Field Evaluation of Rootstocks in the Presence of Huanglongbing

MISSING IMAGE

Material Information

Title:
Field Evaluation of Rootstocks in the Presence of Huanglongbing
Physical Description:
1 online resource (96 p.)
Language:
english
Creator:
Das, Sanghamitra
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Horticultural Sciences
Committee Chair:
Gmitter, Frederick G, Jr
Committee Co-Chair:
Grosser, Jude William

Subjects

Subjects / Keywords:
allotetraploid -- fruitdrop -- greening -- hlb -- huanglongbing -- rootstock
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:
Huanglongbing (HLB or citrus greening) is a serious threat to the citrus industry worldwide. Managing groves where greening exists increases costs for scouting, additional spray applications, tree removal and replanting, nutritional programs etc. All citrus cultivars are susceptible though there is some variation in the degree of susceptibility. The objective of the research was to evaluate different rootstocks, (conventional diploids, somatic hybrids, and allotetraploid sexually derived hybrids) in the field to see if any difference in tolerance may be observed. At McTeer farm, fifteen such rootstocks were evaluated with ‘Sugar Belle’ mandarin hybrid as the scion. Nineteen rootstocks were evaluated at St. Helena west and seven at St. Helena east with the scion being either Vernia or Valquarious. The trees were visually rated for symptom expression on a scale of 0 to 4, and qPCR was used to generate Ct values from symptomatic and asymptomatic leaves in summer and fall 2012. In addition, observations were made on fruit drop. The least severe symptoms were observed on Orange 19 (an allotetraploid hybrid resulting from the cross of Nova mandarin + Hirado Buntan seedling pummelo X Cleopatra mandarin + Argentine trifoliate orange) Orange 19 had least severe symptoms at McTeer as well as at St. Helena. CLEO+CZO had less severe symptoms at St. Helena East. These preliminary results suggest differences among the rootstocks in tree response to HLB, but further work is necessary to validate these conclusions over time.
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 Sanghamitra Das.
Thesis:
Thesis (M.S.)--University of Florida, 2013.
Local:
Adviser: Gmitter, Frederick G, Jr.
Local:
Co-adviser: Grosser, Jude William.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-02-28

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0046018:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 FIELD EVALUATION OF ROOTSTOCKS IN THE PRESENCE OF HUANGLO N GBIN G By SANGHAMITRA DAS A THESIS P RESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR TH E DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Sanghamitra Das

PAGE 3

3 To my parents

PAGE 4

4 ACKNOWLEDG E MENTS I would like to express my sincere thanks and gratitude to Dr. Jude W. Grosser and Dr. Frederick G. Gmitter for giving me this opportunity to work on this project. It was a great learning experience for me. I am also extremely grateful to Dr. Ron H. Brlansky, Dr. Megh Singh for their academic support and valuable advice. My sincere thanks to CRDF for fun ding the project. I am extremely grateful to US Sugar Corporation, Mike Irey, Paula Gadea and her staff for helping me with DNA extraction and Ct value determination. My sincere thanks to all lab mates including postdoctoral research associates, graduate s tudents and OPS employees. Special thanks to Julie Gmitter, Eric Ramjith, Qibin Yu, Gary Barthe for their support in field work. I am grateful to the entire CREC community for their support and friendship. I would also like to thank my family and friends i n India.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGEMENTS ................................ ................................ ................................ ............. 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 9 ABSTRACT ................................ ................................ ................................ ................................ ... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 14 2 LITERATURE REVIEW ................................ ................................ ................................ ....... 15 Huanglon gbing, The Yellow Shoot Disease of Citrus ................................ ............................ 15 Origin ................................ ................................ ................................ ................................ ...... 15 Causal Organism ................................ ................................ ................................ ..................... 16 Symptoms ................................ ................................ ................................ ............................... 17 Transmission ................................ ................................ ................................ ........................... 18 Measures Against the Disease ................................ ................................ ................................ 18 Rootstock Breeding ................................ ................................ ................................ ................ 19 HLB Tolerance ................................ ................................ ................................ ....................... 22 Disease Analysis ................................ ................................ ................................ ..................... 24 Objectives ................................ ................................ ................................ ............................... 26 3 MATERIALS AND METHODS ................................ ................................ ........................... 27 McTeer Farm ................................ ................................ ................................ .......................... 27 St. Helena Farm ................................ ................................ ................................ ...................... 27 Visual Obser vations on Study Trees ................................ ................................ ....................... 28 qPCR and Ct Value Determination ................................ ................................ ......................... 29 DNA extraction ................................ ................................ ................................ ....................... 29 Robot DNA extraction ................................ ................................ ................................ ..... 30 Materials ................................ ................................ ................................ .......................... 30 Methods ................................ ................................ ................................ ........................... 30 Experiment 1: DNA extraction from plant tissue ................................ ............................ 30 Yield Estimation ................................ ................................ ................................ ..................... 31 Fruit Drop Estimation ................................ ................................ ................................ ............. 31 4 MCTEER FARM ................................ ................................ ................................ .................... 37 Results ................................ ................................ ................................ ................................ ..... 37 Visual symptoms ................................ ................................ ................................ .................... 37

PAGE 6

6 PCR Analysis The Ct values ................................ ................................ ................................ 38 HLB Symptomatic leaves ................................ ................................ ................................ ....... 38 Asymptomatic Leaves ................................ ................................ ................................ ............ 39 Biochar Treatment ................................ ................................ ................................ .................. 39 Estimated Fruit Yield and HLB Induced Fruit Drop ................................ .............................. 40 5 ST. HELENA EAST ................................ ................................ ................................ ............... 57 Results ................................ ................................ ................................ ................................ ..... 57 Visual Symptoms (subjective rating scale) ................................ ................................ ............. 57 Mean Ct Values from Symptomatic Leaves ................................ ................................ ........... 58 A comparison of Summer and Fall mean Ct values from symptomatic leaves ...................... 58 Mean Ct values from asymptomatic leaves ................................ ................................ ............ 59 Estimated Fruit Yield and HLB Induced Fruit Drop ................................ .............................. 59 Conclusion ................................ ................................ ................................ .............................. 59 6 ST. HELENA WEST ................................ ................................ ................................ .............. 72 Results ................................ ................................ ................................ ................................ ..... 72 Visual Symptom Assessment ................................ ................................ ................................ 72 The mean Ct values from symptomatic leaves ................................ ................................ ....... 73 Mean Ct values from asymptomatic leaves ................................ ................................ ............ 73 Conclusions ................................ ................................ ................................ ............................. 74 7 CONCLUSION ................................ ................................ ................................ ....................... 89 LIST OF REFERENCES ................................ ................................ ................................ ............... 91 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 9 6

PAGE 7

7 LIST OF TABLES Table page 3 1 Brief description of rootstocks ................................ ................................ .......................... 32 3 2 Rootstocks e valuated at the McTeer trial ................................ ................................ ......... 34 3 3 Rootstocks evaluated at the St. Helena Trial West Section ................................ ........... 34 3 4 Rootstocks evaluated at the St. Helena trial East section ................................ ............... 34 4 1 Mean of visual HLB symptoms at the McTeer trial ................................ ........................ 43 4 2 Results of Wilcoxon sign r ank test ................................ ................................ ................... 45 4 3 Mean Ct values from HLB symptomatic leaves from Sugar Belle trees per rootstock .... 47 4 4 Showing mean Ct values from asymptomatic leaves sampled from Sugar Belle trees on different rootstocks ................................ ................................ ................................ ....... 50 4 5 Comparison of visual symptom means of biochar treated and untreated trees .................. 51 4 6 Comparison of Ct value means from symptomatic leaves of Sugar Belle trees on selected rootstocks treated or untreated with biochar. ................................ ....................... 51 4 7 Comparison of Ct value means from asymptomatic leaves of trees treated with biochar and those untreated ................................ ................................ ............................... 52 4 8 Means of Sugar Belle fruit drop per rootstock in the McTeer Trial ................................ 53 4 9 ................................ ................................ .............................. 55 4 10 ANOVA for fruit drop from Sugar Belle trees ................................ ............................... 55 5 1 Mean visual HLB symptom assessment per rootstock for sweet orange trees in the East section of the St. Helena trial ................................ ................................ ..................... 61 5 2 Result of Wilcoxon signed rank test to compare summer and f all visual mean data. ...... 62 5 3 Me an Ct values from symptomatic leaves per rootstock ................................ .................. 63 5 4 Mean Ct values from asymptomatic leaves per rootstock in the east secti on of the St. Helena trial. ................................ ................................ ................................ ........................ 66 5 5 Mean fruit drop from HLB infected sweet orange trees on various rootstocks in the east section of St. Helena ................................ ................................ ................................ 68 5 6 ................................ ................................ ...... 69

PAGE 8

8 5 7 ANOVA table for the fruit drop data in the east section of the St. Helena trial. .............. 69 5 8 Early HLB infection percentages per rootstock at St. Helena East ................................ .. 70 5 9 Yield and fruit quality data from sweet orange trees on various rootstocks in the east section of the St. Helena trial ................................ ................................ ............................. 71 6 1 Means of visual symptom severity in HLB affected sweet orange trees on various rootstocks in the west section of the St. Helena trial ................................ ......................... 76 6 2 Mean of Ct values from HLB symptomatic leaves from HLB infected sweet orange trees on various rootstocks in the western section of the St. Helena trial .......................... 78 6 3 Mean Ct values from asymptomatic leaves sampled from HLB affected sweet orange trees on various rootstocks ................................ ................................ ................................ 81 6 4 Mean fruit drop for sweet orange trees on different rootstocks in the western section of the St. Helena trial ................................ ................................ ................................ ......... 85 6 5 ....................... 87 6 6 Table showing percentage of infection at St. Helena West ................................ .............. 88

PAGE 9

9 LIST OF FIGURES Figure page 3 1 Example of a Grade 1 tree ................................ ................................ ................................ 35 3 2 Example of a Grade 2 tree ................................ ................................ ................................ 35 3 3 Example of a Grade 3 tree ................................ ................................ ................................ 35 3 4 Example of a Grade 4 tree ................................ ................................ ................................ 36 3 5 Tree with high yield ................................ ................................ ................................ .......... 36 3 6 Tree with medium yield ................................ ................................ ................................ .... 36 3 7 Tree with low yield ................................ ................................ ................................ ........... 36 4 1 Boxplot of visual HLB symptoms on Sugar Belle trees on various rootstocks at the McTeer Trial ................................ ................................ ................................ ...................... 43 4 2 Boxplot of visual HLB symptoms of Sugar Belle trees on various rootstocks at the McTeer Farm, November 2012 ................................ ................................ .......................... 44 4 3 A comparison of the means of visual HLB symptoms of Sugar Belle trees on various roostock for the summer and f all ................................ ................................ ...................... 44 4 4 Checking assumption of normality from HLB symptomatic leaves in summer .............. 45 4 5 Histogram checking assumption of normality from HLB symptomatic leaves in summer ................................ ................................ ................................ ............................... 45 4 6 Boxplot of Ct values from HLB symptomatic leaves from Sugar Belle trees on various roo tstocks ................................ ................................ ................................ .............. 46 4 7 Boxplot of Ct value from symptomatic leaves from Sugar Belle trees on different rootstocks, determined in f all ................................ ................................ ............................. 47 4 8 Comparison of the mean of Ct value from symptomatic leaf samples from Sugar Belle trees on different rootstocks in summe r and f all ................................ .................... 48 4 9 Checking assumption of normality from asymptomatic leaves in summer ..................... 48 4 10 Histogram to check normality from asymptomatic leaves in summer ............................ 49 4 11 Boxplot of Ct value from asymptomatic leaves sampled from Sugar Belle trees on different rootstocks in summer ................................ ................................ .......................... 49

PAGE 10

10 4 12 Boxplot of Ct value from asymptomatic leaves sampled from Sugar Belle trees on different rootstocks in f all ................................ ................................ ................................ .. 50 4 13 Comparison of mean of Ct values from asymptomatic leaves sampled in summer and f all ................................ ................................ ................................ ............................... 51 4 14 Visually estimated yield of Sugar Belle trees on different rootstocks .............................. 52 4 15 Example of a Sugar Belle tree rated as high yield tree ................................ ..................... 52 4 16 Example of a Sugar Belle tr ee rated as medium yield tree ................................ ............... 53 4 17 Example of a Sugar Belle tree rated as low yield tree ................................ ...................... 53 4 18 Boxplot of Sugar Belle fruit drop on different rootstocks in the McTeer trial. ................ 54 4 19 Normality check for untransformed data of fruit drop ................................ ...................... 54 4 20 Normality check for log transformed data of fruit drop ................................ ................... 55 4 21 Figure showing the relationship between fruit drop and yield ................................ ......... 56 5 1 Boxplot of visual symptom per rootstock from HLB affected s weet orange trees in the East section of the St. Helena trial, data from summer assessment ............................. 61 5 2 Boxplot of visual symptom means per rootstock from HLB infected sweet orange trees in the East section of the St. Helena trial, data from f all assessment ........................ 62 5 3 A comparison of visual HLB symptom means per rootstock of sweet orange trees in the East section of the St. Helena trial, comparing summer and f all data ......................... 63 5 4 Boxplot of mean Ct values from HLB symptomatic leaves of sweet orange trees per rootstock, sampled in summer ................................ ................................ ........................... 64 5 5 Boxplot of mean Ct values from HLB symptomatic leaves of sweet orange trees per rootstock, sampled in f all ................................ ................................ ................................ ... 65 5 6 Comparison of mean Ct values from symptomatic leaves of sweet orange trees per rootstock in the east section of the St. Helena trial, summer versus f all data .................... 65 5 7 Boxplot of mean Ct values from HLB asymptomatic leaves of sweet orange trees per rootstock, sampled in summer ................................ ................................ ........................... 66 5 8 Boxplot of mean Ct values from HLB asymptomatic leaves of sweet orange trees per rootstock, sampled in f all ................................ ................................ ................................ ... 67 5 9 Visually estimated yield of HLB affected sweet orange trees on different rootstocks in the east section of the St. Helena trial ................................ ................................ ............ 67

PAGE 11

11 5 10 Boxplot of mean fruit drop for HLB infected sweet orange trees on different rootstocks in the east section of the St. Helena trial. ................................ ......................... 68 6 1 Boxplot of mean visual symptoms per rootstock for HLB affected sweet orange trees in the west section of the St. He lena trial, summer data ................................ .................... 77 6 2 Boxplot of visual HLB symptom means of sweet orange trees on different rootstocks in the Western section of the St. Helena trial; data collected in f all ................................ .. 77 6 3 Comparison of visual HLB symptom means of sweet orange trees on various rootstocks in summer versus f all ................................ ................................ ........................ 78 6 4 Mean Ct values from HLB symptomatic leaves sampled from HLB affected sweet orange trees on various rootstocks in the western section 79 6 5 Ct value from symptomatic leaves during f all (October, 2012) ................................ ....... 80 6 6 Comparison of Summer and Fall mean Ct values from HLB symptomatic leaves s in the western section of the St. Helena trial. ................................ ................................ ......... 80 6 7 Boxplot showing mean Ct values from asymptomatic leaves in s ummer ....................... 82 6 8 Comparison of mean Ct values from s ummer versus f all asymptomatic leaves in the western section of the St. Helena trial. ................................ ................................ .............. 82 6 9 Boxplot showing Ct value from asymptomatic leaves in f all ................................ .......... 83 6 10 Boxplot showing mean fruit drop data for HLB infected sweet orange trees in the western section of the St. Helena trial. ................................ ................................ .............. 84 6 11 Visual estimate of yield from HLB affected sweet orange trees per rootstock in the western section of the St. Helena trial. ................................ ................................ .............. 85 6 12 St. Helena West fruit drop data without log transformation ................................ ............. 86 6 13 St. Helena West fruit drop data with log transformation ................................ .................. 86 6 14 Mean fruit drop and yield at St. Helena West ................................ ................................ ... 87

PAGE 12

12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for th e Degree of Master of Science FIELD EVALUATION OF ROOTSTOCKS IN THE PRESENCE OF HUANGLONGBING By Sanghamitra Das August 2013 Chair: Fred erick G. Gmitter Cochair: Jude W. Grosser Major: Horticultural Sciences Huanglongbing (HLB or citrus greening) is a serious threat to the citrus industry worldwide. Managing groves where greening exists increases costs for sco uting, additional spray applications, tree removal and replanting, nutritional programs etc. All citrus cultivars are susceptible though there is some variation in the degree of susceptibility. The objective of the research was to evaluate different rootst ocks, (conventional diploids, somatic hybrids, and allotetraploid sexually derived hybrids) in the field to see if any difference in tolerance may be hybrid as the scion. Nineteen rootstocks were evaluated at St. Helena west and seven at St. Helena east with the scion being either Vernia or Valquarius. The trees were visually rated for symptom expression on a scale of 0 to 4, and qPCR was used to generate Ct values from symptomatic and asymptomatic leaves in s ummer and f all 2012. In addition, observations were made on fruit drop. The least severe symptoms were observed on Orange 19 (an allotetraploid hybrid resulting from the cross of Nova mandarin + Hirado Buntan se edling pummelo X Cleopatra mandarin + Argentine trifoliate orange) Orange 19 had least severe symptoms at McTeer as well as at St. Helena. CLEO+CZO had less severe symptoms at St. Helena East.

PAGE 13

13 These preliminary results suggest differences among the rootst ocks in tree response to HLB, but further work is necessary to validate these conclusions over time.

PAGE 14

14 CHAPTER 1 INTRODUCTION Huanglongbing has been a devastating disease for the citrus industry. With no known cure, management of the disease has been difficult and expensive. Citrus being grown as a combination plant, the rootstock influences the scion in different ways. With the objective in mind, how genetically different rootstocks, have an influence on the expression of disease, particularly visual symptoms, the Ct value to estimate the bacterial titer and the fruit drop (as HLB affected trees show considerable fruit drop), two trials in ce ntral Florida were studied in the span of a year during summer and fall of 2012. One of the trial s was at the McTeer farm, near Haines C ity in Florida and the other was at St. Helena farm located east of Dundee in Florida. The scion s used in the study were Sugar Belle, at McTeer and Vernia and Valqua r ius at the St. Helena farm. At McTeer the infection was almost one hundred cent percent while at St. Helena the affected trees were a small er percentage of the total number of trees. This study examined only those tress that had been already diagnosed with HLB. The rootstocks were genetica lly different, either diploids, allotet raploid somatic hybrids, or between somatic hybrids). The visual observations and Ct values wer e noted both in summer and fall of 2012. The visual estimation of yield and the fruit drop was also noted. As expected some difference in tree response to HLB emerged from the study.

PAGE 15

15 CHAPTER 2 LITERATURE REVIEW Huanglongbing, The Yellow Shoot Disease of Citrus Florida is the second largest producer of orange juice next to Brazil. Thousands of jobs are created by this agricultural commodity. The total economic impact to the state of Florida is nearly $8.91 billion, and the industry generates more than 75,000 jobs (Hodges and Spreen, 2012) The most important citrus crop in Florida is sweet orange most of which are processed by the juice industry. In recent years however, the disease Huanglongbing (HLB) or citrus greening has had a devastating effect on the industry. Control of the disease increases productio n cost s to a great extent O ne of the most economical ways to manage the disease would be through genetic resistance or tolerance The objective of this proposal is to look at the impacts that genetically distinct rootstocks have on the tolerance of grafted trees to HLB. Huanglongbing is known by many different names, including citrus greening, yellow shoot disease the (Bov, 2006) The most extensive research on HLB in China was carried out by Lin Kung Hsiang from 1941 to 1955 (Bov, 2006) Origin T he origin of HLB may have been in China or India. Beattie and Holford ( 2008) suggests that it possibly spread from V lan ceolata to orange or mandarin trees in south east Africa, then spread to the Indian subcontinent, and then spread from there by Diaphorina citri This review also states that the severity of symptoms depend on the Candidatus L iberibacter form, availability of flush during psyllid activity and presence and severity of CTV strains (Beattie and Holford, 2008) Beattie and Holford (2008) defined sweet orange, mandarin and

PAGE 16

16 grapefruit as highly sensitive, whereas true lemon, rough lemon and trifoliate orange and its hybrids are more tolerant ; t he lime group is considered highly tolerant The disease was first discovered in Brazil in July 2004 (Halbert and Manjunath, 2004) and in Florida it was found in two samples of pummelo in August 2005 in Homestead and Florida City. Causal Organism Saglio, Lafleche and Bove (Saglio et al., 1973) first identified the HLB causal organism from infected orange leaves from So uth A frica, India and Reunion Island, as the phloem limited bact e ria. Candidatus Liberibacter asiaticus ( C Las) which is the form found in Florida (Garnier et al ., 2000) T here are three different species Ca L. asiaticus, Ca L. africanus and Ca L. americ anus which are distinguished by difference s in the sequence of the r operon and environmental tolerances. The first one is heat tolerant, spread by the Asian ci trus psyllid, Diaphorina citri (Capoor et al., 1967) and found in A sia, Indian Ocean islands and now in the America s In Africa, The Ca L. africanus is found at elevations above 700m and is considered heat sensitive. The bacteria are spread by Tryoza erytreae (McClean and Oberholzer, 1965) The third type is found in Brazil (Teixeira et al., 2005) That t he bacteria are Gram negative, was first esta blished in 1984 by the fact that they are sensitive to penicillin (Garnier et al., 1984) They belong to the associated with eukaryotic cells and thrive within arthropod vectors (Jagoueix et al., 1994) The the genome has been sequenced and is of about 1.22 Mbp (Duan et al., 2009) The disease is transmissible by nat ure and it has been experimentally shown that the bacteria can be transmitted from infectious citrus plants to periwinkle through dodder (Garnier and Bove, 1983)

PAGE 17

17 Symptoms The leaf symptoms of the disease include blotchy mottle, yellow and corky vein, green islands, and fruit frequently exhibit a curved cent ral column distorting the shape, aborted seed, and yellow stain in calyx. In later stages, corky veins and leathery lamina appear. Later small upright leaves are seen and often there may be leaf drop and twig dieback as trees decline (Bov, 2006) color and are upright. They can also be seen at the tip of new flushes. The fruits are poorly colored and abnormal coloring starts at t he peduncular end (Bov, 2006) The trees become stunted. The juice of infected fruits is often of poor quality and bitter in taste. The fruits have low sugar and high acid. In advanced cases the root system can also be affected an d there is a general decline of the trees. The symptoms are often confused with nutrient deficiency symptoms. It is difficult to diagnose as there is irregular distribution of pathogen in the tree (Graca, 1991) low concentration of the pathogen, and symptoms are often influenced by the temperature. The most reliable methods of detection include real time PCR and transmission electron microscopy, though at advanced stages, iodine tests may be done to identify starch accumulation. Often there are multi ple infections per tree. Depending on the age of the tree severe infections are seen within one to five years after the onset of the first symptoms (Gottwald, 2010) As the disease progresses there is decrease in yield as fruits drop from the affected branches and quality of fruits becomes inferior. But some fruits may still be harves t able (Gottwald, 2010) The incubation period is much less i n young plantings than in older plantings, and symptoms are expressed much faster in younger plants. In the infected leaves the phloem is blocked due to plugging of sieve pores and the translocation system is blocked due to necrotic phloem. There may be un usual cambial activity and massive accumulation of starch. Foli mo nova and Achor (2010) suggested callose plugging and filamentous material deposition as a cause of phloem plugging

PAGE 18

18 HLB has a long incubation period and there is seasonality of expression. Al though PCR may be used to detect it, it is expensive and labor intensive. Symptom expression is usually more pronounced from the end of s ummer to the beginning of s pring Recently, according to Gottwald, Aubert and Yuan (1989), for every symptomatic tree i n a disease impacted planting, there may be up to 13 asymptomatic trees which will show symptoms over time. There is also anecdotal information suggesting a clustering of diseased trees (Graca, 1991) Transmission Ca. L. asiaticus is spread by the Asian citrus psyllid, Diaphorina citri which is found in Asia, the Indian Ocean islands, and now in America. The presence of D. citri was first reported in Florida and Guadeloupe in 1998 (Halbert and Nez, 2004) The nymphs often concentrate on the upper surface near the midrib causing the leaves to roll. The pop ulation of psyllids vary with the availability of young flush as the young flushes are necessary for the laying of eggs and development of nymphs. Thus there are large psyllid infestations from late s pring to mids ummer when there is abundant young flush. When psyllids infect a plot from outside after visiting infected plots, control becomes difficult, but within a plot, psyllid infection may be reduced by spraying insecticides or removing symptomatic trees which can be inoculum sources (Bassanezi and Gottwald, 2009) The graft transmission is variable and shows that distribution of the pathogen in hosts may be uneven and also depends on the tissue used for inoculum. Measures Against the Disease The damage to citrus groves from HLB is widespread, and the spread is rapid. Some short term protective measures may be taken including psyllid control, covering bud wood sources with insect resistant screens, removing infected tree to reduce inoculum loads, etc.; but for long term, the development of tolerant or resistant varieties is the best and most economical

PAGE 19

19 management option. As a result of HLB, costs of managing gr oves have gone up significantly. Roistacher ( 1996 ) showed that groves must live for at least 10 years to make a profit, but trees affected with greening are unproductive in 5 8 years. Even in groves where there are no symptoms, scouting and efficient psyll id control should be done. Often tree removal is not accep table to many growers and they utilize nutritional programs which help to sustain the productivity of these trees. It has also been shown in one study that when foliar fertilizers are applied, the severity of symptoms are much less than when only soil fertilizers are applied (Haigh et al., 2010) Though all the varieties of citrus are susceptible there may be varied response s in the degree of to lerance and one aspect of managing the disease c ould be to use rootstocks which may show some degree of tolerance. Rootstock B reeding Citrus trees were grown as seedlings until the late 1800s but two important diseases, Phytophthora root rot and citrus triesteza virus changed the way of citrus cultivation (Castle, 2010) ; the use of rootstocks gained prominence, and the budding of scion varieties onto rootstocks gained importance. This method of grafting a scion on to a rootstock combines the good qualities of both to give a productive tree. This method reduces juvenility, and many of the characteristics of the rootstock affect the scion variety. A good rootstock is one which would have adaptability to diffe rent soil conditions, produce good quality fruits, increase fruit size, increase yields, influence the hardiness of the scion variety and have resistance to a number of diseases such as CTV, blight, greening, Phytophthora, Diaprepes etc. Rootstocks should possess nucellar polyembryony to facilitate propagation of true to type rootstock liners. Rootstocks also affect the brix/acid ratio in scion fruit. The main current objectives of using rootstocks are to control tree size, provide tolerance against many di seases, and to improve yield (Grosser and Gmitter, 2010) As the rootstocks have a role in plant water relations and the citrus fruit consist

PAGE 20

20 of sacs filled with juice, the rootstocks have a role in the tas te of fruits (Albrigo, 1977) The most universal and widespread rootstock was sour orange, probably a hybrid of mandarin and pummelo (Nicolosi et al., 2000). Sour orange could be grown on a wide range of soils, the yield and fruit quality are good and trees are tolerant to many diseases especially foot rot and gummosis (Phytophthora) and blight (Grosser, 2004) ; however, most scions especially sweet orange and grapefruit on this rootstock are susceptible to CTV The spread of the brown citrus aphid deepened the CTV/sour orange crisis. T he other rootstocks which gained prominence earlier are the Rough lemon, Volkamer lemon and Cleopatra mandarin (Castle, 2 010) Most of the lemon types are not very cold tolerant but the yield is high with fruit of poorer quality. Rough lemon is also susceptible to foot rot. The fruits are slightly better on Volk and they yield good fruit size, especially for Murcotts. Ra ngpur lime is used in Brazil mainly for its adaptability to saline or calcareous soils and drought tolerance and the fruit quality is similar to that of the rough lemon. Cleopatra mandarin is essentially used for mandarins and the hybrids of mandarin as s cions, but the fruit size is often smaller (Castle, 2010) It is a superior rootstock especially for lemon, grapefruit and sweet orange (Batchelor and Bitters 1952 ) It is also resistant to gummosis and quick decline. Another important rootstock is trifoliate orange ( Poncirus trifoliata ) which is very cold hardy and disease resistant, though it is susceptible to Exocortis virus. The hybrids of trifoliate orange are also very important, such as the citranges and the citrumelos. The citrange is a hybrid of sweet orange and trifoliate orang e of which the Troyer and Carrizo citrange are most popular, the former being popular in California and the latter being popular in Florida. Troyer is very good for Lisbon type of lemons, but other lemons show an incompatibility. Carrizo is good for Valenc ia sweet orange (Castle 2010). The citranges have good disease resistance qualities, with fruits being medium in size but good in quality.

PAGE 21

21 Though Carrizo has been a good choice especially in Florida, it is susceptible to citrus blight (Castle, 2010) Two other important citranges are the Kuharske and C 35. The Kuharske has good tolerance to burrowing nematodes, while C 35 (a hybrid of ruby blood orange and trifoliate orange) is good in t he presence of Diaprepes weevil (Castle, 2010) and also controls tree size One of the disadvantages of using the citranges as rootstock is however their exceptional vigor. The other mentioned group of trifoliate orange hybrids are the citrumelos, a hybrid of grapefrui t and trifoliate orange, of which Swingle (Duncan grapefruit x trifoliate orange) is very important, especially in Florida. This rootstock was released in 1974 and is very resistant to CTV and Phytophthora. The yield on this rootstock is accep table and the juice quality is good. In cold tolerance it is intermediate between sour orange and rough lemon. It is sensit ive to different kinds of soil (Hutchison, 1974) The sweet orange as a rootstock never gained much importance in Fl orida, though it has been used successfully in California. One reason maybe it is not very cold hardy. Trees on sweet orange rootstock are susceptible to nematodes and gummosis but tolerant to CTV, blight, and exocortis. The fruit size is usually smaller o n sweet orange (Bitters, 1961). The grapefruits are not widely used as rootstocks, though the fruit quality is good and the fruits are larger in size, the number of fruits per tree may be smaller. The pummelos are monoembryonic and extremely susceptible to the triesteza virus, making its use limited as a rootstock (Bitters, 1986). The fruit size is large, and the quality is good though yields are low. In recent years another method of breeding is adopted, the process of somatic hybridization, which has been very useful for rootstock breeding. Somatic hybridization is a novel technique by means of which the somatic cells of plants may be combined bypassing the normal sexual hybridization, and an allotetraploid hybrid may be generated. This was first succ essfully done in tobacco (Carlson, 1972) and thereafter has

PAGE 22

22 been adopted in many plants including citrus (Grosser and Gmitter, 2010) In citrus numerous somatic hybrids have been developed by fusio n of protoplasts isolated from embryogenic suspension cultures with leaf derived protoplasts. Though this method does not give rise to new scion cultivars, it has been very successful in producing tetraploid breeding parents and thus facilitating the breed ing of seedless triploid scions, and also for facilitating the breeding of rootstocks at the tetraploid level. Some of the potential rootstocks developed by this method can control tree size as needed for high density plantings, and also for cold protectio n. Some allotetraploid rootstocks also produce high yields. The somatic hybrids may also be used as parents and traditionally bred to form allotetraploid sexual hybrids, which increase the genetic diversity and may have a lot of desirable qualities like d isease resistance, cold hardiness etc. The are sour orange + Rangpur and Nova mandarin + Hirado Buntan pummelo, the latter showing tolerance to the Phytophthor a/Diaprepes complex; both of these produce zygotic seeds (Grosser and Gmitter, 2010) Some somatic hybrids that have been used as pollen parents are SO+CZO (sour orange + Carrizo), Cleopatra + Argentine tri foliate orange, and sour orange + Palestine sweet lime. HLB Tolerance The most HLB susceptible cultivars are sweet oranges, mandarins and tangelos (Albrecht et al., 2012) Moderately HLB susceptible cultivars are grapefruits, lemons, Rangpur lime, and calamondin. Mexican lime and pummelos are considered to be the most HLB toleran t cultivars. In South Africa, a study was conducted to see the effect of rootstock on development of HLB symptoms and it was found that Valencia sweet oranges showed more symptoms when on trifoliate orange than on Empress Mandarin or Troyer citrange. It wa s hypothesized that the trifoliate orange extended the growth flush period exposing it for a longer time to psyllid feeding

PAGE 23

23 (Vuuren and Moll, 1985) In China, a study of Ponkan mandarin on 13 rootstocks did not show any difference (Graca, 1991) In Brazil, a study was conducted to observe the effect of rootstock on development of HLB on sweet orange. The rootstocks used were Rangpur lime, Swingle citrumelo and Su nki mandarin, on which N atal sweet orange was grafted. The incidence of disease was most in trees on Rangpur lime and least on that of Sunki mandarin (Sanches and Augusto, 2010) In another grafted onto Flying Dragon trifoliate orange showed more HLB tolerance as trees did not show symptoms even when surrounded by heavily infected plots. The increased flushes are often influenced by rootstock vigor, which is a source of food for the insect vector; thus low vigor roots tocks such as Flying Dragon could help avoid or red uce opportunities for infection (Sanches and Augusto, 2010) In Australia a study found that sweet orange scions grafted on Limonia acidissima rootstocks showed less severity of symptoms, whereas those grafted on Feroniella lucida were more suscep tible (Haigh et al., 2010) Folimonova et al (2009) studied different citrus genotypes and their responses to HLB. Although there was no correlation between bacteria l titer and severity of diseas e, based on difference in development of symptoms, the genotypes were divided into sensitive, moderately tolera nt, tolerant and variable types. In the sensitive group was placed sweet orange, grapefruit, Nules clementine mandarin, C. hal imii and Minneola t angelo while the pummelo, Cleopatra, C. indica C. amblycarpa and Meiwa kumquat were in the second group with variable responses. The third group of moderately tolerant cultivars consisted of Volk, Swingle, citron, sour orange, C. m a crophylla Sun Chu Sha mandarin, Palestine sweet lime etc.

PAGE 24

24 The most tolerant were found to be Carrizo, Severinia buxifolia Persian lime, Eureka lemon. Progress is being made to solve HLB using the latest biotechnologies and genetic engineering approaches, but this may take 10 to 20 years. Thus, one way of managing the disease in commercial groves would be to use rootstocks which would induce high yields and at the same time lower leaf flushes which would in turn attract less psyllids. Higher incidence of HLB is observed in the marginal areas of impacted orchards and thus several practices have been suggested to reduce the impact of the disease such as usi ng windbreaks including pine, rubber trees or eucalyptus; and psyllid repellants including mango, white guava, or neem; and attractive trees like orange jasmine with systemic insecticides at the edge. In Vietnam it was observed that when guava was intercro pped with citrus, there was a repellant effect of volatiles on Asian citrus psyllid. It was found that DMDS, dimethyl disulfide, a sulfur volatile from crushed guava leaves indeed have a repellent effect on psyllids (Rouseff et al., 2008) Different citrus varieties may have different volatile profiles which may have a differential impact on the behavior of D. citri (Beattie et al., 2010) Disease Analysis The objec tive of present research is to see whether there is any rootstock effect in the expression of HLB symptoms. There are two common measurements of disease, incidence and severity. The measurement of disease incidence is simple, by simply counting the unit (t ree) as diseased or not and then adding them up. It is the proportion of plants diseased out of the total number of trees in the trial on any particular rootstock. Determining disease severity is more challenging and a measure of disease symptoms per sampl ing unit. Severity of disease is often measured by subjective visual assessment. Visual assessment is rapid and useful when hundreds of trees have to be evaluated. There are various ways of visually assessing the data like direct estimation, using standard area diagrams, disease scales or ordinal rating scales. Of the disease

PAGE 25

25 scales the most common is the Hors Fall Barratt scale (Bock et al., 2009) which consists of 12 classes of disease scale which have unequal severity intervals. But perhaps as a general rule equal increment scales may be more preferable than the H B scale. A disease severity scale to quantify severity was developed by Hors Fall and Heuberger in 1942. According to Foolad et al (Foolad et al., 2000) evaluation of plant resistance in the field is more appropriate since reliability of results is greater. Shokes et al (1987) proposed measuring intra rater repeatability with a test retest correlation procedure ( Nutter et al., 1993) The correlation coefficient (r) gives a statistical measure of relationship between repeated assessments of same sampling units by same individual. Least squares regression may be used to determine if there is a statistical relationship between repeated assessments performed by the same individual. When we assess disease, the data may be quantitative, qualitative or both. Usually there i s high variability in collected data, so it is called the random variable which can be discrete or continuous. The data obtained by ordinal disease rating as 0 to 4 should be analyzed by either the non parametric statistical method or parametric generalize d linear model (Madden 2008). One of the advantages of interval scales is that it is quick. Increasing the replication and averaging the data often improved overall agreement with actual data and stabilize variance. When the data are normally distributed, form a continuous pattern and evenly spaced between categories, they may be analyzed using parametric methods like ANOVA but non parametric methods may be used when necessary. Some non parametric tests are Kruskal Wallis test, Spearman correlation, W ilcoxon sign rank test (McDonald, 2009) When assessment is done on same specimen one can see the difference made at two different times or between visual assessment and another form of

PAGE 26

26 assessment (Madden et al., 2007) Irey et al ., (2006) indic ated that visual testing with real time pcr is effective to diagnose more infections than either method alone. Objectives The overall objective of the present research is to determine if the rootstock a ffects the tolerance of the tree to HLB, especially after infection. The collection of rootstocks included in the trial contains diploids, allotetraploid somatic hybrids, and allotetraploid tetrazygs produced from crosses of somatic hybrids; thus a wide r ootstock gene pool is being evaluated regarding the potential to impart tolerance to the scion. This overall objective would be achieved by looking at disease severity based on visual observations, C t values from symptomatic and asymptomatic leaves of the infected trees, and fruit drop, as fruit drop is very common in greening affected trees reducing yield.

PAGE 27

27 CHAPTER 3 MATERIALS AND METHODS McTeer Farm The McTeer site is located in Haines City, Florida, just south of the Haines City high school. The trees were planted in 2009, and have been grown using liquid fertigation. There are six rows of trees, row 1 has 78 trees, rows 2 5 have 79 trees/row, and row 6 has 3 1 trees. About five trees were not included in the evaluation as the scion was different. In total there were 396 trees that were evaluated. The scion used was Sugar Belle, a fresh fruit hybrid from a cross of Clementine and Minneola. Fifteen rootstocks were evaluated including: diploids Flying Dragon trifoliate orange, Rich trifoliate orange, 50 7 trifoliate orange, C 35 citrange (control), Swingle citrumelo (control); somatic hybrids SO+50 7, WGFT+50 7, Changsha + 50 7 and SO+CZO; and tetrazygs Orange 1 Orange 4, Orange 19, White 4, Purple 2 and Purple 4 (2 1). Trees in this trial were nearly 100% infected with HLB St. Helena Farm The St. Helena trial is located just east of Dundee, Florida. This trial consists of two blocks, St. Helena West and St. Helena East. The scions used were Vernia and Valquarius sweet oranges, which have a harvest date between the middle of January and beginning of March, and produce juice with the quality of that from standard Valencia. The 75 rootstocks planted here co nsisted of controls along with selections from the research programs of Dr. Jude Grosser, Dr. Fred Gmitter and Dr. Bill Castle. The trees were planted at three different spacings, 9x20 (240 trees/acre) in the East; and 12x20 (southwestern portion, 181 tre es/acre) and 15x 25 (northwestern portion, 119 trees/acre) in the West. Trees were grown using the high quality rootstocks included conventionally bred diploid rootstocks, allotetraploid somatic hybrids, and

PAGE 28

28 allotetraploid tetrazygs from conventional breeding using somatic hybrid parents. This trial has northeast, a top worked grove to the southeast, and an organic grove to the southwest. Out of 75 rootstocks and approximately 3000 trees, 19 scion/rootstock combinations which were identified by the CREC scouts as being infected based on visual observation of greening symptoms at St. Helena West were evaluated, and likewise 7 scion/rootstock combinations at St. Helena East. A primary goal of this rootstock trial is to develop tree size controlling and precocious bearing rootstocks to facilitate ACPS (Advanced Citrus P roduction Systems). The HLB infected trees selected for the study were all 55 months old. Visual O bservations on S tudy T rees The foliar symptoms of HLB disease (citrus greening) include blotchy mottle, yellow corky vein, green islands and leathery lamina. In the later stages of infection, small upright leaves (rabbit ear leaves), leaf drop and twig die back may be observed. In the advanced symptoms there is a general decline of the tree. The first part of the experiment was to visually categorize a ll the included trial trees on a numerical scale of 0 to 4 where 0 meant apparently no symptoms, 1 for symptoms on up to 25% of the tree, 2 for symptoms on 26 to 50% of the tree, 3 for symptoms on 51 to 75% of the tree, and 4 for symptoms on more than 75% canopy. Trees were evaluated with this scale at both trials. Visual assessment is often used to note the disease severity. The increase in value of the scales indicates the increase in severity of the disease. The advantage of using such a scale is that it may be used to rate a large number of trees in a relatively short time. This method was used both in the s ummer and f all of 2012 to assess the impact of HLB on the test trees.

PAGE 29

29 qPCR and Ct Value Determination The qPCR technique was used to estimate Candidatus Liberibacter asiaticus tite r s in the study trees. Unlike normal PCR where data are collected at the end of the cycle, here the data are collected throughout the cycle. Ct refers to the threshold cycle or the cycl e number at which the fluorescence passes a predetermined threshold. It is used to detect the first amplification. In this experiment the ABI 7300 thermocycler was used and primers of Li et al. (2005) were used to determine the Ct value s Leaf samples from symptomatic and non symptomatic areas of test trees were collected, and PCR samples were run at the Southern Gardens/US Sugar Diagnostic Laboratory. The Ct value s below 30 are considered positive, between 30 and 32 tentatively positive and above 32 as ne gative. The Ct value was determined from asymptomatic as well as symptomatic parts of trees as it is known that in a grove there is more infection than it meets the eye. It has been proposed that for every symptomatic tree there may be 13 asymptomatic trees (Bassanezi and Gottwald, 2009) This gives an opportunity for disease management, where decisions may be t aken about removal of infected trees DNA extraction DNA was extracted using the Bio Sprint DNA plant kit (Qiagen). The technology used in these kits is Mag Attract magnetic particle technology where the leaf material is disrupted to give a fine powd er which is then mixed with a lysis buffer and sedimented by centrifugation. The clear solution is then sent to a 5 well strip for further processing. The DNA is bound to MagAttract magnetic particles in the presence of chaotropic salt. This DNA is further washed with buffer and ethanol and rinsed for further use. This method is slightly modified as follows:

PAGE 30

30 Robot DNA extraction Materials 100 mg tissue ( leaf petioles finely chopped). 2 ml micro centrifuge tubes Stain less steel beads. Bead beater. Repeater or multichannel pipette. Centrifuge. Plates and blocks: 96 well microplate MP and S block Plate covers BioSprint 96 Qiagen robot. Vortex. Reagents DNA Qiagen kit RLT lysis buffer. Isopropanol 91% MagAttract Suspension G. Buffer RPW. Ethanol absolute (96 100%) Distilled water contains 0.02% (v/v) Tween 20 Low salt buffer (AE elution buffer). Methods Protocol provided with Qiagen kit was followed with modifications as follows: Experiment 1: DNA extraction from plant tissue 1. Place 100 mg chopped tissue in 2 ml microcentrifuge tube. Add 1 steel bead. 2. Add 600 ul lysis buffer RLT and homogenize the samples for 1.5 min in the bead beater. 3. Spin down for 20 min at 6000 G. 4. Transfer 200 ul cleared plant lysate into each well of Load Lysate S Block. 5. For next steps to follow the Qiagen BioSprin t DNA Plant Hand book pages 22 thru 24. Following analysis, C t values of all samples were provided by the SG/US Sugar diagnostic lab.

PAGE 31

31 Yield Estimation Formal yield data per rootstock is available from St. Helena both in 2011 and 2012; however, yield from individual trees included in the study was estimated visually with the following subjective ranking scale: High (approximately 2 or more 90 lb boxes), M edium (approximately 1.0 2.0 boxes, or Low (approximately less than 1.0 boxes). At McTeer yield estimation was done in December 2012 and at St. Helena it was done in January 2013. Fruit Drop Estimation At both the St. Helena and Mc T eer trials, fruit dr op was counted and the cumulative average (mean) per rootstock, along with the median for each rootstock were calculated. Fruit drop was counted thrice. At McTeer it was done in August, October and December in 2012. At St. Helena it was done in October, De cember in 2012 and January 2013.

PAGE 32

32 Table 3 1. Brief description of rootstocks Rootstock Description Parents Pink 1802 Pink 1802 produces the smallest trees. These 3 rootstocks were provided by Dr. Bill Castle and developed by retired Argentine rootstock breeder Jose Luis Foguet. They had superior performance with lemon and sweet orange scions in Argentina. Cleopatra mandarin ( Citrus reticulata )X Swingle citru melo ( C. paradisi X P.trifoliata ) Aqua 1803 Cleopatra mandarin ( Citrus reticulata ) X trifoliate orange ( P. trifoliata ) Orange 1804 Cleopatra mandarin ( Citrus reticulata ) X trifoliate orange ( P. trifoliata ) SO+CZO Allotetraploid (4x) somatic hybrids Sour orange( Citrus aurantium ) +Carrizo citrange ( Citrus paradisi X Poncirus trifoliata ) Cleo+CZO Cleopatra mandarin (Citrus reticulata) + Carrizo ( Citrus paradisi X Poncirus trifoliata ) Chang+50 7 Changsha mandarin ( Citrus reticulata Blanco )+ trifoliate 50 7 ( P. trifoliata ) SO+50 7 Sour orange (Citrus aurantium)+ trifoliate 50 7( P. trifoliata ) WGFT+50 7 White grapefruit ( C. paradisi )+ trifoliate 50 7( P. trifoliata ) Orange 2, Orange 4, Orange 13, Orange 16, Orange 18, Orange 19 Allotetraploid (4x) Tetrazygs Nova mandarin ( Citrus reticulata Blanco ) + Hirado Buntan seedling pummelo ( Citrus maxima ) X (Cleopatra mandarin + Argentine trifoliate orange) Green 7 Nova mandarin + Hirado Buntan seedling pummelo X Sour orange + Carrizo Blue 1, Blue 4 Nova mandarin + Hirado Buntan seedling pummelo X Sour orange + Palestine sweet lime ( Citrus limettiodes ) Purple 2, Purple 4 Nova mandarin + Hirado Buntan seedling pummelo X Cleopatra + Sour orange Swingle Controls It is a hybrid of Duncan grapefruit and trifoliate orange ( C. paradisi X P. trifoliata ). It is tolerant of Phytophthora and CTV and also salt and drought tolerant.

PAGE 33

33 Table 3 1. Continued Rootstock Description Parents Volk It is tolerant to CTV but susceptible to Phytophthora. The fruit is better than when grown on rough lemon ( C. volkameriana ) Rough lemon Controls This rootstock is drought tolerant, tolerant of CTV and Phytophthora, produces large fruits but the quality is poor.( C. jambhiri ) C 35 This is a hybrid of trifoliate orange and Ruby blood sweet orange. It is tolerant to Phytophthora and CTV( P. trifoliataXC .sinensis )

PAGE 34

34 Table 3 2 Rootstocks evaluated at the Mc T eer trial (Sugar Belle scion) Rootstock Number of trees Rootstock Number of trees Rootstock Number of trees Orange 1 22 SO+50 7 21 White 4 12 Flying Dragon 54 TF 50 7 40 Changsha+50 7 17 Purple 2 15 Orange 19 15 Orange 4 20 Purple 4 13 Swingle 21 SO+CZO 26 Rich Trifoliate 54 WGFT+50 7 19 C 35 47 Table 3 3. Rootstocks evaluated at the St. Helena Trial West Section (Vernia or Valquarius scion) Rootstock Number of trees Rootstock Number of trees Rootstock Number of trees SO+CZO 15 Aqua 1803 3 Orange 16 3 Orange 1804 3 Orange 4 7 Swingle 6 White 4 8 Orange 18 5 Volk 7 WGFT+50 7 5 Orange 3 4 KCZ 24 Green 7 9 Orange 1 3 Rough Lemon 3 Orange 13 5 Orange 2 3 Orange 19 8 Pink 1802 3 Table 3 4 Rootstocks evalu ated at the St. Helena trial East section (Vernia or Valquarius scion ) Rootstock Number of trees CLEO+CZO 20 Blue 4 4 WGFT+50 7 10 SO+50 7 4 Blue 1 7 Purple 4 6 Changsha+50 7 11

PAGE 35

35 Table 3 5. Visual assessment scale used to estimate HLB impact 0 Apparently n o symptoms 1 Symptoms restricted to very few leaves, up to 25% 2 Symptoms more common, 26 to 50% of canopy 3 Symptoms in majority of canopy, 51 to 75% of canopy symptomatic 4 Entire canopy impacted, > 76% of canopy symptomatic Figure 3 1 Example of a Grade 1 tree Figure 3 2 Example of a Grade 2 tree Figure 3 3 Example of a Grade 3 tree

PAGE 36

36 Figure 3 4 Example of a Grade 4 tree Figure 3 5 Tree with high yield Figure 3 6 Tree with medium yi eld Figure 3 7 Tree with low yield

PAGE 37

37 CHAPTER 4 M C T EER F ARM Results Fifteen different rootstocks were evaluated at the Mc T eer Farm, namely Orange 19, White 4, Orange 4, SO+50 7, Purple 4, Changsha+50 7, SO+CZO, C 35, WGFT+50 7, Orange 1 and Purple 2 ; Swingle, TF 50 7, Rich Trifoliate and Flying Dragon were controls for comparison. The scion used was Sugar Belle a mandarin hybri d. The disease incidence was nearly 100%, based on visual observation and validated by PCR. All trees were evaluated on the basis of visual symptoms, Ct values from both HLB symptomatic and asymptomatic leaves, visual observation of yield, and fruit drop. As a consequence of the diverse genetic makeup of the different rootstocks, differences in the response to HLB infection were expected. Visual symptoms The visual symptoms were graded on a scale of 0 to 4 as described previously, where 0 meant apparently no symptoms and 4 meant maximum symptoms. Visual symptom data are presented in Figure 3 1 Table 3 1 and Figure 3 2 The Kruskal Wallis test was used to determine the significance of differences in the rootstocks. The high Chi Square value of 97 .3984 and low p value of .0001 revealed significant difference among rootstocks during s ummer (May, 2012). A boxplot of visual symptoms revealed that tetrazyg rootstock Orange 19 showed the least visual symptoms in s ummer as well as in the f all of 2012 O ther rootstocks which displayed lower visual symptoms during s ummer were White 4, Orange 4, SO+50 7 and Swingle. In f all Swingle and White 4 had less visual symptoms T he Wilcoxon sign rank test revealed there were no major differences in s ummer and f all rankings. However for most trees, the mean of visual HLB symptoms increased in f all but symptoms rankings were unchanged in case of Swingle and WGFT+50 7, and decreased slightly for TF 50 7, Rich Trifoliate and Flying Dragon. To

PAGE 38

38 measure the differences be tween the s ummer and f all visual results, the Wilcoxon signed rank sum test was used, as there are two nominal variables and one measurement variable. The nominal variables are type of rootstock and the two seasons s ummer and f all and the measurement variable is the ordinal scale 0 to 4. Here the null hypothesis states that the median differences among the groups is zero (McDonald, 2009) It is a non parametric test and used when the distribution may be not normal. In this test a variable difference is computed which is the difference between the two observations and given a rank. The smalles t difference gets the smallest rank as 1, and ties are given average ranks. The ranks are summed up for each direction and the test statistic is given by the smaller rank. The high p value 0.7171 indicates that the null hypothesis may not be rejected and t here is not much difference in the value of s ummer and f all The p value is not significant. Therefore it can be concluded that the differences in the overall s ummer and f all readings are not significantly different from one another. PCR Analysis The Ct values HLB Symptomatic leaves About 5 to 6 leaves with petiole, with and without symptoms were collected from each tree, kept in plastic bags and sent to the US Sugar Corporation Diagnostic Laboratory for DNA extraction and determination of Ct values This was done in s ummer (July, 2012) and f all 2012 (December, 2012) The resulting data were checked for assumptions of normality and it was seen that the assumptions were not met. The histogram is not typically bell shaped as in normal distribution. The data points in the normal probability plot do not f all on the straight line, showing it is not normally distributed. Thus Kruskal Wallis test was performed to see if there were significant differences in the Ct value from symptomatic leaves among differen t rootstocks The Kruskal Wallis test for symptomatic leaves in s ummer revealed a Chi Square value of

PAGE 39

39 46.1697 and p value .0001, showing significant differences. The boxplots revealed most rootstocks showing maximum data with Ct values below 30, except for rootstock Purple 2. Some Purple 2 trees had Ct values above 30. The highest mean Ct value was also seen in Purple 2. Purple 2 had the highest mean in s ummer as well as in f all When the two data sets were compared, it was seen that for most rootstocks the mean Ct value decreased in f all or the bacterial tit er increased (as expected). The Ct mean increased or bacterial tit er decreased in case of Flying Dragon, Orange 4 and SO+CZO. Asymptomatic L eaves The same analysis was performed for non symptomatic leaves (no HLB symptoms) sampled from the same trees. The data for non symptomatic leaves still failed to meet the assumptions of normality. The normality assumption of data is still violated. Thus t he non parametric test Kruskal Wallis test is done to analyze the results to find out if there are any significant differences among the different rootstocks. The Chi square 24.6969 and p value .0377 revealed significant differences in the rootstocks durin g s ummer The box plot revealed that rootstocks Purple 2, Purple 4 and Orange 19 had some samples with Ct values above 30. The highest mean Ct values were also recorded for Purple 2, Purple 4 and Orange 19. In the f all again Purple 2 had many samples with lower bacterial titers with Ct values above 30. Comparing the means for s ummer and f all the trend was for most rootstocks to increase titers of bacteria in f all as the Ct values decreased; however, this was not the case for rootstocks Purple 2 and SO+CZ O, where the Ct value increased and the bacterial titer decreased. Biochar Treatment Another aspect that is being investigated at the McTeer farm is whether or not treatment of trees with biochar has any effect on HLB remediation and expression of HLB

PAGE 40

40 symptoms. Biochar is derived from carbonization of organic matter (southern pine) and helps to improve soil by enhancing nutrient holding capacity. Biochar was applied in February, 2012. ked in the microjet zone by hand. Five trees on Flying Dragon, five on Rich Trifoliate and ten on Swingle were left untreated. The visual symptoms, and Ct values from symptomatic and asymptomatic leaves from treated and untreated trees were compared. The combined data (Tables 3 4, 3 5, and 3 6) suggests that there was no apparent effect of the biochar treatment at this time. Further investigation will be necessary to determine if there is a longer term effect. Estimated Fruit Yield and HLB I nduced Fruit Drop Although formal yield data were not collected at the McTeer farm, each tree was visually assessed for yield as H (High), M (Medium) or L (Low); with percentage mean estimates provided in Figure 4 1 4 Although young trees in this trial were appro ximately 3.5 years old, heavy crops of fruit were set throughout, across all rootstocks. The epidemic HLB infection did not appear to effect fruit set. HLB induced fruit drop was determined for the Sugar Belle trees on the different rootstocks fruit dr op was counted in August, October and December of 2012 and cumulative average was calculated. The harvest date for Sugar Belle was around the third week of December Trees on rootstocks Purple 2, SO+ CZO and SO+50 7 showed the least amount of fruit drop (3 10). Among the control rootstocks, trees on Swingle exhibited moderate fruit drop, but trees on all 3 trifoliate orange selections Rich Trifoliate, TF 50 7 and Flying Dragon had higher rates of frui t drop. Higher fruit drop was also found for trees on Orange 1 and White 4. Overall, the amount of fruit drop in the trial was much lower than expected, and a very low percentage of the total amount of fruit set. When fruit drop was compared with on tree yield, data suggest a higher rate of fruit drop with higher fruit set.

PAGE 41

41 The fruit drop data was log transformed as it did not meet the assumption of normality. be analyzed by A NOVA A NOVA reveals significant difference in fruit drop for different rootstocks At Mc T eer it seems the higher the yield, the higher the fruit drop. The high yielding trees had on an average fruit drop of 36.03, the trees with medium y ield had on an average fruit drop of 25.45 and the trees with low yield an average fruit drop of 10.82. Conclusions In a young field trial of Sugar Belle mandarin hybrid on different test rootstocks where nearly every tree is HLB affected the rootsto ck showing the least severe disease from the cross of two somatic hybrids: [Nova mandarin + Hirado Buntan seedling pummelo] X [Cleopatra mandarin + Argentine tr ifoliate orange]. This result was consistent over the s ummer and f all assessments. During the s ummer assessment, the Ct values from symptomatic leaves of trees on most rootstocks showed values below 30. However, in Purple 2 rootstock, some samples had Ct v alues above 30. In the f all assessment, the bacterial titer had increased in the symptomatic leaves of trees in almost all rootstocks except Flying Dragon, Orange 4 and SO+CZO. This is expected, as high s ummer heat is supposed to reduce bacterial titers ( thermotherapy). There was no change in the average Ct value in the leaves sampled from trees on White 4 from s ummer to f all The asymptomatic leaves from trees on rootstocks Purple 2, Purple 4 and Orange 19 had some samples with Ct values higher than 30 d uring s ummer Again in f all the bacterial titer had increased in the asymptomatic leaves; however, rootstock Purple 2 still had a high median and many samples with Ct values above 30. Unexpectedly, both

PAGE 42

42 SO+CZO and Purple 2 showed an increase in average Ct value during f all probably due to irregular distribution of the pathogen in a tree. Another aspect studied at the McTeer trial was the possible effect of biochar as part of an HLB remediation effort. Biochar was applied (January, 2012) along with su pplemental slow release dry fertilizer as a soil amendment. The entire experiment was treated, excluding five trees each on Flying Dragon, Rich trifoliate and ten trees on Swingle, which received only the slow release fertilizer. In a comparison of biochar treated and untreated trees on the above rootstocks, looking at the visual rating results and average Ct values from both asymptomatic and symptomatic leaves, no benefit from the biochar was observed. However, one of the main features of biochar is to c reate an enhanced environment for beneficial soil microorganisms including mycorrhiza that facilitate nutrient uptake. Thus, it may require more time before a positive benefit can be observed. Evaluation of rootstock differences on HLB induced fruit drop was conducted, and although fruit drop was relatively low, significant differences were observed. Trees on rootstock Purple 2 showed the least amount of fruit drop. Other rootstocks with lower fruit drop were trees growing on Purple 4, SO+50 7 and SO+ CZO. When looking at on tree yield data and fruit drop data together, there was a trend that trees with higher yields showed higher average fruit drop.

PAGE 43

43 Figure 4 1 Boxplot of visual HLB symptoms on Sugar Belle trees on various rootstocks at the Mc T eer Trial May 2012 (Scale 0 4). A) The line identifies the median and the diamond identifies the mean. Table 4 1 Mean of visual HLB symptoms at the Mc T eer trial, in May 2012 and November, 2012 Name Number of trees Summer Mean Fall Mean Orange 19 15 1.47 2.07 White 4 12 2 2.17 Orange 4 20 2.05 2.5 SO+50 7 21 2.1 2.48 Swingle 21 2.1 2.1 Purple 4 13 2.23 2.31 Changsha+50 7 17 2.24 2.29 SO+CZO 26 2.27 2.31 C 35 47 2.36 2.38 WGFT+50 7 19 2.37 2.37 Orange 1 22 2.5 2.55 TF 50 7 40 2.68 2.48 Purple 2 15 2.73 3.21 Rich Trifoliate 54 2.98 2.69 Flying Dragon 54 3.28 2.81

PAGE 44

44 Figure 4 2 Boxplot of visual HLB symptoms of Sugar Belle trees on various rootstocks at the McTeer Farm, November 2012 A) The line identifies the median and the diamond identifies the mean. Figure 4 3 A comparison of the means of visual HLB symptoms of Sugar Belle trees on various roostock for the s ummer (May, 2012) and f all (November, 2012)

PAGE 45

45 Table 4 2. Results of Wilcoxon sign rank test Tests for Location: Mu0=0 Test Statistic p Value Student's t t 0.380586 Pr > |t| 0.7037 Sign M 4.5 Pr >= |M| 0.5746 Signed Rank S 273 Pr >= |S| 0.7171 Figure 4 4 Checking assumption of normality from HLB symptomatic leaves in s ummer Figure 4 5 Histogram checking assumption of normality from HLB symptomatic leaves in s ummer

PAGE 46

46 Figure 4 6 Boxplot of Ct values from HLB symptomatic leaves from Sugar Belle trees on various rootstocks determined in summer (July 2012). A) Here the diamond identifies the mean and the horizontal line identifies the median.

PAGE 47

47 Table 4 3 Mean Ct values from HLB symptomatic leaves from Sugar Belle trees per rootstock determined in summer (July, 2012) and f all (December, 2012) Summer Fall Rootstock Number of trees Mean of Ct value Mean of Ct value Purple 2 15 27.04 24.56 Purple 4 13 25.82 22.76 Rich trifoliate 54 25.09 23.6 Orange 19 15 24.95 23.95 TF 50 7 40 24.81 23.11 SO+50 7 21 24.11 23.1 Orange 1 22 23.87 22.04 Flying Dragon 54 23.87 24.55 Swingle 21 23.64 23.29 SO+CZO 26 23.59 24.26 C 35 47 23.34 21.6 Orange 4 20 23.31 23.43 Changsha + 50 7 17 23.23 22.22 WGFT + 50 7 19 23.22 22.56 White 4 12 22.77 22.68 Figure 4 7 Boxplot of C t value from symptomatic leaves from Sugar Belle trees on different rootstocks, determined in f all (December, 2012) A) Here the horizontal line identifies the median and the diamond identifies the mean.

PAGE 48

48 Figure 4 8. Comparison of the mean of Ct value from symptomatic lea f samples from Sugar Belle trees on different rootstocks in s ummer (July, 2012) and f all (December, 20 12) Figure 4 9 Checking assumption of normality from asymptomatic leaves in s ummer (July 2012)

PAGE 49

49 Figure 4 1 0 Histogram to check normality from asymptomatic leaves in s ummer (July 2012) Figure 4 1 1 Boxplot of C t value from asymptomatic leaves sampled from Sugar Belle trees on different rootstocks in s ummer (July, 2012) A) Here the line is the median and the diamond is the mean.

PAGE 50

50 Table 4 4 S howing mean Ct values from asymptomatic leaves sampled from Sugar Belle trees on different rootstoc ks in s ummer 2012 (July, 2012) and f all 2012 (December, 2012) Name Number of trees Mean of Ct Value( Summer ) Mean of Ct Value( Fall ) Purple 2 14 30.41 32.55 Purple 4 13 28.81 24.67 Orange 19 15 27.68 25.21 TF 50 7 40 27.63 25.37 WGFT+507 19 27.11 24.76 SO+50 7 21 26.98 25.22 Rich trifoliate 54 26.76 25.33 Changsh+50 7 17 26.69 24.88 Orange 1 22 26.36 22.26 Swingle 21 26.22 24.42 C 35 47 26.07 23.07 Orange 4 20 25.66 25.44 Flying Dragon 54 25.54 24.13 SO+CZO 26 25.17 25.85 White 4 12 24.7 24.37 Figure 4 1 2 Boxplot of Ct value from asymptomatic leaves sampled from Sugar B elle trees on different rootstocks in f all (December, 2012)

PAGE 51

51 Figure 4 1 3 Comparison of mean of Ct values from asymptomatic leaves sampled in summer (July, 2012) and f all (December, 2012) from Sugar Belle trees on different rootstocks. Table 4 5 Comparison of visual symptom means of biochar treated and untreated trees of Sugar Belle on selected rootstocks. R ootstock Total trees with biochar without biochar Flying Dragon 54 # trees s ummer f all # trees s ummer f all 49 3.24 2.8 5 3.6 3 Swingle 21 11 2.18 2.18 10 2 2 Rich Trifoliate 54 49 2.89 2.67 5 3.8 2.8 Table 4 6 Comparison of Ct value means from symptomatic leaves of Sugar Belle trees on selected rootstocks treated or untreated with biochar. R ootstock Total trees with biochar without biochar Flying Dragon 54 # trees s ummer f all # trees s ummer f all 49 23.94 24.69 5 23.2 23.18 Swingle 21 11 22.71 22.63 10 24.66 24.02 Rich Trifoliate 54 49 25.28 23.69 5 23.19 22.72

PAGE 52

52 T able 4 7 C ompari son of Ct value means from asymptomatic leaves of trees treated with biochar and those untreated rootstock Total trees with biochar without biochar Flying Dragon 54 #trees s ummer f all # trees s ummer f all 49 25.64 24.27 5 24.63 22.74 Swingle 21 11 26.49 23.33 10 25.84 25.62 Rich Trifoliate 54 49 26.76 25.68 5 25.09 21.91 Figure 4 1 4 Visually estimated yield of Sugar Belle trees on different rootstocks, categorized as high, medium, or low yielding trees, and dead trees. Figure 4 1 5 Example of a Sugar Belle tree rated as h igh yield tree

PAGE 53

53 Figure 4 1 6 Example of a Sugar Belle tree rated as m edium yield tree Figure 4 1 7 Example of a Sugar Belle tree rated as low yield tree Table 4 8 Means of Sugar Belle fruit drop per rootstock in the McTeer Trial Rootstock Number of trees Average fruit drop per tree Purple 2 14 7.36 SO+CZO 26 8 SO+50 7 21 9.81 Purple 4 13 14.85 Swingle 21 24.05 Changsha+50 7 17 25.59 C 35 45 26.22 Orange 4 20 29.65 Orange 19 15 29.87 WGFT+50 7 19 33.68 Rich Trifoliate 54 34.76 White 4 12 35.25 Flying Dragon 54 41.48 Orange 1 22 44.09 TF 50 7 40 54.73

PAGE 54

54 Figure 4 1 8 Boxplot of Sugar Belle fruit drop on different rootstocks in the McTeer trial. Figure 4 19 Normality check for untransformed data of fruit drop from Sugar Belle trees on different rootstocks.

PAGE 55

55 Figure 4 2 0 Normality check for log transformed data of fruit drop from Sugar Belle trees on different rootst ocks. Table 4 Levene's Test for Homogeneity of fruitdroplog Variance ANOVA of Squared Deviations from Group Means Source DF Sum of Squares Mean Square F Value Pr > F group 14 0.1535 0.0110 0.82 0.6434 Error 378 5.0328 0.0133 Table 4 10. ANOVA table for fruit drop from Sugar Belle trees on different rootstocks at the McTeer trial. Source DF Sum of Squares Mean Square F Value Pr > F Model 14 14.31348059 1.02239147 13.23 <.0001 Error 378 29.20887768 0.07727216 Corrected Total 392 43.52235827

PAGE 56

56 Figure 4 2 1 Figure showing the relationship between fruit drop and yield on Sugar Belle trees on different rootstocks in the McTeer trial.

PAGE 57

57 CHAPTER 5 S T. HELENA EAST Results At St. Helena (Dundee, FL), in the East section of the trial, seven rootstocks of previously HLB affected trees were evaluated. The rootstocks were CLEO+CZO, CHANG+50 7, WGFT+50 7, BLUE 4, SO+50 7, BLUE 1 AND PURPLE 4 (Table2 1) The trees were planted in 9X20 (242 trees per acre) spacing. This section of the trial did not have cold protection, due to an inade quate water source. The scions used were Vernia and Valquarius sweet oranges, and replicates were planted in 4 tree rectangles. About 61 trees, known to be HLB positive for more than one year ( less than 10% of the trial trees), were evaluated here for the following parameters: visual HLB symptoms (subjective rating) Ct values from symptomatic and asymptomatic leaves visually estimated yield and fruit drop. Visual Symptoms (subjective rating scale) The subjective visual rating scale described previ ously was used to assess the severity of the HLB infection on the test trees. The Kruskal W allis test on the s ummer data in June 2012 (mean score per rootstock) revealed there were no statistically significant differences in visual symptoms in the seven rootstocks (Chi Square 11.5351, p value 0.0732). However examining the means, it was seen that the lowest (healthiest appearing trees) was for the CLEO+CZO somatic hybrid. The highest symptom mean was for tetrazyg rootstock Purple 4. There were no significant differences in mean visual symptoms in the f all data (December, 2012) as well (Chi square 6.6025, p value 0.3592) However, the lowest mean was again found in the CLEO+CZO somatic hybrid rootstock. As in s ummer data, the highest symptom mean was seen in tetrazyg rootstock Purple 4.

PAGE 58

58 A comparison of s ummer and f all rankings revealed very little difference between the s ummer and f all data (p value 0.0466) Although not statistically significant, the mean symptoms score increased in case of the somatic hybrid rootstock Changsha+50 7, remained unchanged for tetrazyg rootstock Blue 4, and surprisingly decreased for rootstoc ks CLEO+CZO, WGFT+50 7, SO+50 7, BLUE 1 and PURPLE 4. Generally, HLB symptoms are much stronger in the f all Due to nutritional effect s there could be growth of new leaves which may appear to reduce visual symptoms. Mean Ct Values from Symptomatic Leaves To assess the bacterial titers in symptomatic leaves of the HLB infected trees on the various rootstocks, mean Ct values were determined for each rootstock. The Ct values from symptomatic leaves of the seven rootstocks did not show any significant differe nce per rootstocks (Chi square 9.1298, p value 0.1664). The mean and the median were highest for tetrazyg rootstock Blue 4. The somatic hybrid rootstock Cleo+CZO also had some individual samples with higher Ct values. Analysis of the f all mean Ct values pe r rootstock showed significant differences in the rootstocks (Chi square 17.03, p value 0.0092). Both the somatic hybrid rootstock CLEO+CZO and tetrazyg rootstock BLUE 4 had samples with high Ct values, resulting in higher mean values. Th e s e data suggest that rootstock can influence the bacterial titer in the scion, even during the season when bacterial titers are generally higher. A comparison of s ummer and f all mean Ct values from symptomatic leaves A comparison of mean of Ct value in s ummer and f all revealed that for all the rootstocks the bacterial titer increased during f all except for the somatic hybrid WGFT+50 7, where it unexpectedly decreased.

PAGE 59

59 Mean Ct values from asymptomatic leaves HLB aff ected sweet orange trees were sampled in the s ummer and f all for Ct analysis. Data from the s ummer sampling showed significant differences among the mean Ct values for asymptomatic leaves from trees on the various rootstocks (Chi Square 13.74, p value 0.0326). Individual tree Ct values from asymptomatic leaves of rootstocks CLEO+CZO, BLUE 4 and WGFT+50 7 showed some samples with less bacterial titer. The median for these rootstocks is also higher. All the rootstocks showed some samples with less bacterial titer in f all The median Ct values were higher in WGFT+50 7, CLEO+CZO, Blue 4 and Chang+50 7. Again, this is unexpected. Estimated Fruit Yield and HLB Induced Fruit Drop As previously described, a visual estimation of yield was performed on each tree, with tr ees categorized as high, medium and low yielding. The cumulative mean of fruit drop was calculated for the trees on the various rootstocks. For the somatic hybrid CLEO+CZO, there was a linear relationship between yield and fruit drop, that is with higher yield fruit drop was higher but it could not be said for other groups. Examination of the mean fruit drop data per rootstock showed the trees on CLEO+CZO to have the least fruit drop. Fruit drop was significantly lower on this rootstock th an rootstocks CHANG+50 7 and PURPLE 4. The boxplot as well as the Tukey test revealed this difference. thus the data may be analyzed by A NOVA The A NOVA test with F value 5.53 and p value 0.0002 denotes there are significant differences in average fruit drop among the rootstocks. Conclusion Considering all the collected data for the east section of the St. Helena trial, it can be concluded that the somatic hybr id CLEO+CZO (which had the maximum number of samples)

PAGE 60

60 showed a slightly better performance regarding the response to HLB. For this rootstock the mean for visual HLB symptoms was lower in s ummer as well as f all Rootstock Purple 4 on the other hand showed more severe symptoms. BLUE 4 did not show any changes in ranking between s ummer and f all The symptoms decreased for all rootstocks during f all excepted for CHANGSHA+50 7, where they increased. The Ct value from symptomatic leaves showed BLUE 4 having the highest median and mean while CLEO+CZO showed some samples with high Ct values meaning lower bacterial titers. The same trend was also observed during f all with BLUE 4 and CLEO+CZO having some samples with high Ct values. In f all high Ct value s were observ ed in samples from CLEO+CZO, CHANG+50 7, WGFT+50 7 and BLUE 4. It was seen from the mean fruit drop and yield estimates that the highest mean fruit drop was observed in trees which had low yield. T hese trees probably had more fruits initially but with time had excessive fruit drop. Some trees with low yield had low fruit drop (between 0 and 5) but majority had fruit drop above 100, some even above 200. It may be said that fruit drop was more at St. Helena east than the McTeer farm, probably because sweet or ange is very susceptible and affected trees had most of the fruit on ground. This was quite different from what was observed at the McTeer farm and the Sugar Belle mandarin hybrid, where trees with higher yields also had higher numbers of fruit drop. The p erformance data (percentage of infection and lbs solids per box) of the different rootstocks are included in Table 4 8 and 4 9.

PAGE 61

61 Figure 5 1 Boxplot of visual symptom per rootstock from HLB aff ected sweet orange trees in the East section of the St. Helena trial, data from s ummer assessment (June, 2012, the diamond indicates the mean and the horizontal line indicates the median) Table 5 1. Mean visual HLB symptom assessment per rootstock for swee t orange trees in the East section of the St. Helena trial, s ummer (June 2012) and f all (December 2012) assessment, (Scale 0 4) Rootstock Number of trees Summer Mean Fall Mean CLEO+CZO 20 2.3 2.05 CHANG+50 7 11 2.36 2.45 SO+50 7 4 3.0 2.5 WGFT+50 7 10 2.7 2.6 BLUE 4 4 2.75 2.75 BLUE 1 7 3.0 2.83 PURPLE 4 6 3.17 3

PAGE 62

62 Figure 5 2. Boxplot of visual symptom means per rootstock from HLB infected sweet orange trees in the East section of the St. Helena trial, data from f all assessment (Scale 0 4) A) The horizontal line indicates the median and the diamond the mean. Table 5 2. Result of Wilcoxon signed rank test to compare s ummer and f all visual mean data. Tests for Location: Mu0=0 Test Statistic p Value Student's t t 2.012461 Pr > |t| 0.0487 Sign M 4.5 Pr >= |M| 0.0784 Signed Rank S 49.5 Pr >= |S| 0.0466

PAGE 63

63 Figure 5 3 A comparison of visual HLB symptom means per rootstock of sweet orange trees in the East section of the St. Helena trial, comparing s ummer (June 2012) and f all (December 2012) data Table 5 3 Mean Ct values from symptomatic leaves per rootstock in s ummer (August, 2012) and f all (October, 2012) for HLB infected sweet orange trees in the east section of the St. Helena trial. Rootstock Number of trees Mean Ct va lue ( Summer ) Mean Ct value ( Fall ) BLUE 4 4 30.28 26.64 CLEO+CZO 20 27.59 27.28 SO+50 7 4 26.33 26.02 CHANG+50 7 11 23.94 22.47 WGFT+50 7 10 23.53 24.76 PURPLE 4 6 22.7 20.83 BLUE 1 7 22.27 21.2

PAGE 64

64 Figure 5 4. Boxplot of m ean Ct values from HLB symptomatic leaves of sweet orange trees per rootstock, sampled in s ummer (August, 2012), the diamond indicates the mean and the line indicates the median

PAGE 65

65 Figure 5 5 Boxplot of m ean Ct values from HLB symptomatic leaves of s weet orange trees per rootstock, sampled in f all (October, 2012), the diamond indicates the mean and the horizontal line the median Figure 5 6. Comparison of mean Ct values from symptomatic leaves of sweet orange trees per rootstock in the east section of the St. Helena trial, s ummer versus f all data

PAGE 66

66 Table 5 4 Mean Ct values from asymptomatic leaves per rootstock in s ummer (August, 2012) and f all (October, 2012) for HLB infected sweet orange trees in the east section of the St. Helena trial. Rootstock Number of trees Mean Ct value ( s ummer ) Mean Ct value ( f all ) CLEO+CZO 20 39.31 32.79 BLUE 4 4 34.95 32.72 WGFT+50 7 10 31.87 33.17 CHANG+50 7 11 29.79 32.28 PURPLE 4 6 29.56 28.54 SO+50 7 4 29.47 27.00 BLUE 1 7 28.77 29.13 Figure 5 7 Boxplot of m ean Ct values from HLB asymptomatic leaves of sweet orange trees per rootstock, sampled in s ummer (August, 2012), the diamond indicates the mean and the horizontal line indicates the median

PAGE 67

67 Figure 5 8 Boxplot of m ean Ct values from HLB asymptomatic leaves of sweet orange trees per rootstock, sampled in f all (October 2012), the diamond indicates the mean and the line the median Figure 5 9 Visually estimated yield of HLB af fected sweet orange trees on different rootstocks in the east section of the St. Helena trial, categorized as high, medium, or low yielding trees, and trees with no fruit

PAGE 68

68 Figure 5 1 0. Boxplot of mean fruit drop for HLB infected sweet orange trees on different rootstocks in the east section o f the St. Helena trial. A) The diamond indicates the mean and the horizontal line indicates the median Table 5 5 Mean fruit drop from HLB infected sweet orange trees on various rootstocks in the east section of St. Helena east. A) According to the Tukey test Cleo+CZO is significantly different from Purple 4 and Chang+50 7 Rootstock Number of trees Fruit drop CLEO+CZO 20 25.85 BLUE 4 4 36.25 WGFT+50 7 10 70.2 SO+50 7 4 78.25 BLUE 1 7 91.67 PURPLE 4 6 130 CHANG+50 7 11 134.36

PAGE 69

69 Figure 5 1 1. Checking normality of data for fruitdrop from sweet orange trees in the east section of the St. Helena trial. Table 5 Levene's Test for Homogeneity of fruitdroplog Variance ANOVA of Squared Deviations f rom Group Means Source DF Sum of Squares Mean Square F Value Pr > F group 6 0.0466 0.00776 1.07 0.3922 Error 54 0.3918 0.00726 Table 5 7. ANOVA table for the fruit drop data from HLB infected sweet orange trees on various rootstocks in the east section of the St. Helena trial. Source DF Sum of Squares Mean Square F Value Pr > F Model 6 2.54568904 0.42428151 5.53 0.0002 Error 54 4.14068081 0.07667927 Corrected Total 60 6.68636985

PAGE 70

70 Figure 5 1 2 M ean fruit drop among low, medium and high yielding trees at St. Helena East Table 5 8. Early HLB infection percentages per rootstock at St. Helena East, as determined by the CREC scouts 3.5 years after planting (data provided by Jude W. Grosser, UF/CREC) Rootstock Number of trees in trial Infected Percentage Infected SO+50 7 46 4 8.7 PURPLE 4 65 6 9.23 BLUE 1 59 6 10.17 BLUE 4 38 4 10.53 CLEO+CZO 165 20 12.12 WGFT+50 7 60 10 16.67 CHANG+50 7 56 11 19.64

PAGE 71

71 Table 5 9. Yield and fruit quality data from sweet orange trees on various rootstocks in the east section of the St. Helena trial (data provided by Jude W. Grosser and Field Day handouts). Scion Rootstock Lbs Solids/Box Yield Boxes/Tree Yield Boxes/Tree Cumulative Yield 2012 2011 (35 months) 2012 (47 months) Vernia BLUE 1 5.71 0.5 0.84 1.34 Valquarius BLUE 1 4.75 0.5 0.31 0.81 Vernia CH+50 7 5.67 0.4 0.63 1.03 Valquarius CH+50 7 5.64 0.5 0.78 1.28 Vernia CLEO+CZO 5.88 0.5 0.75 1.25 Valquarius CLEO+CZO 5.52 0.4 0.62 1.02 Vernia BLUE 4 5.77 0.4 0.61 1.01 Valquarius BLUE 4 5.07 NS 0.25 0.25 Vernia WGFT+50 7 5.45 0.5 0.25 0.75 Valquarius WGFT+50 7 4.66 0.48 0.51 0.99 Vernia SO+50 7 5.92 0.4 0.48 0.88 Valquarius SO+50 7 4.25 0.4 0.41 0.81 Vernia PURPLE 4 5.23 0.4 0.25 0.65 Valquarius PURPLE 4 4.96 0.3 0.38 0.68

PAGE 72

72 CHAPTER 6 ST. HELENA WEST Results Known HLB aff ected trees on n ineteen different rootstocks were evaluated in the west section of the St. Helena trial, Dundee FL. Rootstocks evaluated were: the diploids Rough lemon, Kuharske Carrizo ( KCZ ) Volk amer lemon Swingle (controls), Pink 1802, Orange 1804, Aqua 1803, and allotetraploid s Green 7, Orange 13, Orange 3, Orange 4 Orange 19, Orang e 18, Orange 2, Orange 1, White 4, Orange 16 and allotetraploid somatic hybrids SO+CZO, WGFT+50 7 (2 1). The scion s used were Vernia and Valquari us sweet oranges The western section of the trial consists of two blocks, the northwestern and southwestern block s, each with different tree spacings: t he northwestern part was planted using a 15x 25 (116 trees per acre) spacing and the southwestern part was planted using 12x 20 (181 trees per acre). The enti re west section of the trial had cold prote ction via elevated microjets. The HLB affe cted trees on the nineteen rootstocks were evaluated on the basis of visual symptoms, Ct values visual estimation of yield and fruit drop. Visual Symptom Assessment As described previously, trees known to be HLB affected for at least one year were visually assessed for HLB symptom severity, using a subjective rating scale. There were significant difference s in the mean visual symptom data per rootstocks in the s ummer data (June, 2012), (Chi square 44.8892, p value 0.0003). The least severe symptoms were found in tetrazyg rootstocks Orange 19, Orange 16, and Orange 3 and the somatic hybrid SO+ CZO. Tetrazyg rootstock Orange 19 also had the least severe symptoms in the McTeer trial with the Sugar Belle scion The m ost severe symptoms were observed in tetrazyg rootstock Orange 2 and diploid Pink 1802 as shown in Table 6 1, Figure 6 1 and Figure 6 2.

PAGE 73

73 From the f all mean visual HLB symptom data ( Table 6 1 ), rootstocks Swingle, Orange 19, Orange 16 and Orange 3 showed the least symptoms and as with the s ummer data, more severe symptoms were seen in rootstocks Orange 2 and Pink 1802. Of the controls, Swingle had lesser HLB symptoms than Kuharske, rough lemon and Volk When a comparison of the s ummer and f all visual rankings was performed ( Figure 6 3), it was seen that for most rootstocks the symptoms increased during f all as expected. However, HLB symptoms remained the same for rootstocks Orange 16, Orange 13, Rough lemon, Orang e 2, Pink 1802 and Orange 1. F all HLB symptoms were less for Swingle, Orange 1804, Orange 18 and KCZ The mean Ct values from symptomatic leaves Analysis of the mean Ct value data from the s ummer leaf sampling indicated no significant differences in the rootstock effect. This was also revealed by the Kruskal Wallis test (Chi square 8.53 p value 0. 93 ). However for the f all data the Kruskal Wallis test revealed significant difference among rootstocks (Chi square 34.83, p value .0099) with Swingle and Orange 2 rootstocks showing increa sed mean Ct values and therefore lower bacterial titer; Swingle and Orange 2 had some individual tree samples with Ct value s higher than 30. Comparison of mean Ct value s from HLB symptomatic leaves in s ummer and f all revealed for most rootstocks the bacte rial titer increased during f all as expected Mean Ct values from asymptomatic leaves Analysis of the mean Ct value s from asymptomatic leaves during s ummer revealed no significant difference s among the rootstocks (Chi Square 20.93, p value 0.18). During s ummer the median was high for rootstocks Orange 19, Aqua 1803, Pink 1802, Orange 2, Volk, Green 7, and Swingle. Ct value s could not be detected in rootstock Orange 16. During f all a number of roo tstocks showed high Ct values, including rootstocks Orange 3 Aqua 1803, Pink 1802, Rough lemon, Orange 19, Orange 2, Volk, Green 7, and Swingle. Data show that for most rootstocks,

PAGE 74

74 the bacterial titer decreased in Fall ; however, increased titers were observed in rootstocks WGFT+50 7, Orange 19, Aqua 1803, Orange 3 and Orange 16. Fruit drop was high at St. Helena west Both KCZ and Rough lemon had trees with either no fruit or very few fruit thus the me an fruit drop was low for both rootstocks Rootstocks White 4 and Orange 16 had very high fruit drop. Moderate fruit drop was seen in rootstocks Pink 1802, Green 7, Orange 13 and Orange 3. The Tukey test revealed a significant difference between KCZ and Orange 16, White 4 and Orange 18 that is fruit drop in KCZ was significantly smaller than the others again at tributed to lack of fruit on KCZ The untransformed data for fruit drop at St. Helena West did not meet assumption of normality as shown by the qq plot. The data were therefore log transformed and the Levene s test done to test equality of variances. If th e p value is less than 0.05 one may conclude there is difference in the variance in the samples. Thus instead of A NOVA NOVA was performed to account for unequal variances and seen that there is significant difference in the fruit drop of different rootstocks (p value 0.0001 ) The mean fruit drop is maximum for trees with medium yield (Figure 6 14). Conclusions At St. Helena West, severity o f symptoms was l ower in rootstocks Orange 19 and Orange 16 both during s ummer and f all Thus, rootstock Orange 19 produced trees with the least amount of HLB symptoms at both the McTeer and St. Helena trials. Surprisingly, the industry standard rootstock Swingle had fewer symptoms during the f all The good performance of Swingle could possibly be due to a lack of a Phytophthora problem at St. Helena. There was no significant difference between Ct values from symptomatic leaves in s ummer al though duri ng f all some of the individual tree samples from rootstocks Orange 2 and Swingle showed less bacterial titer. On comparing the two seasons the bacterial titer increased from s ummer to f all for trees on most

PAGE 75

75 rootstocks, but decreased in case of rootstocks Or ange 2 and Swingle In the asymptomatic leaf data, the mean or median for Ct value was higher than 30 for most rootstocks. Bacteria could not be detected in the asymptomatic leaves of Orange 16 during s ummer however during f all it could be detected. Comparison between two seasons showed in asymptomatic leaves bacterial tite r decreased for most rootstocks, but increased for rootstocks Orange 19, Aqua 1803 and Orange 3. On comparing the mean fruit drop and yield, it was observed t hat the least average fruit drop was in the trees with lowest yield s, whereas highest average fruit drop was observed in trees on rootstocks that exhibited medium yield. The percentage of infection per rootstock is shown in Table 6 6.

PAGE 76

76 Table 6 1 Means of visual symptom severity in HLB af fected sweet orange trees on various rootstocks in the west section of the St. Helena trial, data collected in s ummer (June 2012) and f all (December 2012) Rootstock Number of HLB infected trees Visual Severity Mean ( s um mer ) Visual Severity Mean ( f all ) Orange 19 8 1.5 1.75 Orange 16 3 1.67 1.67 SO+CZO 15 1.73 2.33 Orange 3 4 1.75 2.0 Swingle 6 2 1.5 White 4 8 2.13 2.38 Orange 4 7 2.14 2.57 Orange 13 5 2.2 2.2 Volk 7 2.29 2.43 Green 7 9 2.33 2.56 Orange 1804 3 2.33 2.0 Orange 1 3 2.33 2.33 Orange 18 5 2.4 2.2 Aqua 1803 3 2.67 3.0 Rough Lemon 3 2.67 2.67 WGFT+50 7 5 2.8 3.0 KCZ 24 2.92 2.79 Orange 2 3 3.33 3.33 Pink 1802 3 3.33 3.33 (Scale 0 4 from least to most severely affected)

PAGE 77

77 Figure 6 1 Boxplot of mean visual symptoms per rootstock for HLB af fected sweet orange trees in the west section of the St. Helena trial, s ummer data (June 2012) A) The line is the median and the diamond is the mean. Figure 6 2 Boxplot of visual HLB symptom mean s of sweet orange trees on different rootstocks in the Western section of the St. Helena trial; data col lected in f all (December, 2012). A) The line is the median and the diamond is the mean.

PAGE 78

78 Figure 6 3 Comparison of visual HLB symptom means of sweet orange trees on various rootstocks in s ummer (June, 2012) versus f all (December, 2012) for the western section of the St. Helena trial. A) Numbers in parentheses indicate tree numbers. Table 6 2 Mean of Ct value s from HLB symptomatic leaves from HLB infected sweet orange trees on various rootstocks in the western section of the St. Helena trial, data collected in s ummer (August, 2012) and f all (October, 2012) Rootstock Number of trees Mean Ct value ( s ummer ) Mean Ct value ( f all ) SO+CZO 15 2 3.99 23.55 Orange 19 8 23.47 23.37 Aqua 1803 3 23.18 22.44 Green 7 9 23.17 21.31 Rough lemon 3 23.07 23.45 Orange 1804 3 22.99 20.99 Pink 1802 3 22.97 21.38 Orange 4 7 22.85 21.27 KCZ 24 22.53 22.16 Volk 7 22.47 21.88 White 4 8 22.34 20.66 Orange 3 4 22.22 21.23 Orange 13 5 22.11 20.58 Orange 18 5 21.97 21.17 Orange 16 3 21.87 21.44 WGFT+50 7 5 21.7 20.44 Orange 1 3 21.33 21.57 Orange 2 3 20.99 25.56 Swingle 6 20.77 27.27

PAGE 79

79 Figure 6 4. Mean Ct value s from HLB symptomatic leaves sampled from HLB af fected sweet orange trees on various rootstocks in the western section of the St. Helena trial, data collected during s ummer (August, 2012)

PAGE 80

80 Figure 6 5 Ct value from symptomatic leaves during f all (October, 2012) Figure 6 6 Comparison of s ummer and f all mean Ct value s from HLB symptomatic leaves sampled from HLB infected sweet orange trees on various rootstocks in the western section of the St. Helena trial.

PAGE 81

81 Table 6 3 Mean Ct value s from asymptomatic leaves sampled from HLB af fected sweet orange trees on various rootstock s data collected in s ummer (August, 2012) and f all (October, 2012) Rootstock Number of trees Mean Ct Value ( s ummer ) Mean Ct Value ( f all ) Orange 16 3 Not detected 28.32 Orange 3 4 37.22 31.97 Aqua 1803 3 37.12 36.46 Pink 1802 3 36 37.38 Rough lemon 3 32.95 35.06 Orange 19 8 32.67 32.2 Orange 2 3 32.28 33.47 Volk 7 30.17 31.01 Green 7 9 30.16 35.42 Swingle 6 30.04 36.57 Orange 1 3 28.23 35.53 KCZ 24 27.65 28.27 WGFT+50 7 5 27.43 27.11 SO+CZO 15 26.87 31.52 Orange 4 7 25.86 28.31 Orange 18 5 25.14 29.01 White 4 8 24.89 28.78 Orange 1804 3 24.49 31.45 Orange 13 5 23.99 32.47

PAGE 82

82 Figure 6 7 Boxplot showing mean Ct value s from asymptomatic leaves sampled from HLB af fected sweet orange trees on various rootstocks, data collected in s ummer (August, 2012) Figure 6 8 .C omparison of mean Ct value s from s ummer versus f all asymptomatic leaves sampled from HLB af fected sweet orange trees on various rootstocks in the western section of the St. Helena trial.

PAGE 83

83 Figure 6 9 Boxplot showing Ct value from asymptomatic leaves in f all (October, 2012)

PAGE 84

84 Figure 6 10 Boxplot showing mean fruit drop data for HLB infected sweet orange trees on various rootstocks in the western section of the St. Helena trial. A) Note that R ough lemon and Kuharsk e (KCZ) trees had very little fruit to start with, affecting the result.

PAGE 85

85 Table 6 4 M ean f ruit drop for sweet orange trees on different rootstocks in the western section of the St. Helena trial Name Number of trees Mean of fruit drop Rough Lemon 3 1.33 KCZ 24 8.33 Pink 1802 3 52.67 Green 7 9 63.78 Orange 13 5 75.5 Orange 3 4 75.5 Volk 7 81.43 Orange 4 7 83.29 Aqua 1803 3 89 SO+CZO 15 89.8 WGFT + 50 7 5 96.4 Orange 19 8 99.38 Swingle 6 102.33 Orange 18 5 110.8 Orange 2 3 121.33 Orange 1804 3 130.67 Orange 1 3 137 White 4 8 141.5 Orange 16 3 175.33 Figure 6 1 1 Visual estimate of yield from HLB af fected sweet orange trees per rootstock in the western section of the St. Helena trial. A) Numbers in parentheses indicate the number of trees.

PAGE 86

86 Figure 6 1 2 St. Helena West f ruit drop data without log transformation Figure 6 1 3 St. Helena West f ruit drop data with log transformation

PAGE 87

8 7 Table 6 5. NOVA for St. Helena West fruit drop data Levene's Test for Homogeneity of fruitdroplog Variance ANOVA of Squared Deviations from Group Means Source DF Sum of Squares Mean Square F Value Pr > F type 18 1.8122 0.1007 3.25 <.0001 Error 105 3.2575 0.0310 Welch's ANOVA for fruitdroplog Source DF F Value Pr > F type 18.0000 31.00 <.0001 Error 23.7273 Figure 6 1 4 Mean fruit drop and yield at St. Helena West

PAGE 88

88 Table 6 6 Table showing percentage of infection at St. Helena West (data provided by J.W. Grosser via the CREC scouts, 3.5 years after planting). Rootstock I nfected T otal % Orange 2 3 68 4.41 SO+CZO 15 264 5.68 Orange 19 8 129 6.2 Orange 13 5 50 10 White 4 8 72 11.11 Orange 4 7 63 11.11 Orange 3 4 35 11.43 Orange 16 3 26 11.54 Orange 1 3 24 12.5 Green 7 9 64 14.06 Aqua 1803 3 19 15.79 Orange 1804 3 18 16.67 WGFT+50 7 5 30 16.67 Pink 1802 3 18 16.67 RoughLemon 3 18 16.67 Orange 18 5 27 18.52 Swingle 6 20 30 Volk 7 20 35 KCZ 24 65 36.92

PAGE 89

89 CHAPTER 7 CONCLUSION The use of resistant or tolerant citrus plants could be the most economically feasible way of managing HLB disease. With this in mind, the objective of this field study was to look at the impacts of genetically unique rootstocks on disease in field trials conducted at two farms in Florida, the McTeer farm at Haines City and St. Helena farm near Dundee. The study included various conventional diploids, somatic hybrids and tetrazygs or crosses of two somatic hybrids. The scion was Sugar Belle at McTeer and Ve rnia and Valquarius at St. Helena farm. At the McTeer farm, where the HLB infection was nearly 100 per cent, Orange 19, a tetrazyg rootstock obtained from the cross of two somatic hybrids (Nova mandarin+Hirado Buntan seedling pummelo) X (Cleopatra mandarin +Argentine trifoliate orange) showed less severe symptoms than trees on all other rootstocks in summer and f all of 2012. Though some leaf samples (asymptomatic and symptomatic) from tetr a zyg rootstock Purple 2 showed Ct values higher than 30 indicating low er titers of Liberibacter in the plants the disease symptoms were severe. At McTeer, the yield was high and trees with higher yield showed more fruit drop, although overall fruit drop in the entire trial was quite low. Interestingly, at the other trial si te (St. Helena west section), the tetrazyg rootstock Orange 19 again showed less severity of symptoms. Unlike the McTeer farm however, the fruit drop was higher, with significant fruit drop seen in trees with medium yield. The reason for this difference between McTeer and St. Helena possibly could be that sweet orange scions Vernia and Valquarius (St. Helena scions) are more sensitive to HLB and showed more fruit drop than the mandarin hybrid Sugar Belle (the McTeer scion). At St. Helena (East section), the somatic hybrid rootstock Cleo+CZO (having the largest number of trees in the eastern section), showed a slightly better overall performance. The performance of Swingle citrumelo was better than the other commercial rootstocks included in the trial.

PAGE 90

90 As expected, some differences were observed in the performance of the HLB infected trees on the genetically diverse rootstocks included in the study. However it should be noted that the study was done over a short period of time (less than a year). Lo nger studies of these and other additional field trials along with greenhouse studies will be required to validate any conclusions.

PAGE 91

91 LIST OF REFERENCES 1. Albrecht U, McCollum G, Bowman KD. 2012. Influence of rootstock variety on Huanglongbing disease development in field grown sweet orange ( Citrus sinensis L. Osbeck) trees. Scientia Horticulturae 138 2. Aubert B. 1979. Progress in citrus greening control in ReunionProgres accomplis dans la lutte contre le greening des citrus a la Reunion. Revue Agric ole et Sucriere de l'Ile Maruice 58 3. Aubert B. 1979. Progress made in the control of citrus greening in ReunionProgres accomplis dans la lutte contre le greening des citrus a la Reunion. Revue Agricole et Sucriere de l'Ile Maurice 58 4. Aubert B, Bove JM Etienne J. 1980. Control of citrus greening disease in Reunion. Results and prospectsLa lutte contre la maladie du "greening" des agrumes a l'ile de la Reunion. Resultats et perspectives. Fruits 35 5. Bassanezi RB, Lopes SA, Belasque Junior J, Sposito MB Yamamoto PT, et al. 2010. Huanglongbing epidemiology and implications for disease management Epidemiologia do huanglongbing e suas implicacoes para o manejo da doenca. Citrus Research and Technology 31 6. Bassanezi RB, Montesino LH, Gasparoto MCG, Bergam in Filho A, Amorim L. 2011. Yield loss caused by huanglongbing in different sweet orange cultivars in Sao Paulo, Brazil. European Journal of Plant Pathology 130 7. Bassanezi RB, Montesino LH, Stuchi ES. 2009. Effects of huanglongbing on fruit quality of sw eet orange cultivars in Brazil. European Journal of Plant Pathology 125 8. Batool A, Iftikhar Y, Mughal SM, Khan MM, Jaskani MJ, et al. 2007. Citrus Greening Disease a major cause of citrus decline in the world a review. Horticultural Science 34 9. Bea ttie GAC, Holford P, Mabberley DJ, Haigh AM, Broadbent P. 2008. Australia and huanglongbing. Extension Bulletin Food & Fertilizer Technology Center 10. Belasque Junior J, Barbosa JC, Massari CA, Ayres AJ. 2010. Incidence and distribution of huanglongbin g in Sao Paulo state, BrazilIncidencia e distribuicao do huanglongbing no estado de Sao Paulo, Brasil. Citrus Research and Technology 31 11. Belasque Junior J, Bassanezi RB, Yamamoto PT, Ayres AJ, Tachibana A, et al. 2010. Lessons from Huanglongbing manage ment in Sao Paulo State, Brazil. Journal of Plant Pathology 92 12. Belasque Junior J, Yamamoto PT, Miranda MPd, Bassanezi RB, Ayres AJ, Bove JM. 2010. Huanglongbing control in Sao Paulo State, Brazil Controle do huanglongbing no estado de Sao Paulo, Brasil Citrus Research and Technology 31

PAGE 92

92 13. Boscariol Camargo RL, Cristofani Yaly M, Malosso A, Coletta Filho HD, Machado MA. 2010. Evaluation of different genotypes of citrus to Candidatus Liberibacter asiaticus infection Avaliacao de diferentes genotipos de citros a infeccao por Candidatus liberibacter asiaticus. Citrus Research and Technology 31 14. Bove JM. 2006. Huanglongbing: a destructive, newly emerging, century old disease of citrus. Journal of Plant Pathology 88 15. Bove JM. 2012. Huanglongbing and the future of citrus in Sao Paulo State, Brazil. Journal of Plant Pathology 94 16. Broadbent P, Fraser L, Beattie A, Grylls N, Duncan J. 1979. The Australian citrus dieback problem. Proceedings of the International Society of Citriculture 3 17. Cheema SS, Kapur SP, Chohan JS. 1982. Evaluation of rough lemon strains and other rootstocks against greening disease of citrus. Scientia Horticulturae 18 18. Coletta Filho HD, Carlos EF. 2010. Diagnosis of huanglongbing and detection of associated agents: from disease symptoms to laboratory tools Ferramentas para diagnostico de huanglongbing e deteccao de agentes associados: dos sintomas aos ensaios de laboratorio. Citrus Research and Technology 31 19. Coletta Filho HD, Targon MLPN, Takita MA, Negri JDd, Pompeu J, Jr., et al. 2004. First report of the causal agent of Huanglongbing (" Candidatus Liberibacter asiaticus") in Brazil. Plant Disease 88 20. Duan YP, Zhou LJ, Hall DG, Li WB, Doddapaneni H, et al. 2009. Complete genome sequence of citrus huanglongbing bac terium, Candidatus Liberibacter asiaticus' obtained through metagenomics. Molecular Plant Microbe Interactions 22 21. Folimonova SY, Achor DS. 2010. Early events of citrus greening (Huanglongbing) disease development at the ultrastructural level. Phytopa thology 100 22. Folimonova SY, Robertson CJ, Garnsey SM, Gowda S, Dawson WO. 2009. Examination of the responses of different genotypes of citrus to Huanglongbing (citrus greening) under different conditions. Phytopathology 99 23. Foolad MR, Ntahimpera N, C hrist BJ, Lin GY. 2000. Comparison of field, greenhouse, and detached leaflet evaluations of tomato germ plasm for early blight resistance. Plant Disease 84 24. Garnier M, Bove JM. 1983. Tranmission of the organism associated with citrus greening disease f rom sweet orange to periwinkle by dodder. Phytopathology 73 25. Gottwald TR. 2010. Current epidemiological understanding of citrus huanglongbing. Annual Review of Phytopathology 48

PAGE 93

93 26. Gottwald TR, Aubert B, Xue Yuan Z. 1989. Preliminary analysis of citrus greening (Huanglungbin) epidemics in the People's Republic of China and French Reunion Island. Phytopathology 79 27. Grosser JW, An H, Calovic M, Lee DH, Chen C, et al. 2010. Production of new allotetraploid and autotetraploid citrus breeding parents: foc us on zipperskin mandarins. HortScience 45 28. Grosser JW, Ananthakrishnan G, Calovic M, Serrano P, Chandler JL, et al. 2007. Applications of somatic hybridization and cybridization in scion and rootstock improvement, with focus on citrus. In Acta Horticul turae ed. RE Litz, R Scorza. Number of. 29. Grosser JW, Calovic M, Gmitter FG, Jr. 2012. In vitro breeding facilitates conventional breeding for scion and rootstock improvement in citrus. In Acta Horticulturae ed. D Geelen. Number of. 30. Grosser JW, Gmi tter FG, Jr. 2009. In vitro breeding provides new and unique opportunities for conventional breeding. In Acta Horticulturae ed. RJ Geijskes, P Lakshmanan, A Taji. Number of. 31. Grosser JW, Gmitter FG. 2011. Protoplast fusion for production of tetraploids and triploids: applications for scion and rootstock breeding in citrus. Plant Cell, Tissue and Organ Culture 104 32. Ha l bert S, Manjunath K, Roka F, Brodie M. 2008. Huanglongbing (citrus greening) in Florida, 2008. Extension Bulletin Food & Fertilizer Technology Center 33. Hajivand S, Thohirah Lee A, Kamaruzaman S, Siti Nor Akmar A. 2011. Potential use of selected citrus rootstocks and interstocks against HLB disease in Malaysia. Crop Protection 30 3 4 Halbert SE, Manjunath KL. 2004. Asian citrus psyll ids (Sternorrhyncha: Psyllidae) and greening disease in citrus: a literature review and assessment of risk in Florida. Florida Entomologist 87 3 5 Halbert SE, Nunez CA. 2004. Distribution of the Asian citrus psyllid, Diaphorina citri Kuwayama (Rhynchota: P syllidae) in the Caribbean Basin. Florida Entomologist 87 3 6 Hartung JS, Paul C, Achor D, Brlansky RH. 2010. Colonization of dodder, Cuscuta indecora, by Candidatus Liberibacter asiaticus' and Ca. L. americanus'. Phytopathology 100 3 7 Hors Fall JG, Ba rratt RW. 1945. An improved grading system for measuring plant diseases. Phytopathology 35 3 8 Hors Fall JG, Heuberger JW. 1942. Measuring magnitude of a defoliation disease of tomatoes. Phytopathology 32

PAGE 94

94 39 Jagoueix S, Bove JM, Garnier M. 1994. The phloem limited bacterium of greening disease of citrus is a member of the α subdivision of the Proteobacteria. International Journal of Systematic Bacteriology 44 4 0 Khan IA. 2007. Citrus genetics, breeding and biotechnology x + 369 pp. 4 1 Koizumi M. 19 95. Virus or virus like diseases of citrus in tropical and subtropical zones. JIRCAS International Symposium Series 4 2 Koizumi M, Prommintara M, Linwattana G, Kaisuwan T. 1997. Epidemiological aspects of citrus huanglongbing (greening) disease in Thailan d. JARQ, Japan Agricultural Research Quarterly 31 4 3 Lafleche D, Bove JM. 1970. Mycoplasmas in Citrus affected by greening, stubborn or similar diseases Mycoplasmas dans les Agrumes atteints de greening', de stubborn' ou de maladies similaires. Fruits d'Outre Mer 25 4 4 Li WB, Levy L, Hartung JS. 2009. Quantitative distribution of Candidatus Liberibacter asiaticus' in citrus plants with citrus huanglongbing. Phytopathology 99 4 5 Liang W, Meats A, Beattie GAC, Spooner Hart R, Jiang L. 2010. Conservation of natural enemy fauna in citrus canopies by horticultural mineral oil: comparison with effects of carbaryl and methidathion treatments for control of armored scales. Insect Science 17 4 6 Lin KH. 1956. Observations on yellow shoot of Citrus. Etiological studies of yellow shoot of Citrus. Acta phytopath. sinica 2 4 7 Lopes SA, Bassanezi RB, Belasque J, Jr., Yamamoto PT. 2008. Management of citrus huanglongbing in the state of Sao Pau lo, Brazil. Extension Bulletin Food & Fertilizer Technology Center 48 Mabberley DJ. 1997. A classification for edible Citrus (Rutaceae). Telopea 7 49 Machado MA, Locali Fabris EC, Coletta Filho HD. 2010. Candidatus Liberibacter spp., citrus huanglongb ing agents Candidatus Liberibacter spp., agentes do huanglongbing dos citros. Citrus Research and Technology 31 5 0 Madden LV, Hughes G, Bosch Fvd. 2007. The study of plant disease epidemics xiv + 421 pp. 5 1 Moll JN. 1977. Greening disease: fifty years o f research. Citrograph 62 5 2 Moll JN, Vuuren SPV. 1979. Greening disease in Africa. International Society of Citriculture: Proceedings of the International Society of Citriculture, 1977, Volume 3. 5 3 Muller GW, Yuki VA, Costa AS. 2011. Greening, a poten tial threat to the Brazilian citrus industry Greening: ameaca potencial a citricultura Brasileira. Citrus Research and Technology 32

PAGE 95

95 5 4 Nariani TK, Raychaudhuri SP, Viswanath SM. 1973. Tolerance to greening disease in certain citrus species. Current Science 42 5 5 Onagbola EO, Rouseff RL, Smoot JM, Stelinski LL. 2011. Guava leaf volatiles and dimethyl disulphide inhibit response of Diaphorina citri Kuwayama to host plant volatiles. Journal of Applied Entomology 135 5 6 Planet P, Jagoueix S, Bove JM, G arnier M. 1995. Detection and characterization of the African citrus greening liberobacter by amplification, cloning, and sequencing of the rplKAJL rpoBC operon. Current Microbiology 30 5 7 Roistacher CN. 2011. The Greening/Huanglongbing disease of citrus its dangers and current status worldwide. IOBC/WPRS Bulletin 62 5 8 Rouseff RL, Onagbola EO, Smoot JM, Stelinski LL. 2008. Sulfur volatiles in guava ( Psidium guajava L.) leaves: possible defense mechanism. Journal of Agricultural and Food Chemistry 56 5 9 Shah DA, Madden LV. 2004. Nonparametric analysis of ordinal data in designed factorial experiments. Phytopathology 94 6 0 Shyam S. 1998. Status of citrus decline in India: a review. Agricultural Reviews (Karnal) 19 6 1 Teixeira DdC, Saillard C, Eveillar d S, Danet JL, Costa PId, et al. 2005. Candidatus Liberibacter americanus', associated with citrus huanglongbing (greening disease) in Sao Paulo State, Brazil. International Journal of Systematic and Evolutionary Microbiology 55 6 2 Teixeira DdC, Wulff N A, Lopes SA, Yamamoto PT, Miranda MPd, et al. 2010. Characterization and etiology of the bacteria associated with huanglongbing Caracterizacao e etiologia das bacterias associadas ao huanglongbing. Citrus Research and Technology 31 6 3 Texeira DC, Ayres J, Kitajima EW, Tanaka FAO, Danet L, et al. 2005. First report of a Huanglongbing like disease of citrus in Sao Paulo State, Brazil and association of a new liberibacter species, Candidatus Liberibacter americanus", with the disease. Plant Disease 89 6 4 V uuren SPv, Moll JN. 1985. The influence of the rootstock on greening fruit symptoms. Citrus and Subtropical Fruit Journal 6 5 Zhou LJ, Gabriel DW, Duan YP, Halbert SE, Dixon WN. 2007. First report of dodder transmission of Huanglongbing from naturally inf ected Murraya paniculata to citrus. Plant Disease 91

PAGE 96

96 BIOGRAPHICAL SKETCH Sanghamitra Das was born in Kolkata, India. She received a B achelor of S cience with an honors degree with major in Botany from Lady Brabourne College, Kolkata, India. In 2010, she completed Master of Science with major in Botany from Annamalai University, Chidambaram, India In May 2010, she started the m aster s program in Horticultur al Sciences at University of Florida and completed her degree in August 2013.