and metaphase I, of the 42 PMCs, 40.7 % were observed to have univa lents. The most fre quent configuration observed was 12 bivalents in 31.0% of the cells. Multivalent associations were observed in all PMCs. Bivalent, trivalent and quadrivalent associations we re observed in 92.9%, 28.6%, and 64.3%, respectively. Of the 37 PMCs, the most frequent chro mosome assortment, 12-11, was observed in 43.2% of the cells. Normal assortment, 12-12, wa s observed in 8.1% of the cells. No lagging chromosomes were observed. Assortments with mo re than 12 chromosomes, in either one or both poles, were observed in 8.1% of the cells. These results may be produced by equational division or early sister chromatid sepa ration of univalents during anaphase I. Chromatid disjunction was normal at anaphase II. Four chromatid assortments were observed at anaphase II and telophase II, 12-12-11-0, 12-12-11-9, 12-10-10-10 + 1 excluded chromatid, and 11-10-10-9. Of the PMCs, three cells had at least one viable pollen grain with 12 chromatids. Pollen stainability averaged 6.32% for 170 sporads scored (Table 2-10). Hybrid FL98-95 (2n=2x=24) Data on the meiotic behavior is described in Table 2-16. Of the 68 PMCs observed at diakenesis and metaphase I, 26.5% of the cells had at least one univalent. The most frequent configurations, 4IV+4II, 5I V+2II, and 12II, were observed in 22.1%, 19.1%, and 10.3%, respectively. Multivalent associations were observed. Quadrivalents, trivalents, and bivalents, were observed in 85.3%, 25.1%, and 98.5% of the cells. Of the 37 PMCs, 48.6% were observed to ha ve a 12-11 chromosome assortment. Lagging chromosomes were observed in 5.4% of the PM Cs. Normal chromatid disjunction, 12-12, was present in 27.0% of the cells. Asso rtments with more than 12 chromosomes, in either one or both 79

PAGE 80

poles, were observed in 5.4% of the cells. Thr ee PMCs were observed to have asynchronized stages. Chromatid disjunction at ana phase II was normal. Just one PMCs was observed, with a chromatid assortment of 12-12-11-9. Pollen stai nability averaged 23.66% for 112 sporads scored (Table 2-10). Hybrid FL98-93 (2n=2x=24) Data on the meiotic behavior is described in Table 2-17. One PMCs was observed having two synezetic knots and two NORs, which may be produced by the V. darrowi and V. arboreum genomes acting independently. Of the 47 PMCs, 29.6% had at least one univalent at diakenesis and metaphase I. The most frequent configura tions, 12II and 4IV+4II, were observed in 36.2% and 10.6% of the PMCs. Configurations varied among PMCs. Multivalents were observed in the majority of cells. Quadrivalents, trivalents and bivalents were observed in 59.6%, 8.6%, and 89.0% of the cells respectively. The most frequent chromosome assortment at anaphase I and telophase I, 12-11, was observed in 41.2% of 17 PMCs. Normal chromoso me disjunction, 12-12, was observed in 17.6% of the cells, similar to 11-11 assortment. Of the 17 PMCs, 12-12-11-10 was the most freq uent chromatid assortment observed in 11.8% of the cells at anaphase II and telophase II. One PMC was observed to have misaligned spindles at anaphase II. Five PMCs had excluded and lagging chromatids at telophase II. One PMCs had over twelve chromatids in one of the poles. Probably lagging chromatids were included in the pole during anaphase II. Three PMCs had 25-12-11, 22-11-10 and 20-9-9, assortments. These assortments were probably pr oduced by misaligned spindles at anaphase II. Pollen stainability averaged 2.56% fo r 127 sporads scored (Table 2-10). 80

PAGE 81

Hybrid FL98-53 (2n=2x=24) Data on the meiotic behavior is describe d in Table 2-18. Of the 97 PMCs, 29.8% were observed to have at least one uni valent. The most frequent configurations, 4IV+4II, and 5IV+2II, were observed in 22.7% and 18.6% of the cells. Th e configuration of 12 bivalents was observed in 3.1% of the cells. Chromosome configurations varied from cell to cell. Multivalent associations were present in the majority of the observed PMCs. Quadrivalent, trivalent, and bivalent pairing were observed in 94.8%, 6.1%, and 89.8% of the cells. Of the 29 PMCs at anaphase I and telophase I, 17.2% of the cells were observed to have 12-12 assortment. The most frequent unbalanced assortments, 12-11 and 12-10, were equally observed in 13.8% of the cells, respectively. Lagging chromosomes were observed in one PMCs. Assortments with more than 12 chromosomes, in either one or both poles, were observed in 24.1% of the cells. These results may be produc ed by equational division or early sister chromatid separation of univalents during anaphase I. Misaligned spindles at early an aphase II were observed in one cell. Just one PMCs was observed at anaphase II and te lophase II. The chromatid assortment was 12-10-9-9. Pollen stainability averaged 1.56% for 128 sporads scored (Table 2-10). Hybrid FL98-61 (2n=2x=24) Data on the meiotic behavior is described in Table 2-19. Of the 45 PMCs, 42.2% of the cells had at least one univalent. The most frequent configurati ons, 4IV+4II, and 5IV+2II, were observed in 20.0% and 11.1% of the PMCs. Norm al configuration for diploid species in Vaccinium 12 bivalents, was observed in 8.9% of the cells. Configurations varied among PMCs. Multivalent pairing was observed in the majority of PMCs. Quadrivalent, trivalent, and bivalent pairing was observed in 87.5%, 6.3%, a nd 100% of the cells, respectively. 81

PAGE 82

Of the 29 PMCs at anaphase I and telophase I, 27.6 % of the cells were observed with 1211 chromosome assortment. The chromosome a ssortment, 12-12, was observed in 13.8% of the cells. Lagging chromosomes were observed in four PMCs. Assortments with more than 12 chromosomes, in either one or both poles, were observed in 10.3% of the cells. These results may be produced by equational division or early sister chromatid separation of univalents during anaphase I. No cells were observe d at anaphase II and telophase II. Pollen stainability averaged 0.89% for 112 sporads scored (Table 2-10). Hybrid FL98-64 (2n=2x=24) Data on the meiotic behavior is described in Table 2-20. Of the 94 PMCs, 33.1% of the cells had univalents. The most frequent confi gurations, 4IV+4II, 5IV+2II, and 12II, were observed in 29.8%, 13.8% and 11.7% of the PM Cs. Configuration varied among cells. Multivalent associations were observed in the ma jority of PMCs. Quadrivalent, trivalent, and bivalent, were observed in 84.0%, 13.8%, and 93.6%. Of the 65 PMCs at anaphase I and telophase I, 20.0% and 13.8% of the cells were observed with 11-10 and 12-11 assortment, respectively. Normal disjunction, 12-12, was observed in 9.2% of the cells. Lagging chromoso mes were observed in two PMCs. Assortments with more than 12 chromosomes, in either one or both poles, were observe d in 17% of the cells. These results may be produced by equational divi sion or early sister ch romatid separation of univalents during anaphase I. Misaligned spindles at anapha se II and asynchrony in meiosis II was observed in several PMCs. Of the 10 PMCs, two cells were observed in 11-10-10-9 assortment at anaphase II. One PMC had more than 12 chromatids in one of their poles. Probably lagging chromatids were included in the pole during anaphase II. One PMC had three excluded chromatids. Two PMCs 82

PAGE 83

had 22-9-9 and 19-10-10 assortments. These assortments were probably produced by misaligned spindles at anaphase II. Pollen stainability averaged 3.39% for 96 sporads scored (Table 2-10). Hybrid FL98-165 (2n=2x=24) Data on the meiotic behavior is described in Table 2-21. Of the 32 PMCs, 53.1% of the cells had univalents among their configurations. The most frequent configurations, 4IV+4II and 5IV+2II, were observed in 25.0% and 12.5% of th e cells. Chromosome configuration, 12II, was observed in 6.3% of the cells. Configurations we re highly variable. Quad rivalent, trivalent and bivalent associations were observed in 90.6%, 25.1%, and 81.2%, respectively. Of the 29 PMCs at anaphase I and telophase I, 12-11 and 11-10 assortments were the most frequent and were observed in 34.5% and 17.2% of the cells, re spectively. Normal assortment, 12-12, was observed in 10.3% of the cells. No lagging chromosomes were observed. Assortments with more than 12 chromosomes, in either one or both poles, were observed in 3.4% of the cells. Eleven PMCs were found in anaphase II and telophase II. The most frequent chromatid assortment was 11-10-10-9 with two cells th at were observed. Excluded chromatids were observed in one PMC. One cell had 22-9-9 arrang ement. This assortment was probably produced by misaligned spindles at anaphase II. Misaligned spindles at anaphase II was observed in one PMC. Pollen stainability averaged 1.32% for 152 sporads scored (Table 2-10). Overall results Meiotic analysis of the ten V. darrowi V. arboreum revealed several abnormalities (Figures 2-5, 2-6, 2-7 and 2-8): 1) Presence of two synezetic knot and nucleolar organizing regions (NOR) during meiotic prophase in clone FL98-93. Similar observations were described in V. darrowi V. ashei hybrids by Goldy and Lyrene (1983). In that report, it was postulated that the V. darrowi and V. ashei genomes could sometimes act independently of each other. A 83

PAGE 84

result of this abnormality could be unequal chromosome number in the gametes and/or unreduced gamete formation depending on chromo some segregation and pollen wall formation. 2) Partial desynapsis. Uncoordina ted timing of meiotic events was observed in most of the PMCs of FL98-129. Desynapsis caused th e paired chromosomes to fall ap art before anaphase I. 3) Lagging chromosomes were common at anaphase I. Of the ten hybrids, six were observed to have lagging chromosomes in one or several ce lls. Whether or not lagging chromosomes are included in the end products of meiosis would depend on whether they move to the poles and on whether and where pollen walls forms. 4) Unbala nced chromosome assortments at anaphase I and telophase I. Chromosomes that fail to migrat e could form micronuclei or be included in the end products of meiosis. 5) Chromatid separati on at anaphase I. Early chromatid splitting or chromatid equational division during anaphase I may produce chromosome assortments with more than 12 chromosomes. Lagging chromosomes and univalents may contribute to the chromatid equational division. 6) Asynchrony of meiosis II. Several PMCs were observed to have different products of meiosis at the same stage. 7) Misaligned spindles at anaphase II. This could cause two sets of chromosomes to be in cluded in the same nucleus, producing unreduced gametes. 8) Lagging and excluded chromatids at an aphase II. The ploidy le vel of the pollen grain will vary depending on the number of chromatids that were included. 9) Unbalanced chromatid assortments at anaphase II and telophase II. Th ese could result from the abnormality described before. All these abnormalities can explai n the poor pollen st ainability of the V. darrowi V. arboreum hybrids. In general, presence of multivalent associations during diakenesis and metaphase I may reflect homoeologous associations between V. darrowi and V. arboreum chromosomes, possibly resulting because the chromosome s have no true homologues. Ho moeologous associations could 84

PAGE 85

indicate relatedness between the two genomes th at has been maintained since the species diverged. Reciprocal translocations may ha ve occurred as the two species diverged. Conclusions Crosses of V. arboreum (section Batodendron ) with V. fuscatum and V. darrowi (section Cyanococcus) were easy to make. The F-1 hybrids from these crosses and their reciprocals were highly vigorous. Although most fe rtility indices showed that the crosses were more productive when V. arboreum was the pollen parent, some hybr ids were also obtained with V. arboreum as seed parent. V. darrowi V. darrowi and V. arboreum V. arboreum crosses produced more plump seeds per pollinated flower and more seedlings per pollinated flower than the V. darrowi V. arboreum inter-sectional crosses. Yet, as is so often the case in Vaccinium it is surprising that species so morphologically and taxonomically distinct as V. darrowi and V. arboreum can make hybrids at all, and that so many of the hybrids are vigorous. Pollen stainability of the ten-year-old F-1 ( V. darrowi V. arboreum ) hybrids was low, probably due to the infrequency of normal bivale nt pairing between the two chromosome sets, which were derived from different sections of Vaccinium The F-1 hybrids were highly vigorous, produced a large number of shoots, and flowered heavily, but they produced few fruit. When surrounded by tetraploid highbush cultivars, they produced a few viable seeds (Brooks and Lyrene, 1998b). Now that the surrounding highbush have died, vi able seeds are no longer found. Numerous meiotic abnormalities were found. These included: two synezetic knot and nucleolar organizing regions (NOR) during meio tic prophase I, partial desynapsis, lagging chromosomes at anaphase I, unbalanced chromosome assortments at anaphase I, early chromatid separation at anaphase I, asynchrony of meiosis II, misaligned spindles at anaphase II, lagging and excluded chromatids at anaphase II, unbalan ced chromatid assortments at anaphase II, 85

PAGE 86

increased the percentage of incomplete and unstained tetrad, and production of unreduced gametes. 86

PAGE 87

Table 2-1. Hybridization e xperiments between diploid V. darrowi in section Cyanococcus and diploid V. arboreum in section Batodendron in 2007. V. darrowi was female parent. V. darrowiz V. arboreumy Flowers (No.) Berries (No.) Fruit set (%) PPFx SPFw HPFv 03-421-P 06-770 829 617 74.43 6.241 1.362 1.211 03-418-I 06-772 582 344 59.11 8.915 2.552 0.221 03-405-I 06-774 525 399 76.00 35.273 3.358 3.358 06-740-I 06-773 493 152 30.83 5.122 0.912 0.871 zDiploid V. darrowi, Darrows evergreen blueberry, section Cyanococcus P Florida panhandle race. I Istokpoga race. yDiploid V. arboreum Sparkleberry, section Batodendron. xPPF = number of plump seeds per pollinated flower. wSPF = number of seedlings per pollinated flower. vHPF = number of hybrids per pollinated flower. Table 2-2. Hybridization e xperiments between diploid V. arboreum in section Cyanococcus and diploid V. darrowi in section Batodendron in 2007. V. arboreum was the female parent. V. arboreumz V. darrowiy Flowers (No.) Berries (No.) Fruit set (%) PPFx SPFw HPFv 06-770 03-421-P 564 39 6.91 0.324 0.078 0.078 06-772 03-418-I 552 3 0.54 0.058 0.000 0.000 06-774 03-405-I 524 9 1.72 0.206 0.027 0.027 06-773 06-740-I 526 1 0.19 0.048 0.019 0.017 06-777 06-740-I 51 14 27.45 1.216 0.471 0.471 zDiploid V. arboreum Sparkleberry, section Batodendron yDiploid V. darrowi, Darrows evergreen blueberry, section Cyanococcus P Florida panhandle race. I Istokpoga race. xPPF = number of plump seeds per pollinated flower. wSPF = number of seedlings per pollinated flower. vHPF = number of hybrids per pollinated flower. Table 2-3. Result of inter-specific hybr idization experiments between diploid Vaccinium fuscatum in section Cyanococcus and V. arboreum in section Batodendron in 2007. V. fuscatum was the female parent V. fuscatumz V. arboreumy Flowers (No.) Berries (No.) Fruit set (%) PPFx SPFw HPFv 06-724 06-776 270 53 19.63 2.096 0.133 0.133 06-723 06-770 287 139 48.43 5.583 1.224 1.224 06-722 06-771 166 87 52.41 4.516 1.089 1.089 zDiploid V. fuscatum section Cyanococcus yDiploid V. arboreum Sparkleberry, section Batodendron. xPPF = number of plump seeds per pollinated flower. wSPF = number of seedlings per pollinated flower. vHPF = number of hybrids per pollinated flower. 87

PAGE 88

Table 2-4. Result of inter-specifi c hybridization between diploid V. arboreum in section Batodendron and diploid V. fuscatum in section Cyanococcus in 2007. V. arboreum was the female parent. V. arboreumz V. fuscatumy Flowers (No.) Berries (No.) Fruit set (%) PPFx SPFw HPFv 06-776 06-724 319 55 17.24 0.747 0.006 0.006 06-770 06-723 222 9 4.05 0.284 0.014 0.014 06-771 06-722 38 0 0.00 0.000 0.000 0.000 zDiploid V. arboreum Sparkleberry, section Batodendron. yDiploid V. fuscatum section Cyanococcus. xPPF = number of plump seeds per pollinated flower. wSPF = number of seedlings per pollinated flower. vHPF = number of hybrids per pollinated flower. Table 2-5. Hybridization e xperiments between diploid V. darrowi in section Cyanococcus and diploid V. arboreum in section Batodendron in 2008. V. darrowi was the female parent. V. darrowiz V. arboreum y Flowers (No.) Berries (No.) Fruit set (%) PPFx 06-740-I 06-774 509 402 78.98 13.374 06-716-P 06-774 517 304 58.80 7.488 06-719-I 06-771 539 370 68.65 8.737 06-708-P 06-771 549 315 57.38 4.383 06-718-I 06-735 525 338 64.38 4.946 06-704-P 06-735 336 74 22.02 0.395 zDiploid V. darrowi, Darrows evergreen blueberry, section Cyanococcus P Florida panhandle race. I Istokpoga race. yDiploid V. arboreum Sparkleberry, section Batodendron. xPPF = number of plump seeds per pollinated flower. Table 2-6. Hybridization e xperiments between diploid V. arboreum in section Batodendron and diploid V. darrowi in section Cyanococcus in 2008. V. arboreum was the female parent. V. arboreumz V. darrowi y Flowers (No.) Berries (No.) Fruit set (%) PPFx 06-774 06-740-I 505 36 7.13 0.764 06-774 06-716-P 460 17 3.70 0.398 06-771 06-719-I 147 0 0.00 0.000 06-771 06-708-P 303 0 0.00 0.000 06-735 06-718-I 153 0 0.00 0.000 06-735 06-704-P 10 0 0.00 0.000 zDiploid V. arboreum Sparkleberry, section Batodendron yDiploid V. darrowi, Darrows evergreen blueberry, section Cyanococcus P Florida panhandle race. I Istokpoga race. xPPF = number of plump seeds per pollinated flower. 88

PAGE 89

Table 2-7. Result of intra-speci fic hybridization of diploid V. arboreum clones in 2007. Female Male Flowers (No.) Berries (No.) Fruit set (%) PPFy SPFx 06-770-Az 06-772-A 216 117 54.17 6.099 0.859 06-772-A 06-770-A 511 245 47.95 5.563 0.343 06-774-A 06-773-A 121 4 3.31 0.463 0.099 06-773-A 06-774-A 92 27 29.35 6.775 1.795 06-776-A 06-771-A 71 13 18.31 1.718 0.141 06-771-A 06-776-A 50 25 50.00 5.820 1.166 zA = diploid V. arboreum Sparkleberry, section Batodendron. yPPF = number of plump seeds per pollinated flower. xSPF = number of seedlings per pollinated flower. Table 2-8. Summary of crossi ng results between diploid V. arboreum diploid V. darrowi and diploid V. fuscatum in 2007. Crossz Number of crosses Flowers (No.) Fruit set (%) PPFy SPFx HPFw VA VA 6 1061 33.85b 4.406 NS 0.733 ab -v NS VA VD 5 2217 7.36 c 0.370 0.119 b 0.119 VA VF 3 579 10.65c 0.516 0.010 b 0.010 VD VA 4 2429 63.27ab 11.1101.636 ab 1.132 VD VD 6 1496 74.28a 15.2754.536 a VF VA 3 723 40.16b 4.065 0.815 ab 0.815 zVA = V. arboreum sparkleberry. VD = V. darrowi. VF = V. fuscatum yPPF = number of plump seeds per pollinated flower. xSPF = number of seedlings per pollinated flower. wHPF = number of hybrids per pollinated flower. v = no data. *Similar letters within a column indicates means not significantly different, Tukeys test, =0.05 for SPF. Chi-square, test of independence for fruit set (%), =0.05. NS= indicated means not significantly different within a column, Tukeys test, =0.05. Table 2-9. Weight and s eed count per berry in crosses between diploid V. arboreum diploid V. darrowi and diploid V. fuscatum in 2007. Crossz Number of crosses Berries (No.) Berry weight (g) Large seeds per berry ( ) Small seeds per berry ( ) Total seeds per berry ( ) VA VA 6 120 0.32b 9.95 b 7.91 cd 17.86b VA VD 5 100 0.20c 3.98 b 6.57 d 10.56c VA VF 3 60 0.41a 4.24 b 8.65 cd 12.89bc VD VA 4 80 0.41a 16.97a 19.13 a 36.10a VD VD 6 120 0.43a 21.24a 11.80 bc 33.04a VF VA 3 60 0.12d 17.27a 13.97 b 31.23a zVA = V. arboreum sparkleberry. VD = V. darrowi. VF = V. fuscatum *Similar letters within a column indicates means not significantly different, Tukeys test, =0.05. 89

PAGE 90

Table 2-10. Mean pollen stainabi lity of ten-year-old F-1 ( V. darrowi V. arboreum ) hybrids of the Horticultural Plant Science Unit at Gainesville, Florida. F-1 seedling Sporads (No.) Pollen stainability (%)z 98-146 108 7.87 98-64 96 3.39 98-165 152 1.32 98-53 128 1.56 98-93 127 2.56 98-129 138 9.06 98-115 170 6.32 98-61 112 0.89 98-132 150 2.17 98-95 112 23.66 zPercent stainability was calculated as the number of stained spores per pollen grain (tetrad, triad, dyad and monad), averaged by the to tal number of pollen grains counted. Table 2-11. Mean pollen stainabi lity of ten-year-old F-1 ( V. darrowi V. arboreum ) hybrids and V. darrowi and V. arboreum clones. Taxa Plants (No.) Pollen stainability (%) V. darrowiFlorida panhandle race 8 97.42 a V. darrowi-Istokpoga race 15 94.26 a V. arboreum 3 92.23 a F-1 ( V. darrowi V. arboreum ) hybrids 10 5.88 b *Similar letters within a column indicates means not significantly different, Chi-square, test of independence, =0.05. Table 2-12. Meiotic analysis of PMCs in seedling FL98-146 a V. darrowi V. arboreum hybrid. Late diakenesis and metaphase I PMCs (No.) Configuration 4IV+4II 5IV+ 2II 3IV+6II 2IV+8II 12II 4IV+3II +2I 48 Frequency (%) 20.8 16.7 12.5 12.5 10.4 6.3 Configuration 1IV+9II +2I 3IV+1III +4II+1I 3IV+3II +5I 2IV+4III +2II 2IV+7II +3I Otherz Frequency (%) 6.3 2.1 2.1 2.1 2.1 6.3 Anaphase I and telophase I Assortment 12-12 12-11 12-10 11-11 11-10 11-10 + 3 laggards 18 Frequency (%) 22.2 38.9 5.6 16.7 11.1 5.6 Anaphase II and telophase II Assortment 10-10-11-12 9-10-11-11 2 zOther configurations included 1IV+10II 2III+7II+4I, each with one PMC. 90

PAGE 91

Table 2-13. Meiotic analysis of PMCs in seedling FL98-132 a V. darrowi V. arboreum hybrid. Late diakenesis and metaphase I PMCs (No.) Configurations 3IV+6II 4I V+4II 12II 2IV+8II 5IV+2II 4IV+3II +2I 55 Frequency (%) 18.2 12.7 10.9 7.3 5.5 5.5 Configurations 3IV+5II +2I 2IV+7II +1I 4IV+3II +1I 1IV+4III +4II 1IV+10II Otherz Frequency (%) 5.5 5.5 3.6 3.6 3.6 18.2 Anaphase I and telophase I Assortment 12-10 1211 14-12 13-13 13-12 12-12 16 Frequency (%) 37.5 25.0 6.3 6.3 6.3 6.3 Assortment 10-10 9-9 Frequency (%) 6.3 6.3 Anaphase II and telophase II Assortment 12-1111-10 11-1010-10 13-12-1111 11-11-1110 13-1111-10 13-1110-10 10 Frequency (%) 20.0 20.0 10.0 10.0 10.0 10.0 Assortments 11-1110-10 23-117 Frequency (%) 10.0 10.0 zOther configurations included 5IV+4I, 4IV+4II+1I, 4IV+3II, 3IV+5II, 2IV+3III+3II+2I, 2IV+1III+6II+1I, 2IV+7II+2I, 1IV+5III+2II+1I, 1IV+4III+3II+2I, 1IV+9II+2I, each with one PMC. Table 2-14. Meiotic analysis of PMCs in seedling FL98-129 a V. darrowi V. arboreum hybrid. Late diakenesis and metaphase I PMCs (No.) Configurations 6II+12I 4II+16I 3III+2II +11I 3IV+2II +8I 2IV+2II +12I 1IV+1III+ 2II+13I 55 Frequency (%) 7.3 7.3 5.5 3.6 3.6 3.6 Configurations 1IV+2II +16I 1IV+20I 1III+4II +13I 8II+8I 7II+10I Otherz Frequency (%) 3.6 3.6 3.6 3.6 3.6 50.9 Anaphase I and Telophase I Assortment 11-10 16-15 11-11 10-9 + 2 laggards 5 Frequency (%) 40.0 20.0 20.0 20.0 zOther configurations included 1II+22I with two PMCs, and 4IV+1II+6I, 3IV+5II+2I, 3IV+4II+4I, 3IV+3II+6I, 3IV+12I, 2IV+2III+3II+4I, 2IV+2III+9I, 2IV+1III+13I, 2IV+7II+2I, 2IV+5II+6I, 1IV+2III+2II+10I, 1IV+1III+8II+1I, 1IV+1III+4II+9I, 1IV+10II, 1IV+5II+10I, 3III+1II+13I, 2III+5II+8I, 2III+2II+14I, 2III+1II+16I, 12II, 11II+2I, 9II+6I, 5II+14I, 3II+18I, 2II+20I, 24I, each with one PMC. 91

PAGE 92

Table 2-15. Meiotic analysis of PMCs in seedling FL98-115 a V. darrowi V. arboreum hybrid. Late diakenesis and metaphase I PMCs (No.) Configuration 12II 3IV+6II 3IV+5II +2I 1IV+10II 4IV+3II +2I 2IV+2III +5II 42 Frequency (%) 31.0 7.1 7.1 7.1 4.8 4.8 Configuration 2IV+7I I+2I 1IV+4III +2II+4I 4IV+4II 4IV+2II+ 4I 3IV+4II I Otherz Frequency (%) 4.8 4.8 2.4 2.4 2.4 21.4 Anaphase I and telophase I Assorment 12-11 12-10 11-11 11-10 12-12 10-10 37 Frequency (%) 43.2 10.8 10.8 10.8 8.1 5.4 Assorment 13-12 13-11 15-11 12-9 Frequency (%) 2.7 2.7 2.7 2.7 Anaphase II and telophase II Assorments 12-12-11-10 1212-11-9 12-10-10-10 + 1 E.C y 11-10-10-9 4 Frequency (%) 25.0 25.0 25.0 25.0 zOther configurations included 3IV+2III+6I, 2IV+1III+6II+1I, 2IV+8II, 1IV+4III+4II, 1IV+4III+3II+2I, 1IV+3III+5II+1I, 1IV+2III+6II+2I, 3III+15I, 7II+10I, each with one PMC. y E.C. = excluded chromatid. Table 2-16. Meiotic analysis of PMCs in seedling FL98-95 a V. darrowi V. arboreum hybrid. Late diakenesis and metaphase I PMCs (No.) Configuration 4IV+4II 5IV+2II 12II 3IV+6II 3IV+1III +4II+1I 3IV+1III +2II+5I 68 Frequency (%) 22.1 19.1 10.3 5.9 4.4 4.4 Configuration 3IV+5II +2I 4IV+2II +4I 2IV+3III +3II+1I 1IV+3III +5II+1I 5IV+1III +1I Otherz Frequency (%) 4.4 2.9 2.9 2.9 1.5 19.1 Anaphase I and telophase I Assortment 12-11 1212 12-10 11-10 15-11 14-12 37 Frequency (%) 48.6 27.0 8.1 5.4 2.7 2.7 Assortment 12-10 + 2 laggards 11-10 + 1 laggard Frequency (%) 2.7 2.7 Anaphase II and telophase II Assortment 12-12-11-9 1 zOther configurations included 3IV+2III+2II, 3IV+4II+4I, 2IV+4III+2II, 2IV+2III+5II, 2IV+1III+5II+2I, 2IV+8II, 2IV+7II+2I, 1IV+4III+4II, 1IV+3III+3II+4I, 1IV+8II+4I, 11II+2I, 10II+4I, 9II+5I, each with one PMC. 92

PAGE 93

Table 2-17. Meiotic analysis of PMCs in seedling FL98-93 a V. darrowi V. arboreum hybrid. Late diakenesis and metaphase I PMCs (No.) Configuration 12II 4IV+4II 3IV+6II 3IV+5II +2I 1IV+10II 6IV 47 Frequency (%) 36.2 10.6 8.5 8.5 6.4 2.1 Configuration 5IV+2II 5IV+4I 4IV+2II +4I 4IV+7I 3IV+12I Otherz Frequency (%) 2.1 2.1 2.1 2.1 2.1 17.0 Anaphase I and Telophase I Assortment 12-11 1212 11-11 12-10 11-10 11-9 17 Frequency (%) 41.2 17.6 17.6 11.8 5.9 5.9 Anaphase II and Telophase II Assortment 12-1211-10 13-1010-8 12-1212-10 12-1211-9 + 3 E.C.y 12-1110-10 12-11-10-9 + 1 E.C 17 Frequency (%) 11.8 5.9 5.9 5.9 5.9 5.9 Assortment 12-1111-11 12-1211-7 11-1111-10 12-10-98 11-10-99 + 2 E.C. Otherx Frequency (%) 5.9 5.9 5.9 5.9 5.9 29.4 zOther configurations included 2IV+2III+5II, 2IV+2III+4II+2I, 2IV+8II, 2IV+6II+4I, 1IV+6III+2I, 1IV+4III+2II+4I, 10II+4I, 6II+12I, each with one PMC. yE.C = excluded chromatid. xOther assortments included 11-11-11-9 + 1 E.C., 12-12-9-9 + 3 E.C., 25-12-11, 2211-10, 20-9-9, each with one PMC. 93

PAGE 94

Table 2-18. Meiotic analysis of PMCs in seedling FL98-53 a V. darrowi V. arboreum hybrid. Late diakenesis and metaphase I PMCs (No.) Configuration 4IV+4I I 5IV+2II 3IV+6II 4IV+2II +4I 2IV+8II 5IV+4I 97 Frequency (%) 22.7 18.6 17.5 7.2 5.2 4.1 Configuration 3IV+5II 12II 4IV+2III +2I 4IV+3II +2I 4IV+8I Otherz Frequency (%) 3.1 3.1 2.1 2.1 2.1 12.4 Anaphase I and telophase I Assortment 12-12 12-11 12-10 12-9 11-10 10-10 29 Frequency (%) 17.2 13.8 13.8 6.9 6.9 6.9 Assortment 15-14 14-12 13-12 13-12 + 1 laggard 13-11 Othery Frequency (%) 3.4 3.4 3.4 3.4 3.4 17.2 Anaphase II and telophase II Assortment 12-10-9-9 1 zOther configurations included 1IV+10II with two PMCs, and 3IV+2III+6I, 3IV+5II+2I, 3IV+4II+4I, 3IV+3II+6I, 2IV+5III, 2IV+3III+3II+1I, 2IV+6II+4I, 1IV+1III+5II+7I, 11II+2I, 6II+12, each with one PMC. yOther assortments included 1311 + 1 laggard, 16-11, 12-11 + 1 laggard, 11-11, 11-9, each with one PMC. Table 2-19. Meiotic analysis of PMCs in seedling FL98-61 a V. darrowi V. arboreum hybrid. Late diakenesis and metaphase I PMCs (No.) Configuration 4IV+4II 5IV+2II 4IV+3II +2I 12II 3IV+2III+ 2II+2I 2IV+3III +3II+1I 45 Frequency (%) 20.0 11.1 8.9 8.9 6.7 6.7 Configuration 3IV+6II 2IV+7II +2I 5IV+1II I+2I 4IV+1III +4I 3IV+3II+ 6I Otherz Frequency (%) 4.4 4.4 2.2 2.2 2.2 22.2 Anaphase I and telophase I Assortment 12-11 12-12 11-11 13-12 11-10 + 2 laggards 11-10 29 Frequency (%) 27.6 13.8 10.3 6.9 6.9 6.9 Assortment 15-11 14-13 14-11 12-10 11-10 + 1 laggard Othery Frequency (%) 3.4 3.4 3.4 3.4 3.4 10.3 zOther configurations included 2IV+2III+5II, 2IV+8III, 2IV+5II+6I, 1IV+4III+4II, 1IV+3III+5II+1I, 1IV+2III+5II+4I, 1IV+10II, 1IV+8II+4I, 4III+6II, 3III+5II+5I, each with one PMC. yOther assortments included 11-9, 10-10, 9-9 + 1 laggard, each with one PMC. 94

PAGE 95

Table 2-20. Meiotic analysis of PMCs in seedling FL98-64 a V. darrowi V. arboreum hybrid. Late diakenesis and metaphase I PMCs (No.) Configuration 4IV+4II 5IV+2II 12II 3IV+6II 2IV+3III+ 3II+1I 5IV+4I 94 Frequency (%) 29.8 13.8 11.7 4.3 4.3 3.2 Configuration 3IV+3III +1II+1I 2IV+8II 4IV+3II +2I 2IV+5II I+1I 2IV+1II+ 14I Otherz Frequency (%) 3.2 3.2 2.1 2.1 2.1 20.2 Anaphase I and Telophase I Assortment 11-10 12-11 12-12 10-9 11-11 13-12 65 Frequency (%) 20.0 13.8 9.2 9.2 7.7 6.2 Assortment 12-10 11-9 14-10 12-9 10-8 Other y Frequency (%) 6.2 6.2 3.1 3.1 3.1 12.3 Anaphase II and Telophase II Assortment 11-1010-9 13-11-98 12-1211-11 12-1111-9 11-11-10-9 + 3 E.C.x 11-108-8 10 Frequency (%) 20.0 10.0 10.0 10.0 10.0 10.0 Assortment 9-8-7-7 19-1010 22-9-9 Frequency (%) 10.0 10.0 10.0 zOther configurations included 1IV+10II with two PMCs, and 5IV+1II+2I, 4IV+2III+2I, 4IV+2II+4I, 3IV+2III+3II, 3IV+1III+4II+1I, 3IV+5II+2I, 3IV+4II+4I, 2IV+3III+2II+3I, 2IV+1III+6II, 2IV+1III+3II+6I, 1IV+8II+4I, 1IV+4II+12I, 1IV+2II+16I, 10II+4I, 8II+8I, 7II+10I, 2II+20I, each with one PMC. yOther assortments included 14-11, 14-11 + 1 laggard, 1311, 13-10, 13-9, 12-8, 10-10, 9-8, each with one PMC. xE.C = excluded chromatid. 95

PAGE 96

Table 2-21. Meiotic analysis of PMCs in seedling FL98-165 a V. darrowi V. arboreum hybrid. Late diakenesis and metaphase I PMCs (No.) Configuration 4IV+4II 5IV+2II 5IV+4I 12II 5IV+1III +1I 5IV+1II +2I 32 Frequency (%) 25.0 12.5 9.4 6.3 3.1 3.1 Configuration 4IV+3II +2I 4IV+8I 3IV+3III +1II+1I 3IV+3III +3I 3IV+1III +2II+4I Otherz Frequency (%) 3.1 3.1 3.1 3.1 3.1 25.0 Anaphase I and Telophase I Assortment 12-11 1110 11-11 12-12 13-11 12-10 29 Frequency (%) 34.5 17.2 13.8 10.3 6.9 6.9 Assortment 15-11 10-9 11-9 Frequency (%) 3.4 3.4 3.4 Anaphase II and Telophase II Assortment 13-1212-11 12-1211-11 12-1111-10 13-1211-11 12-1211-10 12-1211-9 11 Frequency (%) 18.2 18.2 18.2 9.1 9.1 9.1 Assortment 12-11-11-11 11-10-10-10 Frequency (%) 9.1 9.1 zOther configurations included 3IV+6II, 3IV+2II+8I, 2IV+4III+2II, 2IV+2III+2II+6I, 2IV+6II+3I, 1IV+3III+5II+1I, 1IV+2III+6II+2I, 10II+4I, each with one PMC. 96

PAGE 97

Figure 2-1. Microphotographs of pollen from two F-1 ( V. darrowi V. arboreum ) hybrids; FL98-165 (left) and FL98-61 (right), 250x. Figure 2-2. Ten-year-old F-1 ( V. darrowi V. arboreum ) hybrid population bushes at University of Florida Horticultural Unit in Gainesvill e, Florida. Plant height approx. 3 m. The subject holding the pole (Dario Chavez) is 1.65 m tall. The plants had not been irrigated or fertilized for the past 5 years. 97

PAGE 98

Figure 2-3. Ten-year-old F-1 ( V. darrowi V. arboreum ) hybrid plant at University of Florida Horticultural Unit in Gainesville, Florida. Heavy flowering in November 2007. 98

PAGE 99

Figure 2-4. One plant from a ten-year-old F-1 ( V. darrowi V. arboreum ) hybrid at the University of Florida Horticultural Unit in Gainesville, Florida. Large number of shoots per plant. The photo shows the lower 1 m of the plant. 99

PAGE 100

c) d) b) a) Figure 2-5.Meiotic analysis at pr ophase I and metaphase I of F-1 ( V. darrowi V. arboreum ) hybrids PMCs. a) FL98-93, prophase I, nor mal: one synezetic knots and nuclear organizing region (NOR) (arrow). b) FL 98-93, prophase I, abnormal: two synezetic knot and two NORs (arrows). c) FL98-129, metaphase I, abnormal: partial desynapsis, 2IV+2II+12I. d) FL98-95, meta phase I, normal: complete synapsis, 12II. 1000x. 100

PAGE 101

c) d) b) a) Figure 2-6. Meiotic analysis at metaphase I, anaphase I and telophase I of F-1 (V. darrowi V. arboreum ) hybrids PMCs. a) FL98-61, metaphase I: 4IV+4II. b) FL98-64, anaphase I. Note lagging chromosomes. c) FL98-115, anaphase I, normal disjunction, 12-12. d) FL98-61, telophase I, abnormal: mi cronuclei present (arrows). 1000x. 101

PAGE 102

c) d) b) a) Figure 2-7.Meiotic analysis at anap hase II and telophase II of F-1 ( V. darrowi V. arboreum ) hybrids PMCs. a) FL98-115, early anaphase II: abnormal asynchrony between the two nuclei. b) FL98-93, early anaphase II, abnormal: misaligne d spindles. c) FL9893, anaphase II, lagging chromatids (arro ws). d) FL98-93,telophase II, normal chromatid disjunction: 12-12-12-11. 1000x. 102

PAGE 103

a) b) Figure 2-8.Meiotic analysis at telophase II of F-1 ( V. darrowi V. arboreum ) hybrids PMCs. a) FL98-93, telophase II, abnormal: misaligned spindles at anaphase II may produce 2n microspores depending on the pollen wall formation, assortment 22-11-10. b) FL98115, telophase II, abnormal: misaligned sp indles at anaphase II may produce 2n microspores depending on the pollen wa ll formation, assortment12-12-11-9. 1000x. 103

PAGE 104

CHAPTER 4 SELF-POLLINATION OF V. DARROWI AND OTHER LOW CHILL BLUEBERRY TAXA Introduction In the genus Vaccinium there are no fundamental steril ity barriers between homoploid members of the same phyletic se ction (subgenus). There are dipl oid, tetraploid, and hexaploid species in the genus. Partial to complete se lf incompatibility and inter-fertility between homoploid species has allowed formation of inter-specific hybrid swarms (Camp, 1942). Adaptive physiological and mor phological differences between homoploid species have been created by geographical and ecological disjunction over time Even the more primitive members in Vaccinium have undergone evolutionary modifications (Camp, 1942). Coville (1921) reported that fruit set after self-pollination in highbush blueberries was less than after cross-pollination. Some effects of cross-pollination versus self-pollination in cultivated blueberry were summarized by Morrow (1943). Other studies in several Vaccinium species gave varying results from self-pollination and cross-po llination experiments. In most cases, partial to complete self incompatibility was present, partic ularly in wild or uni mproved plants. Several variables were used to assess the results of pollination experiments: fruit set, fruit size, days to maturity, number of seeds per berry (Coville, 1921; Merrill, 1936; Bailey, 1938; White and Clark, 1939; Merrill and Johnston, 1940). In seve ral experiments, paired shoots on individual plants were used to compare self -pollination with cross-pollination. In cultivated blueberry, cross-pollination us ually produced earlier ripening berries, and increased berry size and fruit se t. Fewer fully developed seeds were found in self-pollinated berries. Partial to complete self incompatibility, slower pollen tube growth, or collapse of ovules after fertilization could be factors that produced thes e results (Morrow, 1943). 104

PAGE 105

Meader and Darrow (1944) studied the crossing behavior of several rabbiteye varieties. They found differences between fruit set from selfand cross-pollination. Berries from crosspollinated flowers always averaged larger than be rries from self-pollinated flowers. Seed weight per berry for all rabbiteye varieties was greater fo r crossthan for self-pollination. In the same study, several species from different ploidy levels were selfand cross-pollinated. Vaccinium virgatum (4x) was found to be self-ste rile, but gave a fruit set of 1 to 27 per cent when crosspollinated with highbush. Diploid V tenellum Aiton gave a low fruit set of 16 and 29 per cent when selfed, but when crossed with a different clone of the same species, had over 80 per cent fruit set. V. darrowi Camp (2x) gave a fruit set of 15 per cent when selfed. V. darrowi crossed with other diploid species, ( V. pallidum and V. elliottii Chapman), gave a fruit set of 71.5 and 62.2 per cent, respectively. V. darrowi berries formed after pollinating the flowers with V. elliottii pollen ripened earlier than those produced by se lf-pollination. However, V. myrsinites (4x) produced two thirds less fruit set when crosse d with other tetraploid species than it did when self pollinated. Percent fru it set of self-pollinated V. myrsinites averaged 79 per cent. Krebs and Hancock (1990) found that fruit set was significantly greater after crosspollination of V. corymbosum L. cultivars. They proposed that variable selfand crosspollination fertility in cultivated highbush was produced by a continuum of early-acting inbreeding depression (seed abortion), being re gulated by level of z ygotic inbreeding, which depended on the parents that were mated. Reduced self-fertility was attr ibuted to homozygosity for sublethal mutations at loci controlling embryo development, or to loss of heterotic interactions at these loci (Krebs and Hancock, 1991; Hokanson and Hancock, 2000). Earlyacting inbreeding depression was proposed to explain low levels of self-fertility in V. myrtilloides Michaux and V. angustifolium Aiton. 105

PAGE 106

In 1945, Camp described the native Vaccinium species present in Florida: five diploids, five tetraploids, and two hexaploids. From these species, Vaccinium ashei Reade (6x), V. myrsinites Lam. (4x), and V. darrowi Camp (2x), were used as parents to create a highbush blueberry adapted to Florid a (Sharpe and Darrow, 1959). V. ashei Reade (6x), rabbiteye blueberry, crossed with tetraploid northern hi ghbush cultivars produced pentaploid hybrids, which were thought to have no further value as breeding parents (Darrow, et al., 1949; Sharpe, 1953). In addition, V. ashei was crossed with V. darrowi (2x), the plan being to produce tetraploid progeny (Darrow, et al., 1949). Instead, the hybrids were pentaploid due to unreduced gamete production in V. darrowi (Goldy and Lyrene, 1983). V. myrsinites Lam. (4x), Florida evergreen blueberry, was crossed with norther n highbush cultivars (Sha rpe, 1953; Sharpe and Darrow, 1959; Moore, 1965). The F-1 hybrids were described as small, twiggy bushes, with small dark berries. They did not seem promising for use in further breeding. V. darrowi Camp (2x), Darrows evergreen blueberry, was crossed with tetraploid northern highbush cultivars. The progeny was expected to be trip loid and sterile, bu t tetraploid hybrids were selected and backcrossed to highbush cultivars (Sharpe and Darrow, 1959). Unreduced gamete production in V. darrowi and the strong triploid block in Vaccinium permitted this hybridization. In 1986, Lyrene found discrepancies between Camps description of V. darrowi and some populations of V. darrowi in Florida. He found major morphological differences between the Florida panhandle race, the Ocala Forest race, and a highly variable population in south-central Florida (the Lake Istokpoga and Lake Placid areas). Lyrene described three types of V. darrowi in Florida: (1) the Florida panhandle race, a pe tit form with highly glaucous leaves, which matches Camps original description of the speci es; (2) the Ocala Forest race, a tall form with shiny green leaves (glaucous on new growth flushes); and (3) the Istokpoga race, a highly 106

PAGE 107

variable population with short and tall plants. Introgression from a diploid hi ghbush species ( V. fuscatum ) was clearly ocurring in the area. In addition, Lyrene described the presence of an interspecific hybrid swarm between two diploid species, V. darrowi and V. fuscatum in the Ocala National Forest. Plants of the Florida panhandle race of V. darrowi have been never used in breeding. The purpose of this study was to analyze th e crossing behavior of various accessions of V. darrowi and of Floridas other Vaccinium species, and to assess their value in the southern highbush blueberry program. Materials and Methods Hybridization Experiments Vaccinium darrowi selfand cross-pollination The V. darrowi clones available at University of Florid a had been previous ly selected from the wild. Lyrene collected the V. darrowi clones used in this experiment during April and May 2003. Softwood cuttings from the selected plants fr om the wild were rooted in a 1:1 mixture of sphagnum peat and perlite. Rooted cuttings were transplanted to 4-liter pots containing sphagnum peat and perlite (1:1). In January 2007, six V. darrowi clones were selected as parents to compare selfand crosscompatibility wi thin the species (Table 3-1 and Table 1-3, respectively). Two clones were from the Florid a panhandle race and four from the Istokpoga race. In February 2007, each plant was divided into two sections to compare selfand crosspollination (Morrow, 1943). All flowers that had opened previously were removed. Flowers were emasculated before pollination. For self-pollinati on experiments, flowers were pollinated with pollen obtained from the same plant. For cross-pollination, pollen from a different V. darrowi plant was used. Female-male plant combinati ons were randomly assigned. Pollen from the selected male parent was collected on the thum bnail to be used during pollination. To pollinate, 107

PAGE 108

the stigmas of the emasculated flowers were to uched by the thumbnail that contained the pollen previously collected. On each plant, approximately, 250 flowers were self-pollinated and 250 flowers were cross-pollinated. Berr ies were harvested when fully ripe. Berry weight and number of seeds per berry were determined for the first twenty berries ripe from each cross. Seeds were classified as plump or shriveled. Seeds from additional berries were removed using a food blender. Seeds were washed free of pulp and sk ins with water, dried on a bench top at room temperature, and stored in coin envelopes at 5C. Pollination-to-ripening interval and average pollination-to-ripening interval were determined af ter crossand selfpol lination. Pollination-toripening interval for each cross was calculated as the number of days from the first day of pollination to the first day of berry ripening. Average pollination-to-r ipening interval was calculated as the number of days from the mean day of pollination to the mean day of harvesting. Seeds were planted in November 2007 on the top layer of 4-liter pots of sphagnum peat. Pots were kept in a greenhouse with intermittent mist for 2-3 months until germination was completed. Seedlings for each cross were counted. Fruit set percentage, berry weight, number of seeds per berry, number of plump seeds per polli nated flower, pollination-to-ripening interval, average pollination-to-ripening interval and num ber of seedlings per pollinated flower were determined for selfand cr oss-pollinated flowers. In January 2008, in a study similar to that of 2007, five-year-old plants of four V. darrowi Istokpoga clones collected in 2003 were selected as parents for self-pollin ation experiments from germplasm maintained in 8-liter pots in Gaines ville, Florida (Table 3-2). These clones were evergreen and were not chilled. Pollination, frui t harvesting, seed extraction and seed planting were as described above. Approximately, 200 flow ers were self-pollinate d per plant. Fruit set percentage, berry weight, number of seeds pe r berry, number of plump seeds per pollinated 108

PAGE 109

flower, pollination-to-ripening interval and av erage pollination-to-ripe ning interval were determined for self-pollinated flowers. Means for fruit set percentage were separate d using Chi-square t est of independence, with significance level 5%. Means for berry weig ht, number of seeds per berry, number of plump seeds per pollinated flower, number of seedlings per pollinated flower, pollination-to-ripening interval and average pollination-to-ripening inte rval for the different treatments were separated using least square means by Tukeys test, with significance level 5%. Data were subjected to ANOVA by the PROC GLM and PROC FREQ procedur es of SAS (Statistical Analysis System Version 9.1, SAS Institute, Cary, NC). Vaccinium arboreum selfand cross-pollination Lyrene collected open-pollinated berries of V. arboreum in 2003 from the Jennings State Forest, in Clay County, Florida. The seeds fr om these berries were removed using a food blender. The seeds were washed with water, drie d, and stored in coin envelopes at 5C. The seeds were planted in November 2005 on the top layer of 4-liter pots of sphagnum peat. The pots were maintained in a greenhouse in Gainesville, Fl orida. The seedlings were transplanted to a high density nursery at University of Florida Plant Science Unit in Citra, Florida, in May 2006. In November 2006, ten to twenty plants were dug with a root-ball of soil a nd fitted to 8-liter pots using sphagnum peat. The pots were placed in a greenhouse in Gainesville, Florida. In January 2007, two-yea r-old plants of six V. arboreum clones were selected as parents for cross-pollination experiments. Thes e plants were selected from various V. arboreum plants available in a greenhouse in Gain esville, Florida, in 2006. All or iginated from seed gathered from the forest in northeast Florida. In February 2007, all flowers that had previously opened were removed. These plants were divided into thr ee sections to compare crosspollinations with V. darrowi V. fuscatum and V. arboreum. Pollination experiments with other Vaccinium species 109

PAGE 110

were described on chapter II. For cr oss-pollination experiments within V. arboreum new flowers were emasculated and cross-pollinated with pollen from a different V. arboreum plant (Table 27). Pollen was collected on the thumbnail and transferred by touching the stigma to the thumbnail. Approximately, 150 flowers were crosspollinated per plant. Berries were harvested when fully ripe. Berry weights and number of seed s per berry were measured for the first twenty berries that were harvested. Pollination-to-ripen ing interval and average pollination-to-ripening interval were used to compare crossing behavior. Pollination-to-ripening interval measured the number of days from the first day of pollination to the first day of berry ripening. Average pollination-to-ripening interval measured the number of days from the mean day of pollination to the mean day of harvesting. Seeds from additiona l berries were removed using a food blender. Seed were planted as described above. Number of seedlings per cross were calculated after germination was completed. In January 2008, three-year-o ld potted plants of three V. arboreum clones were selected for use in self-pollination experiments (Table 3-3) Pollination, fruit harvesting, seed extraction and seed planting were as described above, the only difference being that the flowers were pollinated with pollen obtained from the same plant. Appr oximately, 200 flowers were self-pollinated per plant. The crossing behavior of V. arboreum was assessed by fruit set percentage, number of seeds per berry, berry weight, pollination-to-ripening interval, average pollination-to-ripening interval, number of plump seeds per pollinated flower and number of seedlings per pollinated flower. Means for fruit set percentage were sepa rated using Chi-square test of independence, with significance level 5%. Means for berry weig ht, number of seeds per berry, number of plump seeds per pollinated flower, number of seedlings per pollinated flower, pollination-to-ripening 110

PAGE 111

interval and average pollination-to-ripening inte rval for the different treatments were separated using least square means by Tukeys test, with significance level 5%. Data were subjected to ANOVA by the PROC GLM and PROC FREQ procedur es of SAS (Statistical Analysis System Version 9.1, SAS Institute, Cary, NC). Southern highbush cultivars ( V. corymbosum ) selfand cross-pollination In November 2006, two-year-old plants of five tetraploid southern highbush selections from the University of Florid a breeding program were dug from a field nursery and potted in 8liter pots of sphagnum peat. These plants were se lected from the field based on homogeneity of size and structure to be used in cross-pollina tion experiments. Each plant was a different genotype of southern highbush, and they are here considered V. corymbosum even when other species are included in the ge netic background (Muoz and Lyre ne, 1984b). The plants were kept in a cooler at 5C with no light until February 2007, when they were brought into a beeproof greenhouse. Open flowers from the plants were removed to avoid undesired pollination. Flowers were emasculated before pollination. Cross-pollination between southern highbush cultivars was tested (Table 1-2). Pollination wa s done as described previously. Approximately 150 flowers were cross-pollinated per plant. Poll en used on each plant was from a different genotype randomly selected. Berries were harvested when fully ripe. For the first twenty berries that were harvested, berry weights and number of seeds per berry were measured. Pollination-toripening interval and average pol lination-to-ripening interval were calculated for each cross. Seeds from the remaining berries were removed using a food blender. Seed planting was as described above in V. darrowi crosses. In November 2007, five southern highbush cultivars ( V. corymbosum ) were selected for self-pollination experiments (Table 3-4). They we re dug with a root ball and transplanted to 8liter pots of sphagnum peat from field nurseries at University of Florid a Plant Science Unit in 111

PAGE 112

Citra, Florida. These clones were kept in a walk -in cooler at 5C with no light until they were brought into a greenhouse in Gaines ville, Florida, in February 2008. Pollination, fruit harvesting, seed extraction and seed planting were as desc ribed above. Pollen used on each plant was from the same plant. Approximately, 100 flowers were self-pollinated per plant. Means for fruit set percentage were separate d using Chi-square t est of independence, with significance level 5%. Means for berry weig ht, number of seeds per berry, number of plump seeds per pollinated flower, number of seedlings per pollinated flower, pollination-to-ripening interval and average pollination-to-ripening inte rval for the different treatments were separated using least square means by Tukeys test, with significance level 5%. Data were subjected to ANOVA by the PROC GLM and PROC FREQ procedur es of SAS (Statistical Analysis System Version 9.1, SAS Institute, Cary, NC). Vaccinium fuscatum self-pollination experiments Camp (1945) described V. fuscatum as usually crown-forming, or sometimes in small colonies, 1.5 to 3 m high. Leaves coriaceous to evergreen to sub-persistent, 1.5 to 2.5 cm wide, 3.5 to 5 cm long. Camp gave the range of V. fuscatum as southern Georgia, south through Florida, to just north of Lake Okeechobee; us ually in sandy flatwoods and bottomlands, or along streams and around lakes, frequent in cut-over land and abandoned areas. Camp was not sure of the chromosome complement of V. fuscatum but believed that most forms, at least in the northern part of the range, were tetraploid. Si nce Camp wrote, the taxon he referred to as V. fuscatum has been found to contain both tetraploid and diploid elements. All plants collected from south of Alachua County, Florida, includin g those studied here, have been found to be diploid. Wild V. fuscatum clones were selected from near Da venport, Florida during April and May 2006 by P. Lyrene and K. Hummer. Softwood cuttings from the selected plants were rooted in a 112

PAGE 113

1:1 mixture of sphagnum peat and perlite. Th e cuttings were kept for 2-3 months under intermittent mist and shade cloth in a greenhouse in Gainesville, Florida. The rooted cuttings were transferred to 4-liter pots. V. fuscatum clones were evergreen and had no chilling. They were maintained in a greenhouse in Gainesville, Florida. In January 2008, two-year-old plants of three diploid V. fuscatum clones were selected as parents for self-pollination experiments (Table 3-5). Twenty to 200 flowers on each plant were self-pollinated using methods previously described. Berry harvest, seed extraction and seed planting were as described before. Fruit set percentage, and number of plump seeds per pollinated flower were used to study the different pollination treatments. Hybrids (V. darrowi V. corymbosum ) selfand cross-pollination One F-1 hybrid from crosses made in 2004 and one F-1 hybrid from crosses made in 2005 were dug from the field nursery at University of Florida Plant Science Unit in Citra, Florida in November 2005 and November 2006, respectively. The F-1 hybrids were potted in 8-liter pots of sphagnum peat and maintained in a greenhouse in Gainesville, Florida. In early December 2006, the F-1 hybrids were placed in a walk-in cooler at 4C with no light. The plants were moved to a greenhouse in February 2007 for cross-pollinatio n experiments (Table 3-6). All open flowers were removed before pollination. Flowers were emasculated before pollination. The stigmas of the flowers were rubbed with the pollen previously collected on the thumbnail. Flowers were pollinated with pollen collected from a different F-1 ( V. darrowi V. corymbosum ) genotype. Approximately, 250 flowers were cross-pollinated pe r plant. Berries were harvested when fully ripe. For the first twenty berries harvested, berry weights and nu mber of seeds per berry were obtained. Pollination-to-ripening interval and av erage pollination-to-ripening interval were calculated for each cross. The seeds from additio nal berries were removed using a food blender. 113

PAGE 114

They were washed with water and dried at r oom temperature on a bench top. Seed planting was as described above in V. darrowi crosses. In February 2008, seventeen F-1 ( V. darrowi V. corymbosum ) hybrids from crosses made in 2006 were selected as parents for self-pollinat ion experiments (Table 3-7). They were dug and transplanted to 8-liter po ts of sphagnum peat from field nurseri es at University of Florida Plant Science Unit in Citra, Florida. These clones we re evergreen and were chilled in the field. The plants were brought into a greenhouse in Gainesv ille, Florida. They we re divided into two sections to compare selfand cross-pollination (cross-pollination experiments will be described in chapter IV). For self-pollination experiment s, pollen used on each plant was from the same plant. Twenty to 250 flowers on each plant were self-pollinated using methods previously described. Fruit harvest, seed extraction a nd seed planting were as described above. Means for fruit set percentage were separate d using Chi-square t est of independence, with significance level 5%. Means for berry weig ht, number of seeds per berry, number of plump seeds per pollinated flower, number of seedlings per pollinated flower, pollination-to-ripening interval and average pollination-to-ripening inte rval for the different treatments were separated using least square means by Tukeys test, with significance level 5%. Data were subjected to ANOVA by the PROC GLM and PROC FREQ procedur es of SAS (Statistical Analysis System Version 9.1, SAS Institute, Cary, NC). Results and Discussion Hybridization Experiments Vaccinium darrowi selfand cross-pollination Fruit set from self-pollination of V. darrowi in 2007 and 2008 was less than 43% (Table 31 and Table 3-2). This compared with greater than 90% fruit set fo r cross-pollination of V. darrowi in 2007 (Table 1-3). Coville (1937) described his first attempts to self-pollinated 114

PAGE 115

highbush cultivars as a failure. Self-pollinated plan ts did not set fruit. Few berries ripened, and those that ripened contained few seed. In crosspollinations among different plants of the same species, he obtained many berries with numerous seeds. Number of plump seeds per pollinated flower (PPF) after self-pollination of V. darrowi in 2007 and 2008 ranged from 0.014 to 3.316, depending on the cl one that was selfed (Table 3-1 and Table 3-2). Number of seedlings per pollinated flower (SPF) after selfing these clones in 2007 ranged from 0.015 to 0.323 (Table 3-1) By contrast, cross-pollinated V. darrowi gave PPF ranging from 3.393 to 37.959 and SPF ranging from 1.232 to 9.828 (Table 1-3). Morrow (1943) found that after self-pollinati on, the number of fully developed seeds can be reduced by partial or complete self-incompatib ility, reduction in the poll en tube growth or the collapse of ovules after fertilization. In another study, V. darrowi gave low fruit set when selfed (15%), but high fruit set when cross-pollinated with other diploid species, V. elliottii and V. pallidum (62.2% and 71.5%, respective ly) (Meader and Darrow, 1944). Mean fruit set for self-pollination of V. darrowi in 2007 and 2008 was significantly lower than fruit set for cross-pollination of V. darrowi in 2007 (P<0.05) (Table 3-8). Pollination-toripening interval and average pollin ation-to-ripening interval for V. darrowi crossand selfpollination experiments were not significantly different (P<0.05). Meader and Darrow (1944) found similar results for V. darrowi selfand cross-pollination expe riments. Their berries ripened at the same time regardless of the different crosses that were made. SPF of V. darrowi after selfpollination was lower than after cross-pollina tion in 2007 (20x higher than self-pollination experiments, P<0.05) (Table 3-8). Simila r results in cultivated blueberry ( V. corymbosum ) were attributed to post-zygotic events, concluding wi th abortion of most self seeds (Krebs and Hancock, 1990). 115

PAGE 116

Mean berry weight of V. darrowi was lower after self-pol lination in 2007 and 2008 than after cross-pollination in 2007 (P<0.05) (Table 3-9) Cross-pollination also increased berry size of three northern highbush cult ivars, Scammel, Weymouth and Dixi (Morrow, 1943). In ten rabbiteye varieties, without exception, crossed berries were larger than selfed berries. In the same study, V. tenellum and V. darrowi cross-pollinated berries were larg er than self-pollinated berries (Meader and Darrow, 1944). The average number of seeds per berry was higher when V. darrowi clones were cross-pollinated compared with self -pollination (P<0.05) (Tab le 3-9). Berry weight and number of seeds per berry were directly asso ciated. Larger berries had more seed (Table 39). For PPF and SPF, V. darrowi self-pollination gave lower values than cross-pollination (P<0.05) (Table 3-9). Reduction in PPF and SPF may have been due to embryo abortion after self-pollination and increase in the number of shriveled seeds (s eeds lacking embryos). In our study, few plants were produced from self-polli nated seeds. Krebs and Hancock (1990) found that few plants were produced from self-pollination seeds of V. corymbosum due to an increased homozygosity of deleterious alleles at loci that are critical to embryo vigor and survival. Coville (1937) found that self-pollination of highbush cultiv ars resulted in few berries. Most seeds were abnormal and lacked embryos, and few useful plants were obtained from self-pollination. Vaccinium arboreum selfand cross-pollination Selfand cross-pollination experiments of V. arboreum are described in Table 3-3 and Table 2-7, respectively. Fruit set after cros s-pollination ( V. arboreum V. arboreum) in 2007 ranged from 3.31% to 54.17% compared with 0% set after se lf-pollination of V. arboreum in 2008. PPF and SPF after V. arboreum cross-pollination ranged from 0.463 to 6.775 and 0.099 to 1.795 respectively, compared with zero values for V. arboreum self-pollination. Self-pollinated V. arboreum flowers dropped a few weeks after po llination. Fruit set, PPF and SPF of V. 116

PAGE 117

arboreum differed substantially after selfand cross-pollination (T able 3-10). Complete selfincompatibility was found in V. arboreum It was noticed that several V. arboreum clones that flowered in a greenhouse in Gainesville, Florida, without access to pollination, set no berries. However, V. corymbosum V. darrowi V. fuscatum and F-1 ( V. darrowi V. fuscatum ) natural hybrids, under the same conditions set a few berries (unpublished data). Southern highbush cultivars ( V. corymbosum ) selfand cross-pollination Fruit set after cross-pollination experiments of southern highbush ranged from 37.10% to 94.26% (Table 1-2), and for selfpollination ranged from 9.77% to 74.07% (Table 3-4). PPF for cross-pollination ranged from 3.079 to 11.500 compared to self-pol lination experiments that ranged from 0.519 to 8.130. Variation for fruit set a nd PPF was attributed to post-zygotic events that caused abortion of most self seeds (K rebs and Hancock, 1990; Krebs and Hancock, 1991). Fruit set, pollination-to-ripening interval, average pollination-to-ripening interval, berry weight, number of large seeds per berry, number of small seeds per berry, total number of seeds per berry and number of plump seeds per berr y, were different betw een crossand selfpollination for V. corymbosum cultivars (Table 3-11 and Table 3-12). Similar results were described by Vander Kloet and Lyrene (1987). They found that intraand interpopulation crosses in V. corymbosum were equally fertile, and self-pollination resulted in reductions in all fertility parameters. Fruit set was 2x higher when V. corymbosum was cross-pollinated than when it was selfpollinated (P<0.05). Number of large seeds per berry and total number of seeds per berry were higher for cross-pollination compared with self -pollination (P<0.05). Cross-pollinated berries had fewer small seeds compared to self-pollin ated berries (P<0.05). Pollination-to-ripening 117

PAGE 118

interval and average pollinati on-to-ripening interval were lo nger when southern highbush cultivars were self-pollinated than when they were cross-pollinated. Vaccinium fuscatum self-pollination experiments Two of the three V. fuscatum clones had a fruit set over 50% after self-pollination (Table 3-5). Variation in fruit set after self-pollination was similar to that of southern highbush cultivars. PPF increased when fruit set increased. In other studies, V. fuscatum cross-pollinated with diploid V. arboreum (Table 2-3 and Table 2-4) and tetraploid V. corymbosum (Table 1-7), produced equal or less fruit set and PPF than V. fuscatum self-pollinated. Hybrids ( V. darrowi V. corymbosum ) selfand cross-pollinated Fruit set after inter-pollination of F-1 hybrids was 59.77% and 98.16% for two crosses made in 2007 (Table 3-6). Fruit set for self-pol lination of F-1 hybrids in 2008 ranged from 0% to 90.31% (Table 3-7). PPF for cross-pollinati on ranged from 8.694 to 21.897, and for selfpollination ranged from 0.000 to 12.221. The F-1 hybrid s used in these pollination experiments had different ranges of pollen stainability as de scribed in chapter I (Fi gure 1-7). Reduced pollen stainability in some of the F-1 hybr ids may have been due to triploidy. Fruit set, pollination-to-ripening interval, average pollination-to-ripening interval, berry weight, total number of seeds per berry, and PPF of F-1 (V. darrowi V. corymbosum ) hybrids, were different for crossand se lf-pollination (Table 3-13 and Tabl e 3-14). The results for the F-1 hybrids were similar to those for th e southern highbush cultivars and V. darrowi clones. Conclusions Fruit set of V. darrowi V. corymbosum V. arboreum and F-1 ( V. darrowi V. arboreum ) hybrids after cross-pollination was higher than after self-pollination. V. arboreum was selfsterile. Different degrees of fruit set and partial to complete self -incompatibility were present in V. darrowi V. corymbosum and their tetraploid F-1 hybrids. Pollination-to-ripening interval and 118

PAGE 119

average pollination-to-ripening interval of southern highbush cultivars and F-1 (V. darrowi V. corymbosum ) hybrids were lower for cro ss-pollination compared w ith self-pollination. No difference was found for pollination-to-ripening in terval and average pollination-to-ripening interval between crossand self-pollinated V. darrowi clones. Average berry weight, number of seeds per berry, and number of plump seeds per pollinated flower of V. darrowi V. corymbosum and F-1 ( V. darrowi V. corymbosum ) hybrids was higher when these species were cross-pollinated than self-pollinated. Differences in fruit set, berry weight, pollination-to-ripening interval, aver age pollination-to-ripening interval, number of seeds per berry, and number of plump seeds per be rry in these experiments were similar to those previously found in Vaccinium 119

PAGE 120

Table 3-1. Results of self-pollinating six V. darrowi clones in 2007. Female Male Flowers (No.) Berries (No.) Fruit set (%) PPF y SPFx 03-421-Pz 03-421-P 315 25 7.94 0.4060.103 03-423-P 03-423-P 264 75 28.41 2.1550.187 03-412-I 03-412-I 254 109 42.91 1.0730.323 03-404-I 03-404-I 242 11 4.55 0.5540.091 03-418-I 03-418-I 262 6 2.29 0.0570.015 03-405-I 03-405-I 230 95 41.30 3.3160.220 zDiploid V. darrowi, Darrows evergreen blueberry, section Cyanococcus P Florida panhandle race. I Istokpoga race. yPPF = number of plump seeds per pollinated flower. xSPF = number of seedlings per pollinated flower. Table 3-2. Results of self-pollinating four V. darrowi clones in 2008. Crossesz Flowers (No.) Berries (No.) Fruit set (%) PPF y 03-404-I 03-404-I 221 1 0.45 0.014 03-419-I 03-419-I 222 25 11.26 0.275 03-405-I 03-405-I 207 51 24.64 0.321 06-660-I 06-660-I 213 34 15.96 0.249 zDiploid V. darrowi, Darrows evergreen blueberry, section Cyanococcus -I Istokpoga race. yPPF = number of plump seeds per pollinated flower. Table 3-3. Results of self-pollinating three V. arboreum clones in 2008. Crossesz Flowers (No.) Berries (No.) Fruit set (%) PPF y 06-774-A 06-774-A 206 0 0.00 0.000 06-771-A 06-771-A 199 0 0.00 0.000 06-735-A 06-735-A 142 0 0.00 0.000 zA = diploid V. arboreum Sparkleberry, section Batodendron. yPPF = number of plump seeds per pollinated flower. Table 3-4. Results of self-pollinating five southern highbush cultivars in 2008. Crossesz Flowers (No.) Berries (No.) Fruit set (%) PPF y Sweet Crisp-H Sweet Crisp-H 87 26 29.89 2.206 01-25-H 01-25-H 65 22 33.85 1.000 01-243-H 01-243-H 107 53 49.53 3.164 03-293-H 03-293-H 108 80 74.07 8.130 00-34-H 00-34-H 133 13 9.77 0.519 z-H = tetraploid V. corymbosum southern highbush blueberry, section Cyanococcus yPPF = number of plump seeds per pollinated flower. 120

PAGE 121

Table 3-5. Results of self-pollinating three V. fuscatum clones in 2008. Crossesz Flowers (No.) Berries (No.) Fruit set (%) PPF y 06-722-I 06-722-I 434 246 56.68 3.166 06-723-I 06-723-I 237 138 58.23 2.301 06-724-I 06-724-I 241 18 7.47 0.112 zDiploid V. fuscatum Diploid Florida swamp highbush, section Cyanococcus .-I Istokpoga area. yPPF = number of plump seeds per pollinated flower. Table 3-6. Result of cross-pollination experiments between F-1 ( V. darrowi southern highbush cultivars) in 2007. Crossz Flowers (No.) Berries (No.) Fruit set (%) PPF y SPFx 07-111-F1 06-107-F1 272 267 98.16 21.8972.909 06-107-F1 07-111-F1 266 159 59.77 8.694 1.837 zF1 = F-1 hybrid (V. darrowi southern highbush cultivars). yPPF = number of plump seeds per pollinated flower. xSPF = number of seedlings per pollinated flower. Table 3-7. Results of selfpollinating 17 F-1 hybrid ( V. darrowi V. corymbosum ) plants in 2008. Crossesz Flowers (No.) Berries (No.) Fruit set (%) PPF y 84-F1 84-F1 191 47 24.61 0.183 127-F1 127-F1 105 63 60.00 4.832 133-F1 133-F1 149 74 49.66 4.617 89-F1 89-F1 165 112 67.88 1.609 94-F1 94-F1 148 10 6.76 0.122 96-F1 96-F1 83 19 22.89 2.141 26-F1 26-F1 12 8 66.67 3.667 57-F1 57-F1 265 214 80.75 12.211 134-F1 134-F1 258 233 90.31 7.863 24-F1 24-F1 249 93 37.35 0.810 29-F1 29-F1 154 113 73.38 4.859 121-F1 121-F1 208 122 58.65 2.304 132-F1 132-F1 109 77 70.64 8.186 135-F1 135-F1 56 0 0.00 0.000 41-F1 41-F1 149 20 13.42 0.544 136-F1 136-F1 12 0 0.00 0.000 129-F1 129-F1 239 120 50.21 2.202 zF1 = F-1 hybrid (V. darrowi southern highbush cultivars). Ea ch number refers to a different F-1 hybrid plant. yPPF = number of plump seeds per pollinated flower. 121

PAGE 122

Table 3-8. Selfand cross-pollination results in Florida V. darrowi races measured by fruit set (%), pollination-to-ripening in terval (pol-ripe interval), average pol-ripe-interval and number of seedlings per pollinated flower Cross type Race Number of crosses Flowers (No.) Berries (No.) Fruit set (%) Pol-ripe intervalz AvPolripe intervaly Seedlings per pol. flower Self 2007 Panhandle 2 579 50.0 18.2b 90.0a 122.0a 0.15b Istokpoga 4 988 55.3 22.8b 98.7a 111.3a 0.16b Self 2008 Istokpoga 4 863 61.0 13.1b 83.0a 94.7a -x Cross 2007 Panhandle 2 531 217.5 82.5a 90.0a 102.0a 4.18ab Istokpoga 4 965 170.7 70.2a 80.7a 94.1a 4.71a zPol-ripe interval= number of days from first pollinated flower to first ripe berry. yAvPol-ripe= number of days from median pollination date to median ripening date. x = no data. *Similar letters within a column indicates means not significantly different. Tukeys test for pol-ripe interval, avpol-ripe interval and seedlings per pol. flower, =0.05.Chi-square, test of independence for fruit set (%), =0.05. Table 3.9. Results of selfand cross-pollination in Florida V. darrowi races measured by berry weight (g), seeds per berry, plump seeds per pollinated flower and seedlings per pollinated flower. Cross type Number of clones tested Berry weight (g) Seeds per berry ( ) Plump seeds per pollinated flower Seedlings per pollinated flower Self-07 6 0.27b 21.05b 1.26b 0.16b Self-08 4 0.28b 20.67b 0.22b -z Cross-07 6 0.43a 33.04a 15.28a 4.54a z = no data. *Similar letters within a column indicates means not significantly different, Tukeys test, =0.05. Table 3-10. Results of six crosses and three self-pollinations in V. arboreum Each cross involved two unrelated V. arboreum parents. Three different V. arboreum seedlings were self-pollinated. Cross type Number of crosses Flowers (No.) Berries (No.) Fruit set (%) Plump seeds per pol. flower Seedlings per pol. flower Cross 2007 6 1061 366 33.85 4.406 0.733 Self 2008 3 547 0 0.00 -z z = no data. 122

PAGE 123

Table 3-11. Results of selfand cross-pollina tion in Florida southe rn highbush clones and seedlings. Cross type Number of crosses Flowers (No.) Berries (No.) Fruit set (%) Pol-ripe intervalz AvPol-ripe intervaly Seedlings per pol. Flower Cross 2007 74 10168 8741 82.8a* 53.7b 61.1b 17.518x Self 2008 5 500 188 39.4b 75.8a 71.9a -w zPol-ripe interval= number of days from first pollinated flower to first ripe berry. yAvPol-ripe= number of days from median pollination date to median ripening date. xMean value just for five crosses. w = no data. *Similar letters within a column indicates means not significantly different. Tukeys test for pol-ripe interval, avpol-ripe interval and seedlings per pol. flower, =0.05. Chisquare, test of independence for fruit set (%), =0.05. Table 3-12. Results of selfand cross-pollina tion for southern highbush cultivars measured by berry weight (g), number of large seeds, number of small seeds, number of seeds per berry and number of plump seeds per pollinated flower.. Cross type Number of clones tested Berry weight (g) Large seeds ( ) Small seeds ( ) Seeds per berry ( ) Plump seeds per pollinated flower Cross-07 5 1.53b 31.71a 19.92b 51.63a 7.914a Self-08 3 1.69a 7.73b 24.13a 31.86b 3.004b *Similar letters within a column indicates means not significantly different, Tukeys test, =0.05. Table 3-13. Results of two crosses a nd 17 self-pollinations involving F-1 ( V. darrowi V. corymbosum ) hybrids. Each of the 17 self-pollin ations used as di fferent F-1 hybrid. Cross type Number of crosses Flowers (No.) Berries (No.) Fruit set (%) Pol-ripe intervalz AvPol-ripe intervaly Seedlings per pol. Flower Cross 2007 2 538 426 78.9a* 58.5b 71.6a 2.373 Self 2008 17 1794 1325 45.5b 76.8a 80.2a -x zPol-ripe interval= number of days from first pollinated flower to first ripe berry. yAvPol-ripe= number of days from median pollination date to median ripening date. x = no data. *Similar letters within a column indicates means not significantly different. Tukeys test for pol-ripe interval, avpol-ripe interval and seedlings per pol. flower, =0.05. Chi-square, test of independence for fruit set, =0.05. 123

PAGE 124

Table 3-14. Results of selfand cross-pollination involving F-1 ( V. darrowi V. corymbosum ) hybrids measured by berry weight (g), seeds per berry, plump seeds per pollinated flower and seedlings per pollinated flower. Cross type Number of clones tested Berry weight (g) Seeds per berry ( ) Plump seeds per pollinated flower Seedlings per pollinated flower Cross-07 2 1.24a 48.34a 15.29a 2.373 Self-08 17 0.94b 22.93b 3.30b -z z = no data. *Similar letters within a column indicates means not significantly different, Tukeys test, =0.05. 124

PAGE 125

CHAPTER 5 FERTILITY AND PLOIDY OF V. DARROWI SOUTHERN HIGHBUSH F-1 HYBRIDS AND THEIR ABILITY TO BACKCROSS TO TETRAPLOID SOUTHERN HIGHBUSH BLUEBERRY. Introduction Numerous hybridization experime nts have been undertaken to study the in trogression of traits from wild Vaccinium species into highbush blueberry cultiv ars. The first studies were part of a USDA research program, with Dr. Frederic k Coville as principal investigator (Coville, 1937). His hybridization experiments began with wild species. From Covilles work, a cultivated blueberry industry grew steadily after 1930, with continuous in troduction of new cultivars by various federal, state, and priv ate blueberry breeding programs. During the 1920s, the Florida blueberry indus try was based on rabbiteye blueberries ( V. ashei ), and most of the plants in commercial pl antings were selected from the wild. After extensive plantings during the 1920s, Florida acreage did not increase significantly for many years, nor was there significan t improvement in rabbiteye cul tivars, even though rapid progress was being made in developing better highbush cultivars for the northern blueberry industry. Camp (1945) studied the native Vaccinium species of Florida. The USDA in Washington was interested in finding genetic material that could be hybridized with northern blueberry cultivars to improve highbush bl ueberry performance in warm climates. Camp identified two species as having potenti al value as parents: V. myrsinites Lam. (4x), and V. darrowi Camp (2x). These, along with V. ashei (6x), constituted the genetic pool that was used to reduce the chilling requirement of northern highbush cultivars to produce southern highbush (Sharpe, 1953). In 1949, a southern highbush blueberry breeding program began at University of Florida with the collaboration of Shar pe in Florida and Darrow of the USDA in Washington, D.C. (Sharpe and Darrow, 1959). V. myrsinites (4x), which is native in Flor ida as far south as Miami, 125

PAGE 126

was crossed with northern highbush cultivars (4x) Forty seedlings were produced, but these showed little horticultural value wh en grown in Gainesville, Florida. V. ashei (6x) was crossed with V. darrowi (2x), with the expectation that the te traploid progeny could be backcrossed to tetraploid northern highbush cultivars. From about 7500 flowers pollinated in V. darrowi V. ashei crosses, 5 hybrids were obtained. Goldy and Lyrene (1983) later made similar crosses, and the hybrids produced were pentaploid, a result of unreduced gametes from V. darrowi. Northern highbush cultivars (4x) were crossed with V. darrowi (2x). From about 1600 pollinations, 31 tetraploid hybrids were selected. It had been expected that the progeny would be triploid and sterile, but apparently, the production of unreduced gametes by V. darrowi and the presence of the strong triploid block in Vaccinium allowed a few tetraploids to be obtained from this hybridization. Using the original hybrids as pa rents, additional hybridi zations were made in 1958. The tetraploid F-1 hybrids ( V. darrowi northern highbush and reci procals) were crossed with northern highbush. About 10000 seedlings were obtained. Seedlings from these crosses broke dormancy well at Gainesville in 1959, indicating that their chilling requirement had been greatly reduced compared to north ern highbush. Plants had light blue fruit that was intermediate in size between the parents (14 to 16 mm di ameter), and had fairly good quality. The first hybrids showed promise as parental material for further crosses and selection (Sharpe and Darrow, 1959; Moore, 1965). The best selections from the southern highbush blueberry program at Gainesville, Florida all had V. darrowi in their pedigree (Sharpe and Sherman, 1971). Undesired characteristics from V. darrowi (i.e. short, twiggy plant structure, small fruit, long fruit-development period) in the southern highbush genetic pool were el iminated by crossing and selection. Acceptable fruit and plant quality were obtained by 1961. 126

PAGE 127

Many crosses between Vaccinium species have been made since the beginning of blueberry breeding, and the degree of success in producing inter-specific hybrids has been reported for many species combinations In most cases, the inte r-specific hybrids were not maintained until maturity to confirm their hybridity and determine their level of fertility. V. darrowi (2x) was described as being able to hybridize in homoploid and heteroploid combinations with species in section Cyanococcus such as V. tenellum (2x), V. elliottii (2x), V. pallidum (2x), V. atrococcum (2x), V. caesariense (2x), V. fuscatum (2x), V. corymbosum (4x), V. australe (4x), V. ashei (6x) and with species in several other sections, such as V. ovatum (2x) (section Pyxothamnus ), V. stamineum (2x) (section Polycodium ), and V. arboreum (2x) (section Batodendron ) (Darrow and Camp, 1945; Lyrene and Ballington, 1986). V. darrowi has been used in bridging crosses to overcome sterility barriers between tetraploid highbush cultivars and several Vaccinium species. The yield of tetraploid hybrids was less than 0.002 per pollinat ed flower for highbush V. elliottii less than 10% of what is typically obtained from highbush V. darrowi crosses (Lyrene and Sherman, 1983). V. darrowi was crossed with V. elliottii, and the F-1 hybrids were crossed with V. corymbosum (4x) (Meader and Darrow, 1944). The diploid F-1 hybrids were easily crossed with tetraploid southern highbush blueberry (Lyrene and Ballington, 1986). V. darrowi V. stamineum ( Cyanococcus Polycodium ) crosses produced F-1 hybrids fertile enough to produce F-2 and BC-1 progenies in North Carolina (Lyrene and Ballington, 1986). In 1991, Lyrene crossed V. darrowi (diploid Cyanococcus) with V. arboreum (diploid Batodendron ). The F-1 hybrids were not difficult to obtain. The BC-1 progeny, in which tetraploid highbush blueberry was the backcro ss parent, was promising, with upright-growing plants and very open flower clus ters (Brooks and Lyrene, 1995). 127

PAGE 128

Most inter-specific crosses that used V. darrowi as a parent have used V. darrowi clones that did not match Camps description of V. darrowi Camp (1942) stated that V. darrowi ocurred in extensive colonies, 0.15-0.40 m. hi gh, from Louisiana to Florida. All V. darrowi clones that have been used in breeding southern highbush cultivars, most importantly clone Fla. 4B, have been much taller. Lyrene (1986) found mo rphological differences among populations of V. darrowi in Florida V. darrowi clones from the Florida panha ndle region were short stature (0.34-0.70 m. high), those from the Ocala Forest region tall (1.10-1.61 m. high), and clones from the Istokpoga region were highly va riable in height. In the Ist okpoga region, in addition to the parent species, there were hybrid swarms involving V. darrowi and V. fuscatum This type of gene flow between short-stature V. darrowi and V. fuscatum which is 3-4m tall, probably accounts for the tall V. darrowi race in the Ocala National Forest (Lyrene personal communication). The following study was conducted to de termine how various F-1 hybrids ( V. darrowi southern highbush cultivars and reciprocals) pe rform in inter-specific crosses with southern highbush blueberry cultivars, and ho w certain traits of interest se gregate in the F-2 populations. Materials and Methods Fertility of F-1 ( V. darrowi V. corymbosum and Reciprocals) Hybrids Pollen stainability In January 2007, two F-1 ( V. darrowi V. corymbosum ) inter-specific hybrids from crosses made in 2004 and three from crosses ma de in 2005 were selected as parents to be backcrossed to southern highbush cultivars. Plants were grown in Citra, Florida in a high density nursery. These plants were selected because they were the only seedlings from these crosses whose vegetative morphology clearly showed they we re hybrids. Plants were placed in a cooler at 5C with no light on Dec. 20, 2006. In late January 2007, they were moved to a bee-proof 128

PAGE 129

greenhouse. Open flowers were collected to de termine pollen stainability. Pollen from dry flowers was stained with 1% aceto-carmine solution. Microphotographs were taken using a Moticam 1000 1.3MPixel microscope digital camera with the Motic Images Plus Version 2.0ML software, mounted on a phase-contrast Leitz micros cope. Pollen stainability was measured as percentage of well-stained pollen grains in norma l tetrads, triads, dyads and monads compared with the numbers of sporads that had only one, tw o, or three stained polle n grains (each sporad normally has four pollen grains). This result was averaged by the tota l of counted sporads. Sterile pollen grains in water or aceto-carmine appear shrive led, unstained, or abnormal in shape, and have varying degrees of pollen in flation (Dermen, 1940). Abnormal pollen grains may result from chromosome pairing abnormalitie s during meiosis. Goldy and Lyrene (1983) described several meiotic abnormalities in hybrids between V. ashei (6x) and V. darrowi (2x). These abnormalities led to a high pe rcentage of shrunken pollen. Similar studies were made in V. corymbosum (4x) V. elliottii (2x) hybrids by Lyrene and Sherman (1983). Both mitotic and meiotic instabilities reduced th e fertility of these hybrids. In November 2007 and January 2008, flowers from one-hundred-eight hybrids, FL03-421P highbush pollen composite, and one addi tional hybrid, FL03-423-P highbush pollen composite, from crosses made in 2006 were coll ected to measure pollen fertility. Dry pollen from the flowers was stained with 1% aceto-carmi ne stain. Pollen stainaibility was determined as described before. Backcrossing Experiments Hybrids ( V. darrowi V. corymbosum and reciprocals) V. corymbosum In November 2006, three-year-old plants of two F-1 hybrids, FL06-105-F1 (highbush FL02-62 V. darrowi FL03-418-I) and FL06-107-F1 (highbush FL03-67 V. darrowi FL03417-I) and two-year-old plants of three F-1 hybrids, FL07-110-F1 ( V. darrowi FL03-405-I 129

PAGE 130

Emerald), FL07-111-F1 ( V. darrowi FL03-404-I highbush FL0106), and FL07-112-F1 ( V. darrowi FL03-419-I Emerald) were selected as parents from crosses made in 2004 and 2005. Plants were placed in a cooler at 5C with no light on Dec. 20, 2006. In late January 2007, the selected F-1 plants were placed in a bee-proof greenhouse. F-1 hybrids were used as female parents in crosses with southern highbush cultivars or advanced selections. In addition, a threeyear-old F-1 hybrid was crossed as male parent with a southern highbush cultivar (Table 4-1). The southern highbush cultivars used in these crossing experiments are considered V. corymbosum even though several speci es are found in their genetic background (Muoz and Lyrene, 1984b). Pollen from the male parents wa s collected on the thumbnail. Flowers were emasculated before pollination. The stigma of the emasculated flower was touched with the pollen on the thumbnail. The F-1 hybrids were di vided into two sections. One section was pollinated with pollen from southern highbush cultiv ars. The other section was pollinated with pollen from other F-1 hybrids or V. ashei Florida Rose (these experiments will be described in the next section, Table 4-2). Approximately, 250 flowers were pollinated per section for each plant. Berries were harvested when fully ripe. Number of seeds per berry for the first twenty ripe berries were counted and classified as plump or shriveled seeds. Berry weight for these berries was recorded. Seeds from additional berries were extracted using a food blender, washed and dried (Moore, 1965). Number of plump seeds per pollinated flower was determined for each cross. Seeds were sown in November 2007 on the top layer of 4-liter pots of sphagnum peat. The pots were maintained in a greenhouse with intermittent mist for 2-3 months. After germination was complete, the total number of seedlings was counted for each cross. Number of seedlings per 130

PAGE 131

pollinated flower was calculated for each cross. Seedlings were grown in a greenhouse until they reached 1-2 cm high. Then they were transferred to plastic trays, 100 seedlings each, filled with sphagnum peat. All the seedlings from each cross were transferred to trays. In May 2008, when the seedlings were 4-8 cm high, they were plante d in a high density nursery in Citra, Florida, with plant-to-plant spacing of 10 cm and row-to-row spacing of 40 cm. In February 2008, seventeen F-1 ( V. darrowi V. corymbosum ) hybrids from crosses made in 2006 were selected as female parents for cross-pollination experiments with southern highbush advanced selections (Table 4-3). The F-1 plants were dug from field nurseries at University of Florida Plant Science Unit in Citr a, Florida and transplant ed to 8-liter pots of sphagnum peat. These clones were evergreen and were chilled in the field. The plants were brought into a greenhouse in Gainesv ille, Florida. Each plant was divided into two sections to compare crossand self-pollination (self-pollinat ion experiments were described on chapter III, Table 3-7). Thirty to 500 flowers on each plant were cross-pollinated using methods previously described. Fruit harvesting and seed extraction was as described above. Means for fruit set percentage were separate d using Chi-square t est of independence, with significance level 5%. Means for berry weig ht, number of seeds per berry, number of plump seeds per pollinated flower and number of seedlings per pollinated flower for the different treatments were separated using least square me ans by Tukeys test, with significance level 5%. Data were subjected to ANOVA by the PROC GLM and PROC FREQ procedures of SAS (Statistical Analysis System Ve rsion 9.1, SAS Institute, Cary, NC). Hybrids ( V. darrowi V. corymbosum ) V. ashei and reciprocals Two F-1 ( V. darrowi V. corymbosum ) hybrids that resulted from crosses made in 2004 and 2005 had lower pollen fertility, and they were selected as male and female parents to be hybridized with hexaploid V. ashei The idea was that the low-fertility hybrids might be triploid 131

PAGE 132

and might, through unreduced gametes, produce hexa ploid hybrids when crossed with hexaploid V. ashei One F-1 ( V. darrowi V. corymbosum ) hybrid with high pollen fertility, from crosses made in 2005 was also used as male and female parent in crosses with hexaploid V. ashei Florida Rose (Table 4-2). These F-1 plants we re dug from the field and placed in a cooler at 5C with no light on Dec. 20, 2006. In late January 2007, they we re placed in a bee-proof greenhouse. The F-1 hybrid plants were divided into two sections. One section was pollinated with pollen from southern highbush cultivars (a s described before). The other section was pollinated with pollen from other F-1 hybrid or V. ashei Florida Rose. Approximately 250 flowers were pollinated for each section. Fruit harvest, seed extraction, seed planting and seedling identification were as described above. The results of pollination treatments were assessed by fruit set percentage, berry weight, number of seeds per berry, number of plump seed s per pollinated flower and number of seedlings per pollinated flower. Means for berry weight, nu mber of seeds per berry, number of plump seeds per pollinated flower and number of seedlings per pollinated flower for the different treatments were separated using least square means by Tukeys test, with significance level 5%. Means for fruit set percentage were separated using Chi-square test of independence, with significance level 5%. Data we re subjected to ANOVA by the PROC GLM and PROC FREQ procedures of SAS (Statistical Analysis Sy stem Version 9.1, SAS Institute, Cary, NC). Hybrids ( V. darrowi V. corymbosum ) V. darrowi and reciprocals Two of the V. darrowi V. corymbosum hybrids, FL07-111 and FL07-112, from crosses made in 2005, had high pollen fertility and were selected as ma le and female parents to be hybridized with diploid V. darrowi (Table 4-4). The two V. darrowi clones used as male and female parents for all the experiments were known to be high 2n egg producers. The goal was to obtain tetraploid plants with 75% of their genes derived from V. darrowi. The F-1 plants were 132

PAGE 133

dug from the field and placed in a cooler at 5C with no light on Dec. 20, 2006. The V. darrowi clones were evergreen and were not chilled. In late Ja nuary 2007, the selected F-1 plants and the V. darrowi clones, were placed in a bee-proof gr eenhouse. Approximately, 500 flowers were pollinated for each plant. Fruit harvest, seed ex traction, seed planting and seedling identification was as described above. The results of pollin ation treatments were assessed by fruit set percentage, berry weight, number of seeds per berry and number of plump seeds per pollinated flower. Results and Discussion Fertility of F-1 ( V. darrowi V. corymbosum and Reciprocal) Hybrids Pollen stainability Pollen stainability of the F-1 ( V. darrowi V. corymbosum ) hybrids averaged 88.06% for 109 hybrids from crosses in 2006 (Table 4-5). This indicated that most of the hybrids were tetraploid rather than triploid. A few plants ha d low pollen stainability, probably due to triploidy or to meiotic abnormalities due to failures in chromosome pairing (Goldy and Lyrene, 1983; Lyrene and Sherman, 1983). Crosses made in 2004 and 2005, produced two F-1 hybrid that had pollen stainability values lower than 21% (Tab le 4-6). Pollen staina bility of the F-1 ( V. darrowi V. corymbosum ) hybrids from 2006 was highly variable (F igure 1-7). Pollen stainability 0-50% was found in six plants of the F-1 hybrids (5.5%). These plants had shriveled, abnormal pollen shapes which failed to stain with 1% aceto-carmine and had different degrees of pollen inflation as described by Dermen (1940) (Figure 1-8a). Pollen stainability 50.1-100% was found in 103 plants (94.9%) of the F-1 hybrid population (Figure 1-8b). The te traploid F-1 hybrids should be easy to backcross to the tetrap loid gene pool. The triploid F1 hybrids can be backcrossed to hexaploid V. ashei through unreduced gamete production of the triploid plants. In this way, 133

PAGE 134

genes from V. darrowi and V. corymbosum can be moved to the V. ashei gene pool (Ehlenfeldt and Vorsa, 1993). Hybrid FL07-110-F1 ( V. darrowi V. corymbosum ) was found to have a high number of 2n pollen sporads and unstained tetrads when studied under the microscope (Figure 4-1a and Figure 4-1b). Meiotic abnormalities during pollen formation produced the many abnormal tetrads (unstained sporads) and th e 2n pollen sporads. FL07-110-F1 was probably triploid. Backcrossing Experiments Hybrids ( V. darrowi V. corymbosum ) V. corymbosum Backcrossing allowed the introgression of genes from V. darrowi into the southern highbush genetic pool. Fruit set from backcros sing ranged from 3.15% to 97.78% (Table 4-1). Low fruit set, 3.15% and 36.30%, were for crosses when the F-1 hybrids FL07-110 and FL06105 were used as seed parents. These two F-1 hybrids had low pollen stai nability (as described before). The other four backcrosses had fruit set from 49.29% to 97.78%. The F-1 hybrids used in these backcrosses had normal pollen stainabi lity. Number of plump seeds per pollinated flower (PPF) and number of seedlings per pollinated flower (SPF) were positively correlated with pollen stainability of the F-1 hybrids (Table 4-1). Fruit set values for F-1 ( V. darrowi V. corymbosum ) hybrids pollinated with pollen from highbush cultivars in 2008 ranged from 0.32% to 86.85% (Table 4-3). Pollen stainability for the F-1 hybrids was highly variable. These resu lts were similar to those in 2007. Fruit set in crosses between F-1 (V. darrowi V. corymbosum ) hybrids and southern highbush cultivars was not different from that in V. darrowi V. darrowi and V. corymbosum V. corymbosum crosses (P<0.05) (Table 4-7). Th e diploid tetraploid crosses, V. darrowi V. corymbosum and reciprocals, as expected, had much lower fruit set than the homoploid crosses (Table 4-7). These results are consistent with the findings of previous studies (Table 1-14). 134

PAGE 135

Even thought PPF ranged from 0.465 for tetraploid highbush diploid V. darrowi to 15.295 for the F1 F1 and V. corymbosum V. corymbosum crosses, the means were not significantly different due to lo w replication and high error vari ance (Table 4-7). SPF was higher for highbush highbush than for any other cross, but F1 F1 gave enough seedlings to grow large progeny populations if desired. Berry weight was highly variable depending on th e female parent (Table 4-8). The largest berries were produced by highbush parents and the smallest by V. darrowi Berry weight of the F-1 hybrids was intermediate, and probably too sm all for a commercial blueberry cultivar. Large seeds per berry were highest for highbush highbus h crosses, lowest for the interploid crosses, and intermediate for crosses between highbush a nd the F-1 hybrids. Seed production in highbush F-1 hybrids was sufficient to produce large backcross populations. Hybrids ( V. darrowi V. corymbosum ) V. ashei and reciprocals Cross-pollination experiments of the F-1 ( V. darrowi V. corymbosum ) hybrids V. ashei and reciprocal crosses are described in Table 4-2. Pollen stainability was high for FL07-112-F1 and lower for FL06-105-F1 and FL07-110-F1. Fruit set was 16.30% for FL06-105-F1 and 1.79% for FL07-110-F1 when crossed as seed parents with V. ashei FL07-112-F1 fruit set was 33.20%. PPF and SPF were proportional to fruit set. SPF was very low for all of these crosses, but highest for the cross with FL07-112. It is not know whether these V. darrowi highbush hybrids were triploid, tetraploid, or aneuploid. Some of these hybrids crossed with V. ashei could produce hexaploid, pentaploid, or near pentaploid progeny. Lyrene (1997) described several crosses between diploid V. darrowi and tetraploid V. corymbosum made in the southern highbush blueberry breeding program at University of Florida. These crosses produced some hybrids that had reduced male and female fertility, and were probably triploids. Triploid plants were used in crosses with hexaploid blueberries to 135

PAGE 136

produce fertile hexaploid hybrids that may be useful in rabbiteye breeding (Ehlenfeldt and Vorsa, 1993). Hybrids ( V. darrowi V. corymbosum ) V. darrowi and reciprocals Previous backcrosses of F-1 V. darrowi V. corymbosum hybrids have been made to southern highbush, with the intention of obtaining comme rcial cultivars with highbush characteristics. The objective of this study was to produce tetraploids wi th most of the genes from V. darrowi The results of F-1 ( V. darrowi V. corymbosum ) hybrids V. darrowi and reciprocals are described in Table 4-4. V. darrowi clones used in these crosses produced high frequency of 2n egg when crossed with southern highbush cultiv ars. This had been determined in highbush V. darrowi crosses made in 2006 and 2007. The F-1 hybrid s had high pollen stainability and were probably tetraploids. Fruit set wa s lower when F-1 hybrids were us ed as seed parent compared with their reciprocals. In crease of fruit setting for V. darrowi F-1 hybrids may have been due to 2n egg production by V. darrowi or might indicate a high er level of parthenocarpy. The preliminary results indicate that backcrosses to V. darrowi can be successful. However, the past experiments, PPF had not b een a good estimate of the number of hybrid seedlings obtained, and until the seeds are plan ted (Nov. 2008) and the seedlings are examined, we will not know how many tetraploid b ackcross hybrids have been produced. Fruit set and PPF of F-1 hybrids V. darrowi and reciprocals were not different than from southern highbush V. darrowi and reciprocals (P<0.05) (Table 4-9). Fruit set and PPF for the homoploid crosses, F-1 ( V. darrowi V. corymbosum ) V. corymbosum were higher than for the heteroploid crosses, F-1 (4x) V. darrowi (2x) and reciprocals, and V. corymbosum (4x) V. darrowi (2x) and reciprocal crosses. 136

PAGE 137

Mean berry weight was highly variable depending on the fema le parent (Table 4-10). The number of large seeds per berry in the homoploid crosses, F-1 ( V. darrowi V. corymbosum ) (4x) V. corymbosum (4x), was higher than for the heteroploid crosses, F-1 (4x) V. darrowi (2x) and reciprocals, and V. corymbosum (4x) V. darrowi (2x) and reciprocal crosses (P<0.05). The number of large seeds per berry was co rrelated with fruit set for each cross. Conclusions F-1 ( V. darrowi V. corymbosum ) hybrid pollen stainability was highly variable among clones. F-1 hybrids with high pollen stainability were probably tetr aploids, and F-1 hybrids with low pollen stainability were probably triploids. Microscopic examinati on of FL07-110 F-1 pollen showed that, a V. darrowi V. corymbosum hybrid with low pollen stainability, produced a high frequency of 2n pollen sporads and a high number of unstained tetrads. Backcrossing the F-1 ( V. darrowi V. corymbosum ) hybrids to the southern highbush genetic pool was easy. Fruit set for these cros ses were variable, and lower fruit set was associated with F-1 hybrids with lower pollen fert ility. These results indicate that clones with high pollen stainability we re probably tetraploid and those with lower pollen stainability were probably triploids. Number of plump seeds per pollinated flower and number of seedlings per pollinated flower for each backcross were directly correlated with the fruit set percentage and pollen stainability of the F1 hybrids used in the cross. Fruit set of the backcrosses, ( V. darrowi V. corymbosum ) V. corymbosum was not different from that in V. darrowi V. darrowi and V. corymbosum V. corymbosum crosses. Heteroploid crosses between V. darrowi and V. corymbosum gave lower fruit set. Similar results were obtained for number of plump seeds per pollinated flower and number of seedlings per pollinated flower. 137

PAGE 138

Crosses of the F-1 (V. darrowi V. corymbosum ) hybrids V. ashei gave few seeds. Results differed somewhat depending on the percent pollen staining of the V. darrowi V. corymbosum parent that was used. FL07-110-F1, whic h had low pollen stainability and was probably triploid, produced a higher PPF and SP F than FL07-112-F1, which had high pollen stainability and was probably tetraploid. Fruit set differences between F-1 ( V. darrowi V. corymbosum ) hybrids V. darrowi and reciprocal crosses was attributed to the relative rates of production of 2n egg and 2n pollen in the V. darrowi clones. Varying rates of parthenocarpy could also have affected fruit set. The degree of success of the F-1 ( V. darrowi V. corymbosum ) hybrids diploid V. darrowi and their reciprocals was similar to th e analogous southern highbush cultivars V. darrowi crosses and their reciprocals. Fruit set and number of plump seeds per pollinated flower for F-1 hybrids southern highbush cultivars were higher than for heteroploid crosses, F-1 (4x) V. darrowi (2x) and reciprocals, and V. corymbosum (4x) V. darrowi (2x) and reciprocal crosses. Overall, the F-1 hybrids with hi gh pollen fertility can be backcr ossed easily to tetraploid plants (southern highbush cultivar s). These results indicate that the F-1 hybrids with high pollen stainability, which included most of the F-1 populations, were tetraploid. In addition, the F-1 V. darrowi V. corymbosum hybrids that had the highe st pollen staining were ju st as successful as tetraploid highbush cultivars in backcrosses to diploid V. darrowi Even thought the number of seedlings obtained was low, the F-1 hybrids with lower pollen stainability were successfully backcrossed to the hexaploid V. ashei Florida Rose. 138

PAGE 139

Table 4-1. Result of hybridization of F-1 ( V. darrowi southern highbush cultivars) hybrids with southern highbush cultivars in 2007. Crossz Flowers (No.) Berries (No.) Fruit set (%) PPF y SPFx 06-105-F1 Jewel-H 281 102 36.30 0.452 0.050 07-111-F1 95-12-H 270 264 97.78 23.1665.160 06-107-F1 95-12-H 282 179 63.48 11.4112.773 07-110-F1 Jewel-H 254 8 3.15 0.035 0.012 07-112-F1 Jewel-H 264 229 86.74 13.2254.328 02-16-H 07-112-F1 280 138 49.29 6.810 1.925 z-F1 = F-1 hybrid (V. darrowi southern highbush cultivars). -H = highbush cultivars. yPPF = number of plump seeds per pollinated flower. xSPF = number of seedlings per pollinated flower. Table 4-2. Result of recipro cal hybridizations of F-1 ( V. darrowi southern highbush cultivars) hybrids with V. ashei Florida Rose in 2007. Crossz Flowers (No.) Berries (No.) Fruit set (%) PPFy SPFx 06-105-F1 Florida Rose-R 270 44 16.30 0.137 0.044 07-110-F1 Florida Rose-R 279 5 1.79 0.079 0.039 07-112-F1 Florida Rose-R 244 81 33.20 1.278 0.131 Florida Rose-R 06-105-F1 211 0 0.00 0.000 0.000 Florida Rose-R 07-110-F1 259 4 1.54 0.158 0.012 Florida Rose-R 07-112-F1 289 5 1.73 0.080 0.007 z-F1 = F-1 hybrid (V. darrowi southern highbush cultivars). R = hexaploid, V. ashei, Florida Rose rabbiteye. yPPF = number of plump seeds per pollinated flower. xSPF = number of seedlings per pollinated flower. 139

PAGE 140

Table 4-3. Result of hybridization of F-1 ( V. darrowi southern highbush cultivars) hybrids with southern highbush cultivars in 2008. F1 hybrids were the female parents. Crossesz Flowers (No.) Berries (No.) Fruit set (%) PPF y 84-F1 03-293-H 134 35 26.12 0.239 127-F1 00-59-H 73 59 80.82 11.993 133-F1 Emerald-H 289 237 82.01 18.964 89-F1 00-59-H 366 294 80.33 3.617 94-F1 Emerald-H 120 60 50.00 0.900 96-F1 00-59-H 123 95 77.24 13.714 26-F1 Emerald-H 47 38 80.85 12.862 57-F1 Emerald-H 376 311 82.71 15.056 134-F1 Emerald-H 576 459 79.69 9.256 24-F1 00-59-H 363 294 80.99 3.858 29-F1 00-59-H 327 284 86.85 12.772 121-F1 03-293-H 229 161 70.31 4.848 132-F1 00-59-H 329 235 71.43 6.260 135-F1 Emerald-H 315 1 0.32 0.003 41-F1 Emerald-H 215 182 84.65 13.432 136-F1 Emerald-H 65 3 4.62 0.031 129-F1 00-59-H 542 345 63.65 8.113 z-F1 = F-1 hybrid (V. darrowi southern highbush cultivars). H = highbush cultivars. yPPF = number of plump seeds per pollinated flower. Table 4-4. Result of recipro cal hybridizations of F-1 ( V. darrowi southern highbush cultivars) hybrids with diploid Vaccinium darrowi high 2n egg producersin 2008. Crossesz Flowers (No.) Berries (No.) Fruit set (%) PPF y 07-111-F1 06-660-I 528 47 8.90 0.254 07-112-F1 06-660-I 293 13 4.44 0.188 03-421-P 07-111-F1 507 166 32.74 0.362 03-421-P 07-112-F1 491 165 33.60 0.673 zF1 = F-1 hybrid (V. darrowi southern highbush cultivars). I = V. darrowi Istkopoga race. P = V. darrowi Florida panhandle race. yPPF = number of plump seeds per pollinated flower. Table 4-5. Mean pollen st ainability of F-1 ( V. darrowi V. corymbosum ) hybrids selected from crosses made in 2006. The population wa s planted in Citra, Florida. F-1 (V. darrowi V. corymbosum) hybrids Plants (No.) Pollen stainability (%)z 2004 crosses 2 56.97b 2005 crosses 3 71.75ab 2006 crosses 109 88.06a zPercent stainability was calculated as the number of stained spores per pollen grain (tetrad, triad, dyad and monad), averaged by the to tal number of pollen grains counted. *Similar letters within a column indicates means not significantly di fferent, Chi-square, test of independence, =0.05. 140

PAGE 141

Table 4-6. Pollen stainability of five F-1 ( V. darrowi V. corymbosum ) hybrids from crosses made in 2004 and 2005. F-1 hybrid Sporads examined (No.)zPollen stainability (%) y 06-107 142 99.41 06-105 136 14.52 07-111 106 95.52 07-112 119 98.32 07-110 111 21.40 zEach sporad normally contains four pollen grains. yPercent stainability was calculated as the number of stained pollen grains pe r sporad (tetrad = four, triad = three, dyad = two, and monad = one), averaged by the total number of sporads examined. Table 4-7. Crossing behavior of F-1 hybrids ( V. darrowi southern highbush), V. darrowi and southern highbush cultivars in 2007. Crossesz Number of crosses Flowers (No.) Fruit set (%) PPFy SPFx HPFw F1 F1 2 538 78.97a 15.295N S 2.373 b -v N S F1 HB 5 1351 57.49ab 9.658 2.464 b HB F1 1 280 49.29b 6.810 1.925 b HB HB 5 652 78.06a 7.914 17.520 a HB VD 10 5145 15.68d 0.465 0.114 b 0.114 VD HB 10 5866 29.54c 1.422 0.112 b 0.076 VD VD 5 1496 74.28a 15.2754.536 ab zF1= F-1 hybrid (V. darrowi southern highbush). HB = southern highbush. VD = V. darrowi. yPPF = number of plump seeds per pollinated flower. xSPF = number of seedlings per pollinated flower. wHPF = number of hybrids per pollinated flower. v = no data. *Similar letters within a column indicates means not significantly different. Tukeys test for SPF, =0.05. Chi-square, test of independence for fruit set (%), =0.05. NS= indicated means not significantly different within a column, Tukeys test, =0.05. Table 4-8. Weight and s eed count per berry in crosses of F-1 hybrids ( V. darrowi southern highbush), V. darrowi and southern highbush cultivars in 2007. Crossz Berries (No.) Berry weight (g) Large seeds per berry ( ) Small seeds per berry ( ) Total seeds per berry ( ) F1 F1 40 1.24 c 18.15 bc 30.18 b 48.33 a F1 HB 100 1.04 c 13.14 cd 11.15 d 24.28 bc HB F1 20 1.99 a 11.80 cd 43.30 a 55.10 a HB HB 100 1.53 b 31.71 a 19.92 c 51.63 a HB VD 200 1.12 c 3.13 de 12.46 d 15.59 c VD HB 200 0.33 d 8.71 de 23.63 bc 32.33 b VD VD 120 0.43 d 21.24 b 11.80d 33.04b zF1= F-1 hybrid (V. darrowi southern highbush). HB = southern highbush. VD = V. darrowi. *Similar letters within a column indicates means not significantly different, Tukeys test, =0.05. 141

PAGE 142

Table 4-9. Crossing behavior of F-1 hybrids ( V. darrowi southern highbush), V. darrowi and southern highbush cultivars in 2008. Crossesz Ploidy Numbery Flowers (No.) Fruit set (%) Plump seeds per pollinated flower F1 HB 4x-4x 17 3507 64.86 a 7.995 a F1 VD 4x-2x 2 721 6.67 c 0.221 b HB VD 4x-2x 12 4873 7.61 c 0.082 b VD F1 2x-4x 2 998 33.17 b 0.517 b VD HB 2x-4x 15 5122 33.80 b 1.404 ab zF1= F-1 hybrid (V. darrowi southern highbush cultivars). HB = southern highbush. VD = V. darrowi. yNumber of crosses per type. *Similar letters within a column indicates means not significantly different. Tukeys test for plump seeds per pollinated flower, =0.05. Chi-square, test of independence for fruit set (%), =0.05. Table 4-10. Weight and s eed count per berry in crosses of F-1 hybrids ( V. darrowi southern highbush), V. darrowi and southern highbush cultivars in 2008. Crossz Berries (No.) Berry weight (g) Large seeds per berry ( ) Small seeds per berry ( ) Total seeds per berry ( ) F1 HB 304 1.16 b 16.06 a 13.47 bc 29.54 b F1 VD 33 0.84 c 3.18 c 15.45 b 18.64 c HB VD 125 1.40 a 1.68 c 7.21 d 8.89 d VD F1 40 0.32 d 2.57 c 8.85 cd 11.43 d VD HB 277 0.27 d 9.31 b 30.05a 39.36a zF1= F-1 hybrid (V. darrowi southern highbush cultivars). HB = southern highbush. VD = V. darrowi. *Similar letters within a column indicates means not significantly different, Tukeys test, =0.05. 142

PAGE 143

a) b) Figure 4-1. Microphotographs of shrunken and shriveled pollen (arrow), and plump and well stained 2n sporads (arrow head) of FL07-110 F-1 ( V. darrowi V. corymbosum ) hybrid. a) and b) show two di fferent microscope fields from the same pollen sample; 250x. 143

PAGE 144

CHAPTER 6 CROSSING VALUE OF V. DARROWI COLCHICINE-DERIVED TETRAPLOIDS IN CROSSES WITH SOUTHERN HIGHBUSH. Introduction Various techniques have been used to overc ome crossing barriers between species in Vaccinium Tetraploids have been produced by crossi ng diploid and tetraploid species. The production of unreduced gametes and the exis tence of a strong triploid block in Vaccinium allowed the production of tetr aploids from 4xx crosses a nd reciprocals (Sharpe and Shoemaker, 1958; Sharpe and Darrow, 1959; Moore, 1965). Tetraploid seedlings produced using only species from section Cyanococcus have generally been easy to intercross and backcross with tetraploid highbush cultivars. Tetraploids have been produced extensively using diploid V. darrowi in the southern highbush blueberry breeding program. It wa s originally thought that crossing V. darrowi (2x) with northern highbush cultivars (4x) would pr oduce sterile triploid progeny (Sharpe and Darrow, 1959). Instead, from 1600 pollinations, Sharpe and Darrow obtained 31 tetraploid hybrids, along with an unknown number of triplo ids. The F-1 hybrids were fertile and were easily intercrossed and backcrossed to te traploid northern hi ghbush cultivars. Camp (1945) studied the native blueberry spec ies present in Florida. He stated that V. darrowi was found in extensive colonies, 0.15-0.40 m high, from Louisiana to Florida. Lyrene (1986) found V. darrowi populations in which the plants did not correspond to Camps description He described three populations: (1) the Fl orida panhandle race (short stature, 0.340.70 m high), (2) the Ocala Forest race (tall stature, 1.10-1.61 m high), and (3) the Istokpoga race (highly variable, including s hort and tall plants). The V. darrowi Istokpoga race and the Ocala Forest race have been used extensively in breeding southern highbush cultivars. 144

PAGE 145

To date, most of the tetrap loid hybrids obtained using V. darrowi have been obtained by exploiting V. darrowi s tendency to produce 2n gametes. Colchicine-derived tetraploids offer an alternative method of producing plants that will easily cross with te traploid species of interest. F-1 hybrids between cultivars and wild plants normally have low agronomic value but may provide traits that are required in species being cultivated. Blakes lee (1937) found that colchicine treatment of plants disrupts spindle formation during mitosis, resulting in cells with doubled chromosome number. The colchicine technique wa s widely tested and was rapidly adopted for doubling chromosome numbers in various plant species. The first cytological studies of colchicine-d erived plants showed that the effect of colchicine on plant cells was different from the cytological effects on cells that were doubled by temperature treatments with high, low and ra pidly changing temperature. In temperature treatments, the chromosomes clumped during m itotic metaphase. When meiosis was disrupted by colchicine, the chromosomes were in a resting st age, (primarily metaphase I or metaphase II). The chromosomes did not move to the poles, and a doubled cell was produced. In the colchicine treatment, cell division into sister cells during mi tosis is prevented. Chromosomes split into sister chromosomes but remained together, producing cells that had the doubled number of chromosomes (Dermen, 1940). Plant tissue treated with colchicine show s hypertrophy around the gr owing points. Some external structural changes can be found in part s of the plant and even in the whole plant. Chimeras may be produced. Morphological changes become apparent in flowers, leaves, nodes, shoots, and soon. Dermen and Bain (1944) found in cranberry that some runners, when treated with colchicine, showed larger stems with shorter internodes, and thicker, longer, broader, darker green leaves and larger stomata than normal. Th ree types of colchicinederived plants were 145

PAGE 146

found depending on the layers that were affected: (1 ) LI-layer or epidermal tetraploids, with all internal tissues unchanged, (2) LII la yer or internal tetraploids, with all internal tissues tetraploid but with the epidermis unchanged, a nd (3) LI and LII layers or tota l tetraploids, with all tissues tetraploid. Colchicine-treated plants must produce gametes with twice the normal chromosome number if they are to be useful in further breeding. Stomata length and pollen diameter are used to detect changes in ploidy levels in LI and L II layers, respectively. Cha ndler and Lyrene (1982) found that diploid species in Vaccinium have shorter guard cells th an hexaploid species. Guard cell length could be used in Vaccinium to detect changes in LI ploidy levels after treatment with colchicine. In blueberries, plants with doubled chromo some number have been produced using colchicine. A decaploid seedling (10x) was prod uced at Beltsville, Maryland by treating with colchicine a pentaploid s eedling obtained by crossing V. ashei (6x) with northern highbush (4x) (Moore, 1965). In-vitro culture has been used to treat plant tissues with colchicine. Lyrene and Perry (1982) treated shoot-tip cult ures of rabbiteye blueberry ( V. ashei 6x) cultivars and V. elliottii (2x). Tissue-culture-grown shoot tips were pl aced on Knops medium containing agar and 0.001 percent to 0.200 percent colchicine for eight weeks. Variation in shoot thickness in shoot colonies that re-grew from treat ed stems was used as a screen ing method to identify induced polyploids. Five clones, two V. ashei and three V. elliottii, were selected and grown. All the selected clones had doubled chromosomes. In another experiment, shoot-tip cultures of V. darrowi V. elliottii and F-1 hybrids ( V. darrowi V. elliottii ) were treated with colchicine dissolved in liquid modified Knops medium at concentrations from 0 percent to 0.20 percent for 6 to 72 hours. Explants were then washed and placed on colchicine-free solid nutrient medium for shoot proliferation. Shoot th ickness was used as a screeni ng method to find doubled colonies. 146

PAGE 147

Within the colonies that had thick shoots, 70 percent had increased guard cells length and a higher ploidy level (Perry and Lyrene, 1984). Within Vaccinium section Cyanococcus, colchicine-derived plants can be readily crossed with species of the same ploidy level. Dweikat and Lyrene (1991) studied the crossing behavior of a synthetic autotetraploid of V. elliottii (2x) produced by colchicine treatment. The colchiploid, V. elliottii clone, FL519, was crossed with tetr aploid southern highbush cultivars. When diploid V. elliottii was crossed, as a pollen parent, with tetraploid cultivars, 0.01 seedlings per pollinated flower were obt ained. When autotetraploid V. elliottii FL519 was used as a pollen parent, 3.86 seedlings per pollinated flower were obtained. Theref ore, doubling of chromosomes reduces the crossing barrier be tween diploid and tetraploid Cyanococcus. Dermen (1940) noted a correlation between nuclear volume and cellular volume. If cell volume increases after colchicine treatment, changes will become apparent in certain plant parts, such as leaf, flower, fruit, seeds, etc. Measurem ents of stomata and pollen grains can be used to detect polyploid changes in treated plants. The purpose of this study was to analyze the crossing behavior of autotetraploid V. darrowi (colchicine-derived) with tetraploid southern highbush cultivars and diploid V. darrowi and to assess the value of the colchicine-derived 4x V. darrowi in southern highbush blueberry breeding. Materials and Methods Colchicine Treatment In November 2006, open-pollinated seed from V. darrowi FL03-405 (Istokpoga race) was treated with aqueous colchicine. Dry V. darrowi seeds were divided in two lots of 6.35 g each. One seed lot was assigned to each of the following colchicine treatments: (1) Seed imbibed in 0.2% colchicine solution in distilled water for 96 hours, and (2) Seed imbibed in 0.2% colchicine 147

PAGE 148

solution in distilled water for 96 hours, followed by one subsequent treatment, which is described below. For the first part of both treatments, colchicine (0.2 g pe r 100ml) was dissolved in distilled water by shaking for 4-5 min. Each lot of 6.35 g of dried seeds was imbibed in 100 ml of aqueous colchicine solution for 96 hours w ith aeration provided by shaking the flask approximately three times each day. The seeds were planted in two 4-liter pots of 1:1 mix of sphagnum peat and perlite. One seed lot was plan ted directly on top of sphagnum peat in the 4liter pot. The second seed lot was planted on top of sphagnum peat filled in a plastic container with holes at the bottom. This plastic container was put into the 4-liter pot filled with sphagnum peat until covered. Two days afte r they were planted, the seeds from one of the pots received a second colchicine treatment (the seed lot with a plastic bowl). This pot was immersed in water until the water level reached the treated seeds that were located in a plastic container within the pot. Aqueous colchicine solution ( 0.1%) was poured over the seeds, wh ich were in a thin layer of wet peat in the plastic container. By immersing the pot in water, it was possible to keep the added colchicine in the top laye r of peat that contained the seeds. The colchicine was kept around the seeds for one hour. The seeds were th en removed from the plastic container and planted directly on the 4-liter pot filled with sphagnum peat. The pots were maintained in a greenhouse in Gainesville, Florida until germination was completed. In January 2007, cotyledons from the plante d seeds began to emer ge. From this point, seedlings that presented irregula r morphology were selected and tr ansplanted to a plastic tray filled with 1:1 sphagnum peat and perlite. Seedlings that showed changes in cotyledon size, coloration, shoot length, shoot morphology, shoot diameter, abnormal growth, or large size, were selected as possible colchiploi ds. One-hundred selected seedlings were transplanted for each 148

PAGE 149

treatment from a total of approximately 2000 seedli ngs that survived each treatment. Trays and pots were maintained for 4-5 months in a greenhous e. As the seedlings became larger, those that continued to show the characteristics described above were maintained. Seedlings that had normal appearance and morphology were removed. From the 200 seedlings (100 from each treatm ent) that had been selected earlier and transplanted to trays, 25 from treatment one and 19 from treatment two were re-selected in October 2007. These seedlings were transplanted to 4-liter pots filled with 1:1 sphagnum peat and perlite. Seedlings were maintained in the greenhouse until they flowered. Stomata and Pollen Screening In January 2008, selected seedlings started to flower. Flowers and l eaves were collected from the colchicine-treated plants. The lower surf ace of three leaves per plant, was covered with fingernail polish to obtain a dry print of the l eaf lower surface, which was removed and placed on a microscope slide for measurement. One or tw o drops of water were put on top of the threeleaf replicate and a microscope cover glass wa s slowly pressed over th e top of the fingernail polish prints on the microscope slide. Fifty to 80 stomata were examined per genotype. Pollen from dry flowers was stained with 1% aceto-carmi ne solution to measure tetrad diameter. Thirty to 100 pollen tetrads were examined per genotyp e. Stomata length and pollen tetrad diameter measurements were obtained using microphot ographs taken with a Moticam 1000 1.3MPixel microscope digital camera with the Motic Imag es Plus Version 2.0ML software, mounted on a light phase-contrast Leitz microscope, 250 a nd 400 magnification. The standard measurement used in this procedure was obtai ned by calibrating the microscope digital camera to a microscope micrometer slide. Four groups of plants were cl assified after screening: (1) Non-doubled: plants with normal stomatal guard cells and normal-si ze pollen tetrads (2x-2x-?), (2) Pe riclinal chimeras with large 149

PAGE 150

stomatal guard cells but pollen tetrads of normal size (4x-2x-?), (3) Periclinal chimeras with large pollen tetrads but stomatal guard cells of normal size (2x-4x?), and (4) Tetraploid plants, with both pollen tetrads and stomatal guard cells large (4x-4x-?). Stomata length and pollen diameter means were analyzed with Tukeys test, significance level of 5%. A chi-square test of independenc e, significance level of 5%, was used to test whether the two treatments differed in the ratio s of colchicine-derived plants produced per treatment. Data were subjected to ANOVA by th e PROC GLM and PROC FREQ procedures of SAS (Statistical Analysis System Version 9.1, SAS Institute, Cary, NC). Morphological Studies of Co lchiploid Derived Plants In January 2008, ten non-doubled V. darrowi clones, one periclinal chimera with doubled LI layer, three periclinal chimeras with doubled LII layer and one tetraploid plant with doubled LI and LII layers were identified after co lchicine treatments (Table 5-1). These V. darrowi clones were used to search for flower, leaf and berry morphological characters that distinguished each type of plant. Five samples were measured for each clone. Leaf and flower characteristics were measured using a Traceable Carbon Fiber Calip er from Fisher Scientific, Pittsburg, PA. Leaf characteristics measured for the four groups were leaf length and leaf width. Flower characteristics measured were corolla length, corolla width, corolla aperture, pedicel length, peduncle length, bracteole lengt h, bracteole width, and diam eter of the pedicel base. In addition, berry weight was measured. To produce the berries, the flowers of the tetraploid plants and periclinal chimeras with doubled LII layer had been pollinated with pollen from colchicine-derived tetraploid V. arboreum The flowers of the diploid V. darrowi and periclinal chimeras with doubled LI layer had been pollinated with diploid V. darrowi pollen. Twenty berries were measured for each group. M eans for leaf, flower a nd berry characteristics were separated using least squares means by Tukeys test, with significance level 5%. Data were 150

PAGE 151

subjected to ANOVA by the PROC GLM procedur e of SAS (Statistical Analysis System Version 9.1, SAS Institute, Cary, NC). Inter-Specific Hybridization Experiments Vaccinium corymbosum (4x) colchicine-derived V. darrowi (4x) In March 2008, after screening and dete cting colchicine-der ived tetraploid V. darrowi plants, they were used in crosses with te traploid southern hig hbush cultivars. Four V. darrowi clones were selected as male pa rents and eight highbush clones were used as females (Table 52). These southern highbush cultivars were a dvanced selections from the Florida breeding program, and they are here considered V. corymbosum although other species are present in their genetic background (Muoz and Lyrene, 1984b). Of the three V. darrowi clones, one was tetraploid with doubled LI and LII layers, and th ree were periclinal chimeras with doubled LII layer and non-doubled LI layer (Table 5-2). Pollen was collected from the V. darrowi colchiploids. Pollen was dehisced onto the thumbnail and transferred by rubbi ng the stigma of the emascu lated flower on the thumbnail. Thirty to 360 flowers were pollinated per plant, depending on the flowers available. Berries were harvested when fully ripe. Seeds from the first twen ty ripe berries were c ounted and classified as plump or shriveled. The remaining berries were harvested, and seeds were extracted using a food blender (Moore, 1965). The seeds were washed with water and dried. They were stored in coin envelopes at 5C. Colchicine-derived V. darrowi (4x) colchicine-derived V. arboreum (4x) F-1 hybrids from crosses between diploid V. darrowi and diploid V. arboreum were easy to obtain. Both parents were diploid and the hybr ids were highly vigorous. However, F-1 hybrids have low fertility, probably due to poor chro mosome pairing resulting from chromosome 151

PAGE 152

structural differentiation between the two sections of the genus (Cyanococcus and Batodendron ) (Lyrene, 1991; Brooks and Lyrene 1995; Brooks and Lyrene, 1998b). Dermen (1940) explained how the sterility of F-1 hybrids from some wide crosses can be overcome by chromosome doubling. Doubling is most likely to restore fertility if the reason for sterility in the diploid F1 hybrids is non-homology of th e parental chromosomes. Four colchicine-derived V. darrowi plants, oneplant with double d LI and LII layers (4x-4x?), and three periclinal chimeras with doubled LII layer and non-doubled LI layer (2x-4x-?), were used as female parents in crosses with one colchicine-derived tetraploid V. arboreum (Table 5-3). Amphidiploids are expected to be produced because both parents are autotetraploids and the two have structurally distinct genomes. Pollen from a colchicine-derived tetraploid V. arboreum plant FL06-730) was collected and preserved at 4C. Pollen was placed on the stigmas of emasculated flowers of the colchicine-derived V. darrowi plants. From 24 to 234 flowers were pollinated per female plant, depending on the flowers available. Berries were harvested when fully ripe. Seeds from the first twenty berries that ripened for each cross were counted and classified as plump or shrivele d. Berry weight was also recorded. The remaining berries were harvested, and seeds were extracted using a food bl ender. The seeds were washed with water and dried. They were stored in coin envelopes at 5C. Vaccinium darrowi (2x) colchicine-derived V. darrowi (4x) Four colchicine-derived V. darrowi plants, oneplant with double d LI and LII layers (4x-4x?), and three periclinal chimeras with doubled LII layer and non-doubled LI layer (2x-4x-?), were used as male parent s in crosses with diploid V. darrowi (Table 5-4). Two plants of V. darrowi clone FL06-660-I (Istokpoga race) were used as female parents for all the experiments. FL06-660-I was used because it produced a high nu mber of 2n egg gametes when crossed with tetraploid southern highbush cultivars (Table 1-9). The plan was to produce tetraploid V. darrowi 152

PAGE 153

progeny by uniting unreduced eggs from diploid V. darrowi with reduced pollen from tetraploid V. darrowi Each female plant was pollinated with a composite of pollen from two colchicinederived V. darrowi plants. Pollen was placed on the stig mas of emasculated flowers of the diploid V. darrowi plants. Approximately 250 flowers were pollinated for each plant. Berries were harvested when fully ripe. Seeds from the first twenty berries that ripened for each cross were counted and classified as plump or shri veled. Berry weight was also recorded. The remaining berries were harvested, and seeds were extracted using a food blender. The seeds were washed with water and then dried. They we re stored in coin envelopes at 5C. Control Crosses Vaccinium darrowi (2x) V. corymbosum (4x) and reciprocals In February 2008, twelve two-year old tetraploid southern hig hbush cultivars were selected as female parents in crosses with diploid V. darrowi clones collected by P. Lyrene and K. Hummer in 2006. Of the V. darrowi clones, three belonged to the Istokpo ga race and nine to the Florida panhandle race (Table 1-6). Flowering, pollination, harvesting and seed extraction were as previously stated. In a parallel study, two-year ol d plants of fifteen diploid V. darrowi clones, collected by P. Lyrene and K. Hummer in 2006, were selected as female parents in crosses with fifteen tetraploid southern highbush cu ltivars (Table 1-12) Of the V. darrowi clones, six were from the Istokpoga race and nine from the Florida pa nhandle race. Flower ing, pollination, berry harvesting, and seed extraction were as described before. Vaccinium darrowi (2x) V. arboreum (2x) and reciprocals In January 2008, six V. darrowi clones collected in 2006 by P. Lyrene and K. Hummer (described before) were selected as parents for crosses with V. arboreum clones. Three V. darrowi clones were from the Florida panhandle r ace and three from the Istokpoga race. These 153

PAGE 154

V. darrowi clones were chosen from the germplas m collected in 2006 and grown during 2007. For V. darrowi V. arboreum crosses and reciprocals, approximately 500 flowers were pollinated per plant (Table 2-5 and Table 2-6, re spectively). Pollination, harvesting and seed extraction were as described above. Statistical Analysis The results of pollination treatments were assessed by fruit set percentage and number of plump seeds per pollinated flower. Means for numb ers of plump seeds per pollinated flower for different treatments were separated using least squares means by Tukeys test, with significance level 5%. Means for fruit set percentage were se parated using Chi-square test of independence, with significance level 5%. Data were subj ected to ANOVA by the PROC GLM and PROC FREQ procedures of SAS (Statistical Analys is System Version 9.1, SAS Institute, Cary, NC). Cytogenetics of Colchicine-Derived Tetraploids In March 2008, flower buds at various devel opmental stages were collected from one periclinal V. darrowi chimera with large pollen tetrads (FL8-403) that was identified following treatment with aqueous colchicine solution in Ga inesville, Florida, in 2006. Flower buds were fixed in a 3:1 solution of glacial acetic acid abso lute ethanol. Samples were kept in fixative until the pigments were removed and then stored in fixative at 5C. The flower buds were then assessed for stage of development by cutting open a random floret from the several floret buds in one axillary bud, largest or smallest, and squash ing two anthers on a micr oscope slide in 45% acetic acid. The slide was analyz ed with a phase-contrast Leitz microscope (250 and 400) to determine whether the pollen mother cells were undergoing meiosis. When a meiotic bud was found, the additional eight anthers were digested for 3 hours at room temperature in a 5% solution (diluted in citrate buffer at pH 6.0) of cell wall degrading enzyme complex from Aspergillus sp. (Viscozyme from Novoz ymes Corp.). After digestion, the anthers were stored 154

PAGE 155

in 70% ethanol solution at 5C. For chromosome counting, anther s were imbibed for 20 min at room temperature in 45% acetic ac id. Four anthers were used to prepare each slide. Each anther was placed in one corner on the microscope slid e area that could be covered by one cover slip. Each anther was divided into two or three portions to produce an optimal squashing. Anthers were macerated in 45% acetic acid on the microscope slide. The cover slip was then placed on the slide. A medium-light tappi ng with the head of a pen was done over the cover slip at the points where the anthers were covered. Then the slide was heated lightly over a flame for one second and allowed to cool. Exce ss acetic acid was removed from the slide. The edges of the microscope cover glass were sealed with nail polish. Cells were observed in a phase-contrast Leitz microscope at 250 and 400 magnifica tions. Microphotographs were taken using Moticam 1000 1.3MPixel microscope digital camera with the Motic Images Plus Version 2.0ML software. Chromosome pairing during meiosis was studied in 3 metaphase I cells of colchicinederived V. darrowi clones FL08-403. The number of univalent s, bivalents, triv alents, and other chromosome associations at metaphase I was re corded. Pollen mother cells were studied at anaphase I and anaphase II to detect any meiotic abnormalities that might be occurring. Results and Discussion Results of Colchicine Treatment Abnormal growth and morphological changes were found in plants grown from seeds treated with colchicine. Morphol ogical abnormalities included changes in coloration, shoot length, shoot morphology, and shoot diameter. An unusual growth habit characterized by reduction or loss of apical dominance, was found in two clones (Figures 5-1 and 5-2). Increases in leaf size (area) and shoot diameter were also found in some treated plants (Figure 5-3). No distinctive morphological differenc es were found by which plants of different ploidy could be 155

PAGE 156

reliably distinguished. It was always necessary to study stomata and pollen size. In cranberry, Dermen and Bain (1944) found that plants treated with colchicine often appeared abnormal in shape, or showed sections of epidermis with enlarged stomatas. Stomata and Pollen Screening Data Stomata length and pollen diameter of V. darrowi plants grown from colchicine-treated seed were measured. Stomata and pollen measurements of three V. darrowi clones collected by P. Lyrene and K. Hummer in 2006 were performe d. All 44 treated seedlings plus three untreated clones, collected by P. Lyrene and K. Hummer, we re used to calculate the population mean value ( ). Stomata length and pollen diameter were higher in some colchicine treated plants (Table 55). Unusually large stomates were found for si x genotypes, whose stomates averaged from 22.70% to 39.01% longer than the population mean. The mean was based on 47 plants, four of which had large stomates. Had these four plants been excluded in calculating the mean, the four would have deviated even farther from the mean. Unusually large pollen was found in five seedlings, the increases ranging from 15.45% to 30.19% over the population mean. These increases were visually di stinctive by microscope, 250 and 400 magnification, when comparing plants with normal and unusual stomata and pollen size (Figure 5-4 and Figure 5-5). Several measurements per genotype were made. Box plots from each genotype were created to compare and confirm increments in stomata leng th and pollen diameter (Figure 5-6, and Figure 5-7). Two genotypes, VD-28 and VD-38, had large stomata and normal pollen size and were classified as periclinal chimeras with doubled LI layer. Two genotypes, VD-17 (FL08-400) and VD-26 (FL08-401), had large stomata and large po llen size and were classified as doubled plants. In addition, three genotypes, VD-36 (FL08-402), VD-40 (FL08-403) and VD-43 (FL08404), had normal stomata and large pollen size and we re classified as periclinal chimeras with 156

PAGE 157

doubled LII layer. These results were confirmed with the box plot representations for each genotype. The clones described previously were above the population mean. Doubled plants and periclinal chimeras with doubled LII layer can be used easily as parents in homoploid crosses with tetraploid plants. In our study, stomata length and pollen diamet er measurements were an efficient way to screen for polyploid changes after colchicine treatment. Induced polyploidy increased stomata and pollen size by 15.45% to 39.01 % above the population means ( ) respectively. Both measurements were necessary to identify the vari ous types of periclinal chimeras that can be produced. Screening by pollen diameter alone should be sufficient to identify clones that would breed like tetraploids. No difference was found between the two colchici ne treatments in the rate of production of tetraploid plants, but this is not surprising given the small numbe r of tetraploid plants produced (Table 5-6). Treatment 2, which included a second colchicine application, produced six colchiploid plants, and Treatment 1, which had only the initial seed soak, produced one. From the six colchiploid plants in tr eatment two, four were easy to use as parents with tetraploid plants. No differences were det ected in the types of colchiploi ds produced by the two treatments. Analysis of Morphological Characteri stics of Colchiploid-Derived Plants No distinctive macroscopic differences were found in vegetative characters between types of colchicine-derived plants. Colchicine-derived tetraploid V. darrowi plants (LI + LII doubled) were different from periclinal chimeras (either LI or LII doubled but no t both) and from diploid V. darrowi plants for leaf width, leaf length, coroll a length, and bracteole length (Table 5-1). Pedicel length and bracteole width were not different between groups. None of these characteristics could be used to visually distinguish between fully -doubled seedlings and seedlings in which only the LII layer is doubled. 157

PAGE 158

Corolla width, corolla aperture and pedicel ba se diameter were larger for doubled plants and periclinal chimeras with doubled LII layer compared to diploid pl ants and periclinal chimeras with doubled LI layer. Corolla width, coro lla aperture and pedicel base diameter can be used for visual selection of functional colchicine-derived V. darrowi plants that can be used in further tetraploid breeding. Berry weight measurements were highly variable between groups. Similar results were found by Dermen and Ba in (1944). They found that cranberry after colchicine treatments presented highl y variable morphological changes. In our study, stomata and pollen screening were by far the best method for identifying colchicine-derived tetraploid V. darrowi plants, but the flower char acteristics described above can be used to pre-select plants with a high probability of being fully doubled or periclinal chimeras that have a doubled LII layer. Inter-Specific Hybridization Fruit set of V. corymbosum (4x) colchicine-derived V. darrowi (4x) ranged from 30.81% to 75.63% (Table 5-2). Fruit set of V. corymbosum (4x) V. darrowi (2x) ranged from 0.27% to 46.04% (Table 1-6). Fruit set of V. darrowi (2x) V. corymbosum (4x) ranged from 2.76% to 64.04% (Table 1-12). Pollen from tetraploid V. darrowi, when used to pollinate tetraploid highbush, gave higher fruit set, berry weight, to tal number of seeds pe r berry and number of plump seeds per berry (PPF) than pollen from diploid V. darrowi (Table 5-7). Crosses between diploid and tetraploid V. darrowi gave 2.54 PPF. It is yet to be s een whether large populations of tetraploid V. darrowi can be obtained from these crosses. The two V. darrowi -2x V. darrowi 4x crosses are shown individually in Table 5-4. Those crosses were made using a V. darrowi clone that was a high 2n egg producer, FL06-660-I, as the female parent Two different pollen composites were used in each of these crosses, and each composite was from two distinct tetraploid V. darrowi clones. Fruit set was 33.82% and 54.34% for the two crosses, and PPF was 158

PAGE 159

1.786 to 3.283 (Table 5-4). Fruit set and PPF for V. darrowi (2x) V. corymbosum (4x) crosses was highly variable (Table 1-12). In bot h type of crosses, 2n egg production by V. darrowi presumably allowed these hybridizations to occur. Mean fruit set, total nu mber of seeds per berry and PPF were not different when diploid V. darrowi was used as female parent in crosses with colchicine-derived V. darrowi (4x) and V. corymbosum (4x) (Table 5-7). Fruit set of V. darrowi V. darrowi crosses ranged from 51.22% to 101.60% (Table1-3). Fruit set and PPF were 1.6x and 6x, re spectively, higher when diploid V. darrowi was used in V. darrowi V. darrowi crosses compared to crosses with colchicine-derived tetraploid V. darrowi as male parent. This reduction in cross-ability is consistent with previous experience with heteroploid Vaccinium crosses. The relatively high productivity of the V. corymbosum (4x) colchicine-derived V. darrowi (4x) is also consistent with the V. darrowi parent being tetraploid. It is expected that a large number of F-1 hybrids will be obtained when the V. corymbosum (4x) colchicine-derived V. darrowi (4x) seeds are planted in November 2008. Production of a large tetraploid population of V. darrowi through V. darrowi (2x) colchi cine-derived V. darrowi (4x) crosses will permit sele ction of the best tetraploid V. darrowi plants for crossing with tetraploid hi ghbush. Tetraploid V. darrowi has never before been reported. Breeding V. darrowi at the tetraploid level could pr oduce ornamental blueberry clones. Fruit set for colchicine-derived V. darrowi (4x) colchicine-derived V. arboreum (4x) ranged from 15.81% to 60.92% (Table 5-3). Their analogues at the diploid level had a fruit set from 22.02% to 79.98% (Table 2-5). Fruit set fo r both types of crosses was highly variable. Mean fruit set and total number of seeds per berry for crosses between V. darrowi (2x) V. arboreum (2x) were higher than for their analogues at the tetraploid level (Table 5-8). Crossing success for V. arboreum (2x) V. darrowi (2x) was lower than for either the reciprocals or for 159

PAGE 160

their analogues at the tetraploid leve l. Crosses between colchicine-derived V. darrowi and colchicine-derived V. arboreum were easy to make, similar to the analogous crosses at the diploid level (Table 5-8). Cytogenetics of Colchicine-Derived Tetraploid Chromosomes associations at metaphase I of colchicine-derived tetraploid FL08-403 included univalents, bivalents, trivalents and quadrivalents (Figure 58). Chromosome number for FL08-403 averaged 48 (Table 5-9). Chromo some doubling by colchicine was confirmed by cytogenetic studies of clone FL08-403. No PMCs were found in anaphase I or anaphase II. It is expected that other colchicine-derived plants have similar chromosome pairing behavior at metaphase I. Conclusions No consistent macro-morphological differen ces were found to be associated with chromosome doubling after colchicine application in V. darrowi clones. Morphological changes in coloration, shoot length, shoot morphology, and shoot diameter, can be used to pre-select for possible colchicine-derived plants. Stomata and pollen screening were both effi cient and distinctive indicators of doubled plants and periclinal chimeras that can be used in tetraplo id highbush breeding. Stomata length and pollen tetrad diameter both increased substantially after chromosome doubling by colchicine. Several types of plants were produced by colchicine treatments -plants with doubled LI and LII layers, periclinal chimeras with doubled LI layer, and periclinal chimeras with doubled LII layer. No distinctive differences in vegetative morphology were found between types of colchicine-derived plants. Corolla width, corolla ap erture and pedicel base diameter can be used as a pre-selection method for the detection of do ubled LII layer plants that can be used in tetraploid breeding. 160

PAGE 161

Crosses between colchicine-derived V. darrowi (4x) plants and southern highbush cultivars were highly successful in comparison with 4x-2x and 2x-4x crosses using diploid V. darrowi and southern highbush cultivars. In addition, V. darrowi (2x) colchicine-derived V. darrowi (LII=4x) crosses were similar to heteroploid cros ses in section Cyanococcus Production of a V. darrowi tetraploid seedling progeny through unilate ral sexual polyploidization will permit selection of desirable V. darrowi tetraploid seedlings and may lead to selection of tetraploid V. darrowi plants with high ornamental value. Crosses between colchicine-derived V. darrowi and colchicine-derived V. arboreum were easy to make. The plan is to produce an amphidi ploid with high vigor and fertility compared with hybrids obtained at the diploi d level, which have high vigor but little or no fertility. These hybrids might be genetic resources for breeding better bush characteristic s in highbush cultivars. Cytogenetic studies of FL08-403 confirmed that it was tetraploid (2n=4x=48). Stomata and pollen screening were a reliable and efficient method for identif ying colchicine-doubled plants. Hybridization experiments and chromosome counts using colchicine-treated plants confirmed the results found by stomata and pollen screening. St omata and pollen measurements, hybridization experiments and cytogenetic studies of the colchicine-derived pl ants gave consistent results. 161

PAGE 162

Table 5-1. Leaf and corolla char acteristics of ten diploid, four periclinal chimeras, and one tetraploid V. darrowi seedlings grown from colchicine-treated seed. Seedlings were approximately two-year-old when measured. Typez VD VDI VDII VDI& II No. clones 10 1 3 1 No. reps per clone y 5 5 5 5 Leaf characteristics (mm) Length 14.23 (0.45) 13.24 (1.43) 15.57 (0.83) 23.20 (1.43) b* b b a Width 5.58 (0.15) 5.36 (0.46) 6.19 (0.27) 8.04 (0.46) b b b a Flower characteristics (mm) Corolla length 4.84 (0.07) 4.74 (0.23) 4.94 (0.14) 5.72 (0.23) b b b a Corolla width 4.11 (0.09) 3.96 (0.28) 5.72 (0.16) 5.58 (0.28) b b a a Corolla aperture 1.76 (0.07) 1.46 (0.21) 2.79 (0.12) 3.10 (0.21) b b a a Pedicel length 6.18 (0.23) 7.14 (0.72) 5.79 (0.41) 6.14 (0.72) a a a a Peduncle length 7.12 (0.29) 9.58 (0.92) 4.36 (0.53) 6.30 (0.92) ab a c bc Bracteole length 2.20 (0.10) 1.84 (0.30) 1.94 (0.17) 3.14 (0.30) b b b a Bracteole width 1.21 (0.05) 1.16 (0.15) 1.19 (0.09) 1.58 (0.15) a a a a Pedicel base diameter 0.56 (0.02) 0.54 (0.06) 0.93 (0.03) 0.82 (0.06) b b a a Berry characteristics Weight (g) 0.55 (0.05) 0.16 (0.08) 0.49 (0.03) 0.48 (0.07) ax bw av au zVD = diploid V. darrowi (2x-2x-2x) VDI = Periclinal chimera with large stomatal guard cells but pollen tetrads normal size (4x-2x-?). VDII = Periclinal chimera with large pollen tetrads but stomatal guard cells normal size (2x-4x-?). VD I&II = Both pollen tetrads and stomatal guard cells large (4x-4x-?). yA replication consisted on one leaf or one flower measured on the subject plant. x1 plant, 20 berries. w1 plant, 7 berries. v3 plants, 54 berries. u1 plant, 10 berries. VDII and VDI&II berries resulted from cross-pollinations with pollen from colchicine-derived V. arboreum (4x-4x-?). VDI and VD berries resulted from cross-pollinations with pollen from diploid V. darrowi. *Similar letters within a row indicates means not significantly different, Tukeys test, =0.05. Values within the table = Mean (standard error). 162

PAGE 163

Table 5-2. Result of inter-speci fic hybridization between sout hern highbush cultivars and colchicine-derived V. darrowi clones in 2008. Highbush was used as the female parent. Highbushz Colchiploid V. darrowiy Flowers (No.) Berries (No.) Fruit set (%) Plump seeds per pollinated flower 02-106 40-VDII 359 206 57.38 4.393 04-103 40-VDII 119 90 75.63 7.796 03-161 17-VDI&II 26 18 69.23 0.231 00-211 17-VDI&II 383 118 30.81 1.523 05-323 36-VDII 53 30 56.60 4.585 00-211 36-VDII 249 173 69.48 9.322 03-291 43-VDII 74 51 68.92 2.873 00-211 43-VDII 312 185 59.29 6.129 zTetraploid V. corymbosum southern highbush. yColchicine-derived V. darrowi. VDII Periclinal chimera with doubled LII layer and non -doubled LI layer (2x-4x-?). VDI&II plant, with doubled LI and LII layers (4x-4x-?). The numbers attached to the clone names identify specific clones. Table 5-3. Result of inter-specific hybr idization between co lchicine-derived V. darrowi and colchicine-derived V. arboreum clone in 2008. Colchicine-derived V. darrowi were used as the female parent. Colchiploid V. darrowiz Colchiploid V. arboreumy Flowers (No.) Berries (No.) Fruit set (%) Plump seeds per pollinated flower 17-VDI&II 06-730-VAII 176 31 17.61 0.563 40-VDII 06-730-VAII 87 53 60.92 5.719 36-VDII 06-730-VAII 234 37 15.81 0.389 43-VDII 06-730-VAII 24 14 58.33 4.501 zColchicine-derived V. darrowi VDII Periclinal chimera with doubled LII layer and nondoubled LI layer (2x-4x-?). VDI&II plant, with doubled LI and LII layers (4x-4x-?). yColchicine-derived V. arboreum VAII (4x-4x-?). Numbers attached to the clone names identify specific clones. Table 5-4. Result of hybrid ization between diploid V. darrowi clones and colchicine-derived V. darrowi clones in 2008. Diploid V. darrowi was used as the female parent. V. darrowi z Colchiploid V. darrowiy Flowers (No.) Berries (No.) Fruit set (%) Plump seeds per pollinated flower FL06-660-I 40+43 VDcomp. 275 93 33.82 1.786 FL06-660-I 17+36 VDcomp. 265 144 54.34 3.283 zDiploid V. darrowi, Florida evergreen blueberry. I = Istokpoga race. yColchicine-derived V. darrowi. Clone 36, 40, 43 = periclinal chimeras with doubled LII layer and non-doubled LI layer. Clone 17 = plant with doubled LI and LII layers. VDcomp = V. darrowi composite. 163

PAGE 164

Table 5-5. Screening measurements of stomata length and pollen diameter for V. darrowi seedlings grown from colchicine-treated seed. Plantz Stomata length (um) Pollen diameter (um) Mean Standard error [x-( )]/x (%)y Mean Standard error [x-( )]/x (%)y 03402-I 15.25 0.21 -6.56 52.90 0.29 13.64 03404-I 15.07 0.20 -7.68 40.09 0.29 -12.41 03405-I 17.09 0.19 4.70 44.59 0.29 -3.25 clone1 15.16 0.18 -7.13 45.03 0.31 -2.36 clone10 15.46 0.20 -5.27 50.05 0.54 7.85 clone11 17.15 0.19 5.11 43.76 0.41 -4.94 clone12 15.08 0.18 -7.58 44.04 0.29 -4.37 clone13 16.31 0.19 -0.07 44.61 0.39 -3.21 clone14 13.51 0.20 -17.24 42.50 0.34 -7.50 clone15 16.37 0.27 0.33 44.94 0.29 -2.55 clone16 16.91 0.20 3.61 45.45 0.27 -1.50 clone17 22.69 0.21 39.01 54.64 0.41 17.17 clone18 15.31 0.21 -6.17 46.40 0.34 0.42 clone19 16.11 0.21 -1.28 45.18 0.28 -2.06 clone2 15.53 0.25 -4.86 45.09 0.33 -2.23 clone20 15.53 0.20 -4.82 50.43 0.32 8.61 clone21 15.46 0.20 -5.24 45.00 0.38 -2.42 clone22 15.77 0.22 -3.39 43.43 0.27 -5.62 clone23 16.91 0.18 3.64 44.07 0.35 -4.31 clone24 16.36 0.19 0.26 42.87 0.36 -6.74 clone25 16.00 0.18 -1.97 44.20 0.29 -4.04 clone26 20.02 0.23 22.70 53.79 0.37 15.45 clone27 17.12 0.18 4.88 44.40 0.31 -3.64 clone28 20.89 0.22 28.02 42.70 0.31 -7.10 clone29 16.65 0.18 2.00 43.72 0.32 -5.03 clone3 15.60 0.20 -4.39 40.05 0.32 -12.47 clone30 15.23 0.22 -6.66 47.33 0.27 2.32 clone31 15.73 0.18 -3.62 48.54 0.33 4.78 clone32 16.01 0.22 -1.91 45.69 0.26 -1.02 clone33 16.26 0.17 -0.34 43.00 0.26 -6.47 clone34 15.27 0.19 -6.46 44.87 0.34 -2.68 clone35 17.59 0.20 7.78 43.91 0.26 -4.64 clone36 15.13 0.19 -7.29 61.04 0.48 30.19 clone37 17.23 0.19 5.59 44.14 0.28 -4.17 clone38 21.71 0.20 33.05 45.95 0.34 -0.49 clone39 15.33 0.20 -6.06 45.93 0.59 -0.53 clone4 16.76 0.20 2.67 43.89 0.32 -4.67 clone40 15.82 0.20 -3.08 59.70 0.37 27.47 clone41 14.38 0.19 -11.89 164

PAGE 165

Table 5.5. Continued clone42 17.01 0.21 4.23 46.61 0.30 0.85 clone43 15.38 0.19 -5.75 59.10 0.44 26.25 clone44 14.87 0.17 -8.89 42.31 0.27 -7.89 clone5 14.93 0.18 -8.52 44.61 0.34 -3.21 clone6 16.53 0.16 1.30 42.98 0.34 -6.52 clone7 15.90 0.19 -2.57 44.27 0.38 -3.90 clone8 15.81 0.17 -3.10 41.57 0.34 -9.40 clone9 15.03 0.21 -7.92 45.44 0.28 -1.53 Population mean ( ) 16.32 0.20 0.00 46.19 0.34 0.00 zGrey highlight indicate plants with an increase of more than 20 percent in stomata length or more than 15 percent in pollen diameter compared with the population mean ( ). yPercentage of increase or decrease from the population mean ( ). Table 5-6. Colchicine doubling rate after aqueous treatments in V. darrowi seed planted. Treatmentz Seed (g) Plants (No.) 1st selection (No.)y 2n d selection (No.)y Doubled plants (No.)x Rate doubled (%) T1 6.35 ~2000 100 25 1 0.05 NS T2 6.35 ~2000 100 19 6 0.30 zT1 = seed imbibed in 0.2% aqueous colchicine solution for 96 hours. T2 = seed imbibed in 0.2% aqueous colchicine solution for 96 hours + re-t reatment after planting with 0.1% aqueous colchicine solution for 1 hour. ySelection of seedlings w ith abnormal morphology. xIncludes plants with both LI and LII doubled and periclin al chimeras with either LI or LII doubled. NS= indicates means not significantly different within a column, Chi-square, test of independence, =0.05. Table 5-7. Result of crossing colchicine-derived V. darrowi plants with other Vaccinium species in 2008. Cross typez Number of crosses Flowers (No.) Berries (No.) Fruit set (%) Weight (g)y Total seeds ( )x PPFw HB VD-4x 8 1575 871 60.92a 2.18a 46.62a 4.61a HB VD-2x 12 4875 258 7.61c 1.40b 8.89c 0.08b VD-2x VD-4x 2 540 237 44.08ab0.61c 33.73b 2.54ab VD-2x HB 15 5122 2304 33.80b 0.27d 39.36ab 1.40b zHB = tetraploid southern highbush. VD-4x = colchicine-derived tetraploid V. darrowi. VD-2x = diploid V. darrowi. yTwenty berries per cross. xSeeds per berry. wPPF = number of plump seeds per pollinated flower. *Similar letters within a column indicates means not significantly different. Tukeys test for weight, total seeds and PPF, =0.05. Chi-square, test of independence for fruit set, =0.05. 165

PAGE 166

Table 5-8. Results of crossing V. darrowi with V. arboreum at the tetraploid and diploid levels in 2008. Cross typez Crosses (No.) Flowers (No.) Berries (No.) Fruit set (%) Weight (g)y Total seeds ( )x PPFw VD-4x VA-4x 4 521 135 38.17b 0.47a30.62b 2.793ab VD-2x VA-2x 6 2975 1803 58.37a 0.28b 52.38a 6.554a VA-2x VD-2x 6 1578 53 1.80c 0.24b 21.05c 0.194b zVD-4x = colchicine-derived tetraploid V. darrowi. VA-4x = colchicine-derived tetraploid V. arboreum VD = diploid V. darrowi. VA = V. arboreum yTwenty berries per cross. xSeeds per berry. wPPF = number of plump seeds per pollinated flower. *Similar letters within a column indicates means not significantly different. T ukeys test for weight, total seeds and PPF, =0.05. Chi-square, test of indepe ndence for fruit set (%), =0.05. Table 5-9. Chromosome associations at metaphase I for colchiploid-derived V. darrowi plant FL08-403. Cell No. Chromosome association in PMCs at metaphase I Total No. chromosomes Univalents Bivalents Trivalents Quadrivalents 1 8 8 4 3 48 2 6 13 4 1 48 3 3 5 5 5 48 166

PAGE 167

Figure 5-1. Tetraploid V. darrowi seedling (Clone 43) after treatment with colchicine. The stake is 20 mm wide. 167

PAGE 168

Figure 5-2. Tetraploid V. darrowi seedling (Clone 40) after treatment with colchicine. The stake is 20 mm wide. 168

PAGE 169

Figure 5-3. Tetraploid V. darrowi seedlings (Clone 17) after treatment with colchicine. The stake is 20 mm wide. 169

PAGE 170

20 um 20 um Figure 5-4. Stomata microphotographs Normal LI layer (top) a nd doubled LI layer (bottom). 250 170

PAGE 171

40 um 40 um Figure 5-5. Pollen microphotographs. Normal LII layer (top) and doubled LII layer(bottom) 250. 171

PAGE 172

5 10 15 20 25 30 Clones b 5 10 15 20 25 30Clones a Figure 5-6. Box plots, a) and b), of stomata length measurements for V. darrowi colchicine treated plants. The height of the box tells 75 th percentile and the bottom 25 th percentile. Line within th e box tells median. External lines, upper level maximum and lower level minimum. 172

PAGE 173

30 35 40 45 50 55 60 65 70 75Clones b 30 35 40 45 50 55 60 65 70 75Clones a Pollen diameter (um) Figure 5-7. Box plots a) and b), of pollen diameter measurements for V. darrowi colchicine treated plants. The height of the box tells 75 th percentile and the bottom 25 th percentile. Line within th e box tells median. External lines, upper level maximum and lower level minimum. 173

PAGE 174

Figure 5-8. Meiotic metaphase I in colchicine-d erived FL08-403 (2n=4x=48) (Clone 40). Left: 8 I, 8 II, 4 III and 3 IV. Right: 3 I, 5 II, 5 II and 5 IV. 400. 174

PAGE 175

LIST OF REFERENCES Arora, R., L.J. Rowland, and K. Tanino, K. 2003. Induction and release of bud dormancy in woody perennials: a science comes of age. HortScience. 38(5):911-921. Bailey, J.S. 1938. The pollination of the cultivated blueberry. Proc. Amer. Soc. Hort. Sci. 35:7172. Baados, M.P. and B. Strik. 2006. Manipulation of the annual growth cycle of blueberry using photoperiod. Acta Hort. 715:65-71. Baptista, M.C., P.B. Oliveira, L. Lopes-da-Fonseca, and C.M. Oliveira. 2006. Early ripening of southern highbush blueberries under mild winter conditions. Ac ta Hort. 715:191-196. Blakeslee, A.F. 1937. Methods of doubling of chromosomes in plants by treatment with colchicine. J. Hered. 28:393-411. Bretagnolle, F. and J.D. Thompson. 1995. Tansley Review No. 78. Gametes with the somatic chromosome number: mechanism of their fo rmation and role in the evolution of autopolyploid plants. New Phytol. 129:1-22. Brooks, S.J. and P.M. Lyrene. 1995. Characteristics of sparkleberry Blueberry hybrids. Proc. Fla. St. Hort. Soc. 108:337-339. Brooks, S.J. and P.M. Lyrene. 1998a. Derivatives of Vaccinium arboreum Vaccinium section Cyanococcus: I. Morphological charact eristics. J. Amer. Soc. Hort. Sci. 123(2):273-277. Brooks, S.J. and P.M. Lyrene. 1998b. Derivatives of Vaccinium arboreum Vaccinium section Cyanococcus: II Fertility and fertility parameters. J. Amer. Soc. Hort. Sci. 123(6):9971003. Camp, W.H. 1942. On the structure of populations in the Genus Vaccinium Brittonia 4(2):189204. Camp, W.H. 1945. The North American blue berries with notes on other groups of Vacciniaceae Brittonia 5(3):203-275. Chandler, C.K. and P.M. Lyrene. 1982. Relatio nship between guard cell length and ploidy in Vaccinium HortScience 17, 53-54. Coville, F.V. 1921. Directions for blueberry culture. U.S.D.A. Bul. 974. Coville, F. 1937. Improving the wi ld blueberry. USDA Yearbook. 559-574 pp. Darrow, G.M. and W.H. Camp. 1945. Vaccinium hybrids and the development of new horticultural material. Bul. Torrey Bot. Club. 72(1):1-21. Darrow, G.M., H. Dermen, and D.H. Scott. 1949. A tetraploid blueberry from a cross of diploid and hexaploid species. J. Her.40:304-306 pp. 175

PAGE 176

Dermen, H. 1940. Colchicine polyploi dy and technique. Bot. Rev. 6:599-635. Dermen, H. and H.F. Bain. 1944. A general cytohi stological study of colchicine polyploidy in cranberry. Ame. J. Bot. 31(8):451-463. Dweikat, I.M. and P.M. Lyrene. 1991. Induced tetraploidy in a Vaccinium elliottii clone facilitates crossing with cu ltivated highbush blueberry. J. Amer. Soc. Hort. Sci. 116(6):1063-1066. Ehlenfeldt, M.K. and N. Vorsa. 1993. The genera tion, evaluation and uti lization of hexaploid progeny from 3x x 3x crosses of highbush blueberry: germplasm transfer and 2n gametes in blueberry. Acta Hort. 346:95-102. Goldy, R.G. and P.M. Lyrene. 1983. Meiotic abnormalities of Vaccinium ashei V. darrowi hybrids. Can. J. Genet. Cytol. 26:146-151. Hokanson, K. and J. Hancock. 2000. Early-acting inbreeding depression in three species of Vaccinium (Ericaceae). Sex Plant Reprod. 13:145-150. Krebs, S.L. and J.F. Hancock. 1990. Early-acting inbreeding depression and reproductive success in the highbush blueberry, Vaccinium corymbosum L. Theor. Appl. Genet. 79:825-832. Krebs, S.L. and J.F. Hancock. 1991. Embryonic genetic load in the highbush blueberry, Vaccinium corymbosum ( Ericaceae). Amer. J. Bot. 78(10):1427-1437. Lyrene, P.M. 1986. Variation within Vaccinium darrowi blueberry in Florida. HortScience. 21(3):512-514. Lyrene, P.M. 1991. Fertile derivatives from Sparkl eberry Blueberry crosses. J. Amer. Soc. Hort. Sci. 116(5):899-902. Lyrene, P.M. 1997. Value of various taxa in breedi ng tetraploid blueberries in Florida. Euphytica 94:15-22. Lyrene, P.M. and J.R. Ballington. 1986. Wide hybridization in Vaccinium HortScience 2(1):5257. Lyrene, P.M. and J.L. Perry. 1982. Production and se lection of blueberry pol yploids in vitro. J. Hered. 73:377-378. Lyrene, P.M. and W.B. Sherman. 1983. Mitoti c instability and 2n gamete production in Vaccinium corymbosum V. elliottii hybrids. J. Amer. Soc. Hort. Sci. 108(2):339-342. Meader, E.M. and G.M. Darrow. 1944. Pollination of the rabbiteye blueberry and related species. Amer. Soc. Hort. Sci. 45:267-274. Megalos, B.S. and J.R. Ballington. 1988. Unreduced pollen frequencies ve rsus hybrid production in diploid-tetraploid Vaccinium crosses. Euphytica 39: 271-278. 176

PAGE 177

Merrill, T.A. 1936. Pollination of the highbush bl ueberry. Mich. Agr. Exp. Sta. Tech. Bul. 151. Merrill, T.A. and S. Johnston. 1940. Further obs ervations on the pollination of the highbush blueberry. Proc. Amer. Soc. Hort. Sci. 37:617-619. Moore, J.N. 1965. Improving highbush blueberrie s by breeding and selec tion. Euphytica 14: 3948. Morrow, E.B. 1943. Some effects of cross-pollina tion experiments versus self-pollination in cultivated blueberries. Proc. Amer. Soc. Hort. Sci. 42:469-472. Multiple authors. 1997. The Brooks and Olmo Regi ster of Fruit and Nut Varieties. ISHS publication. Muoz, C.E. and P.M. Lyrene. 1984a. In vitro at tempts to overcome the cross-incompatibility between V. corymbosum L. and V. elliottii Champ. Theor. Appl. Genet. 69:591-596. Muoz, C.E. and P.M. Lyrene. 1984b. Reproductive incompatibility barriers in crosses between Vaccinium corymbosum and V.elliottii. Can. J. Bot. 63:1987-1996. Perry, J.L. and P.M. Lyrene. 1984. In vitro induction of tetraploidy in Vaccinium darrowi V. elliottii, and V. darrowi x V. elliottii with colchicine treatment. J. Amer. Soc. Hort. Sci. 109(1):4-6. Qu, L. and J.F. Hancock. 1995. Nature of 2n gamete formation and mode of inheritance in interspecific hybrids of diploid Vaccinium darrowi and tetraploid V. corymbosum Theor. Appl. Genet 91:1309-1315. Qu, L., J.F. Hancock, and J.H. Whallon. 1998. E volution in an autopol yploid group displaying predominantly bivalent pairing at meio sis: genomic similarity of diploid Vaccinium darrowi and autotetraploid V. corymbosum ( Ericaceae). Amer. J. Bot. 85(5):698-703 Rieseberg, L.H. 1995. The role of hybridization in evolution: Old wine in new skins. Amer. J. Bot. 82(7):944-953. Rousi, A. 1966. The use of North-European Vaccinium species in blueberry breeding. Acta Agriculture Scandina via Suppl. 16:50-54. Sharpe, R.H. 1953. Horticultural development of Fl orida blueberries. Proc. Fla. St. Hort. Soc. 66:188-190. Sharpe, R.H. and G.M. Darrow. 1959. Breeding bluebe rries for the Florida climate. Proc. Fla. St. Hort. Soc. 71:308-311. Sharpe, R.H. and W.B. Sherman. 1971. Breedin g blueberries for low-chilling requirement. HortScience. 6 (2):145-147. 177

PAGE 178

Sharpe, R.H. and J.S. Shoemaker. 1958. Developmen t of temperate-climate fruits for Florida. Fla. St. Hort. Soc. 811:294-300. Spann, T.M., J.G. Williamson, and R.L. Darnell. 2003. Photoperiodic effects on vegetative and reproductive growth of Vaccinium darrowi and V. corymbosum interspecific hybrids. HortScience 38(2):193-195. Stelly, D.M. and S.J. Peloquin. 1983. Methods to detect 2n female gametophytes in Solanums Amer. Potato J. 60:821. Stockton, L.A. 1976. Propagation and autecology of V. arboreum and its graft compatibility with V. ashei M.S. Thesis, Texas A&M Univ., College Station. Vander Kloet, S.P. and P.M. Lyrene. 1987. Self -incompatibility in diploid, tetraploid, and hexaploid Vaccinium corymbosum Can. J. Bot. 65:660-665. Wenslaff, T. and P.M. Lyrene. 2003a. The use of mentor pollination to facilitate wide hybridization in blueberry. HortScience 35(1):114-115. Wenslaff, T. and P.M. Lyrene. 2003b. Unilateral cross compatibility in Vaccinium elliottii V. arboreum an intersectional blueberry hybrid. Euphytica 131:255-258. White, E. and J.H. Clark. 1939. Some results of se lf-pollination of the highbush blueberry at Whitesbog, New Jersey. Proc. Amer. Soc. Hort. Sci. 36:305-309. 178

PAGE 179

BIOGRAPHICAL SKETCH Dario Chavez, born in Riobamba, Ecuador, in 1983, is the youngest son of Carlos Chavez and Ximena Velasquez. His three older brothers taught him severa l lessons of life that allowed him to look for new ideas and goals. Since he was 6 years old, he studied under the Salesians, but like many of us, he was undecided about what was next after high school. His father, a farmer by heritage and passion, ta ught him to love farming, yet Dario never thought about relating the farm to a field of st udy. Alex, his older brot her, challenged him to apply to Zamorano University. This well-know n agricultural school in Honduras has the educational basis of learning by doi ng. His studies started in 2002 and in four years he earned his bachelor of science in ag ricultural science and production. In Zamorano, Dario made new friends from all over Latin America. After graduation, he worked at Ohio State University for six months as a short-term scholar. His work was suspended with his desired opportunity to pursue a masters degree in plant breeding and genetics, where he met Dr. Lyrene at the University of Florida. Dario met several people who changed his perc eption of life. He also met many who have taught him to love and have passion for what they do: his father his mother, his brothers and finally his mentor, Dr. Lyrene. Darios ne w goal will be to start his Ph.D. studies. 179