<%BANNER%>

Field Performance of Southern Highbush Blueberry Cultivars (Vaccinium Corymbosum) Obtained from Micropropagation and Sof...

Permanent Link: http://ufdc.ufl.edu/UFE0045078/00001

Material Information

Title: Field Performance of Southern Highbush Blueberry Cultivars (Vaccinium Corymbosum) Obtained from Micropropagation and Softwood Cuttings in Two Florida Locations
Physical Description: 1 online resource (111 p.)
Language: english
Creator: Marino, Silvia R
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: culture -- growth -- propagation -- tissue -- vegetative -- yield
Horticultural Sciences -- Dissertations, Academic -- UF
Genre: Horticultural Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The study was conducted at two locations with different average chill hour accumulation: Citra, FL (420-540 chill hours) and Haines City FL, (110-210 hours/year). In both locations, ‘Emerald’, ‘Jewel’ and ‘Primadonna’ were planted in April 2010.Whole plants were harvested at planting, after the first and second growing seasons and after the first harvest. Plant canopy volume was evaluated from June to November in 2010 and 2011 and flower buds were counted at the end of both years at both locations. Fruit yield, size and quality were evaluated in Citra and Haines City in 2010 and in Citra only for 2011. Propagation effect on plant volume varied within cultivars and between the first and second growing seasons. Nineteen months after planting, propagation effect was not significant for any of the cultivars at either one of the locations. Micropropagated 'Emerald' and 'Jewel’ plants had more canes and total shoots at planting and after the first and second growing seasons, with no significant effect on total shoot number for 'Primadonna' at any point during the study. Plant dry weight was greater for 'Emerald' and 'Jewel' TC than for SW after one season but for the second season plant dry weight was greater for TC 'Jewel', but not for Emerald, compared to SW plants. Micropropagation resulted in significantly greater yield for 'Emerald' and 'Jewel' plants at both locations during the first harvest season, without any effect of propagation type on yield for 'Primadonna', at either location. Fruit yield was not significantly affected by propagation method during the second harvest season. Reduced average berry weight was recorded for TC'Emerald' at the beginning of first harvest season, in Haines City, without any differences between propagation types during the rest of the season or at any point during the harvest in Citra. Fruit size for the other two cultivars was not affected by propagation method. Berry weight was not affected by propagation type during the second harvest.
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 Silvia R Marino.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Williamson, Jeffrey G.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0045078:00001

Permanent Link: http://ufdc.ufl.edu/UFE0045078/00001

Material Information

Title: Field Performance of Southern Highbush Blueberry Cultivars (Vaccinium Corymbosum) Obtained from Micropropagation and Softwood Cuttings in Two Florida Locations
Physical Description: 1 online resource (111 p.)
Language: english
Creator: Marino, Silvia R
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: culture -- growth -- propagation -- tissue -- vegetative -- yield
Horticultural Sciences -- Dissertations, Academic -- UF
Genre: Horticultural Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The study was conducted at two locations with different average chill hour accumulation: Citra, FL (420-540 chill hours) and Haines City FL, (110-210 hours/year). In both locations, ‘Emerald’, ‘Jewel’ and ‘Primadonna’ were planted in April 2010.Whole plants were harvested at planting, after the first and second growing seasons and after the first harvest. Plant canopy volume was evaluated from June to November in 2010 and 2011 and flower buds were counted at the end of both years at both locations. Fruit yield, size and quality were evaluated in Citra and Haines City in 2010 and in Citra only for 2011. Propagation effect on plant volume varied within cultivars and between the first and second growing seasons. Nineteen months after planting, propagation effect was not significant for any of the cultivars at either one of the locations. Micropropagated 'Emerald' and 'Jewel’ plants had more canes and total shoots at planting and after the first and second growing seasons, with no significant effect on total shoot number for 'Primadonna' at any point during the study. Plant dry weight was greater for 'Emerald' and 'Jewel' TC than for SW after one season but for the second season plant dry weight was greater for TC 'Jewel', but not for Emerald, compared to SW plants. Micropropagation resulted in significantly greater yield for 'Emerald' and 'Jewel' plants at both locations during the first harvest season, without any effect of propagation type on yield for 'Primadonna', at either location. Fruit yield was not significantly affected by propagation method during the second harvest season. Reduced average berry weight was recorded for TC'Emerald' at the beginning of first harvest season, in Haines City, without any differences between propagation types during the rest of the season or at any point during the harvest in Citra. Fruit size for the other two cultivars was not affected by propagation method. Berry weight was not affected by propagation type during the second harvest.
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 Silvia R Marino.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Williamson, Jeffrey G.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0045078:00001


This item has the following downloads:


Full Text

PAGE 1

1 FIELD PERFORMANCE OF SOUTHERN HIGHBUSH BLUEBERRY CULTIVARS ( VACCINIUM CORYMBOSUM ) OBTAINED FROM MICROPROPAGATION AND SOFTWOOD CUTTINGS IN TWO FLORIDA LOCATIONS By SILVIA ROSA MARINO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Silvia Rosa Marino

PAGE 3

3 To my par ents, who always believed in me

PAGE 4

4 ACKNOWLEDGMENTS I thank my parents for their love and support and the values they thought me which I try to live by I sincerely thank my advisors, Dr. Jeffrey G. Williamson and Dr. James W. Olmstead, for the ir guidance support and encouragement. I acknowledge my third committe e member Dr. Phillip Harmon I express gratitude to all the students and members of the blueberry breeding program that hel ped during the course of this research I am deeply grateful to David Norden for his help with the d estructive sampling experiment and all throughout the study. I want to thank Dr. Darnell for help ing with the carbohydrate analysis and for the use of her lab oratory. I want to acknowledge James Colee the statistician consultant, for his help with data analysis. I want to acknowledge the s taff at Citra research center as well as Mixon family farms staff in Haines City. I am also grateful for my friends and the people I met in Gainesville who made my life as a graduate student a little more fun.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ ........ 11 LIST OF ABBREVIA TIONS ................................ ................................ ........................... 12 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 LITERATURE REVIEW ................................ ................................ .......................... 15 Blueberry Industry ................................ ................................ ................................ ... 15 Worldwide Blueberry Production ................................ ................................ ...... 15 U.S. and Florida Blueberry Industry ................................ ................................ 15 Blueberry Consumption in the United States ................................ .................... 16 Taxonomy and History of Cultivated Blueberry ................................ ....................... 17 History and Status of Blueberry Cultivation in Florida ................................ ............. 19 Vegetative Propagation of Blueberry and other Fruit Crops ................................ .... 21 Propagation of Blueberries ................................ ................................ ............... 21 Effect of Propagation Method on Vegetative and ................................ ............. 25 Reproductive Growth of Blueberry and other Small Fruit Crops ....................... 25 2 EFFECT OF PROPAGATION TYPE ON VEGETATIVE GROWTH OF THREE SOUTHERN BLUEBERRY CULTIVARS ( VACCINIUM CORYMBOSUM L.) OBTAINED FROM MICROPROPAGATION AND SOFTWOOD CUTTINGS ......... 35 Introduction ................................ ................................ ................................ ............. 35 Materials and Methods ................................ ................................ ............................ 37 Locations and Experimental Design ................................ ................................ 37 Cultural Practices ................................ ................................ ............................. 38 Vegetative Field Measurements ................................ ................................ ....... 39 De structive Sampling ................................ ................................ ....................... 40 Root Measurements ................................ ................................ ......................... 42 Statistical Analysis ................................ ................................ ............................ 42 Results ................................ ................................ ................................ .................... 43 Plant Volume during the First Growing Season. ................................ ............... 43 Plant Volume during the Second Growing Season. ................................ .......... 44 Number of Major Canes, Lateral Branches and Total Shoots .......................... 44 Initial Destructive Sampling (DS0) ................................ ................................ .... 45 Field Destructive Sampling One (DS1) ................................ ............................. 45 Destructive Sampling after First Harvest (DS2) ................................ ................ 47

PAGE 6

6 Destructive Sampling after Second Growing Season (DS3) ............................. 48 Root Measurements ................................ ................................ ......................... 49 Discussion ................................ ................................ ................................ .............. 51 Plant Volume, Size, We ight and Branching Data ................................ ............. 51 Root Measurements ................................ ................................ ......................... 54 Conclusions ................................ ................................ ................................ ............ 56 3 PLANT TISSUE CARBOHYDRATE CONTENT OF SOUTHERN HIGHBUSH CULTIVARS ( VACCINIUM CORYMBOSUM L.) OBTAINED FROM MICROPROPAGATION AND ROOTED CUTTINGS ................................ ............. 67 Introduction ................................ ................................ ................................ ............. 67 Materials and Methods ................................ ................................ ............................ 68 Results ................................ ................................ ................................ .................... 71 Soluble Sugars at Planting ................................ ................................ ............... 71 Soluble Sugars after the First Harvest ................................ .............................. 72 Plant Starch at Planting ................................ ................................ .................... 73 Plant Starch after First Harvest ................................ ................................ ........ 74 Total Plant Carbohydrate Content ................................ ................................ .... 75 Discussion ................................ ................................ ................................ .............. 75 Conclusions ................................ ................................ ................................ ............ 77 4 EFFECT OF PROPAGATION METHOD ON REPRODUCTIVE GROWTH, YIELD AND FRUIT QUALITY OF SHB CULTIVARS (VACCINIUM CORYMBOSUM L.) OBTAINED FROM MICROPROPAGATION AND SOFTWOOD CUTTINGS ................................ ................................ ....................... 82 Introduction ................................ ................................ ................................ ............. 82 Materials and Methods ................................ ................................ ............................ 84 Flower Bud Number ................................ ................................ ......................... 84 Full Bloom Date ................................ ................................ ................................ 85 Yield ................................ ................................ ................................ ................. 85 Berry Weight ................................ ................................ ................................ ..... 85 Fruit Quality ................................ ................................ ................................ ...... 86 Statistical Analysis ................................ ................................ ............................ 86 Results ................................ ................................ ................................ .................... 87 Flower Buds ................................ ................................ ................................ ..... 87 Full Bloom Date ................................ ................................ ................................ 88 Yield ................................ ................................ ................................ ................. 88 Berry Weight ................................ ................................ ................................ ..... 89 Fruit Quality ................................ ................................ ................................ ...... 90 Discussion ................................ ................................ ................................ .............. 92 Effect of Propagation Method on Flowering ................................ ...................... 92 Effect of Propagation Method on Reproductive Growth and Yi eld .................... 92 Berry Weight ................................ ................................ ................................ ..... 94 Fruit Quality ................................ ................................ ................................ ...... 94 Conclusions ................................ ................................ ................................ ............ 95

PAGE 7

7 5 SUMMARY AND CONCLUSIONS ................................ ................................ ........ 100 APPEND IX : ADDITIONAL TABLES ................................ ................................ ............ 103 LIST OF REFERENCES ................................ ................................ ............................. 105 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 111

PAGE 8

8 LIST OF TABLE S Table page 2 1 Effect of cultivar and propagation method on plant volume during the first growing season (June to November 2010). Means of both locations. ................ 61 2 2 Effect of cultivar and propagation method on plant volume at two locations in 2011. ................................ ................................ ................................ .................. 61 2 3 Effect of cultivar and propagation method on plant volume in September and October 2011. Means of both locations. ................................ ............................. 61 2 4 Effect of cultivar and propagation method on the number of canes/plant from planting until end of second growing season. ................................ ..................... 62 2 5 Effect of cultivar and propagation method on plant size and dry weights at planting, April 2010. ................................ ................................ ............................ 63 2 6 Effect of location and propagation method on leaf and cane dry weights in November/December 2010. ................................ ................................ ................ 63 2 7 Effect of cultivar and propagation method on crown dry weight and root:shoot ratio at two locations in November/December 2010. ................................ .......... 63 2 8 Effect of cultivar a nd propagation method on number of canes, lateral branches and total shoots per plant, width, volume and shoot and plant dry weights in November/Decemer 2010. Means of both locations. ......................... 64 2 9 Effect of cultivar and propagation method on height, volume and dry weights in June 2011. Means of both locations. ................................ .............................. 64 2 10 Effect of location and propagation method on plant width, leaf dry weight and root:shoot ratio in June 2011. ................................ ................................ ............. 64 2 11 P values from ANOVA sliced by cultivar, for cane and total shoot number, cane, total shoot, top, root, and plant dry weights in November 2011. ............... 64 2 12 Effect of cultivar and propagation method on number of canes and total shoots, and cane, total shoot, top, root, and plant dry weights in November 2011. Means of both locations. ................................ ................................ ........... 65 2 13 Effect of cultivar, propagation method and location on leaf dry weight in November 2011. ................................ ................................ ................................ 65 2 14 Effect of cultivar and propagation method on total length of roots 2.5 3 mm in diameter (L6) and roots with diameters greater than 3 mm (L7) at DS1. Means of both locations. ................................ ................................ ..................... 65

PAGE 9

9 2 15 Effect of propagation method on total root length (m), total surface area (m 2 ), length of roots 0 0.5 mm (L1), 0.5 1 (L2), 1 1.5 (L3), 1.5 2 (L4), 2 2.5 (L5), 2.5 3 (L6) and > 3 mm in diameter (L7) at Citra, DS2. Means of both locations. ................................ ................................ ................................ ............ 66 3 1 Effect of cultivar and propagation method on sugar concentration and total sugar content in April 2010. ................................ ................................ ................ 79 3 2 Effect of cultivar and propagation method on sugar concentration and total sugar content in June 2011. ................................ ................................ ............... 79 3 3 Effect of propagation method on total root s ugar content and crown and leaf sugar concentration and in June 2011. ................................ ............................... 79 3 4 Effect of cultivar and propagation method on starch co ncentration and total starch content in April 2010. ................................ ................................ ............... 80 3 5 Effect of cultivar and propagation method on starch concentration and total starch content in June 2011. ................................ ................................ ............... 81 3 6 Effect of cultivar and propagation method on plant total carbohydrate in April 2010 and Ju ne 2011. ................................ ................................ .......................... 81 4 1 Effect of cultivar and propagation method on average number of flower buds per cane and total flower buds per plant during the first and second reproductive seasons. Means of both locations. ................................ ................. 97 4 2 Effect of cultivar and propagation method on perce ntage of full bloom at Citra on 4 March 2011 and Haines City on 14 February 2011. ................................ ... 97 4 3 Effect of cultivar and propagation method on yield during the second harvest season (2012). Data from Citra, FL only. ................................ ............................ 98 4 4 Effect of cultivar, propagation method and locat ion on berry weight. Data recorded at Citra on 30 April, and at Haines City on 25 April, 2011. ................... 98 4 5 Effect of cultivar and propaga tion method on berry weight at four points during the first harvest season. Means of both locations. ................................ ... 99 4 6 Effect of cultivar and propagation method on fruit pH, total soluble solids and titratable acidity at three points during the first harvest season. Means of both locations. ................................ ................................ ................................ ............ 99 4 7 Effect of cultivar and propagation method on soluble sugar content and titratable acidity. Fruit harvested from Citra on 22 May and 1 June, 2011. ......... 99 A 1 P values from ANOVA for plant volume in 2010 ................................ ............... 103

PAGE 10

10 A 2 P values from ANOVA for plant volume in 2011 ................................ ............... 103 A 3 Effect of cultivar and location on flower bud number in 2011. Means are averages of propagation methods. ................................ ................................ ... 103 A 4 Effect of cultivar and location on percentage of full bloom on 4 March (Citra) and 14 February (Haines City). Means of propagation methods. ..................... 104 A 5 Effect of cultivar and location on fruit yield (g) in 2011. Means of propagation methods. ................................ ................................ ................................ ........... 104

PAGE 11

11 LIST OF FIGURE S Figure page 2 1 Picture of the field in Citra at planting. April, 2010. ................................ ............ 57 2 2 Picture of the field in Haines City at planting. April, 2010. ................................ .. 57 2 3 Leaves were removed from canes and shoots prior to collecting dry weights. ... 58 2 4 Picture shows major canes and lateral branches after leaves were removed. ... 58 2 5 Whole root systems were washed over a 2 mm. sieve ................................ ....... 59 2 6 Measuring the width of the root system prior to harvesting plants in Citra (November, 2011). ................................ ................................ .............................. 59 2 7 E xa mple of root branches selected for scanning on root systems from November/December 2010 and June 2011. ................................ ....................... 60 2 8 Picture illustrates how roots were divided in eighths to obtain the subsamples used for scanning after the second growing season. ................................ .......... 60 2 9 Effect of cultivar and propagation method on total shoot number per plant. Means of both locations. ................................ ................................ ..................... 62 4 1 Effect of cultivar and propagation method on total yield in 2011. Means of both locations. ................................ ................................ ................................ .... 98

PAGE 12

12 LIST OF ABBREVIATION S FB Flower buds DW Dry weight LSM Least Square Means SHB Southern highbush blueberry SW Softwood cuttings T Tons TC Tissue culture, tissue culture derived Treat Treatment

PAGE 13

13 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science FIELD PERFORMANCE OF SOUTHERN HIGHBUSH BLUEBERRY CULTIVARS ( VACCINIUM CORYMBOSUM ) OBTAINED FROM MICROPROPAGATION AND SOFTWOOD CUTTINGS IN TWO FLORIDA LOCATIONS By Silvia Rosa Marino December 2012 Chair: Jeffrey G. Williamson Major: Horticultural Sciences The study was conducted at two locations with different average chill hour accumulation: Citra, FL (420 540 chill hours) and Haines City FL, (110 210 hours/year). plan ts were harvested at planting, after the first and second growing seasons and after the first harvest. Plant canopy volume was evaluated from June to November in 2010 and 2011 and flower buds were counted at the end of both years at both locations. Fruit y ield, size and quality were evaluated in Citra and Haines City in 2010 and in Citra only for 2011. Propagation effect on plant volume varied within cultivars and between the first and second growing seasons. Nineteen months after planting, propagation eff ect was not significant for any of the cultivars at either one of the locations. planting and after the first and second growing seasons, with no significant effect on total sh oot number for 'Primadonna' at any point during the study. Plant dry weight was greater for 'Emerald' and 'Jewel' TC than for SW after one season but for the second

PAGE 14

14 season plant dry weight was greater for TC 'Jewel', but not for Emerald, compared to SW pla nts. Micropropagation resulted in significantly greater yield for 'Emerald' and 'Jewel' plants at both locations during the first harvest season, without any effect of propagation type on yield for 'Primadonna', at either location. Fruit yield was not sig nificantly affected by propagation method during the second harvest season. Reduced average berry weight was recorded for TC 'Emerald' at the beginning of first harvest season, in Haines City, without any differences between propagation types during the re st of the season or at any point during the harvest in Citra. Fruit size for the other two cultivars was not affected by propagation method. Berry weight was not affected by propagation type during the second harvest season.

PAGE 15

15 CHAPTER 1 LITERATURE REVIEW Blueberry Industry Worldwide Blueberry Production Since the 1970s blueberry cultivation and total world production have increased considerably. In 2010, total world production was approximately 341,000 t (Brazelton, 2011). The United States is and has been the largest blueberry producer since 1970. In 2011, US production of 214,000 t of cultivated blueberries more than doubled that of Canada, which ranks second in world blueberry production. Poland, Lithuania, Germany, Romania, Netherlands, Ukraine, Sweden and New Zealand, in decreasing order, complete the list of 10 top blueberry producers (USDA, 2012). U.S. and Florida Blueberry I ndustry Although M ichigan and Maine are still the biggest blueberry producers in the U.S., their combined share of total production during 2008 2010 declined to around 40%. In contrast, other states such as Georgia, Washington, Oregon, North Carolina, New Jersey, California and Florida increased their production. In some states this increase was due to more harvested area, other states saw an increase in yield/acre and some state increases in production were due to both increased acreage and yield (USDA, 2012). Harvested ar ea in Florida has shown an upward trend since 2004, reaching 1538 hectares in 2011. Florida yields have also increased since 2007, with a record high in 2011, Although yield per acre in Florida is low in comparison to other leading states, it is similar to that in Georgia, the largest blueberry producer in the Southeastern U.S (USDA, 2012). In 2010, the value of Florida blueberry production reached just over $47 million. ( Geisler, 2012 ; USDA, 2012)

PAGE 16

16 In the last decade, blueberry production has been the faste st growing industry among all temperate fruit crop crops in Florida (Williamson and Crane, 2010). Florida blueberry production increased from 1.27 million kg in 2000 to 7.99 million kg in 2010 (USDA, 2011). Almost all Florida blueberry production is destin ed for the fresh market, which has higher prices than fruit for the process industry. In addition to this, Florida blueberries that are harvested from 1 April to 20 May fill a market window during which the state is the only producer of blueberries (Willia mson et al, 2012). For the past three years, seasonal average grower price for Florida blueberries ranged from at least $2.27 per kg in 2008 to approximately $1.36 per kg in 2010, while US average seasonal price was between $0.59 and $0.68 per kg (USDA, 20 12). Blueberry C onsumption in the United States Blueberry per capita consumption in the US has increased during the last two decades from 0.077 kg in 1991 to 0.5 kg in 2010. However, fresh consumption increased more than frozen blueberry consumption over t he last decade and by 2002 per capita fresh blueberry consumption exceeded that of frozen (House and Wysocki, 2012). Fresh blueberry consumption exhibited an upward trend since the 1990s when it averaged 0.11 kg/year, and by 2010 it had increased to 0.5 kg /year. Continued increase in fresh blueberry demand led to increase in fresh market production, exceeding that for processed use starting in 2002 (Geisler, 2012). U.S. fresh production in 2010 was 110 million kg, which more than tripled production from 20 00. Production of blueberries for the processed industry also reached its high in 2010, with nearly 76.5 million kg. In 2010 total crop value reached $640.7 million, second to strawberries among berry crops in the U.S. (Geisler, 2012).

PAGE 17

17 Imports and exports. Blueberry fresh imports reached almost 80 million kg in 2010. From 2008 to 2010 around 47% of the blueberries consumed fresh in U.S. were imported, mainly from Canada, Chile and Argentina. Despite the continued rise in domestic production during recent ye ars and the contribution of fresh blueberries from the southern hemisphere this does not seem enough to meet the continuously increasing demand for blueberries in the United States (House and Wysoki, 2012). Frozen blueberry imports also rose, averaging 60% of the local consumption, during the same time. Currently, the United States exports more fresh blueberries than frozen, with a peak of 36 million kg in 2011, while frozen blueberry exports for the same time averaged 16 million kilos (AGMRC, 2012). Taxon omy and History of Cultivated B lueberry The cultivated blueberry belongs to the Vaccinium genus within the Ericaceae family. One distinctive characteristic of the family is a requirement for acidic soils. The Vaccinium genus comprises around 400 species; a pproximately 25% are native to North America, many of which are cultivated for their fruit. Blueberry relatives include the cranberry ( V. macrocarpon Aiton ), the lingonberry ( V. vitis idaea L.), small cranberry ( V. oxycoccos L.), and deerberry ( V. stamineu m L.), to name a few. The highbush blueberry was first described by Linnaeus in 1753 from material collected in eastern North America and has had a complicated taxonomic history. Chapman (1987) recognized five species and three varieties of highbush blu eberry, but Camp (1945) characterized 12 different Vaccinium species under this designation (Vander Kloet, 1980). Blueberry cultivation did not start until the end of the 19th century (Vander Kloet, 1988). Improvement of the wild blueberry did not start u ntil 1908, when Dr. F. V.

PAGE 18

18 Coville, after spending two years studying the cultural requirements of the blueberry, selected the first wild highbush blueberry used for breeding in 1908. It produced large berries, reaching over a 2.5 cm in diameter, the flesh was firm and juicy and had New Hampshire in which the plant was found. The second blueberry selected by Coville Vaccinium Angustifolim ) (Co ville, 1937). In 1911, Coville initiated a hybridization program using these and other wild blueberries as parents project and offered cash prizes for highbush blueberry plants that produced the largest berries, which she would plant in her farm in Whitesbog, New Jersey (Vander Kloet, 1988). Intercrosses between some of these wild selections gave origin to the first released in 1920. At the time of Dr introduced. Dr. George M. Darrow, a blueberry breeder for the United States Department of Agriculture (Beltsville, Maryland) initiated a coope rative program of seedling and selection testing with several state experiment stations and private growers. The program aid in the development of varieties adapted to different environments. Other states joined the program soon thereafter. Since then, sev eral blueberry varieties have been developed from breeding programs at different universities, including New Jersey, Michigan, North Carolina, Arkansas, Mississippi and Florida (Lyrene and Moore, 2006). In 1992, J. Moore (1994) conducted a survey of the bl ueberry plantings in North America. He reported that major highbush blueberry producers, in order of decreasing

PAGE 19

19 acreage, were Michigan, New Jersey, North Carolina, Oregon, Arkansas, Washington and New York, while the two major rabbiteye producing states we re Georgia and Florida. Wild lowbush blueberry production was centered in Maine, Quebec and Nova Scotia and southern highbush was found in large plantings only in Florida at that time. (Moore,1994). Currently commercially grown blueberries can be divided i nto five major cultivated groups based primarily on the location they are best adapted to: 1) lowbush (primarily V. angustifolium Ait., with lesser contributions from V. myrtilloides Michx., and V. boreale Hall and Aald.); 2) highbush ( V. corymbosum L.); 3 ) half highbush (hybrids between highbush and lowbush blueberries); 4) rabbiteye ( V. virgatum ); and 5) southern highbush (hybrids between highbush, rabbiteye, and several wild southern species with low chilling requirement (Debnath, 2007). History and Stat us of Blueberry C ultivation in Florida Cultivation of rabbiteye blueberry in Florida began in the 1890s, in the western Florida panhandle. These first plantings were established with native rabbiteye plants transplanted from the woods to cultivated fields (Williamson, 2004). Although this may have been the first attempt to cultivate blueberries in the world, plantings were soon abandoned because of poor fruit quality and marketing problems. Despite some small plantings being established in 1960s for local consumption, rabbiteye blueberries were not planted in Florida for commercial harvest, packing, or shipping because they ripen later than highbush blueberries grown in North Carolina (Williamson and Lyrene, 2004a). In 1949 professor Ralph Sharpe at the Uni versity of Florida began breeding blueberries, peaches and other temperate fruit crops. The goal in the blueberry program

PAGE 20

20 was to obtain cultivars that would ripen early and have a low chilling requirement. These factors would allow production in areas with mild winters and longer growing seasons than where northern highbush blueberries thrive (Williamson, 2004). During the early 1950s, Sharpe identified Vaccinium darrowii an evergreen blueberry that grew abundantly in Florida as far south as Lake Placid, a s a potential contributor of low chill genes to be incorporated into V. corymbosum. The V. darrowii clone Fla 4B, which is in the pedigree of most southern highbush cultivars, was collected near Winter Haven, FL (Lyrene, 1998). Crosses between V. darrowii and V. virgatum and V. darrowii and Sharpe and Sherman (Lyrene, 1998).). A third cult later. Fruit from these cultivars ripened a month earlier than the earliest rabbiteyes grown in Florida at that time (Williamson and Lyrene, 2004b). commercial hectares in 1973 to 1538 hectares in 2011. In 1989, around two thirds of the acreage was planted with rabbiteye varieties, and southern highbush acreage comprised one third of the Florida acreage (Crocker, 1989). Although total Florida blueberr y acreage decreased from 1989 to 2000 (from 852 to 631 hectares), southern highbush cultivation increased by 23% while rabbiteye planted area decreased 56% (Williamson, 2000). Most rabbiteye plantings have been replaced with southern highbush cultivars, wh ich are currently the preferred blueberry species grown commercially in Florida. Most of the state blueberry production is shipped fresh and a small portion, mostly rabbiteye, is sold locally (Williamson et al., 2012).

PAGE 21

21 Although southern highbush varieties have been planted along the Gulf Coast from South Georgia to east Texas and in California, the market period from April 1 to May 10 is still available almost exclusively to Florida growers, with fruit prices usually high until late May when Georgia and Nor th Carolina harvests have begun. (Williamson and Lyrene, 2004) Vegetative Propagation of Blueberry and other Fruit C rops Vaccinium species are heterozygous, and plants produced from seed are generally different from their parents. Additionally seedlings ex perience a juvenile period during which they will not produce flowers (Griffin, 1989). Therefore, blueberries, as well as cranberry, lingonberry and other species in Vaccinium are vegetatively propagated to maintain the genetic and phenotypic characterist ics of the cultivar of interest and achieve fruit bearing quickly (Debnath, 2007). Sexual propagation of these species is only used extensively in breeding programs which often cross different cultivars and species to obtain hybrid plants. Most woody spec ies are vegetatively propagated by means of hardwood, semi hardwood and softwood cuttings. Others, such as the black raspberry (Rubus occidentallis) blackberry ( Rubus fruticosus ), boysenberry ( Rubus ursinus ), lemon ( Citrus x limon ) and rhododendron spp can also be produced from leaf bud cuttings. These are comprised of a leaf blade, petiole and a short piece of the stem with the axillary bud attached. Each node can be used as a cutting and therefore this method becomes very valuable when stock material i s limited. (Hartman,1975). Propagation of B lueberries Plant material for propagation by hardwood cuttings is taken during the dormant season. Late winter before any bud swell or early winter in cold places where low

PAGE 22

22 temperatures could damage the wood is th e optimal time to take the cuttings. These buds at the tip, and 0.6 to 1.27 cm in diameter. The cuttings are grouped in bundles with their terminal end up in moist saw dust bark or peat in plastic bags and stored at 1 to 9 C until time for sticking them in the propagation bed. Mist is not required when rooting hardwood cuttings (Mainland, 2006). Propagation by means of softwood cuttings starts with collection of plant ma terial tissue lignification so they can be stuck in the propagation bed. Cuttings should be kept moist and refrigerated until they can be stuck in trays for rooting. A com monly used rooting media for propagation under cover is a mix of 50% peat and 50% perlite, which provides a well drained, porous substrate, appropriate for root development (Mainland, 2006). Intermittent mist is needed for propagation with softwood and sem i hardwood cuttings. Shading also helps reduce evaporation from the leaf cuttings, which is essential until a root system capable of absorbing enough water to avoid desiccation has developed. Rooting may take up to several weeks and rooting percentages var y with the species and cultivar, time of the year, and age of the mother plant (Douglas 1966; Lyrene, 1981). Once a secondary fibrous root system has developed, to which the rooting medium clings in a ball, the new plants can be dug and potted. Increased r ooting percentage may be obtained by treating cuttings with rooting hormones, such as indole butyric acid (Hartman, 1975).

PAGE 23

23 Propagation by semi hardwood cutting is a process similar to that of softwood cuttings, but requires more mature material from curre more difficult as cuttings mature and this technique is not commonly used (Mainland, 2006) propagation in 1910. Tissue culture, in vitro propagation or micropropagation refer to the clonal propagation of plants from cells, tissues or organs, in a controlled, artifici al environment, in glass or plastic culture vessels, using aseptic techniques and a defined medium (Donnelly and Vidaver, 1988). Nickerson (1978) made the first attempt at micropropagating blueberries. At that time, lowbush blueberry was vegetatively prop agated by stem or rhizome cuttings and Nickerson used micropropagation as a method to reduce the time required to build up stock of selected, superior, clones. Although twelve micro shoots were excised after eight weeks of culture, only two rooted and both plantlets subsequently died (Nickerson, 1978). In the early 1980s rabbiteye blueberries, which were the predominant blueberry commercially grown in southeast United States, were propagated by softwood and hardwood cuttings. Plants could be produced at low cost but several years were needed to produce enough plants from a newly released cultivar to plant significant areas and some cultivars had low rooting percentage (Miller et al., 2004), or did not respond to root inducing growth regulators (Nickerson, 19 78, Meiners et al 2007) ex vitro. Lyrene (1980) started to micropropagate rabbiteye and highbush blueberries as a way to speed up the production of large quantities of plants of the new lines and cultivars from the

PAGE 24

24 University of Florida breeding program. M icropropagation has the potential to produce large numbers of plants in a short period of time. Isutsa et al. (1994) developed a protocol that could allow a plant propagator to produce over 1000 plants/m2 in 2 months and four months later these plants coul d be between 30 60 cm tall and suitable for field planting (Isutsa et al, 1994). Another advantage of micropropagation is that it produces disease free plants and production can take place year round, without seasonal differences (Meiners et al, 2007; Mill er et al., 2004). In addition, micropropagated plants grow more uniformly and develop more shoots than plants propagated from cuttings (Mainland, 2006). Different micropropagation techniques have been used to micropropagate blueberry. These include shoot t ip culture (Lyrene, 1980), node or meristem culture (Gajdosova et al, 2006), direct (adventitious shoot regeneration from leaves, Dweikat and Lyrene, 1988; Meiners et al 2007) or indirect shoot organogenesis (Read 1993). The regeneration pathway choice d epends on species, efficiency, and proliferation rate of the method chosen and final goal of the protocol. Extensive work has been done in developing micropropagation protocols for blueberry and some of its relatives, such as lingonberry and cranberry and for other small fruit crops, such as strawberries and blackberries (Anderson 1980; Brisette et al. 1990; Debnath and McRae 2001; Debnath 2008, Frett and Smagula 1983, Gajdosova et al. 2006; Jaakola et al. 2002; Ostrolucka et al. 2007, Reed and Abdelnour E squivel, 1991; Wolfe et al. 1983). However, this review will focus on studies that evaluated growth and development of small fruit crops as affected by the propagation method,

PAGE 25

25 with emphasis on vegetative and reproductive development of blueberry propagated in vitro or by means of rooted cuttings during the first two years on the field. Effect of P r opagation Method on V egetative and Reproductive Growth of Blueberry and other Small Fruit C rops Changes in growth habit and morphology of micropropagated plants h ave been reported for many fruit species. Swartz et al. (1981) reported increased vigor and runner production of strawberry tissue cultured (TC) plants compared to cutting derived plants (SW) plants. Reproductive growth was also affected by propagation met hod, as micropropagated plants had higher yields, due to increased number of berries/unit. However, decreased mean fruit weight for tissue culture derived plants was also seen (Swartz et al., 1981). Enhanced vegetative growth during the first years in the field was also reported for tissue culture propagated blackberries. Swartz et al. (1983) studied the effect of propagation method and number of subcultures on morphology and yield of five cultivars and one advanced selection of thornless blackberries. The cultivars were the selection was SI US 68 6 17. Tissue cultured plants were propagated from 5 th 6 th and 7 th consecutive proliferation subcultures and planted on 31 May 25 June and 18 July of 1979, respectively. Cutting derived plants were planted with each subculture. Longer and wider leaflets were observed in micropropagated blackberries of cultivars ls/plant during the first season of vegetative growth. There was no significant effect of propagation type on he second year, TC plants of all

PAGE 26

26 significantly more flowers/inflorescence than SC plants. There were no significant differences in flower bud break or number of inflorescences per meter of canes due to propagation method. Yield of TC plants (g/plant), however, was significantly higher for three cultivars and when all genotypes were analyzed tog ether. The second harvest season berry weight of tissue cultured plants was significantly lower than cutting derived Gustavsson (2000) compared field performance of micr opropagated lingonberry plants with those derived from stem cuttings and evaluated weight and fruit yield, accumulated plant growth, and rhizome derived daughter plants. Although some fruit set in the SC plants from stem cuttings, TC plants did not flower the first year. The second and third year after field planting, yield of TC derived plants was significantly higher than for SC plants, without any effect on average fruit weight during either year. In addition, TC plants produced significantly more rhizom e derived daughter plants and had higher accumulated growth harvested the third year after planting. Survival in the field was 97% for TC plants and 83% for SC plants, and was attributed by the authors to more branching of the TC plants. Similar results, w ith micropropagation resulting in more vigorous plants, have been reported with blueberries. Grout et al. (1986) evaluated vegetative and reproductive performance of half branches per plant a nd branch length of micropropagated and rooted cutting derived plants were recorded, from week 27 to week 34 1982. Number of branches in TC plants

PAGE 27

27 increased from four to six, while in SC plants they remained the same. Tissue cultured plants increased nearl y 50% in size, while plants derived from cuttings did not show a significant increase in size. The effect of chilling method and pruning on vegetative and reproductive growth of plants derived from both tissue culture and leaf bud cuttings was also evaluat ed. The field work was conducted at three sites: Grand Rapids, St. Paul and Becker (MN), and field planting of 82 week old plants took place between 25 May 25 and 1 June, 1983. The chilling method and pruning (or not) treatments did not affect basal branch number of tissue cultured plants. Four months after planting, tissue cultured plants had more lateral branches than stem cutting derived plants at St. Paul and Grand Rapids. There were no significant differences in the number of branches/plant at Becker. The number of flower buds/plant at St. Paul and Grand Rapids was twice for tissue cultured plants compared to plants derived from rooted cuttings. The difference in flower buds/plant was a consequence of the larger number of laterals or shoots/plant, since the flower buds/lateral was not significantly different between propagation types. There was no significant difference in flower bud number for the first reproductive season at Becker (Grout et al., 1986). Read et al. (1988, 1989) reported fruit yield of the fields at Grand Rapids, (MN) from 1984 through 1986 for the same plants. During the first two harvests, TC plants had twice the fruit yield of cutting derived plants, and during the third year the yield increase observed on the tissue culture plants w as around 50% greater. Although the data was not presented, the authors report that fruit size and quality were not affected by propagation method. Number of basal branches of the micropropagated plants was significantly higher than on cutting produced pl ants, for the same period; more than

PAGE 28

28 double the first year, 50% higher during the second, and almost double during the third year of field growth (Read et al., 1989). Flower buds per plant during the same period (1983 1985) were also significantly higher o n micropropagated plants. There were almost four times as many flower buds per plant in 1983, about twice in 1984 and almost double the number produced in cutting derived plants in 1985 (Read et al. 1989). The effect of the developmental stage of the sto ck plant (for cuttings and explants from tissue culture) rooting percentage, and growth of the propagated plant were also evaluated in the same study. Explants taken from plants in the vegetative state rooted better after micropropagation than did cuttings taken from the same plants. Additionally, when explants and cuttings were taken from plants in the reproductive state (with flowers or flower buds), percent rooting of tissue culture micro cuttings decreased (but not significantly), and leaf bud cuttings did not root at all (Read et al., 1989). Long term effect of propagation type on vegetative and reproductive growth of the half Shieklh et al. (1996), using the Becker r eviewed work (Grout et al., 1986, Read et al., 1988 and 1989). At Becker, yield was not significantly affected by the propagation method during any year. Berry weight of TC plants was significantly lower in 1987 and 1989. Berry weight of all years combined was not significantly affected by propagation method. The authors reported no significant effect on height and spread (average diameter within and between rows), which were measured in 1994. In 1987, plants from the original location at Grand Rapids had t o be dug and moved to a new site, where TC and SC plants were planted in a completely randomized design, with four replications and four plants/block. Management practices were similar to the Becker location. From

PAGE 29

29 1989 to 1994, there was no significant ef fect of propagation method on winter injury, bloom or crop rating, or plant height. Plants were also rated for vigor (based on number and shoot length), and TC plants were more vigorous in 1994 or when all years were analyzed together. Plant spread, measur ed in spring, was larger in TC plants for all years except in 1992 and 1994, and fall width of tissue cultured plants was larger from 1990 to 1994. Total yield of TC plants was significantly higher in 1989, 1994, and when all years were combined, without a ny effect of propagation method on average berry weight. Albert et al (2009) conducted a study with half Estonia where they also evaluated the effect of propagation method on growth under field conditions. Along with plants ob tained from softwood cuttings, two types of TC field was established in 2003, and shoo t numbers per plant, height and plant widths were recorded from 2004 to 2006. For the first three years, plants micropropagated from Conversely, plants from cuttings produced s ignificantly more shoots/plant during the first two years or when all years were analyzed together. Their results, however, differ from the work previously carried out in Minnesota, where micropropagated 'Northblue' plants produced more basal and lateral b ranches during the first season of field growth (Grout et al., 1986) There have been few reports studying the effect of propagation method on growth and development of highbush blueberry. Dweikat and Lyrene (1988) evaluated young

PAGE 30

30 hybrid blueberry plants ( V accinium corymbosum x V. elliottii ) from different propagation methods, but they did not evaluate reproductive growth as plants were only a few months old. The most recent study that evaluated vegetative and reproductive growth of TC highbush blueberry pla nts was conducted in Poland (Litwinczuk et al., 2005). The authors compared growth and fruit characteristics of highbush blueberries derived from rooted cuttings with plants propagated in vitro from axillary or adventitious shoots or from an 11 year old cu lture with a mixture of both. The field was established in May 2001. During the first three years, TC plants produced significantly more and longer shoots than SC plants and were more uniform. In the first reproductive season, SC plants had a higher percen tage of flowering (near 60 %) than TC plants (22 %) and had more inflorescences/plant. Ten berries/plant were collected the second year to evaluate fruit characteristics. Berry weight of SC plants was significantly higher than for TC plants, and among the TC treatments, the smallest berries were produced on plants obtained from the 11 year old culture. (Litwinczuk et al, 2005). Three rabbiteye cultivars propagated by TC and SC were recently evaluated in from semi hardwood cuttings and TC plants were planted in August, 2009. Plant height, diameter, and number of shoots from the base of the plant were recorded in July 2010. Fruit was harvested from 19 November 2010 to 2 0 January 2011 and fresh weight and number of fruits per plant were taken. There was no significant difference in plant yield among propagation methods. Micropropagated plants had higher initial vegetative growth, evidenced by taller plants with more and thicker shoots. There was no significant difference between the cultivars propagated by SC, but among

PAGE 31

31 micropropagated plants 'Woodard' had the highest number of shoots, followed by 'Bluegem' and 'Briteblue' Maine (US) only produces wild blueberries and the se lowbush fields often have incomplete coverage. Frett and Smagula (1983) developed a protocol for in vitro shoot production of lowbush, but it was not until 2000 that growth of micropropagated lowbush plants under field conditions was evaluated. Morrison et al. (2000) hypothesized that it would be possible that juvenile characteristics of the micropropagated plants would result in increased rhizome production, which could facilitate plant spread and coverage of bare areas of the field. The study found tha t after one year in the field, micropropagated plants showed more vigorous rhizome growth (higher number and length) than plants from SW but there was no significant difference in stem dry weight, flower bud number or berry weight. After a second year of f ield growth, however, micropropagated plants had 37% more flower buds, and more than three times as many rhizomes as plants from SW. Although rhizome production (total rhizome DW) was significantly higher for TC plants, there was no significant difference in stem dry weight or leaf area per plant. The same group also conducted an experiment to compare seedlings with TC and SW plants. After six months in the greenhouse (from August to January) TC and SC plants averaged the same number of stems and a similar number of branches/plant; both had significantly more than seedlings. Stem cutting derived plants had more flower buds than TC plants after six months in the greenhouse and seedlings did not produce any flower buds. After one season of field growth, seedli ngs had produced three times the number of rhizomes as TC and SW plants. Rhizome production measured in dry weight, however, was not significantly different among the

PAGE 32

32 treatments, and stem dry weight of TC plants was the lowest, without any significant diff erence between the other two propagation types (Morrison et al. 2000). In Canada, Jamieson and Nickerson (2003) conducted a sixteen year study to evaluate the effect of propagation type on fruit yield, plant spread and rhizome production of three clones o f lowbush blueberry. Two advanced selections from the breeding programs of the University of Maine and the Atlantic Food and Horticulture Research Center (ME3 and K74 13, respectively) and a natural selection (K206) were chosen. The field, established in 1 985, included plants from open pollinated seeds, TC and SW plants. After one year of field growth, seedlings had the greatest number of stems/plant and branches/stem, followed by TC plants and SW plants. There were no significant differences between genoty pes. Rhizome production/plant in the second year followed the same trend between propagation types, but K 206 produced more rhizomes than the other genotypes. After five years in the field, plant diameter (measured at ground level) was largest for seedling s, intermediate for micropropagated and smallest for cutting derived plants. The authors report that SW plants produced few stems, TC plants were intermediate, and seedlings had twice as many stems as SW by 1990. Although stems from SC plants were fewest, they were the longest and had the most flower buds, for all genotypes. There was no significant effect of propagation type on fruit yield for the first year (1987). Yield of K206 was the lowest for all years, and yields of TC plants of this genotype were t he lowest of all treatments. Cutting derived plants of ME3 achieved good yields by 1992, having higher yields than TC plants in 1994, 1997 and 2001. However, TC plants of the third genotype, K74 13 had higher yields than SC and

PAGE 33

33 seedling plants with the exc eption of 1994, when they yielded the lowest, and 1997 when their yield was intermediate, and seedlings had the best yield. When yield was expressed as kg fruit/area of the row, TC plants had the lowest yield and there was a significant interaction between propagation type and genotype. Average berry weight (50 berries) was usually lower for TC plants, with SC being the highest or intermediate, depending on the year or genotype. Two major issues stand out after reviewing previous literature on this subject : 1) the research comparing field performance of Vaccinium species resulting from micropropagation with traditional vegetative propagation is limited and showed contradictory results. In some studies, micropropagation resulted in more vigorous plants and/o r better reproductive growth or yield, while in other cases, micropropagated plants flowered later or resulted in smaller plants or there was no significant effect of propagation method. 2) All previous work that evaluated vegetative and reproductive field growth of micropropagated and cutting derived highbush blueberry ( Vaccinium corymbosum L.) plants was conducted in regions with climates different than Florida, such as cold winters and shorter growing seasons. As a consequence, the impact of propagatio n type on establishment and performance of southern highbush cultivars has not been studied. Anecdotal differences in vegetative growth between micropropagated and cutting derived plants have been reported by growers and researchers in Florida, but thes e have not been well documented. As micropropagation is becoming a much more commonly utilized method by nurseries and commercial blueberry propagators, the need for this study became apparent. Our objectives were to evaluate vegetative and

PAGE 34

34 reproductive g rowth of three southern highbush blueberry cultivars obtained from micropropagation and rooted cuttings grown in two locati ons in Florida. The hypothesis wa s that plants de rived from micropropagation would grow more vigorously and produce more shoots than plants from rooted cuttings, resulting in higher fruit yield during the first two years of field establishment, and tha t berry weight and fruit quality would not be affected by propagation method.

PAGE 35

35 CHAPTER 2 EFFECT OF PROPAGATION TYPE ON VEGETATIVE GROWTH OF THREE SOUTHERN BLUEBERRY CULTIVARS ( VACCINIUM CORYMBOSUM L.) OBTAINED FROM MICROPROPAGATION AND SOFTWOOD CUTTINGS Introduction Blueberries have traditionally been vegetatively propagated by softwood or hardwood cuttings. Although plants can be produced at low cost, cuttings from some cultivars have low or very low rooting percentages (Miller et al., 2006), and it may take up to s everal years to generate enough stock of newly released cultivars so that significant acreage can be planted. In contrast, micropropagation has the potential to produce large numbers of plants more quickly than propagation by rooted cuttings. A plant prop agator, starting from micro shoot culture, can produce up to 1000 plants large enough for field planting, in less than six months (Isutsa et al., 1994). Another drawback of propagating plants by rooted cuttings is the increased risk of diseases (Cline and Schilder 2006) such as Phytophthora root rot, stem blight ( Botryosphaeria spp.), Phomopsis canker ( Phomopsis vaccini ), bacterial leaf scorch ( Xylella fastidiosa) and viral diseases, which can spread through infected softwood cutting in propagation beds. ( Krewer and Cline, 2012). Although changes in the growth habit of micropropagated Vaccinium and other small fruit crops have been reported, information on field performance of TC plants is limited, and results are variable and contradictory (Debnath, 2007). Increased vigor and runner production of TC strawberries have been reported (Swartz et al. 1981). Micropropagated lingonberry ( Vaccinium vitis idaea L. cv. Sanna) plants produced more rhizomes and had significantly higher plant dry weight by the third yea r after planting.

PAGE 36

36 Growth of half evaluated with variable results. Grout et al. (1986) and Read et al. (1988, 1989) reported that for the first two years, TC plants had significantly more latera l branching than cutting derived plants, while Albert et al (2009) reported that cutting derived plants were taller and produced significantly more shoots than TC plants during the first two ly and more uniformly during the first three years in the field (Litwiczuk et al. 2005). Conversely, propagation method did not have a significant effect in the number of shoots or plant hould be noted that much of the previous research comparing vegetative growth has been done in climates with colder winters and shorter growing seasons than in Florida. In addition to overall growth measurements, several researchers have studied the effe ct of micropropagation on rooting ability of blueberries and other Vaccinium species. Meiners et al. (2007) found that 'Ozarkblue' (blueberry) and 'Red Pearl' (lingonberry) in vitro shoots planted directly in soils rooted significantly better than cuttings from mature field grown plants. Morrison et al (2000) reported that micropropagation resulted in increased rhizome number and rhizome dry weight in lowbush blueberry. Lyrene (1981) reported 100 percent rooting for in vitro cuttings and for tissue culture derived plants (3 months old), compared to 54% rooting for softwood cuttings obtained from mature rabbiteye plants. However, high levels of cytokinins used to promote shoot initiation during the tissue culture process has been shown to inhibit rooting abi lity and subsequent survival and root growth (Valero Aracama et al., 2010) To evaluate size of different root systems, root mass is the typical measure used. However, root mass by

PAGE 37

37 itself cannot accurately illustrate root functions involved in plant soil w ater relations. Therefore, total root length, surface area and branching patterns have been reported to affect nutrient uptake (Costa et al, 2000). The objective of this experiment was to evaluate the effect of propagation type on above and below ground vegetative growth parameters of three blueberry cultivars in two locations during the first two years of field establishment. The hypothesis wa s that plants derived from micropropagation will have more shoots and larger size than plants from rooted cutting s during the first two years after field planting. Materials and Methods Locations and Experimental Design The study was conducted at two locations with different average chill hour accumulation. The northern site, at the University of Florida Plant Scien ce Research 540 chill hours (hours below 7 C received by February 10 in 75% of winters) ines 210 hours/year. Three southern highbush cultivars that are commercially grown in both locations ement of treatments (3 cultivars x 2 propagation methods) with five replications was arranged in a completely randomized block design. Each replication used 20 plant plots. The field at Citra was planted on 15 and 16 April, 2010 on row beds of Arredondo sa nd (from the Entisol order), which were amended with pine bark according to standard industry practices. The native soil had characteristics of good drainage, low

PAGE 38

38 available water holding capacity, less than 5% slope, and a sand profile depth of 165 cm (US DA, SCA, 2012). Plots were planted in three rows and treatments within each block were randomly assigned across the three rows to minimize effects that might result from the slight slope of the field. Distance between rows was 3 m and plant spacing within the row was 0.75 m (4620.77 plants/ha). Beds 91 cm wide were covered with black woven polypropylene ground cover fabric (104 g/m 2 ) (Figure 2 1). The Haines City field was planted on 23 April 23, 2010 on Candler sand to which 250 to 300 yards of pine bark/a cre in 3 m wide bands were laid and tilled in. Weed control was achieved by covering the beds with woven plastic mat, with hand weeding as needed and herbicides along the edge. Cultural P ractices At the Citra planting, water was applied through a single dr ip tube which delivered approximately 374.16 liters/min/ha (40 gal/min/acre) in three cycles that totaled about 45 to 60 minutes/day. In Haines City, plants were irrigated with two rows of drip tube, delivering about 561.24 liters/min/ha (60 gal/min/acre). Irrigation was applied in five to eight daily cycles that totaled between 75 to 120 minutes/day. Overhead irrigation was used in both locations for freeze protection as necessary. Both locations utilized fertigation for nutrition management and acid in jection for maintenance of low soil pH. At Citra a complete liquid fertilizer (10 2 5 with micronutrients) was injected at a rate of 134 kg N / ha (120 lb/acre), from planting until the end of September 2010. The second year 100.8 kg N/ha (90 lb/acre N) w as applied from end of January until middle of September, 2011. Fertilizer was injected at a rate of 1,000 ppm for seven minutes during each 15 20 min irrigation event. In Haines City a complete 6 6 3 liquid fertilizer was injected once a day in spring and fall and three

PAGE 39

39 times/day during summer flushes for a total of 224 kg N/ha/year (200 lb N/acre). At the Citra planting, sulfuric acid (38 percent) was injected continuously during irrigation cycles at approximately 200 ppm. This rate was determined by titr ating the high pH well water (approximately 7.8) with 38 % sulfuric acid to a point where the bicarbonates were neutralized at approximately ph 5.7 to 5.8. This eliminated the buffering capacity of the well water. Both the fertilizer and the sulfuric acid were injected with Tima Mix rite proportional injection pumps. Fertilization, irrigation system and the pumps were controlled with a Sterling 12 irrigation controller. Pest and disease control followed standard practices at each location. In row weed cont rol was accomplished by using woven fabric row covers, and row middles were spot treated as needed for weed control from March through September. Hydrogen Cyanamide (Bud Pro, Dormex) was applied at the Haines City field at 1.25%, using 655 l/ha (70 gall ons per acre) on December 20, 2010. Dormancy breaking compounds were not used at the Citra field during the study. During the first growing season plants were not pruned at either location, since this would have masked differences in growth during the esta blishment year. However, following commercial standard practices, plants were pruned the second year of the study. In Citra, plants were pruned a week after the first harvest season was completed. Plants to be used for the destructive sampling were flagged and not pruned In Haines City plants were manually pruned after harvest (end of May), which consisted of removing the old flower spikes and any new flush was allowed to grow. Vegetative Field M easurements At both locations, ten plants in the center of ea ch plot were used as subsamples for vegetative field data. From July to November in 2010 and 2011 at Citra, and July to

PAGE 40

40 October 2010 and 2011 in Haines City, average plant height and width in both directions (spread within and between rows) were measured a t monthly intervals. These data were used to calculate plant volume using the formula of an elliptical cone: Plant Canopy Volume 2 .h)/6. (NASA, 2010). Where h is height, d is diameter (average width) Additional measurements were taken of the plant s selected for reproductive growth evaluation (Chapter 4). Two representative plants per plot were flagged, and the number of major canes and total shoot number per plant were recorded. Major canes were defined as stems arising from the first 12 cm above t he soil line with a minimum diameter of 7 8 mm. Shoots that were shorter than 5 cm and had a diameter of less than 2 mm were not counted. Data from Haines City were taken on 10 December 2010 and at Citra on 12 and 14 January 2011. For all measurements, dat a from the subsample plants were averaged and used for statistical analysis. Destructive Sampling Whole plants were harvested at biologically significant points during the study: at the beginning of the experiment, after the first season of vegetative gro wth, after the first harvest, and after the second growing season. With the exception of the initial destructive sampling, one plant per plot was used for dry weights. Initial destructive sampling (DS0). At planting, ten representative plants of each culti var and propagation type combination were measured for the same characteristics as described above. Subsequently, plants were divided into roots, shoots and leaves (Fig. 2 3 and Fig. 2 4). Plant root systems were washed free of soil over a table with wide mesh metal top using a hose and a nozzle with different pressure settings as

PAGE 41

41 shown in Figure 2 5. Loose root pieces and root ends were washed over 2.18 mm and 3.175 mm sieves to reduce root losses. All plant parts were placed on paper bags and dried separately at 60 C for at least two weeks, until constant weights were achieved. Destructive sampling after the first growing season (DS1). On 22 November (Haines City) and 3 December (Citra), one plant from each plot was removed by excavating an area approximately 30 cm deep and as wide as half the distance to the next p lant on either side Roots were immediately separated from the tops approximately at the soil line. Tops were placed in black 39 gallon trash bags to minimize loss of leaves. Roots were placed in plastic bags and placed in a cooler at 4 C until they were washed. In the laboratory, leaves were removed from the plants, put in paper bags, and placed in drying oven at 60 C within 48 hours of destructive sampling. Stems arising 12 cm fr om the soil line with a minimum diameter of 7 mm were recorded as major canes. The total number of shoots with a minimum of 2 mm diameter and 5 cm in length were counted and dried separately from the major canes. These shoots are referred to as lateral br anches, and the summation of the lateral shoots and the major canes comprises the total number of shoots per plant (Figure 2 4). At this point, plants had a distinguishable crown, which was also dried separately from other plant parts. After washing the r oots free of soil, they were dried as described above. Destructive sampling after first harvest (DS2 ). Plants were excavated, dissected, and dried as described for DS1. In Citra, plants were excavated on 8 June, 2011 and in Haines City on 14 June.

PAGE 42

42 Destruc tive sampling after second growing season (DS3). Plants in Citra were excavated on 9 November (Figure 2 6) and in Haines City on 28 November 2011. Plant samples were treated as previously described for DS1 and DS2. Root Measurements A portion of each roo t system from plants obtained in DS1 and DS2 were sampled by randomly selecting one or more of the root branches at their origins and including all associated subtending roots (Figure 2 7). The total sample size approximated 10 15% of the whole root system At DS3 the root subsample was taken by placing four thin metal bars over the whole root system to divide it into eight sections and randomly selecting one of these sections (Figure 2 8). Subsequently, samples were fully submerged in water in a white plas tic container and the smaller pine bark pieces were manually removed. Samples for scanning were placed wet in plastic bags to prevent root shrinkage and stored in a 4 C cooler until scans were done, using the program WinRHIZO Pro (Regent Instruments Inc. Quebec, Canada). Root architecture was not measured at DS3 from samples at Haines City. Statistical Analysis Statistical analysis was done with SAS software version 9.2 (SAS Institute Inc., Cary, NC) using PROC GLM. Analysis of Variance (ANOVA) was perf ormed first, with cultivar and propagation type as main effects. Treatment means were separated using level of significance. In cases where the overall effect of propagation was not significant but interactions between main effects were significant (cultivar by propagation, location by propagation, or location by cultivar by propagation) the slice statement was used. This was done to detect if the lack of propagation effect over all cultivars was due to one

PAGE 43

43 cultivar affected by propagation method in the opposite direction than the other ones, causing the overall means of propagation methods to be statistically not different from one another. This statement produces a separate ANO VA for each cultivar, without having to analyze the data for each cultivar separately, which would reduce the degrees of freedom of the analysis (Littell et al., 2002). PROC GLIMMIX was used to analyze number of branches and total shoot tips in the second destructive sampling because of shoot number and berry yield were determined using the PROC CORR procedure. Results Plant Volume during the F i rst Growing S eason. Plan t volume data from June, July, and November 2010 were transformed using the logarithmic function to approximate a normal distribution. Throughout the first season, plants in Citra were larger than in Haines City Although location effect was significant, there was no significant interaction involving location and propagation method; and therefore, data represent treatment means across both locations I n June, July and November 2010 TC plants had larger plant canopy volume than SW plants. Additionally, signif icant interactions of propagation type and cultivar were observed throughout the s eason (Table A 1 larger than for cutting derived plants from June through November, while the opposite larger volume in June, July and November, with no significant differences between propagation types during the rest of the first growing season (Table 2 1).

PAGE 44

44 Plant Volume during the Second Growi ng S eason. Main effect of propagation method and significant propagation by cultivar interactions were observed from June until October 2011 (Table A 2). Propagation method did not have a significant effect on plant volume in November ( p =0.064). S ignifica nt location by cultivar by propagation interaction resulted in larger plant volume of Emerald' and Jewel' TC plants and reduced plant volume of 'Primadonna' TC plants, compared to SW plants in Haines City from June to August 2011 (Table 2 2). However, in Citra TC 'Jewel' had larger plant volume only in June, without any difference in plant volume between propagation types in July and August (Table 2.2). In Citra, propagation oint during the season From September to November interactions involving location and propagation method were not significant, while in September and October a significant cultivar by propagation interaction was observed. In September and October, 'Emeral d' and 'Jewel' plant volume was greater for TC than SW plants, averaged across locations. Conversely, there was no significant effect of propagation type on plant volume of 'Primadonna' plants for the same months (Table 2 3). Number of Major Canes, Lateral Branches and Total S hoots At planting, TC plants of all cultivars had more major canes than plants from rooted cuttings (Table 2 more lateral branches and total shoots, without any significa nt difference between 4 and Figure 2 9). At the time of the first flower bud count (Dec. 2010 and January 2011) and at DS3 (November 2011), TC plants of all cultivars had more canes (Table 2 4) while only 9). In June 2011

PAGE 45

45 total shoot number among prop agation types. In December 2011 TC plants of all cultivars had significantly more canes than plants obtained from rooted cuttings ( Table 2 4 ) and 'Jewel' TC plants had more total shoots than SW plants (Figure 2 9) Propagation method did not have a significant effect on number of lateral branches or tota ates in either year Initial Destructive Sampling (DS0) Data for plant volume, number of canes per plant and root, shoot, leaf and total plant dry weights did not follow a normal distribution and were tr ansformed using the logarithmic function. At planting (DS0), propagation type had a significant effect on height, width, plant volume, number of canes, number of lateral branches and root, shoot, leaf and plant dry weights (Table 2.5). Interactions betwee n cultivar and propagation type were significant for all the abo ve mentioned variables. Height and root and shoot dry weights of TC plants were higher than for cutting derived plants for each of the cultivars studied. Initially, average width and plant vo plants propagated by cuttings while the opposite 5). Propagation effect on leaf dry weight varied among cultivars: was greater in 'Jewel' TC and reduced in 'Primadonna' TC plant compared to SW plants, with no significant difference between propagation types for 'Emerald'. Field Destructive Sampling One (DS1) At the end of the first season of field growth (DS1, Nov. 2010), there were no significant effects of propagation type on plant height plant volume or root dry weight for any of the cultivars at either location. Conversely, plant canopy width, number of

PAGE 46

46 canes per plant, and leaf, cane and plant dry weights along with root:shoot ratios were significantly affected by propagation type. Significant location by propagation interactions were observed for leaf and cane dry weights, with TC resulting in greater leaf and cane dry weights in Citra while propagation method did not significantly affect the same varia bles in Haines City (Table 2 6). Significant location by cultivar by propagation interactions were observed for root and crown dry weight s and root:shoot ratio, where TC plants of all cultivars had greater crown dry weight in Citra with no difference betwe en propagation types in Haines City (Table 2 7). In Citra root dry weight was greater for 'Primadonna' TC compared to SW plants, while the opposite happened in Haines City. There were no significant differences in root dry weight between propagation types for 'Emerald' and 'Jewel' at either location. Also, 'Emerald' and 'Jewel' TC had reduced root:shoot ratio with no difference between propagation methods for 'Primadonna' in Citra or for any of the cultivars in Haines City. Treatment means averaged across locations were compared for number of canes, lateral branches, total shoot number, volume and shoot and plant dry weights because there were no location by propagation interactions for these variables (Table 2 8). There was a significant interaction betwe en propagation type and cultivar for plant canopy width, number of lateral branches total shoots and shoot and plant dry weights. significant increases in number of canes, lat eral branches, total shoot number, plant compared to SW plants. However, propagation type had no significant effect on number of canes or shoots, volume or dry weights of 8).

PAGE 47

47 Destructive Sampling after First H arvest (DS2) Data collected in June 2011 for leaf, shoot, and cane dry weights were transformed using the logarithmic function to approximate a normal distribution. Analysis of variance show ed a significant effect of propagation method on height, width, volume, number of canes, lateral branches and total shoot tips, and dry weights of leaves, canes, shoots and whole plants, but the main effect of propagation type on crown and root dry weights ( p =0.0507) was not significant. The interactions between propagation method and cultivar were significant for all variables shown except root:shoot ratio. Treatment means across locations for those variables that did not show a location by propagation int eraction are shown in Table 2 9. 'Emerald' and 'Jewel' TC plants had greater root dry weight than SW plants, while the opposite was obse rved on 'Primadonna' Micropropagation significantly increased plant height and volume number of lateral branches and cane shoot derived plants without a significant effect on size or dry Crown dry weight of 'Emerald' plants was no t affected by propagation method, while TC 'Jewel' plants had greater drown dry weight and 'Primadonna' TC plants had reduced crown dry weight than SW plants (Table 2 9). After the first harvest season, significant location by propagation interactions were observed for width, leaf dry weight, and root:shoot ratio. Averaged across cultivars, TC plants had a larger mean diameter in Haines City, with no difference between propagation types in Citra (Table 2 10). The opposite was observed for root:shoot ratio, where TC plants had smaller root:shoot ratios in Citra with no significant difference in Haines City. S ignificant locat ion by propagation interaction resulted in TC plants having

PAGE 48

48 greater leaf dry weight in both locations, with treatment differences being more marked at Citra than at Haines City (Table 2 10). Micropropagation resulted in significantly greater number of canes (Table 2 4), lateral branches (Table 2 8) and total shoots of 9). Location effect varied among measurements when averaged across cultivar and propagation type: plants were taller, wider and of larger volume in Citra, but leaf, above ground plant and total plant dry weights were higher in Haines City. The number of major canes, lateral branches, or total shoots showed no differences in treatment means (all cultiv ar by propagation method combinations) between locations or significant interaction with propagation method. Destructive Sampling after Second Growing S eason (DS3) Leaf and root dry weight and root:shoot ratio data were transformed using the logarithmic fu nction to approximate a normal distribution, and analysis of variance of the transformed data showed no significant main effect of propagation method No significant differences were found for plant height or for crown dry weight of plants derived from eit her propagation method. However, significant effects of propagation type were observed for plant width and volume, number of canes, lateral branches, and shoot number, and shoot, cane, and plant dry weights and root:shoot ratio. TC plants had larger mean d iameter (SW= 92.22 cm vs TC= 99.12 cm) and greater plant volume (SW= 569.5 dm 3 vs TC= 664.2 dm 3 ). Location by propagation interactions were not significant for total number of canes, lateral branches or total shoots/plant, or cane total shoot, root and plant dry weights. Therefore, data were averaged across locations. Micropropagation resulted in more canes for all three cultivars and increased cane, total shoot, and plant dry weights

PAGE 49

49 11 a nd 2 12). 'Emerald' and 'Jewel' root dry weight was not affected by propagation type but TC 'Primadonna' plants had reduced root dry weight than SW plants. Micropropagation also resulted in effect on the other two cultivars (Figures 2 9). A significant location by cultivar by propagation type interaction was observed for leaf dry weight: while there was no effect of propagation type on any cultivar in Citra, af dry weight than SW plants, 'Jewel' TC plants had reduced leaf dry weight compared to SW plants and there was no effect of propagation type on 'Primadonna' in Haines City (Table 2 13). Root M easurements For root architecture samples taken at DS1, sample size at Citra ranged from 7 to 26%, and averaged 10%, of the whole root system, expressed as a percentage of total root dry weight. At Haines City, the average sample size was 16% of the root system, and ranged between 7 to 28%. Scanned roots were divided into ten classes based on diameter, with each class boundary equal to 0.5 mm. Average root diameter (mm) was also obtained from the scanning. For statistical analysis roots with diameters equal to or bigger than 3 mm were grouped together. Total root lengt h, total surface area and length of roots 0 0.5 mm (L1) and 0.5 to 1 mm (L2) were transformed using the logarithmic function to approximate a normal distribution. Analysis of variance using PROC GLIMMIX showed a significant location effect for total root l ength (Citra= 970.7 m vs Haines City=650 m), total surface area (Citra=14656 cm 2 vs Haines City= 1276.6 cm 2 ) and length of roots smaller than 2.5 mm in diameter. There was no effect of location on length of roots with diameter equal to or

PAGE 50

50 larger than 2.5 m m. Total root length and average root diameter and length of the different categories were not significantly affected by propagation type and there were no significant interactions that involved propagation type. Significant propagation effect resulted in increased root surface area in the TC plants compared to SW plants (TC= 8571.67 cm 2 vs SW= 7361.01 cm 2 P =0.0430),. Significant cultivar by propagation interactions for total root length of roots 2.5 to 3 mm and larger than 3 mm in diameter resulted in re plants compared to SW, while there was no significant effect of propagation types on the other two cultivars.(Table 2 14). At DS2, root samples were scanned from both Citra and Haines Cit y. However, due to excessive root breakage while excavating the plants in Haines City, only data from Citra is presented. Root samples scanned ranged from 6 to 21% and averaged 13% of whole root system, expressed as percentage of dry weight. Overall, TC significantly increased total root length, total root surface area and root length of all the diameter based categories analyzed (Table 2 15). Cultivar by propagation interactions were not significant for any of the variables evaluated at this point. At DS3, roots were only sampled at Citra. Root samples scanned ranged from 4 to 10% and averaged 7.1% of whole root system, expressed as a percentage of total root dry weight. Data for length of roots 0 0.5 mm in diameter were transformed to approximate a n ormal distribution using the logarithmic function. Propagation did not have a significant effect on total root length, total surface area, or on the root lengths of the diameter based categories. None of the variables analyzed showed a significant propagat ion by cultivar interaction.

PAGE 51

51 Discussion Plant Volume, Size, Weight and Branching D ata Differences in plant volume due to propagation method were larger at the beginning of the study but tended to decrease towards the end of each growing season. By the end of the second growing season (November 2011), plant volume was not significantly different among propagation types for any of the cultivars in the study (Table 2 2). However differences in plant volume observed in October 2011 among propagation types were primarily due to a larger average diameter of TC plants, since plant height was not significantly different among propagation types at this point or at the end of the first growing season. These results agree with those of Albert et al. (2009), who report ed that plants derived from TC after several cultures had larger diameter. Our results for plant size disagree with those of Souza et al. (2011) who reported increased plant height of micropropagated rabbiteye one year after field planting. In addition to soil type and climate, which varied between locations, differences in management practices may have contributed to the differences in plant growth observed between the two locations. Examples of those management practices include irrigation and nutrient ma nagement, harvest practices, and pruning time and strategy. Throughout the study, higher annual N and irrigation rates were applied in Haines City. Management practices in Haines City should have resulted in more vegetative growth. However, the converse w as observed for most of the study. In June and July of the second year, plants at Citra had reduced plant volume compared to plants in Haines City (averaged across all levels of the other factors), and there was a significant location by cultivar by propag ation interaction. One possible explanation for these differences is

PAGE 52

52 that only two plants per plot were harvested (for data) in Citra, while in Haines City, because the field was part of a commercial operation, all plants were harvested regularly and there fore did not suffer the same stress due to the fruit load. Moreover, in Citra fruit yields were much higher than in Haines City (Chapter 4) adding further to the stress brought on by young plants carrying a heavy crop load, which often results in reduced v egetative growth. Several plants died, presumably from stem blight ( Botryosphaeria spp. ) which was favored by the high stress levels from carrying a heavy crop load. This is also supported by the data from the destructive sampling after harvest, where leaf shoot, and whole plant dry weights, were greater in Haines City, even though plants at that location had been pruned prior to the experiment. Conversely, leaf, total shoot and plant dry weights were larger in the northern location at the previous samplin g date (end of first growing season, 2010). Pruning differences between locations may have also contributed to the differences in plant volume between the two locations. Plants in Haines City were pruned earlier and less aggressively than at the Citra loc ation. In Citra, the plants were pruned two weeks before starting measurements in June 2011, as opposed to four weeks prior to measurements in Haines City. Moreover, differences in pruning style may have also play a role in the location effect on canopy vo lume. In Citra topping and pruning was done manually, and canes that did not grow upright and were near the crown were removed. At Haines City, pruning consisted mostly of topping the plants with much lighter cane removal. This can also explain the interac tion between treatments and location: where plants had recently been pruned, there were no differences between propagation types, but in Haines City, where plants had time to

PAGE 53

53 grow after summer pruning, and before measurements were taken, we observed larger (Table 2 2). Differences in location effect and location interaction between dates were also observed in data from the three destructive samplings done during the study. At the end of the first growing season, there was a significant location by propagation interaction from rooted cuttings, while in the Haines City the differences d ue to propagation types were numerically smaller and not significant. The same interaction was observed from the samples collected at the end of the second season. All treatments (cultivar x propagation combination) had less leaves in Haines City at the en d of both seasons. This may be attributed to unusually warm and wet weather prior to the last destructive sampling which was favorable for the development of leaf spot diseases that caused early defoliation in Haines City (Table 2 6 and Table 2 13). Althou gh location by propagation interaction was also significant after the first harvest, overall, TC resulted in higher leaf dry weight at both locations, with the difference being larger in Citra. Similar results were observed in lingonberry (Debnath, 2005), where TC plants had more leaves than plants from cuttings, when averaged over three cultivars and three growing seasons in the greenhouse. Micropropagation resulted in greater plant dry weight at the end of the second year in the field, which agrees with r esults obtained with lingonberry by Gustavsson (2000) who reported higher accumulated plant growth of TC plants after three years in the field.

PAGE 54

54 canes and total shoots than plan ts from cuttings, but there was no significant effect on blackberry done by Swartz et al. (1983), who also reported that two cultivars had more laterals when plants we re obtained by TC, while there was no effect of propagation method on laterals per plant in others. By the end of the second season, TC plants averaged across all cultivars had more canes and total shoots than plants propagated from cuttings. These resul ts agree with Litwinczuk et al. (2005) and Grout et al (1986) who also reported that TC blueberry plants had more basal branches (in our study reported as canes) or produced more lateral branches (shoots) than plants from rooted cuttings during their first three years of field growth. The present study also agrees with findings in lowbush blueberry, where Jamieson and Nickerson (2003) observed that TC plants of three cultivars had more stems per plant and branches per stem than cutting derived plants after one year of field growth. Root M easurements Even when the overall effect of propagation method on total root length, total surface area and total root length of all the diameter based categories was significant, multiple comparison analysis showed no sig explanations to why differences between treatment means were not statistically significant. It may be possible that the size of the sca nned sample was too small. Araujo et al (2004) reported that for bean roots, scanning near 15% of the whole root system

PAGE 55

55 (excluding taproot and nodules) expressed as percentage of dry weight, provided good estimates of root characteristics. In our study roo t sample size (excluding crown) for the first sampling date averaged 10 percent (in Citra) and 16 percent (for Haines City) of the root dry weight but the second and third sampling dates averaged 13% and 7%, respectively. It is possible that a larger sampl e size was needed to detect differences between propagation methods. Although care was taken to minimize root loss while plants were excavated and roots were washed, some fine root loss was observed. It may be possible plants derived from TC, because they were more root bound, suffered greater root loss during washing. This hypothesis is supported by the fact that at the end of the first growing season it was still possible to observe the original root ball of the tissue culture plants that comprised the ro ot system at planting. Another factor to consider concerns the diameters of the root categories analyzed by the software and how well they relate to the root characteristics of the genus Vaccinium Traditionally, roots are classified as absorptive or cond uctive based on their diameters. The short lived, absorptive, roots would be less than 1 to 2 mm in diameter (Jackson et al, 1997). In Vaccinium absorptive roots are extremely fine, and can be less than 50 100 m in diameter (Valenzuela Estrada, 2008). Th e smallest diameter root classification identified with the software used in this study was up to 500 m. It is possible that differences in the proportion of small diameter absorptive roots among the propagation types could not be detected with the method ology and software used in this study.

PAGE 56

56 Conclusion s Micropropagated 'Jewel' plants had larger plant volume than plants derived from softwood cuttings, while the opposite was observed for 'Primadonna', the first growing season. In June, July and November of the second growing season, plant volume was not affected by propagation method in Citra. However, in Haines City, 'Emerald' and 'Jewel' TC plants had larger plant volume from June to August 2011. Conversely, larger plant volume was recorded in Haines City for SW compared to TC 'Primadonna' plants during that time. By the end of the second growing season there were no differences in plant volume between propagation types at any of the locations. Micropropagation resulted in greater number of canes, lateral b ranches and total shoots per plant for 'Emerald' and 'Jewel' after the first and second growing seasons, with no significant effect on total shoot number for 'Primadonna'. Micropropagated 'Emerald' and 'Jewel' had significantly greater plant dry weight aft er one season in the field but after the second season, plant dry weight was greater only for TC 'Jewel' plants, when compared to plants from softwood cuttings. However, TC 'Primadonna' plants had more canes per plant than softwood cutting derived plants, after the first and second growing seasons. Although there were critical differences in management practices between the two locations used in this study, the field performance of the three cultivars used were similar in Citra and in Haines City. Future tr ials should be conducted in several locations to study the effect of propagation method on vegetative growth of southern highbush blueberry cultivars under different environmental conditions.

PAGE 57

57 Figure 2 1. Picture of the field in Citra at planting. April 2010 (Photo by Silvia Marino ). Figure 2 2. Picture of the field in Haines City at planting. April, 2010 (Photo by Silvia Marino ).

PAGE 58

58 Figure 2 3 L eaves were removed from canes and shoots prior to collecting dry weights ( Photo by Silvia Marino ). Figure 2 4. After leaves were removed, the number of major canes and shoot tips were recorded, and shoots from each major cane were cut and bagged separately ( Photo by Silvia Marino ).

PAGE 59

59 Figure 2 5 Roots were washed using a 2 mm. sieve ( Photo by Silvia Marino ) Figure 2 6 Measuring the width of the root system prior to harvesting plants in Citra (November, 2011) ( Photo by Silvia Marino ).

PAGE 60

60 Figure 2 7 Arrows showing example of root branches. Representative root branches from roots systems harvested after first growing season were sampled for root scanning ( Photo by Silvia Marino ). Figure 2 8 Picture illustrates how roots were divided in eight h s to obtain the subsamples used for scanning after the second growing season ( Photo by Silvia Marino ).

PAGE 61

61 Table 2 1. Effect of cultivar and propagation method on p lant volume during the first growing season ( June to November 2010 ) Means of both locations. Volume (dm 3 ) Treat J une J uly A ugust S eptember O ctober N ovember ESW 7.42 13.87 ** 23.66 NS 38.69 NS 37.98 NS 39.93 *** ETC 10.50 25.14 35.57 62.75 58.34 63.70 JSW 2.17 *** 7.06 *** 12.33 ** 36.19 *** 36.63 *** 41.70 *** JTC 16.14 29.18 40.08 94.69 92.33 107.54 PSW 31.87 ** 74.11 *** 99.47 *** 171.95 ** 163.59 168.29 NS PTC 20.45 45.89 63.13 127.05 128.51 133.85 NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types within a cultivar, according to P values from ANOVA sliced by cultivar Table 2 2 Effect of cultivar and propagation method on p lant volume at two locations in 2011 NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types within a cultivar, according to P values from ANOVA sliced by cultivar. z Means followed by the same letters within a column indicate no differences Tukey's HSD test, 0.05. Table 2 3 Effect of cultivar and propagation method on plant volume in September and October 2011. Means of both locations. Volume (dm 3 ) Treatment September October ESW 331.16 388.00 ETC 446.23 500.89 JSW 445.84 516.44 JTC 547.29 627.97 PSW 537.34 NS 631.19 NS PTC 497.16 572.57 NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types within a cultivar, according to P values from ANOVA sliced by cultivar. Treatment means averaged across both locations. Volume (dm 3 ) Location Treat June July August November Citra ESW 108.14 NS 160.68 NS 264.31 NS 490.32 ab z ETC 116.09 183.60 311.24 543.74 ab JSW 117.74 194.73 NS 318.17 NS 580.97 ab JTC 170.57 239.37 400.41 689.11 ab PSW 160.18 NS 264.26 NS 415.75 NS 780.34 a PTC 177.59 269.18 460.40 803.83 a Haines ESW 161.51 *** 192.15 *** 259.62 ** 388.29 b City ETC 259.29 320.17 395.90 502.10 ab JSW 161.07 *** 219.22 ** 333.63 483.10 ab JTC 263.95 325.87 440.48 647.11 ab PSW 221.74 306.46 386.15 700.92 ab PTC 175.12 239.35 288.76 696.29 ab

PAGE 62

62 Table 2 4 Effect of cultivar and propagation method on the n umber of canes/plant from planting until end of second growing season Treat ment April 2010 Dec ember 2010 Jan uary 2011 June 2011 November 2011 December 2011 ESW 1 .0 c z 5 .0 c 5.5 c 7.3 b 8.1 c ETC 2.3 a 8.4 ab 10.7 a 12.1 a 13.7 a JSW 1 .0 c 5.7 c 5.3 c 7.6 b 8.6 c JTC 2.7 a 9.7 a 10.6 a 13.5 a 14 a PSW 1.5 b 7.5 b 7.7 bc 8.4 b 8.5 c PTC 2.2 a 9.5 a 9.3 ab 11.2 a 11.3 b z Means followed by the same letters within a column indicate no differences Tukey's HSD test, Figure 2 9 Effect of cultivar and propagation method on total shoot number per plant. Means of both locations. z Within each date, same letters indicate no differences Tukey's HSD test, NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types within cultivar according to P values from sliced ANOVA

PAGE 63

63 Table 2 5 Effect of cultivar and propagation method on p lant size and dry weights at planting, April 2010. Treat Height (cm 3 ) Width (cm) Volume (dm 3 ) Root DW (g/plant) Shoot DW (g/plant) Leaf DW (g/plant) Plant DW (g/plant) ESW 16.6 d z 15.0 c 6.14 c 1.85 c 1.88 c 2.08 b 5.81 c ETC 34.2 b 19.9 b 10.38 b 3.17 b 4.89 ab 2.88 b 10.94 b JSW 17.2 d 10.4 d 2.78 d 1.06 d 1.1 0 d 0.76 c 2.92 d JTC 43.6 a 22.5 b 12.98 b 7.03 a 5.93 a 2.57 b 15.53 a PSW 23.7 c 28.3 a 20.33 a 2.96 b 3.99 b 5.14 a 12.09 b PTC 37.6 b 21.4 b 11.61 b 4.89 a 6.01 a 2.06 b 12.96 ab z Means followed by the same letters within a column indicate no differences Tukey's HSD test, Table 2 6 Effect of location and propagation method on l eaf and cane dry weight s in November/December 2010. Location Propagation Leaf DW (g) Canes DW (g) Citra SW 110.68 b z 132.23 b TC 145.39 a 184.33 a Haines SW 62.01 c 69.38 c City TC 65.84 c 79.05 c z Means followed by the same letters within a column indicate no differences Tukey's HSD test, Treatment means averaged across cultivars Table 2 7 Effect of cultivar and propagation method on crown dry weight and root:shoot ratio at two location s in November /December 2010. Location Treat Root DW (g) Crown DW (g) Root:shoot ratio Citra ESW 19.32 NS 114.54 0.59 ** ETC 19.94 92.58 0.29 JSW 22.86 NS 98.72 0.54 ** JTC 15.52 78.34 0.25 PSW 31.32 90.00 *** 0.31 NS PTC 28.7 135.22 0.45 Haines ESW 90.08 NS 15.16 NS 0.65 NS City ETC 94.26 15.72 0.55 JSW 74.92 NS 10.36 NS 0.65 NS JTC 82 14.46 0.51 PSW 107.1 ** 14.72 NS 0.62 NS PTC 61.44 14.56 0.51 NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types within a cultivar according to P values from sliced ANOVA.

PAGE 64

64 Table 2 8 Effect of cultivar and propagation method on number of canes, la teral branches and total shoots per plant, width, volume and shoot and plant dry weig hts in November /Decemer 2010. Means of both locations. Treat Major Canes z Lateral Branches Total shoots Plant width (cm) Shoot DW (g) Plant DW (g) ESW 5.1 c 31.40* 36.50* 38.5 35.73 308.39* ETC 8.5 a 45.30 53.80 49.3 57.80 419.41 JSW 6.2 bc 46.00* 52.20* 37.2 ** 43.95 274.26** JTC 8.8 a 59.20 68.00 49.5 65.36 407.90 PSW 6.6 abc 43.70 NS 50.30 NS 60.5 NS 66.85 NS 446.54 NS PTC 7.8 ab 32.30 40.10 52.45 52.28 366.47 z Means followed by the same letters within a column indicate no significant differences at NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types within a cultivar according to P values from sliced ANOVA. Table 2 9 Effect of cultivar and propagation method on h eight, volume and dry weights in June 2011 Means of both locations. Treat Height (cm) Volume (dm 3 ) Lateral branches Canes DW (g) Shoot DW (g) Root DW (g) Crown DW (g) Plant DW (g) ESW 69.0 d z 120.45 b 90.0 ** 100.61 b 72.93 bc 104.02 a 45.68 NS 486.59 bc ETC 77.4 bc 224.76 a 143.9 175.88 a 126.74 a 138.44 bc 51.88 784.85 a JSW 72.2 cd 83.55 b 75.5 *** 66.03 c 55.92 c 73.48 c 28.33 *** 398.97 c JTC 85.8 a 252.78 a 190.8 201.12 a 140.35 a 123.97 ab 55.15 892.92 a PSW 88.1 a 235.65 a 88.7 NS 141.6 ab 80.80 b 138.82 a 67.54 ** 579.22 b PTC 82.2 ab 245.64 a 96.0 138.6 ab 77.46 bc 92.31 bc 48.06 513.62 bc z Means followed by the same letters within a column indicate no significant differences Tukey's HSD NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types within a cultivar, according to P values from sliced ANOVA. Table 2 10 Effect of location and propagation method on plant width, leaf dry weight and root:shoot ratio in J une 2011. Treat Width (cm) Leaf DW (g) R/S ratio Citra SW 79.7 a z 59.52 c 0.72 a Citra TC 85.6 a 129.71 b 0.45 b H aines City SW 55.6 c 108.26 b 0.33 c Haines City TC 69.9 b 159.86 a 0.25 c z Means followed by the same letters within a column indicate no significant differences Tukey's HSD Means averaged across cultivars. Table 2 11 P values from ANOVA sliced by cultivar, for cane and total shoot number, cane, total shoot, top, root, and plant dry weights in November 2011 Cultivar Cane number Total shoot number Canes DW Total Shoots DW Top plant DW Log Root DW Plant DW E <.0001 0.0899 0.0006 0.0039 0.0037 0.4597 0.0226 J <.0001 0.0001 <.0001 <.0001 <.0001 0.1889 0.0009 P 0.0021 0.4478 0.4195 0.6381 0.5662 0.0181 0.7922

PAGE 65

65 Table 2 12 Effect of cultivar and propagation method on number of canes and total shoots and cane, total shoot, top, root, and plant dry weights in November 2011. Means of both locations. Treat Cane number Tot al s hoot number Canes DW (g) Total Shoots DW (g) Top plant DW (g) Root DW (g) Plant DW (g) ESW 7.3 *** 144.5 NS 430.67 *** 701.41 ** 889.02 ** 375.9 0 NS 1441.4 0 ETC 12.1 177.6 614.12 930.14 1162.2 0 425.3 0 1758 .00 JSW 7.6 *** 190.9 *** 380.92 *** 681.82 *** 862.87 *** 338.76 NS 1337.45 *** JTC 13.5 283 .0 652.21 1037.97 1262.7 402.06 1816.83 PSW 8.4 ** 148.9 NS 423.53 NS 669.71 NS 888.89 NS 414.45 1487.7 0 NS PTC 11.2 154.1 463.77 705.13 940.22 330.97 1452.29 NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types within a cultivar according to P values from sliced ANOVA. Table 2 13 Effect of cultivar, propagation method and location on leaf dry weight in November 2011. Location Treatment Leaf DW (g) Citra ESW 295.04 NS ETC 329.14 JSW 288.68 NS1 JTC 400.02 PSW 330.3 NS PTC 347.3 Haines ESW 80.18 ** City ETC 134.98 JSW 73.42 JTC 49.44 PSW 108.06 NS PTC 122.88 NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types within a cultivar, according to P values from sliced ANOVA. NS1: P =0.0651 Table 2 14 Effect of cultivar and propagation method on total length of roots 2.5 3 mm in diameter (L6) and roots with diameter s greater than 3 mm (L7) at DS1. Means of both locations. Treat Total length 6 (m) z Total Length 7 (m) ESW 6.08 NS 12.94 NS ETC 6.42 13.63 JSW 4.48 NS1 10.93 NS1 JTC 626.37 14.12 PSW 7.19 16.26 PTC 5.26 10.99 NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types within a cultivar according to P values from sliced ANOVA. NS1 Significant at P < 0.10. z Roots were scanned and images obtained were an alyzed with WinRHIZO Pro (Regent Instruments Inc.,Quebec, Quebec, Canada )

PAGE 66

66 Table 2 15. Effect of propagation method on total root length (m), total surface area (m 2 ), length of roots 0 0.5 mm (L1), 0.5 1 (L2), 1 1.5 (L3), 1.5 2 (L4), 2 2.5 (L5), 2.5 3 (L6) and > 3 mm in diameter (L7) at Citra, DS2. Means of both locations. Prop agation Total Length (m) Total Surface Area (m 2 ) Total L1 (m) Total L2 (m) Total L3 (m) Total L4 (m) Total L5 (m) Total L6 (m) Total L7 (m) SW 1155.4 z 1.762 881.23 145.50 57.41 26.95 13.87 0.81 1.68 TC 1534.3 2.369 1169.85 194.27 75.27 34.41 18.48 1.06 2.38 ** z Roots were scanned and images obtained were analyzed with WinRHIZO Pro (Regent Instruments Inc.,Quebec, Quebec, Canada ) NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 Treatment means averaged across both locations

PAGE 67

67 CHAPTER 3 PLANT TISSUE CARBOHYDRATE CONTENT OF SOUTHERN HIGHBUSH CULTIVARS ( VACCINIUM CORYMBOSUM L.) OBTAINED FROM MICROPROPAGATION AND ROOTED CUTTINGS Introduction Seasonal variation in nonstructural carbohydrates have been well described in several fruit crops, such as apple ( Malus domestica ), grape ( Vitis vinifera ), pear ( Pyrus spp ), pecan ( Carya illinoinensi s ), sour cherry ( Prunus cerasus ) and walnut ( Juglans regia) (Loescher et al., 1990). Storage carbohydrates are particularly important in woody plants, which depend on these reserves for winter survival (Loescher et al., 1990). In several decid uous fruit crops, where flowering takes place before vegetative growth resumes after dormancy, the early stages of reproductive growth are highly dependent on carbohydrate reserves (Loescher et al., 1990). In rabbiteye ( Vaccinium. virgatum ), and highbush b lueberry ( V. corymbosum ), reproductive growth may precede vegetative bud break or occur at the same time, primarily dependent on the genotype (Darnell and Birkhold, 1996). Photoperiod and temperature also affect timing and intensity of reproductive growth of some highbush and rabbiteye cultivars (Darnell and Davies, 1990; Spann et al., 2003; Spann et al., 2004). Darnell (1991) reported that more flower buds were initiated in 'Beckyblue' rabbiteye blueberry when plants were exposed to short days during the f all, while photoperiod did not affect reproductive growth of 'Climax'. In the same study, carbohydrate translocation to stem tissue was increased under short days for 'Beckyblue', suggesting stem carbohydrates may play a role in the enhanced flower bud num ber observed for this cultivar (Darnell, 1991). Spann et al. (2004) evaluated the effect of photoperiod and temperature on reproductive growth and carbohydrate storage of the southern highbush (SHB)

PAGE 68

68 genotype 'Misty'. Under short days a correlation between reduced carbohydrate concentration and decreased flower bud initiation was observed, suggesting that low carbohydrate levels may be hindering reproductive growth under certain conditions (Spann et al., 2004). There is no research available that compares ca rbohydrate production of TC with SW in any Vaccinium species. However, it is possible that differences in vegetative or reproductive growth could be related to differences in carbohydrate availability at one or more phenological stages. Our objective was t o test whether differences in vegetative and reproductive growth would parallel differences in carbohydrate content between propagation types. Materials and Methods The locations, cultivars, field and experimental design, and management practices were as p reviously described in Chapter 2. At planting and after the first harvest (April 2010 and June 2011, respectively) five plants per treatment were collected from the plot at the Plant Science Research and Education Unit at Citra, FL (Citra) and processed fo r sugar and starch assays to determine total non structural carbohydrate levels. Preparation of dried plant material for carbohydrate analysis was as described for destructive sampling in Chapter 2. Briefly, whole plants were separated into roots, shoots and leaves at the first sample date; by the second sampling date, crown tissue could be separated from roots, shoots, and leaves. Before dry weights of the plants harvested in June 2011 were obtained, one representative major cane per plant was selected for carbohydrate analysis. All plant parts were placed in paper bags and dried separately at 60 C for at least two weeks, until constant weights were achieved. Dried plant parts were first ground in a Thomas Wiley Mill (Model 4, Arthur H. Thomas

PAGE 69

69 Company, Philadelphia, PA) to pass through a 2 mm mesh screen. A random sample of the tissue obtained was ground again to pass through a 20 mesh (1.27 mm) screen on a smaller Wiley Mill. Soluble sugars were extracted by shaking 25 mg of ground tissue in one mL of 80% ethanol (v/v) for 20 minutes. Samples were centrifuged for 10 minutes at 4200 rpm, the supernatant was recovered, and the ethanol extraction was repeated. The pellet that resulted from the ethanol extraction was stored at 20C for starch analysis. Sup ernatant from both extractions was combined and total volume was determined. Clarification of the sample was necessary to remove color: tissue extracts were treated with 25 mg activated charcoal and spun at 13000 rpm for one minute. Several samples still s howed color and a second clarification step was performed with the following changes: samples were incubated with charcoal at room temperature for 1 2 hours, and spun at 13,000 rpm for four minutes prior to collecting the supernatant. Clarified sample extr acts were stored at 20 C until they could be treated for sugar analysis. Total soluble sugars were measured using the phenol sulfuric assay protocol (Dubois, 1956; Buysse and Merckx, 1993) as modified by Chaplin and Kennedy (1994). Additionally, adding the sulfuric acid before the phenol consistently gave better results with glucose solutions of known concentrations, and that modification was adopted for this study. In this spectrophotometric assay, several samples showed color immediately after adding s ulfuric acid. This absorbance change prior to the phenol addition was considered due to compounds in the sample other than sugars. Therefore, a control sample without addition of phenol was run with all samples that developed color upon

PAGE 70

70 addition of sulfuri c acid and was subtracted from the readings obtained when all reagents were added. Starch content of the tissue samples was evaluated by suspending the insoluble fraction that resulted from the ethanol extraction in 2 mL of 0.2 N KOH. Samples were boiled for 30 minutes, cooled to room temperature, and 1 mL of 1M acetic acid at pH 4.5 was added. Starch tissue was digested by adding 100 units of Rhizopus amyloglucosidase (Sigma Chemical Co., St Louis, MO) previously dissolved in 1 mL calcium acetate buffer a t pH 4.5. The enzyme had been dialyzed overnight before use. After incubating for 6 hours at 37C in a dark shaking water bath, samples were centrifuged for 15 minutes at 5000 rpm, the supernatant was decanted and the volume was recorded. The pellets were digested a second time and each digestion was evaluated separately. Samples were clarified by adding 1 mL of the supernatant to 25 mg activated charcoal, incubated for 2 4 hours, spun at 13,000 rpm and approximately 0.5 mL of clear sample was recovered. Th e phenol sulfuric assay, as described above, was used to measure the amount of glucose resulted from each digestion. Data from each digestion was added for analysis. Samples were read at 490 nm in a spectrophotometer (Shimadzu UV 1800), against reference c urves obtained with standard glucose solutions, for both sugar and starch data. Statistical analysis was done with SAS software version 9.2 (SAS Institute Inc., Cary, NC) using PROC GLM. Analysis of Variance (ANOVA) was performed first, with cultivar and p ropagation type as main effects. Treatment means were separated using significance. PROC GLIMMIX was used to analyze starch data from roots collected

PAGE 71

71 after the first harvest bec ause of unbalanced data. In cases where the overall effect of propagation was not significant but interactions between main effects were significant (cultivar by propagation, location by propagation, or location by cultivar by propagation) the sliced state ment was used. This was done to detect if the lack of propagation effect over all cultivars was due to one cultivar affected by propagation method in the opposite direction than the other ones, causing the overall means of propagation methods to be statist ically not different from one another. This statement produces a separate anova for each cultivar, without having to run the data for each cultivar separately, which would reduce the degrees of freedom of the analysis (Littell et al., 2002). Results Sol uble S ugars a t P lanting Total carbohydrate data for root, shoot and leaves were transformed using the logarithmic function to approximate the normal distribution. Means were separated based on Least Square Means (LSM) of the transformed variables. At plan ting, TC had greater root sugar concentration than SW plants (TC= 5.00 vs. SW= 3.98) and higher sugar content (TC=29.50 vs. SW=5.1). Additionally, significant cultivar by propagation interaction was observed: 'Emerald' and 'Primadonna' TC roots had greater sugar concentration while 'Jewel' TC plants had reduced root sugar concentration, compared to SW plants (Table 3 1). However, TC plants of all cultivars had greater root sugar content (Table 3 1). Increased shoot sugar concentration on 'Emerald' and 'Prim adonna' TC shoots was found, while, propagation had no significant effect on sugar concentration of 'Jewel' shoots. Total shoot carbohydrate content was significantly greater in TC plants, compared to SW plants (SW= 21.354 mg glucose equivalent vs. TC= 76.69 mg glucose equivalent ). Overall, propagation did not have a

PAGE 72

72 significant effect on leaf sugar concentration but there was a significant cultivar by propagation interaction which resulted in higher leaf sugar concentration in 'Jewel' plants, w ith no significant effect on the other two cultivars. Micropropagated plants had greater leaf sugar content, compared to plants from cuttings. Propagation effect and cultivar by propagation interaction for leaf sugar content were both significant. 'Emerald leaf sugar content was not affected by propagation (p=0.055), while 'Jewel' sugar content was higher and 'Primadonna' sugar content was lower in TC than in SW plants (Table 3 1). A significant propagation effect on whole plant soluble sugar content and a significant interaction between cultivar and propagation type were observed at planting: 'Emerald' and 'Jewel' TC plants had significantly more total soluble sugars than SW plants, while propagation did not have a significant effect on total soluble suga r content of 'Primadonna' plants (Table 3 1). Soluble Sugars after the First H arvest Data for leaf tot al sugar concentration data and total sugar for crowns, shoots and leaves were transformed using the logarithmic function to approximate the normal distri bution. There was increased sugar concentration in crowns of TC compared to SW plants ( Table 3 3 ). A significant cultivar by propagation interaction was observed for crown sugar content which was significantly higher in 'Jewel' and Emerald' TC compared t o SW plants, while there was no significant difference in crown sugar content between propagation types for 'Primadonna' (Table 3 2) Root sugar content was not significantly affected by propagation types when all cultivars were averaged together. However, there was a significant cultivar by propagation interaction: 'Emerald' root sugar

PAGE 73

73 concentration was not affected by propagation type, 'Jewel' root concentration was reduced in TC compared to SW plants, and 'Primadonna' TC roots had significantly higher su gar concentration than SW roots (Table 3 2). TC plants had significantly greater root sugar content than SW plants when all cultivars averaged together (Table 3 3). A significant propagation effect and a cultivar by propagation interaction were seen for sh oot sugar concentration with no differences in 'Emerald' or 'Jewel' shoots, while 'Primadonna' TC had higher shoot sugar concentration than SW plants (Table 3 2). Shoot sugar content was significantly greater in 'Jewel' TC compared to SW plants, while prop agation method did not have a significant effect in the other two cultivars. Overall, micropropagation resulted in reduced leaf sugar concentration compared to plans from softwood cuttings (Table 3 3). Additionally, propagation effect and cultivar by propa gation interactions were significant for leaf sugar content with 'Emerald' and 'Jewel' having greater leaf sugar content while 'Primadonna' was not affected by propagation method (Table 3 2). Propagation had a significant effect on whole plant sugar and re sulted in significantly greater sugar content T C plants compared to SW plants (SW= 6412 mg glu eq vs TC= 12891 mg glu eq). Plant Starch at P lanting Leaf concentration and root and shoot and whole plant total starch data were transformed with the logarithmi c function and leaf starch content was transformed with the square root function to approximate a normal distribution. Significant propagation and cultivar by propagation interaction were observed for root concentration. 'Emerald' and 'Primadonna TC plant s had higher root starch concentration while 'Jewel' TC plants had reduced concentration compared to that of

PAGE 74

74 SW plants (Table 3 4). Cultivar by propagation interaction was not significant for root starch content and micropropagated plants had greater root starch content (TC: 119.45 mg vs SW:24.73 mg). TC plants had higher shoot starch concentration compared to SW plants (TC= 16.6 vs. SW= 8.72) and there was a significant cultivar by propagation interaction: shoot starch concentration was higher in 'Emerald' and 'Primadonna' while propagation method did not affect 'Jewel' plants. Although the strength of propagation effect varied among cultivars, s hoot starch content was higher in TC plants of all cultivars (Table 3 4). TC plants had higher leaf starch concen tr ation than SW plants (TC= 15.79 vs. SW= 10.36 mg glucose equivalent/g dry weight ), but the interaction between cultivar and propagation method was not significant. There was increased starch content in leaves of 'Emerald' and 'Jewel' TC plants with no si gnificant effect of propagation method on leaf starch content of 'Primadonna' (Table 3 4). Plant starch content was higher in TC compared to SW plants for all cultivars (Table 3 4). Plant Starch after First H arvest Crown starch concentration and content were greater in 'Emerald' and 'Jewel' TC plants with no significant effect of propagation type for 'Primadonna' (Table 3 5). 'Emerald' TC roots had higher starch concentration and greater starch content, while the ef fect of propagation type on the other two cultivars was not significant (Table 3 4). Micropropagated plants of all cultivars had higher shoot starch concentration (Table 3 4). Shoot starch content was greater in 'Emerald' and 'Jewel' TC plants with no sign ificant effect of propagation method on shoot starch content of 'Primadonna'. Propagation had no significant effect on leaf starch concentration. However, total leaf starch content of 'Emerald' and 'Jewel' TC plants was higher than SW plants, with no

PAGE 75

75 signi ficant effect of propagation method on 'Primadonna'. The overall propagation effect on whole plant starch content was significant with TC 'Emerald', 'Jewel' and 'Primadonna' plants having greater plant starch content than SW plants (Table 3 6). Total Plant Carbohydrate C ontent Total carbohydrate content at planting and in June 2011 was higher in TC plants than in SW plants for all cultivars in the study (Figure 3 1 and 3 2).Cultivar by propagation was not significant at either sampling date. No correlation was found between the difference in plant dry weight and the difference in carbohydrate content between propagation types for each cultivar at either one of the two points that plant tissue samples were assayed for carbohydrate content. Discussion In gene ral, differences in carbohydrate content (sugar or starch) were a result of differences in concentration between propagation types. However, we also observed differences in total sugar or starch content of a particular plant part even when there was no eff ect of propagation method on concentration (sugar or starch) for a particular plant part. This was the case for 'Jewel' shoots and 'Primadonna' leaves, where there was no significant effect of propagation method on sugar concentration but the total sugar c ontent of the organ was affected by propagation type at planting (Table 3 1). This could be attributed to large differences in leaf and shoot dry weights at that time (Table 2 5, chapter 2). Occasionally, propagation type had the opposite effect on concent ration than what was observed for total carbohydrate content. For example, roots 'Jewel' TC plants at planting had sugar and starch concentrations that were for total su gar content. Because roots do not photosynthesize, and they depend on

PAGE 76

76 carbohydrates received from other plant parts (stems, leaves), this may have resulted in a 'dilution' of sugar and starch in 'Jewel' TC roots, because mean root DW of TC plants was 7 tim es the mean root DW of SW 'Jewel' plants at that point (Chapter 2, table 2 5). Another explanation for the reduced sugar and starch concentration observed in 'Jewel' TC roots is that actively growing shoots and newly expanded leaves may have been much str onger sinks for carbohydrates than leaves and shoots of SW plants, due the their greater dry weight (Chapter 2). 'Emerald' and 'Primadonna' TC plants also had greater root dry weight at planting and therefore the two fold increase in root starch concentrat ion observed at planting (Table 3 4) was not expected. Although genes involved in storage reserve synthesis can be upregulated by sugars (Koch, 1996) which are present in the micropropagation media, this is not likely the reason of increased starch concen tration in our study as roots formed in vitro are short lived and often blueberries are rooted ex vitro. The cause for the observed increased starch concentration in 'Emerald' and 'Primadonna' TC roots at planting remains unclear and may need further inves tigation. Since micropropagation media often contains a source of sugar (Beyl, 2005; Lyrene, 1980) it is not unusual for leaves formed in culture to have reduced photosynthetic capacity (Zimmerman, 1987). However, by the time we received the TC plants ple nty of new leaves had formed and therefore photosynthetic capacity would likely have been regained. Whole plant sugar or starch content (Table 3 4 and 3 5) differences between propagation types cannot be explained exclusively by differences in whole pl ant dry weight. For example, 'Primadonna' TC plants had significantly more starch content at

PAGE 77

77 planting despite a lack of propagation effect on plant DW (Chapter 2). This may be due to differences in allocation of carbohydrates between TC and SW 'Primadonna plants at planting: greater dry weight of tissues with increased starch and sugar concentraction (root and shoot) was found in TC compared to SW plants, which may help explain the higher whole plant carbohydrate content even though there was no significa nt effect of Conclusions 'Emerald' and 'Jewel' TC plants had greater sugar and starch content than SW plants, at planting and after the first harvest. However, propagation method did not affe ct whole plant sugar content of 'Primadonna' at planting or total starch accumulation and carbohydrate content after the first harvest. We did not observe a correlation between whole plant sugars, starch or carbohydrate content at planting and flower bud number (data not shown). This can be attributed to the fact that flower buds are initiated later in the season and analysis of samples collected at the end of the growing season may provide helpful insight on the relationship between flower bud initiation and number and carbohydrate availability on plants derived from softwood cuttings and tissue culture. No correlation was found between plant dry weight and plant carbohydrate content at planting or after the first harvest. This may be explained by increase d concentration (sugar, starch or both) of one or more plant parts in TC plants at planting and after harvest. This increase in carbohydrate concentration, along with differences in plant dry weight resulted in much larger differences in carbohydrate conte nt between propagation types that what was observed for dry weight. The reason for this increased sugar/starch

PAGE 78

78 concentration observed in some plant parts of TC compared to SW plants is unclear and warrants further investigation.

PAGE 79

79 Table 3 1 Effect of cultivar and propagation method on sugar concentration and total sugar content in April 2010 Mg glu eq/g DW total glu (mg) Roots Shoots Leaves Roots Shoots Leaves Whole plant ESW 2.56 9.52 ** 15.84 NS 5.13 c z 14.36 cb 19.33 NS1 38.82 *** ETC 4.73 16.31 19.11 16.47 b 77.48 a 37.21 131.167 JSW 6.38 9.46 NS 15.04 4.48 c 9.95 c 7.34 *** 21.77 *** JTC 4.31 11.48 19.73 27.92 ab 62.25 ab 46.72 136.89 PSW 3.00 *** 6.81 *** 18.20 NS 5.70 c 39.75 abc 88.56 *** 134.01 NS PTC 8.92 15.60 14.17 44.38 a 90.34 a 29.98 164.70 NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types for each cultivar according to P values from sliced ANOVA. NS1 : p value=0.055 z Means followed by the same letters within a column indicate no differences, Tukey's HSD, P < 0.05. Table 3 2 Effect of cultivar and propagation method on sugar concentration and total sugar content in June 2011 Mg glu eq/g DW total glu (mg) Roots Shoots Crown Shoots Leaves ESW 8.27 NS 14.74 NS 785.35 ** 2271.81 ab 2748.67 ** ETC 8.80 16.61 1252.1 5067.54 a 5132.79 JSW 9.66 13.79 NS 481.83 *** 1488.46 b 2905.40 *** JTC 7.69 16.00 1803.4 5441.25 a 7080.02 PSW 6.43 *** 11.42 *** 814.61 NS 2758.30 ab 2527.87 NS PTC 10.8 22.79 918.07 5742.59 a 3106.80 z Means followed by the same letters within a column indicate no significant differences, Tukey's HSD, P < 0.05. NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types for each cultivar according to P values from sliced ANOVA. Table 3 3 Effect of propagation method on total root sugar content and crown and leaf sugar concentration and in June 2011 Propagation Root Total glu eq (mg) Crown (Mg glu eq/g DW) Leaf (Mg glu eq/g DW) SW 818.12 b z 15.83 b 45.86 a TC 1042.47 a 23.63 a 39.55 b z Means followed by the same letters within a column indicate no significant differences, Tukey's HSD test, P < 0.05.

PAGE 80

80 Table 3 4 Effect of c ultivar and propagation method on starch concentration and total starch content in April 2010 Mg glu eq/g DW total glu (mg) Roots Shoots Roots Shoots Leaves Whole plant ESW 14.94 ** 7.57 *** 30.01 b z 12.60 *** 11.74 bc 54.35 c ETC 30.11 25.13 91.40 a 122.67 33.96 a 248.04 a JSW 26.57 12.75 NS 18.39 b 12.62 *** 5.17 c 36.18 c JTC 12.70 10.02 81.98 a 54.31 35.33 a 171.62 ab PSW 13.07 *** 5.85 ** 25.78 b 33.45 43.99 a 103.23 b PTC 38.16 14.39 184.96 a 84.62 29.84 ab 299.42 a z Means followed by the same letters within a column indicate no significant differences, Tukey's HSD test, P < 0.05. NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types for each cultivar according to P values from sliced ANOVA.

PAGE 81

81 Table 3 5 Effect of cultivar and propagation method on starch concentration and total starch content in June 2011. Mg glu eq/g DW total glu (mg) Crown Roots Shoots Crowns Roots Shoots Leaves Whole plant ESW 11.17 c 10.00 bc z 6.20 bc 482.70 b 962.51 924.75 b 1115.44 c 3485.41 bcd ETC 33.57 a 24.80 a 15.44 a 1943.18 a 3555.72 *** 4641.41 a 2673.21 a 12814.0 0 a JSW 5.90 c 8.83 bc 2.14 d 171.07 c 623.93 NS1 237.64 c 1166.71 c 2199.36 d JTC 25.69 ab 14.07 ab 5.58 bc 1637.56 a 1427.00 1960.21 ab 2513.22 ab 7537.98 ab PSW 4.52 c 4.00 c 3.55 cd 344.52 b 482.18 NS 1054.17 b 1170.06 c 3062.58 cd PTC 14.38 bc 11.10 abc 9.91 ab 698.60 ab 1156.03 2485.28 ab 1447.52 bc 5887.88 bc z Means followed by the same letters within a column indicate no significant differences, Tukey's HSD test P < 0.05. NS,*,**,*** Nonsignificant or significant at P < 0.05, 0.01 or 0.001 between propagation types for each cultivar according to P values from sliced ANOVA. NS1 : p value=0.0971 Table 3 6 Effect of cultivar and propagation method on plant total carbohydrate in April 2010 and June 2011. Plant Carbohydrate (mg) Propagation April 2010 June 2011 SW 129.45 b z 9158.9 b TC 383.95 a 21576 a z Means followed by the same letters within a column indicate no significant differences, Tukey's HSD test P < 0.05.

PAGE 82

82 CHAPTER 4 EFFECT OF PROPAGATION METHOD ON REPRODUCTIVE GROWTH, YIELD AND FRUIT QUALITY OF SHB CULTIVARS (VACCINIUM CORYMBOSUM L.) OBTAINED FROM MICROPROPAGATION AND SOFTWOOD CUTTINGS Introduction The effects of propagation method on reproductive growth have been reported for many fruit species. Swartz et al. (1981) reported tissue culture derived (TC) strawb erry plants had higher yields, due to increased number of berries/unit. Berry weight, however, decreased by 26% in TC plants (Swartz et al., 1981). Conversely, there was no significant effect of propagation method on yield of five cultivars and one advance d selection of blackberry during the first harvest season after planting. During the second season, yields of 'Smoothstern' and the advanced selection, SI US 68 6 17, were significantly higher from TC plants, as was the yield of all the cultivars combined. Berry weight of TC plants was significantly lower than for plants from softwood cuttings for all Varied results have been reported with regards to the effects of propagation method on reproductive growth of blueberry and its Vaccinium relatives. Gustavsson and Stanys (2000) observed that although TC lingonberry plants did not flower during the first year, in the second and third years after planting yield of TC plants was significant ly higher than for plants obtained from softwood cuttings, without any effect on average fruit weight. These results disagree with those reported by Jamieson and Nickerson (2003) for three lowbush genotypes. Yield was significantly affected by propagation method in 1988, 1990 and 1994, and there was a significant propagation by cultivar interaction. Yield of genotypes ME3 and K206 were highest for seedlings,

PAGE 83

83 intermediate for cutting derived plants and lowest for TC plants, but the opposite was reported for K71 13 in 1988 and 1990. Berry weight was usually lower for TC plants. Souza et al. (2011) evaluated reproductive growth and fruit quality of three rabbiteye cultivars propagated by TC and rooted cuttings and reported no significant effect of propagation type on fruit yield or diameter. However, average berry weight of fruit from TC 'Briteblue' plants was significantly lower than berry weight for plants obtained from rooted cuttings, with no significant difference for the other two cultivars. There was no significant difference in soluble sugars content between propagation types for 'Bluegem' and 'Woodward', but fruit from TC 'Briteblue' plants had significantly higher levels of soluble sugars than fruit from cutting derived plants. Conversely, titratable a cidity of fruit from TC plants was significantly higher for 'Woodward', with no effect of propagation type on titratable acidity for the other two cultivars. Reproductive performance of half bud cuttings was studied during the first year after field planting by Grout et al. (1986) in three locations: Grand Rapids, St. Paul, and Becker (Minnesota). Micropropagated p lants had twice the number of flower buds as plants obtained from softwood cuttings in Grand Rapids and St Paul, with no significant difference between propagation types in Becker. Read et al. (1988, 1989) later reported fruit yield from the same planting. In Grand Rapids, TC plants produced twice as much fruit as plants from cuttings during the first two harvests, with a yield increase around fifty percent during the third year. Fruit size and quality were not affected by propagation method. The same plant s were evaluated from 1988 to 1994, and total yield of TC plants at Grand Rapids was significantly greater than plants from cuttings in 1989, 1994 and when all years were

PAGE 84

84 combined. However, at Becker, yield was not significantly affected by propagation me thod in any year. Mean berry weight of TC plants was significantly lower in 1987 and 1989, but when all years were combined, there was no significant effect of propagation method (El Shiekh et al. 1996). These results differ from those reported for highbus h 'Herbert', by Litwinczuk et al. (2005), who reported that more plants from cuttings flowered the first year and had more inflorescences and higher berry weights than plants from TC. Previous research shows contradictory results and most of the work was d one in areas with cold winters and shorter growing seasons than in Florida. Therefore, the present research was conducted to evaluate reproductive growth of three southern highbush blueberry cultivars obtained from TC and rooted cuttings during the first t wo years of field establishment in two Florida locations. Our hypothesis was that flower bud number and yield would be greater for TC plants than for those from cuttings, with no significant difference on average berry weight or fruit quality. Since previo us research showed that propagation effect may vary among cultivars of the same species we tested for interaction between propagation method and the three cultivars. Materials and Methods The locations, cultivars, field and experimental design, and m anagement practices were as previously described in Chapter 2. Flower Bud N umber On 10 December 2010 (Haines City) and 12 14 January 2011 (Citra), two representative plants per replicate were selected and average plant height and width, number of major canes and total number of shoots and flower buds per plant were

PAGE 85

85 recorded. At the beginning of the second reproductive season (December 2011) one representative plant per replicate was selected, two to four representative major canes were flagged, and total flower buds per cane were recorded, along with total number of canes per plant. The percentage of the total plant volume represented by the sampled canes was visually estimated to be between 25 50% and was used to estimate average number flower buds per cane and total flower buds per plant. Full Bloom D ate Five plants per replicate were evaluated twice in February and March, 2011 to estimate propagation effect on date of 50% bloom during the first reproductive season after field establishment. At each da te, a visual estimation was made for the percentage of flower buds on each plant that had mostly open florets, or were at petal fall (Spiers, 1978). Fruit were not included in the evaluation, as they were considered to be from a fall bloom and not importan t for commercial harvest. Yield The same plants selected for flower bud evaluation were harvested every three to five days from 28 March until 2 June in Haines City and from 4 April until 6 June in Citra. The majority of the harvested fruit was at the mat ure blue stage, but overripe berries or berries that were starting to mature were also collected in order to minimize fruit losses due to wind or rain between harvests. Fruit was harvested in plastic buckets, placed in res alable polyethylene bags, and kept in a portable cooler with ice until they were weighed with a Scout PRO 4001 scale (Ohaus Corporation, Pine Brook, NJ). Berry W eight Fruit weight was determined from 25 representative berries from each harvested plant. I n Haines City, average berry weight was recorded on 25 April and 2, 9, and 16

PAGE 86

86 May, 2011. Fruit size in Citra was evaluated on 30 April, and 4, 12, 22 and 26 May. Data from the first four dates were analyzed together and data taken at Citra on 26 May were a nalyzed separately, since in Haines City several of the yield data plants had less than 25 berries. Fruit Quality Fruit total soluble solids (TSS or Brix) and titratable acidity (citric acid equivalent) were evaluated at three points during harvest at Hai nes City (18 April, 2 May and 9 May 2011) and five times at Citra (14 April, 4 May, 12 May, 22 May and 1 June 2011). Fruit samples were frozen at 20C until they were processed. Samples were removed from the freezer, allowed to thaw and a 250 mL beaker wa s filled up to the 100 mL mark with whole fruit. A hand held blender was used to homogenize the fruit, which was centrifuged at 11,000 rpm for 25 minutes and the supernatant was filtered through cheesecloth into a 25 mL tube. The extracted juice was frozen at 20C until it was used to measure pH, soluble solids, and titratable acidity (TA). Total soluble solids content was measured using a digital refractometer (Pocker Refractometer Pal 1, Atago Tokyo). Total acidity as a percentage of citric acid was o btained by diluting 6 mL of juice with 50 mL distilled water and titrating it with 0.1 N NaOH to pH 8.2. pH and total acidity was measured using an autotitrator (Mettler Toledo Titrator Dl15). Statistical Analysis Analysis of V ariance was performed using SAS software version 9.2 (SAS Institute Inc., Cary, NC). Means were separated using the PDIFF option of the LSMEANS where the overall effect of propagation was not signifi cant but interactions between main effects were significant (cultivar by propagation, location by propagation, or

PAGE 87

87 location by cultivar by propagation) the slice statement was used. This was done to detect if the lack of propagation effect over all cultivar s was due to one cultivar affected by propagation method in the opposite direction than the other ones, causing the overall means of propagation methods to be statistically not different from one another. This statement produces a separate ANOVA for each cultivar, without having to run the data for each cultivar separately, which will reduce the degrees of freedom of the analysis and flower bud number and yield wer e determined using the PROC CORR procedure. Results Flower B uds Flower bu d data from the first season were transformed using the logarithmic function to approximate a normal distribution. T he t otal number of flower buds (FB) per plant was significantly hi gher in Citra than in Haines City during the first season of reproductive growth (Citra=291.5 vs HC= 166.2, P =< 0.0001). L ocation by propagation and location by cultivar by propagation interactions were not significant and treatment means across locations are shown. A cultivar by propagation interaction was found for the first season. 'Jewel' and 'Primadonna' TC plants had reduced mean flower buds/cane with no significant differences between propagation types for 'Emerald' (Table 4 1). Micropropagation resu lted in increased flower buds per plant for 'Emerald' with no significant effect of propagation method on flower bud number for 'Jewel' or 'Primadonna' (Table 4 1). During the second reproductive season, average flower bud number per cane was not signific antly affected by propagation method and interactions between cultivar and propagation method were not significant. Data for total flower buds per plant were

PAGE 88

88 transformed using the square root function to approximate a normal distribution. Location had no s ignificant effect on total flower buds per plant but there was a significant location by cultivar interaction: in Haines City, 'Emerald' and 'Jewel' had more flower buds per plant than 'Primadonna' but there were no significant differences among cultivars in Citra (Table A 3). Total number of flower buds per plant was significantly higher for TC than for SW plants (TC=1229 vs SW=868, P =0.0008). Cultivar by propagation method interaction was also significant : 'Emerald' and 'Jewel' TC plants had significantly more flower buds per plant than plants from softwood cuttings, with no significant difference for Primadonna' (Tables 4 1). Full Bloom Date In 2011, bloom was evaluated twice in Citra and three times in Haines City. The date that was closest to full bloom (50 percent fully open flowers) was 14 February in Haines City and 4 March in Citra. Propagation method did not have a significant effect on percent full bloom, and there was no cultivar by propagation interaction (Table 4 2 ). However, location significan tly affected full bloom date, and bloom percentages were higher in Citra. There was a significant location by cultivar interaction: In Citra, the percentage of fully open flowers was similar for 'Emerald' and 'Jewel' but significantly lower for 'Primadonna but there were no significant differences among cultivars in Haines City (Table A 4 ) Yield During the first harvest season, berry yield was significantly higher in Citra than in Haines City, across all levels of the other main effects. Micropropagation resulted in higher fruit yields (TC=1916 g vs SW=1402 g, P<0.0001). There was also a significant location by cultivar interaction: 'Emerald' and 'Jewel' had higher berry yield in Citra than

PAGE 89

89 in Haines City but there was no significant difference in yield b etween locations for 'Primadonna' (Table A 5). Location by propagation and location by cultivar by propagation interactions were not significant. Therefore, data presented are averages of both locations (Figure 4 1). Moreover, TC 'Emerald' and 'Jewel' had 42 and 58 percent greater yields when compared to plants derived from softwood cuttings. 'Emerald' and 'Jewel' had higher yields than 'Primadonna' over all levels of the other factors (Table A 6) A correlation analysis was performed between yield and can e, total shoot number number and the Pearson Correlation coefficient between total shoot number and yield was 0.94 ( P< 0.0001) in Citra and 0.897 ( P= 0.0184) in Haine s City for the first reproductive season. For 'Emerald', yield correlated better with total buds per plant: the correlation coefficient was 0.86 ( P= 0.0013) in Citra and 0.915 ( P= 0.0002) in Haines City. A p ositive correlation between yield and shoot number was also found, with correlation coefficients of 0.72 ( P = 0.0192) in Citra and 0.844 ( P= 0.0021) in Haines City. total shoot number, or bud number per plant. Conversel y, in Haines City yield of 'Primadonna' was positively correlated with flower bud number per plant (r= 0.845, P =0.002) and total shoots per plant (r=0.748 P =0.013). There was no significant effect of propagation method on fruit yield for any of the cultiva rs during the second reproductive season (Table 4 3 ). Berry W eight Location significantly affected berry weight and resulted in larger berries in Citra than in Haines City on harvest dates one and two. T here was a significant location by

PAGE 90

90 propagation interaction: in Citra, berry weight was not affected by propagation method while in Haines City TC plants had reduced berry weight when all cultivars were analyzed together (Table 4 4) There was a significant cultivar by propagation interaction for berry weight on the second harvest date: 'Emerald' TC plants had significantly smaller berries than those produced on plants obtained from softwood cuttings. Propagation method had no significant effect on berry weight on the other two cultivars (Table 4 5 ). On harvest date three, only cultivar had a significant effect on berry weight: 'Emerald' had larger berries than 'Primadonna' or 'Jewel', across both propagation methods and locations ( E merald = 2.29 g, J ewel = 1.74 g, P rimadonna = 1.61 g ). On the last date that berry weight was evaluated in both locations, the data showed no significant effect of propagation method (Table 4 5 ). Additionally, berry weight recorded at Citra on 26 May was not affected by propagation type nor were the interactions between main effe cts significant. Fruit Quality On harvest date one, fruit produced on plants from TC had lower pH than fruit from SW plants (SW= 3.27 vs TC=3.21, P =0.0102) On harvest date two, propagation effect and cultivar by propagation interaction on fruit pH were no t significant. The main effect of propagation method on fruit pH at harvest date 3 was not significant, but there was a significant cultivar by propagation method interaction: fruit from 'Jewel' micropropagated plants had significantly lower pH than fruit produced on plants from softwood cuttings but there was no significant difference among propagation methods for the other two cultivars. (Table 4 6 ). Micropropagation resulted in significantly greater total soluble solids content compared to fruits from so ftwood cutting plants (TC= 8.54 and SW=8.20

PAGE 91

91 P =0.0352 ). Also, differences between propagation types for titratable acidity and sugar:acid ratio were not significant (data not shown), and neither were the interactions between main effects for these variable s. For harvest date two t here was a significant cultivar by propagation interaction on TSS which was significantly higher on fruit from TC plants of 'Jewel' and 'Primadonna', but there was no significant difference between propagation methods for 'Emeral d' fruit (Table 4 6 ). A dditionally, a significant cultivar by propagation method interaction was found for titratable acidity :t itratable acidity was significantly higher for 'Jewel' fruit from TC plants compared to fruit produced on SW plants. Micropropag ation resulted in higher soluble sugar content on harvest date three (SW=9.23 vs TC= 9.92, p = 0.0321) Propagation effects on fruit titratable acidity and sugar:acid ratio were not significant. There was significant location by cultivar by propagation inte raction on titratable acidity at the third harvest date with titratable acidity at Citra being significantly higher for fruit from TC 'Jewel' plants, while in Haines City propagation method had no effect on titratable acidity on any of the cultivars in the study (Table 4 6 TA data only from Citra). On 22 May and 1 June fruit collection was only made at Citra. Propagation effects on pH and titratable acidity on date 4 and 5 and on titratable acidity and sugar:acid ratio on date 5 were not significant, a nd neither was the interaction between propagation type and cultivar. Conversely, there was a significant effect of propagation method and significant propagation by cultivar interaction on sugar content on date four and five and on sugar/acid ratio on da te four. Fruit from 'Primadonna' TC had higher fruit soluble sugar content and sugar/acid ratio, with no significant difference between propagation

PAGE 92

92 types for the other two cultivars for date 4 'Jewel' and 'Primadonna' fruit from date five had higher suga r levels when produced on TC plants, while at that point there was no difference in 'Emerald' fruit quality between propagation methods (Table 4 7 ). Additionally, micropropagation resulted in greater soluble sugar content at both dates. Discussion Effect o f Propagation Method on F lowering After evaluating bloom twice at each location to determine the date of 50% full bloom, we observed no significant differences among propagation type on the percentage of bloom at a specific date during the first year of re productive growth. Our findings differ from those observed by Smolarz et al (1997) who reported that TC 'Bluecrop' plants flowered earlier four out of seven years with no significant difference in flowering time the other three. However, these results disa gree with those of Litwinczuk et al. (2005), who reported that only 22% of TC and 60% of softwood cutting derived in April, flowers were removed to encourage vegeta tive growth but it was noted that TC plants of all three cultivars and softwood cutting derived 'Primadonna' plants had flowers at planting. Effect of P ropagation Method on R epr oductive Growth and Y ield In the present study, TC resulted in increased flower bud number per plant for the first two years and increased yield for the first harvest season. Similar results were reported on half highbush 'Northblue', which had increased flower bud number and yield for the first two harvests on plants obtained from t issue culture compared to those obtained from softwood cuttings (Grout et al., 1986, Read et al., 1989). However, results from our study on yield of 'Emerald' and 'Jewel' disagree with those of Smolarz et al.

PAGE 93

93 (1997) and Souza et al. (2011) who reported no significant differences on yield among cultivars. We also observed no significant effect of propagation type on flower bud number or yield of 'Primadonna' which suggests t hat the response may be cultivar dependent and evaluation of other cultivars may be needed. Location had a significant effect on flower bud number during the first year but not during the second reproductive season. A possible explanation is that flower bu ds were evaluated in Citra one month later than in Haines City during the first year and in the second year both locations were evaluated during the first two weeks of December (2011). It may be possible that flower buds develop and initiate more slowly a t Citra and by evaluating earlier than the first year we may not have captured all flower buds at Citra. number of shoots per plant, respectively. Although total average flower buds per cane did not differ between propagation methods, total flower of buds per plant was greater in 1), suggesting that this is a result of increased number of canes and total shoots per plant in TC plants compared to plants from softwood cuttings (Table 2 4 and Figure 2 9). Similar results were reported by Grout et al. (1986), who attributed the greater number of flower buds per plant to greater lateral branching in TC plants. Despite the increase in flower bud numb er in 'Emerald' and 'Jewel' TC plants compared to SW plants, fruit yield was not significantly affected by propagation method during the second harvest season. This could be a consequence of late freezes

PAGE 94

94 occurred on 12 and 13 February, 2011, which may hav e affected some data plants more than others, depending on the ir location within the field. Another possible explanation could be reduced floret number per flower bud and or reduced fruit set of the TC pl ants compared to the SW plants. Berry W eight We obse rved no significant difference due to propagation method in berry weight of 'Jewel' and 'Primadonna'. Similar results were obtained with lowbush by Morrison et al. (2000), who reported no difference in berry weight of two clones when propagated by tissue c ulture or softwood cuttings. However, these results disagree with results of reported that plants derived from tissue culture had reduced berry weight compared to plants p ropagated from cuttings. In our study we did observe reduced berry weight of TC 'Emerald' at two points during the first harvest season. Similar results were obtained with rabbiteye blueberries by Souza et al. 2011, who reported reduced berry weight of TC 'Briteblue' without significant differences due to propagation method on the other two cultivars. For both studies reduced berry weight as a result of TC was observed in the cultivar with the largest berries, which suggests that it may not be feasible to o btain significantly higher yields in cultivars with very large berries without reduced berry weight. Fruit Q uality In the present study, the greater total soluble solids on fruit from TC plants was observed in mid and late season in 'Jewel' and at the end of the harvest period in 'Primadonna' (Table 4 6 and 4 7 ). During this period, increased titratable acidity of TC 'Jewel' was also observed. Because reproductive growth is concomitant with vegetative

PAGE 95

95 growth, it is possible that limited carbohydrate resources were adversely affecting fruit quality during the first half of the season, since it was noted that plants did not produce leaves well during early berry development. Additionally, the highest frui t load is early in the season, before mature berries began to be harvested. However, by the end of the harvest period, plants had abundant new leaves which may explain the higher sugar content of fruit on TC 'Jewel', which had significantly higher leaf dry weights in June 2011. Coincidentally, 'Jewel' berries ripened later than the other two cultivars. Conclusions Micropropagated 'Emerald' had more flower buds per plant than plants derived from softwood cuttings during the first season of reproductive grow th. Fruit yield s of 'Jewel' and 'Emerald', TC plants were significantly higher than softwood cutting derived plants, but there was no significant effect of propagation method for 'Primadonna'. A positive correlation between yield and total number of shoots was observed for 'Emerald' and 'Jewel' in Citra and Haines City. 'Jewel and 'Primadonna' berry weight was not significantly affected by propagation type. Conversely, reduced average berry weight was recorded for TC 'Emerald' at the beginning of harvest i n Haines City, without any difference between propagation types the rest of the season or in Citra. Regarding fruit quality, greater soluble solids content on fruit from TC 'Jewel' and 'Primadonna' were observed at mid or late season in the first harvest. In addition, increased titratable acidity was mea sured in 'Jewel' fruit from TC plants compared to SW plants. During the second reproductive season, TC resulted in increased flower buds per plant in 'Emerald' and 'Jewel', with no significant effect of prop agation type on 'Primadonna'. We attributed this difference to a greater number of total shoots per plant,

PAGE 96

96 since there were no significant differences in average flower buds per major cane. Although 'Emerald' and 'Jewel' TC plants had more flower buds than SW plants, f ruit yield was not significantly affected by propagation method during the second harvest season. These results suggest that plant response to growth regulators used during in vitro propagation may be cultivar dependent and several other culti vars need to be evaluated under field conditions to determine which propagation is most advantageous for a specific cultivar. Moreover, cultivars should be evaluated during the first two to three years in the field and in several locations before recommend ations can be made.

PAGE 97

97 Table 4 1 Effect of cultivar and propagation method on a verage number of flower buds per cane and total flower buds per plant during the first and second reproductive season s Means of both locations. Cultivar Propagation method Total FB 2010 11 z Average FB /cane 2010 11 Total FB Dec. 2011 Average FB/cane Dec. 2011 E SW 160.1 *** 31.3 NS 821.9 *** 131.2 ab z E TC 264.8 31.1 1472.2 140.9 a J SW 218.4 NS 37.6 1000.0 *** 132.8 a J TC 264.0 27.1 1514.3 144.0 a P SW 255.7 NS 34.0 *** 782.1 NS 116.9 ab P TC 210.0 21.3 699.7 72.4 b NS *** Non significant or significant at P < 0.001 between propagation types f or each cultivar, according to P values from sliced ANOVA in Table 4 2. z Flower buds were evaluated on 9 December 2010 in Haines City and 12 14 January 2011 in Citra. y Means followed by the same letters within a column indicate no differences, Tukey's HSD test, Table 4 2 Effect of cultivar and propagation method on p ercentage of full bloom at Citra on 4 March 2011 and Haines City on 14 February 2011. Location Cultivar Propagation Percentage of full bloom Citra E SW 67.2 ab z E TC 66.4 ab J SW 71.2 a J TC 67.6 ab P SW 54.4 bcd P TC 55.2 bc Haines City E SW 52.4 cd E TC 49.3 cd J SW 41.2 d J TC 49.4 cd P SW 46.8 cd P TC 49.80 cd z Means followed by the same letters within a column indicate no differences, Tukey's HSD test,

PAGE 98

98 Figure 4 1 Effect of cultivar and propagation method on t otal yield in 2011. Means of both locations. NS, *** Non significant or significant at P <0.001 between propagation types f or each cultivar, according to P values from sliced ANOVA. Table 4 3 Effect of cultivar and propagation method on yield during the second harvest season (2012). Data from Citra, F L only Cultivar Propagation Yield (g) E SW 3656.52 ab z E TC 4790.08 a J SW 2444.00 bc J TC 2212.06 c P SW 1115.96 c P TC 1465.70 c z Means followed by the same letters within a column indicate no significant differences Tukey's HSD test, Table 4 4 Effect of cultivar, propagation method and location on berry weight. Data recorded at Citra on 30 April, and at Haines City on 25 April, 2011. Location Propagation BW (g) Citra SW 2.12 a z TC 2.2 4 a Haines City SW 2.19 a TC 1.87 b z Means followed by the same letters within a column indic ate no significant differences, Tukey's HSD test,

PAGE 99

99 Table 4 5 Effect of cultivar and propagation method on b erry weight at four points during the first harvest season. Means of both locations. C ultivar Propagation BW (g) Date 2 BW (g) Date 3 BW (g) Date 4 BW (g) Date 5 y E SW 2.38 ** 2.25 a z 1.81 a 1.54 a E TC 2.23 2.32 a 1.76 a 1.50 ab J SW 1.90 NS 1.79 b 1.59 b 1.32 ab J TC 1.8 6 1.70 b 1.56 b 1.29 ab P SW 1.62 NS 1.69 b 1.39 c 1.19 ab P TC 1.6 8 1.52 b 1.37 c 1.11 b Date 2: 2 May in Haines City and 4 May in Citra. Date 3: 9 May in Haines City and 12 May in Citra. Date 4: 16 May in Haines City and 22 May in Citra. z Means followed by same letters within a column indi cate no significant differences, Tukey's HSD test, y Data recorded from Citra on 26 May. NS,*** Nonsignificant or significant at P < 0.001 between propagation types within a cultivar, according to P values from sl iced ANOVA. Table 4 6 Effect of cultivar and propagation method on fruit p H, total soluble solids and titratable acidity at three points during the first harvest season Means of both locations. Treat TSS Date 2 z TA Date 2 PH Date 3 TA Date 3 y ESW 9.81 NS 0.63 NS 3.29 NS 0.77 NS ETC 9.53 0.61 3.32 0.69 JSW 8.54 *** 0.72 *** 3.22 0.78 *** JTC 9.57 0.85 3.13 1.05 PSW 8.08 ** 0.28 NS 3.78 NS 0.31 NS PTC 8.83 0.30 3.80 0.32 z Date 2: fruit collected on 2 May in Haines City and 4 May in Citra. Date 3: 9 May in Haines City and 12 May in Citra. y This column shows data only from Citra as at that date they were no significant differences in titra t able acidity between propagation types in Haines City. NS,*** Nonsignificant or significant at P < 0.001 between propagation types for each cultivar, according to P values from sliced ANOVA. Table 4 7 Effect of cultivar and propagation method on s oluble sugar content and titratable acidity F ruit harvested from Citra on 22 May and 1 June 2011. Treat ment TSS:TA Date 4 TSS Date 4 TSS Date 5 TSS:TA Date 5 ESW 17.33 NS 9.67 NS 10.75 NS 25.80 b z ETC 15.69 9.15 9.83 27.38 b JSW 14.62 NS 9.38 NS 9.76 24.09 b JTC 14.41 10.49 12.28 26.64 b PSW 32.78 ** 8.45 *** 7.66 34.30 ab PTC 47.78 11.37 9.94 48.83 a NS,*** Nonsignificant or significant at P < 0.001 between propagation types for each cultivar, according to P values from sliced ANOVA. z Means followed by same letters within a column indicate no significant differences Tukey's HSD test,

PAGE 100

100 CHAPTER 5 SUMMARY AND CONCLUSIONS Blueberries have traditionally been vegetatively propagated by softwood or hardwood cuttings. Although plants can be produced at low cost, cuttings from some cultivars do not root easily and it may take up to several yea rs to generate enough stock of newly released cultivars so significant acreage can be planted. Additionally certain fungal, bacterial and viral diseases can be spread throughout the propagation bed when infected material is used. Converse ly mi cropropagation has the potential to produce large numbers of healthy plants more quickly than propagation by rooted cuttings. C hanges in the growth habit of micropropagated Vaccinium spp and other small fruit crops have been reported However, much of the previous research had been done in climates with colder winters and shorter growing seasons than in Florida. Therefore, t he present study was developed to evaluate vegetative and reproductive growth of southern highbush blueberry cultivars obtained from m icropropagation and softwood cuttings under Florida field conditions that enable their direct comparison. The study was conducted at two locations chosen for their different average chill hour accumulation: Citra, FL (420 540 chill hours) and Haines City FL, (110 210 were wer e planted in April 2010 at both locations. At both locations, 'Jewel' TC plants had larger plant volume than SW plants, while the opposit e was observed for 'Primadonna' throughout the first growing season (June to November 2010) Differences in plant volume between propagation methods were less evident during the second growing season, and limited to Haines City from July to

PAGE 101

101 August 2011. By the end of the second growing season there were no differences in plant volume between propagation types at either of t he locations. Micropropagation resulted in greater number of canes, lateral branches and total shoots per plant for 'Emerald' and 'Jewel' after the first and second growing seasons, with no significant effect on total shoot number for 'Primadonna' at any p oint during the study Micropropagated 'Emerald' and 'Jewel' had significantly higher plant dry weight after one season in the field and after the second season higher plant dry weight was recorded for TC 'Jewel', compared to SW plants Conversely there w as no significant effect of propagation type on plant dry weight for 'Primadonna' at either one of the sampling dates. 'Emerald' and 'Jewel' TC plants had greater sugar and starch content and higher carbohydrate availability at planting and after the first harvest. However, propagation method did not affect whole plant sugar content of 'Primadonna' at planting or total starch accumulation and carbohydrate pool after the first harvest. Propagation effect on reproductive growth varied among cultivars and between the first and the second reproductive seasons. At both locations 'Emerald' TC had more flower buds per plant than SW plants during the first season of reproductive growth. During the second reproductive season, TC resulted in increased flower buds per plant in 'Emerald' and 'Jewel', with no significant effect of propagation type on 'Primadonna'. Although TC plants of these two cultivars had greater flower flower bud number the seco nd growing season, propagation effect on fruit yield was not significant for any of the cultivars.

PAGE 102

102 Micropropagation resulted in significantly greater yield for 'Emerald' and 'Jewel' plants, without any effect of propagation type on yield for 'Primadonna', at either one of the locations. A positive correlation between yield and total number of shoots was observed for 'Emerald' and 'Jewel' in Citra and Haines City. R educed average berry weight was recorded for TC plants compared to SW plants early in the se ason in Haines City but not in Citra. At the beginning of May, 'Emerald' TC plants had reduced berry weight compared to SW plants at both locations without a significant effect of propagation method the rest of the season In Citra, greater soluble solids content in fruit from TC 'Jewel' and 'Primadonna' were observed at mid or late season in the first harvest. In addition, increased titratable acidity was measured in fruit from 'Jewel' TC compared to SW plants.

PAGE 103

103 APPENDIX ADDITIONAL TABLES Table A 1. P values from ANOVA for plant volume in 2010 Source June July August September October November REP(LOC) 0.9742 0.5515 0.7495 0.9378 0.7602 0.0479 CV <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 PROP <.0001 0.0004 0.8302 0.1675 0.1326 0.0035 CV*PROP <.0001 <.0001 <.0001 0.0001 0.0007 0.0004 LOC <.0001 <.0001 0.0014 <.0001 <.0001 <.0001 LOC*CV 0.0326 0.0307 0.0011 0.0007 0.0001 0.0472 LOC*PROP 0.2257 0.3776 0.4233 0.1458 0.0754 0.8197 LOC*CV*PROP 0.3728 0.3837 0.6242 0.679 0 0.5022 0.4143 Table A 2. P values from ANOVA for plant volume in 2011 Source June July August September October November REP(LOC) <.0001 <.0001 0.0037 0.07 00 0.0072 0.0613 CV 0.1393 0.0042 0.0015 0.0003 <.0001 <.0001 PROP 0.0002 0.0034 0.0046 0.0243 0.0496 0.0644 CV*PROP 0.0008 0.0015 0.0112 0.0281 0.0195 0.4404 LOC <.0001 0.0005 0.5407 0.0065 0.0002 0.0579 LOC*CV 0.0174 0.054 0 0.0038 0.0099 0.0047 0.9653 LOC*PROP 0.1858 0.2239 0.7933 0.8702 0.8415 0.7164 LOC*CV*PROP 0.0051 0.02 00 0.0317 0.1472 0.0678 0.8799 Table A 3 Effect of cultivar and location on flower bud number in 2011. Means are averages of propagation methods. Location Cultivar Total FB Citra E 1135.9 a J 1039.4 a P 1023.3 a Haines City E 1158.2 a J 1474.9 a P 458.5 b z Means followed by same letters within a column indicate no significant differences, Tukey's HSD test,

PAGE 104

104 Table A 4 Effect of cultivar and location on percentage of full bloom on 4 March (Citra) and 14 February ( Haines City ) Means of propagation methods. Location Cultivar Percentage of full bloom Citra E 66.8 a J 69.4 a P 54.8 b Haines City E 50.8 bc J 45.3 c P 48.3 bc z Means followed by same letters within a column indicate no significant differences, Tukey's HSD test, Table A 5. Effect of cultivar and location on f ruit yield (g) in 2011. Means of propagation methods. Location Cultivar Yield (g) Citra E 2438.5 a J 2221.6 a P 1487.8 bc Haines City E 1171.6 bc J 1587.5 b P 1046.9 c z Means followed by same letters within a column indicate no significant differences, Tukey's HSD test, Table A 6 Effect of cultivar on f ruit yield (g) in 2011. Means of both locations and propagation methods. Cultivar YIELD E 1805.1 a J 1904.5 a P 1267.3 b

PAGE 105

105 LIST OF REFERENCES Abolins ,M. and L. Gurtaja. 2006. Vaccinium spp. production techniques in Latvia. Proc. of VIII Int. symp. on Vaccinium culture Acta Hort. 715:185 190. Albert, T., M. Starast, K. Karp, H. Kaldmae, E. Vool and T. Paal. 2009. The influence of propagation method on growth of the half Acta Hort (ISHS) 812:141 146. Anderson, W. C. (1980). Tissue culture propagation of red and black raspberries, Rubus idaeus and Rubus oc cidentalus Acta Hort 112: 30 31. Araujo, A.P., A.M. Fernandes, F.Y.Kubota, F.Costa Brasil and M. G. Teixeira. 2004. Sample size for measurement of root traits on common bean by image analysis. Pesq Agropec Bras 39: 313 318. Beyl, C.A. 2005. Gettin g started with tissue cultur e: M edia preparation, sterile technique, and laboratory equipment, p. 19 37. In : R.N. Trigiano and D.J. Gray (eds). Plant development and biotechnology. CRC Press, Boca Raton. Brissette, L., L. Tremblay, and D. Lord. 1990. Micropropagation of lowbush blueberry from mature field grown plants. HortScience 25:349 351. Chaplin, M.F. and J.F. Kennedy. 1994. Carbohydrate analysis: A practical approach. 2 nd ed. IRL Press, Oxford. Cline, W. and Schilder, A. 2006. Identification and Control of blueberry Diseas es, p. 115 138. In: N.F. Childers and P.M. Lyrene (eds.). Blueberries for growers, gardeners and promoters. N.F. Childers Horticultural Publications, Gainesville, Fla. Coville, F. V. 1937. Improving the wild blueberry p 559 574. In 1937 yearbook of a gric ulture. U.S. Gvt. Printing Office, Washington, D.C. Costa, C., Dwyer, M. L., Hamilton, R. I., Hamel, C., Nantais, L. and Smith, D. L. 2000. A sampling method for measurement of large root systems with scanner based image analysis. Agron. J. 92 :621 627 Crocker, T. E. and L. Willis. 1989. Survey of southern highbush and rabitteye blueberries in Florida. Proc. Fla. State Hort. Soc. 102:204 206. Darnell, R. L. 1991. Photoperiod, carbon partitioning, and reproductive development in rabbiteye blueberry. J. A mer. Soc. Hort. Sci. 116:856 860. Darnell, R. L. and K. B. Birkhold. 1996. Carbohydrate contribution to fruit development in two phenologically distinct rabbiteye blueberry cultivars. J. Amer. Soc. Hort. Sci. 121:1132 1136.

PAGE 106

106 Debnath S. C. 2005. Morphologica l development of lingonberry as affected by i n vitro and ex vitro propagation methods and sour ce propagule. HortScience, 40 : 760 763. Debnath, S.C. 2007. Propagation of Vaccinium in vitro. I nterna tional Journal of Fruit Science 6 :47 71. Debnath, S.C. 2007 Influence of indole 3 butyric acid and propagation method on growth and development of in vitro and ex vitro derived lowbush blueberry plants. Plant Growth Regul. 51: 245 253. Debnath, S. C. and K. B. Mc Rae. 2001. An efficient in vitro shoot propagati on of cranberry ( Vaccinium macrocarpon Ait.) by axillary bud proliferation. In Vitro Cell. Dev. Biol. Plant 37:243 249. Dweikat, M. and P.M. Lyrene. 1988. Adventitious shoot production from leaves of blueberry cultured in vitro HortScience 23:629. Donnelly, D.J. and W. E. Vidaver. 1988. G losary of plant tissue culture. In: Advances in plant science series Dioscorides Press. El Shiekh, A., D.K. Wildung, J.J. Luby, K.L. Sargent, and P.E. Read. 1996. Long term effects of propagation by tissue culture or softwood single node cuttings on growth 121:339 342. Frett, J. and Smagula, M. 1983. In vitro shoot production of lowbush blueberry. Can. J. Plant Sci. 63:467 472. Geis ler, M. 2012. Blueberries profile. Agricultural marketing resou rce center. University of Iowa. 2 October 2012. < http://www.agmrc.org/commodities__products/ fruits/blueberr ies profile/ > Grout, J.M., P.E. Read and D. K. Wildung. 1986. Influence of tissue culture and leaf bud propagation on the growth habit of 'Northblue' blueberry. J.Amer.Soc.Hort.Sci. 111: 372 375. derived by micropropagation vs. stem cuttings. HortScience 35 :742 744. Hartmann, H.T. and D.E. K ester. 1975 Techniques of propagation by cuttings, p. 271 313. In: Plant propagation : principles and practices. Prentice Hall, Englewood Cli ffs, N.J. House, L. and Wysocki, A. 2012. Blueberry Consumption. The Vegetarian Newsletter, Issue 574. University of Florida, Horticultural Sciences Department. 2 October 2012. < http://ho s.ufl.edu/newsletters/vegetarian/issue no 574 >

PAGE 107

107 Isutsa, D. K. and Pritts, M. P. 1994. Rapid propagation of blueberry plants using ex vitro rooting and controlled acclimatization of mic ropropagules. HortScience 29 :1124 1126. Jaakola, L., A. Tolvanen, K. Laine and A. Hohtola. 2002. Micropropagation of bilberry and lingonberry. (ISHS) Acta Hort. 574:401 403. Jamieson, A. R. and Nickerson, N.L. 2003. Field performance of lowbush blueberry propagated by seed, stem cuttings and micropropagation. Acta Hort. 6 26:423 428. Koch, K.E. 1996. Carbohydrate modulated gene expression in plants. Annual. Rev. Plant. Physio. Plant. Mol. Biol. 47:509 540. Krewer G. and B. Cline. 2003. Blueberry propagation s uggestions. 14 October 2012. < http://www.smallfruits.org/Blueberries/production/03BlueberryPropagation Suggestions.pdf > Littell, R.C., W. W. Stroup and R. J. Freund. 2002. Sas for linear models. SAS Institute. 14 Octob er 2012. < http://books.google.com/books?id=JhxKC8ewcdAC& dq=slicing+proc+glm&source=gbs_navlinks_s > Litwiczuk, W., Szczerba, G. and Wrona. D. 2005. Fiel d performance of highbush blueberries ( Vaccinium corymbosum tissue cu lture. Scientia Horticulturae 106: 162 169. Loescher, W., McCamant, T. and Keller, J.D. 1990. Carbohydrate reserves, translocation, and stora ge in woody plant roots. HortScience 25 : 274 281. Luby, J.J., Wildung, D.K., Stushnoff, C., Munson, S. T., Read, P.E. and Hoover, E. E. : 1240 1242. Lyrene, P.M. 1978. Blueberry callus and shoot tip culture. Proc. Fla. State Hort. Soc. 91:171 172. Lyrene, P.M. 1980. Micropropagation of rabbiteye blueberries. HortScience 15:80 81. Lyrene, P.M. 1998. Ralph Sharpe and the Florida blueberry breeding program. Proc. 8th North Amer. Blueberry Res. and Ext. Worker Conf. 1998. Lyrene, P. M. and Moore, J. N. 2006. Blueberry breeding, p. 38 47. In: N.F. Childers and P.M. Lyrene (eds.). Blueberries for growers, gardeners and promoters. N.F. Childers Horticultural Publications, Gainesvill e, Fla. Mainland, C. M. 2006. Propagation of Blueberries, p. 49 58. In: N.F. Childers and P.M. Lyrene (eds.). Blueberries: For growers, gardeners and promoters. N.F. Childers Horticultural Publications, Gainesville, Fla.

PAGE 108

108 Miller, S.A., Rawnsley, E.K., Geo rge, J. and Patel, N. 2006. A comparison of Blueberry Propagation techniques used in New Zealand. Acta Hort. (ISHS) 715:397 402 Moore, J.N. 1965. Improving highbush blueberries by breeding and selection. Euphytica 14: 39 48 Morrison, S., J.M. Smagula, an d W. Litten. 2000. Morphology, growth, and rhizome development of Vaccinium angustifolium Ait. seedlings, rooted softwood cuttings, and micropropaga ted plantlets. HortScience 35 :738 741. Nickerson, N.L. 1978. In vitro shoot formation in lowbush blueberry s eedling explants. HortScience 13:698. Read, P.E., C.A. Hartley, J.G. Sandahl, and D.K. Wildung. 1988. Field performance of in vitro propagated blueberries. Proc. Intl. Plant Prop. Soc. 37:450 452. Read, P.E., D.K. Wildung, and C.A. Hartley. 1989. Field per formance of in vitro 194. Reed B. M., Abdelnour Esquivel A. (1991). The use of zeatin to initiate in vitro cultures of V accinium species and cultivars. HortScience 26: 1320 1322. Rowland, L.J. and E.I Ogden. 1992. Use of a cytokinin conjugate for efficient shoot regeneration from leaf sections of highbush blueberry. HortScience 27:1127 1129. Sedgley, M. and A.R. Griffin. 1989. Floral initiation and development, p. 11 51. In: Sexual reproduction of tre e crops, Serres, R., J. Klueh, and E. Stang. 1993. Influence of source propagule on rhizome production from lingonbe rry cuttings. Acta Hort. 346:178 182. Serres, R. and B. McCown. 1994. Rapid flowering of microcultured cranberry plants. HortScience 29:159 161. Serres R. A., Pan S., McCown B. H., Stang E. J. (1994). Micropropagation of several lingonberry cultivars. Fruit Varieties Journal 48: 7 14. Smagula, J.M. and J. Harker. 1997. Cranberry micropropagation using a lowbush blueberry medium. Acta Horticult urae 44:343 347. Smolarz, K. and D. Chlebowska. 1997 Growth vigour and yielding of highbush blueberry cv. Bluecrop propagated from semi woody cuttings and in vitro. J. Fruit Ornment. Fruit Res.: 53 60. Spann, T.M., J.G.Williamson, and R.L. Darnell. 2003. Photoperiodic effects on vegetative and reproduct ive growth of Vaccinium darrowi and V. corymbosym interspecific hybrids. HortScience 38:192 195.

PAGE 109

109 Spann, T.M., J.G.Williamson, and R.L. Darnell. 2004. Photoperiod effects on growth and carbohydrate storage in southern highbush blueberry interspecific hybri d. J. Amer. Soc. Hort. Sci. 129 : 294 298. Spiers, J.M. 1978. The effect of stage of bud development on cold injury in rabbiteye blueberry. J. Amer. Hort. Sci. 103:452 455. Swartz, H.J., G.J. Galletta and R. H. Zimmerman. 1983. Field performance and phenoty pic stability of tissue culture propagated thornless blackberries. J. Amer. Soc. Hort. Sci. 108:285 290. USDA NASS. 2011. Noncitrus Fruits and Nuts 2010 Preliminary Summary. United States Department of Agriculture. National Agricultural Statistics Service. Cornell Univ. 25 February 2011. < http://usda.mannlib.cornell.edu/MannUsda/viewDocumentInfo.do?documentID=1 765 > USDA NASS. 2012. Noncitrus Fruits and Nuts 2011 Summa ry. United States Department of Agriculture. National Agricultural Statistics Service. Cornell Univ. 31 October 2012. < http://usda.mannlib.cornell.edu/MannUsda/viewD ocumentInfo.do?documentID=1 113 > USDA NRC. 2012. Soil survey of Marion Country area, Florida. 14 October 2012. < http://soildatamart.nrcs.usda.gov/manuscripts/FL608/0/Marion.pdf > Valenzuela Estrada, L.R., V. Vera Caraballo, Ruth, L.E. and D. M. Eissenstat. 2008. Amer. Journal of botany 95: 1506 1514. Valero Aracama, C., M.E. Kane, S.B. Wilson and N.L. Philman. 2010. Substitution of benzyladenine with meta topolin during shoot multip lication increases acclimatization of difficult and easy t o acclima tize sea oats ( Uniola paniculata L.) gen otypes. Plant Growth Regul. 60: 43 49. Vander Kloet, S.P. 1988. The Genus Vaccinium in North America. Res. Branch Agric. Can. Publ. 1828. Res. Branch, Agr. Canada, Canadian Govt. Publ. Centre, Ottawa. Williamson, J. G., P. M. Lyrene and E. P. Miller. 2000. A survey of blueberry acreage in Florida. Proc. Fla. State Hort. Soc. 113:24 25 Williamson, J.G. and P. M. Lyrene. 2004a. Blueberry var ieties for Florida. University of Florida, Gainesville, FL. 31 October 21012.< https://edis.ifas. ufl.edu> Williamson, J. G. and P. M. Lyrene. 2004b. T he Florida blueberry industry: A decade of growth. Proc. Fla. State Hort. Soc. 117: 234 235.

PAGE 110

110 Williamson, J.G., J. W. Olmstead and P. M. Lyrene. 2012. Florida's commercial blueberry industry. University of Florida, Gainesville, FL. 31 October 21012. < http://edis.ifas.ufl. edu > Williamson, J .G. and J.H. Crane. 2010. Best management practices for temperate and tropical/subtr opical fruit crops in Florida: C urrent practices and future c hallenges. HortTechnology 20 :111 119. Wolfe, D.E., P. Eck, and C. Chin. 1983. Evaluation of s even media for micropropagation of highbush blueberry. HortScience 18:703 705. Zimmerman, R.H. 1987.Mic ropropagation of woody plants: P ost tissue culture aspects. Acta Hort. 227:489 499.

PAGE 111

111 BIOGRAPHICAL SKETCH Silvia Rosa Marino was born in Buenos Aires, Argentina. She is the youngest daughter of Rosa Perez and Alberto Oscar Marino. She graduated from the Faculty of Agronomy (University of Buenos Aires) with a degree of a gricultural e ngineer. Silvia came to Unit ed States in 2001 and was living in South Florida when she decided to return to sch ool, in 2008. She started working with Dr. Jeffrey Williamson and Dr. James Olmstead towards her m aster of s cience at the Department of H orticultural S ciences at the Univers ity of Florida in August 2009 Silvia wants to continue working with fruit crops.