1 STUDIES ON SOME FACTORS AFFECTIN G FLOWER BUD INDUCTION IN SWEET ORANGE ( Citrus sinensis OSBECK): COLD, DROUGHT AND REMOVAL OF TERMINAL BUDS By EDUARDO J. CHICA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007
2 2007 Eduardo J. Chica
3 ACKNOWLEDGMENTS I thank all the persons who enriched my acad emic and personal experience through my program, most of them also cont ributed in various ways to the completion of this project. Thank you for your patience, understand ing, support, and the countless things I learned from you.
4 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................3 LIST OF TABLES................................................................................................................. ..........6 LIST OF FIGURES................................................................................................................ .........7 ABSTRACT....................................................................................................................... ..............8 CHAPTER 1 LITERATURE REVIEW.......................................................................................................10 Citrus Flowering............................................................................................................... ......10 General Characteristics of Citr us Flower Buds and Shoots............................................10 Citrus Flower Bud Induction...........................................................................................12 External (environmental) factors regu lating flower bud induction in citrus............12 Internal factors regulating flower bud induction in citrus........................................13 Pruning and Flowering.......................................................................................................... .15 2 COMPARISON OF COLD AND DROUGHT FLOWER BUD INDUCTION IN ORANGE ( Citrus sinensis osbeck)........................................................................................17 Introduction................................................................................................................... ..........17 Materials and Methods.......................................................................................................... .18 Growth Chamber/Greenhouse Experiment I...................................................................18 Growth Chamber/Greenhouse Experiment II..................................................................19 Field Experiments............................................................................................................20 Results........................................................................................................................ .............22 Growth Chamber/Greenhouse Experiments....................................................................22 Field Experiments............................................................................................................23 Discussion..................................................................................................................... ..........24 3 EFFECT OF LATE FALL TIPPING OF SUMMER FLUSH ON SUBSEQUENT FLOWER BUD INDUCTION IN FLORIDA ORANGES....................................................39 Introduction................................................................................................................... ..........39 Materials and Methods.......................................................................................................... .40 Results........................................................................................................................ .............42 Discussion..................................................................................................................... ..........43 4 CONCLUSIONS....................................................................................................................54 LIST OF REFERENCES............................................................................................................. ..56
5 BIOGRAPHICAL SKETCH.........................................................................................................60
6 LIST OF TABLES Table page 2 Analysis of variance table of five flow ering variables of 4-year-old Valencia orange trees after exposure to diffe rent flower inductive conditions................................31 2 Growth Chamber Experiment II. Number of 4-year-old potted Valencia orange trees that flowered after expos ure to inductive conditions................................................33 2 Accumulation of days under cool and drought flower bud induction conditions for citrus in seasons 2005 and 2006 in Lake Alfred, Florida......................................36 3 Types of sprouts formed in late-fall tipped vigorous summer shoots of Hamlin and Valencia sweet orange trees under field conditions in two successive seasons..............49 3 Comparison of five flowering parameters between late-fall su mmer shoots tipped at bud position 4 (from the apex) and bud positions >4 of intact summer shoots Hamlin and Valencia sweet orange tr ees, under field conditions (season 2006 07)............................................................................................................................ ..........53
7 LIST OF FIGURES Figure page 2 Main effect plots of time under induction, temperature and water deficit on five flowering variables of potted 4-yea r-old Valencia orange trees.....................................30 2 Growth Chamber Experiment I. Interac tion plots Temperature x Water Deficit for five flowering variables of 4-year-o ld potted Valencia orange trees..............................32 2 Effect of winter drought on five flowering variables of field-grown citrus trees..............34 2 Effect of extra-drought on five flowering variables of field-grown Valencia orange trees (Season 2005)......................................................................................................35 2 Distribution of days under natural cool i nduction for citrus and available soil water content in non-irrigated pl ots in seasons 2005 and 2006.......................................37 2 Effect of two levels of drought on five fl owering variables of fi eld-grown Valencia sweet orange trees (Season 2006)................................................................................38 3 Primary flowering intensity parameters of late-fall tipped vigorous summer shoots of Hamlin and Valencia sweet orange tr ees under field conditions on two successive seasons........................................................................................................................ .......48 3 Distribution of different t ypes of sprouts on late-fall tipped and intact summer shoots of Hamlin and Valencia sweet ora nge trees, under field conditions............................50 3 Distribution of sprouts of late-fall-tipped and intact summer shoots of Hamlin and Valencia sweet orange trees on different bud positions (counting from the apex after tipping), under field conditions.................................................................................51 3 Percent distribution of different types of spring sprouts on the first 4 nodes of latefall tipped and intact vigorous summer s hoots of Hamlin and Valencia sweet orange trees under field conditions (season 200607).......................................................52
8 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science STUDIES ON SOME FACTORS AFFECTIN G FLOWER BUD INDUCTION IN SWEET ORANGE ( Citrus sinensis OSBECK): COLD, DROUGHT AND REMOVAL OF TERMINAL BUDS By EDUARDO J. CHICA August 2007 Chair: L. Gene Albrigo Major: Horticultural Science The citrus flower bud induction process in humid subtropical climates is different than in mediterranean climates in that inductive cool te mperatures in subtropical climates are slightly higher. In addition, winter time low temperatures start accumulating later in the fall and scarce winter rainfall may provide flow er-induction by water stress in subt ropical climates. Fruit yield in a given season is a function of several fact ors including flowering intensity. For this reason, agricultural practices to increase flowering intens ity are expected to increase yields to some degree. The combined effect of low ambient temp eratures and drought, and the effect of the late fall removal of terminal buds on flow ering intensity of sweet orange ( Citrus sinensis Osbeck) trees were studied. A significant interaction betw een ambient temperature and drought was detected in an experiment using potted trees. In this interac tion, drought caused an increase in flowering intensity at marginal inductive temperatures ( 23/18C, day/night) when co mpared with irrigated trees, whereas a negative effect of drought on flow ering intensity was obse rved at near optimal inductive temperatures (15/12C) Field grown trees subjected to water stress throughout the winter also flowered more in tensely than well watered contro ls, indicating that jointly both
9 stimuli increase flower induction in sweet or ange trees under Florid as Humid subtropical climate. In another experiment, late fall tipped shoots flowered less intensely than intact shoots. However, late-fall tipping increased sprouti ng and flowering of buds at bud positions along the shoot that otherwise would not have sprouted or flowered. Indirect ly it is concluded that shoots tipped in the fall flowered more intensely th an shoots tipped after bud differentiation starts (winter), suggesting that late fall hedging of commercial orchards may have less negative impact on flowering intensity than at other times of the year. Also, late fall tipping increased or maintained the leafiness of the inflorescences formed and the numb er of vegetative sprouts when compared with intact shoots. Thus, even though flowering intensity was reduced, percentage of fruit-set may increase because leaf ier inflorescences set more fruit than inflorescences with less number or leaves.
10 CHAPTER 1 LITERATURE REVIEW Citrus is commercially grown primarily in re gions with mild winter s located between 20 and 40 of latitude in both hemispheres. Climat es within these regions have a great influence on the physiology of citrus flowering. More than 70 % of the worlds total production of citrus for processing is grown in humid s ubtropical climates in Florida (US) and So Paulo (Brazil). Humid subtropical climates are characterized by cool and dry winter s (McKnight and Hess, 2000), which create unique conditions for flower bud induction in citrus. The purpose of this section is to review the main f eatures of citrus flowering as th ey relate to humid subtropical climates. Citrus Flowering Citrus flowering is highly influenced by envi ronment. Typically a single strong flush of flowers is produced each year following prol onged exposure to natural floral inductive conditions (Cassin et al., 1969). Two major environmental cues i nduce flowering in citrus: low non-freezing temperatures and water stress (Cassin et al., 1969). Winter low ambient temperatures are the major floral inductive stim uli for citrus grown in mediterranean climates, while in tropical climates, it is water stress (Davenport, 1990). Humid subtropical climates are unique in that it is possible for both floral inductive stimuli to be present simultaneously during the citrus flower bud induction season, thus crea ting possibilities for inter active effects on flower induction (Albrigo et al., 2004). General Characteristics of Citrus Flower Buds and Shoots Citrus seedlings start flowering after the completion of a juven ile stage that may last from 5 to 13 years depending on the cultivar (Davie s and Albrigo, 1994). Buds in newly formed shoots of adult trees require 6 weeks to beco me competent to flower (Albrigo, unpublished
11 data). Unlike those of temperate trees, citrus floral buds are not visible, even microscopically, until shortly before bloom (Abbot, 1935; Lord and Eckard, 1985). Flower bud differentiation starts with sepal initiation and ends with the form ation of carpels, in a process that lasts about 2 weeks (Lord and Eckard, 1985). Under California me diterranean-like conditions, anthesis occurs 3 months after this point, and flower buds are macroscopically noticeable two months prior to anthesis (Lord and Eckard, 1985). These times can be shorter in humid subtropical climates where late-winter and spring temperatures are higher than in mediterranean-like climates (Valiente and Albrigo, 2000). In tr opical climates, where the main inductive stimulus is water deficit, anthesis occurs about one month after the first precipitati on that replenishes adequate soil moisture to start grow th (Cassin et al., 1969). In mediterranean and humid subtropical clim ates, the spring flush accounts for almost all of the flowers produced in a given year (Dav enport, 1990). In tropical climates the major reproductive flush normally occurs after the dry season ends (Cassin et al., 1969), but if the region is wet and not prone to prolonged drought, trees may beco me ever-bearing (Reuther and Ros-Castao, 1969). Lemons and limes normally produce flowers throughout the year even in mediterranean climates, with specific peaks of flowering coinciding with those of other citrus species (Erickson, 1969). The spring flush is composed of vegetative and flowering shoots (Moss, 1969). Flowering shoots may be of five types depending on their l eafiness (Moss, 1969). Leafy inflorescences are commonly associated with higher fruit set in sweet oranges (Moss, 1970, Sauer, 1954). Inflorescence leafiness is inversely proportional to the duration of repr oductive temperatures and proportional to the temperature of induction a nd date of bud break (A lbrigo and Galn-Saco, 2004; Moss, 1969; Sauer, 1954). Flowering shoots are formed primarily in buds from the
12 previous years spring and summer flush (S auer, 1951; Valiente and Albrigo, 2004). Summer shoots tend to produce greater number of flowers per shoot (Valiente a nd Albrigo, 2004). Buds nearest to the apex of the shoot tend to become induced first and most frequently (Sauer, 1954; Valiente and Albrigo, 2004). Citrus Flower Bud Induction Though it has been widely studied, the flower induction process in citrus has not been satisfactorily explained. Flor al induction is the process by which competent vegetative meristems become committed to flower (McDanie l et al., 1992). There is no argument about the inductive role of low temperature and water st ress but very little is known about how these stimuli actually trigger the flowering response in citrus. From other species, it is known that a complex genetic network integrates internal and environmental cues that work as signals of conditions that favor reproductive success and triggers flowering (i.e. seed/fruit formation, cross pollination; Araki, 2001; Putterill et al., 2004). This section reviews the influence of external and internal factors regulating citrus flower induction. External (environmental) factors regula ting flower bud induction in citrus Citrus trees are photoperiod in sensitive (Lenz, 1969; Moss, 1969) and undergo floral induction primarily by exposure to cold temper atures or drought unde r natural conditions (Borroto and Rodrguez, 1977; Cassin et al., 196 9; Moss, 1969). Drought and cold temperatures can be considered as both qualitative and quantitative stimuli since the flowering response depends on the intensity of the stimulus (qualita tive: temperature range / water stress severity) and time of exposure (quantitative) (Moss, 1969; Southwick and Davenport, 1986). Even so, their inductive optima and thresholds are not clearly defined (Kra jewsky and Rabe, 1995). Flower buds can be induced in containerized pl ants by as little as 2 weeks of exposure to limiting water conditions and induction is more intense under severe water stress (Leaf water
13 potential -3.5MPa; Southwick and Da venport, 1986). Under field c onditions, periods between 55 and 70 days under moderate water stress are co mmonly regarded as optimal (Albrigo et al., 2004; Cassin et al., 1969). Inductive temperatures are c onsidered optimal between 10 C and 15 C (Moss, 1969), with a lower and upper limit of 5 C and 19 C, respectively (Garca-Luis et al., 1992; Moss, 1969; Valiente and Albrigo, 2000, 2004). Flowering intens ity is proportional to the accumulation of hours at inductive temperatures, an d induction can be observed after 2 weeks of daily exposure to the cold stimulus (Moss, 1969). Exposure to temperatures above 23 C for more than 5 days may interrupt the flower induction process and promote floral differentiation (Valiente and Albrigo, 2000). It is not clear whether water stress and low temperatures induce flowering by the same physiological mechanism. The number of investiga tions characterizing the flowering response of citrus to low temperatures is extensive (Garca-Luis et al., 1989; Lenz, 1969; Moss, 1969, 1971a, 1973b, 1976; Southwick and Davenport, 1986; Valient e and Albrigo, 2004), but drought stress effects on flowering are not as we ll characterized (Borroto and Rodrguez, 1977; Cassin et al., 1969; Koshita and Takahara, 2004; Nir et al., 1972; Southwic k and Davenport, 1986). Although results vary, field experiments indicate that effects of these stimuli might be additive under certain conditions (Albrigo et al., 2004). It has been proposed that cessation of vegetative growth could be by itself the condition required for flower bud inducti on (Monselise, 1985). Internal factors regulating flower bud induction in citrus The role of internal factors in citrus flow er bud induction has been regarded mainly as indirect (Krajewsky and Rabe, 1995). As opposed to external factors that actually induce flowering, internal factor such as plant hormones, carbohydrates, crop load, or nitrogen
14 metabolism are believed to modulate the character istics of the generative flush but may not actually induce flowering. Gibberellins are the only group of plant growth regulators that have a demonstrated effect on citrus flowering by reduci ng and delaying bud sprouting (Cooper and Peynado, 1958; GarcaLuis et al., 1986; Koshita and Takahara, 2004; M onselise and Halevy, 1964). This line of evidence is supported by reports of gibberellin synthesis inhibitors enhancing flowering (Davenport, 1990). However, effect s of gibberellins on flowering are not expressed unless trees are subjected to cold or drought induction tr eatments (Davenport, 1990), thus indicating a secondary role on flower bud induction. Environmentally induced hormonal responses ha ve been proposed to explain the common flowering response (Monselise, 1985). Low levels of GAs were found in citrus leaves during cold and drought induction conditions (Koshita et al., 1999; Koshita and Takahara, 2004), which is consistent with the inhibitory effect of GA on flowering. Conversel y, higher levels of ABA and auxins were found simultaneously (Koshita et al., 1999; Koshita and Takahara, 2004). Earlier work reported little involvement of the former two hormones on flowering control (Krajewsky and Rabe, 1995). Reduced flowering intensity is associated with crop load in the previous season (Goldschmidt and Golomb, 1982; Moss, 1971b; Syvertsen et al., 2003; Valiente and Albrigo, 2004). The reason for the inverse relationship be tween crop load and flowering has not been established (Valiente and Albri go, 2004). Some workers have sugge sted a role for carbohydrates to explain crop load effects on flowering (Gol dschmidt and Golomb, 1982), but other reports indicate a less relevant role of carbohydrates on flowering (G arca-Lus et al., 1988). An
15 alternative explanation for the e ffect of crop load on flowering is associated with increased gibberellin concentrations in fruit-be aring branches (Koshita et al., 1999). Similarly to gibberellins, th e accumulation of ammonia (NH3-NH4 +) in citrus leaves after exposure to low temperatures or drought stress has been proposed as an alternative mechanism for flower induction (Lovatt et al ., 1988). The proposed role of NH3-NH4 + is as a precursor for the biosynthesis of polyamines which have b een associated with floral promotion and morphogenesis in other species (Edwards 1986). Studies so far indicate that NH3-NH4 + only enhances the already established flowering ability and does not initiate flowering (Lovatt et al., 1988). Pruning and Flowering Commercial citrus are pruned periodically to control tree size and architecture so that productivity is maintained over time. Hedgi ng and topping are the most common pruning practices in commercial citrus orchards in humid subtropical climates. These two pruning practices non-selectively cut por tions of branches and promote growth of late ral buds (Davies and Albrigo, 1994). Pruning stimulates the growth of a new flush of shoots, but the nature and vigor of this flush varies depending on the time of the year, cultivar and severity of pruning (Bacon and Bevington, 1978). In medite rranean-like Australia, light pruning in spring or early summer is regarded as the most effective in tensity and pruning times (Bevington, 1980). Late winter or spring pruning nor mally is practiced after the risks of freezes has passed, and just before the spring flush (Davies and Albrigo, 2004). Pruning at this time of the year avoids both freezing damage to the new flush a nd removal of flowers and young fruits. However, spring pruning likely removes flow er buds that have been alrea dy induced (Davies and Albrigo, 2004). Yield reductions from this timing have been reported for spring pruning in Australia (Bacon and Bevington, 1978; Bacon, 1981). Pruning early in the summer would avoid the loss of
16 induced flower buds, but reduce the number of young fr uit, thus decreasing yields in that season (Bacon and Bevington, 1978). Pruning at later times in Australia, re duced yields the most when compared with other pruning times by delaying flushing until next spring and causing a mainly vegetative flush (Bevington, 1981). A mostly vege tative spring flush s uggest that the buds remaining in the pruned shoots did not respond to flower inductive temperatures during the winter for some reason. Another report indicates that pruning in fall may also stimulate a flush just prior to winter that might be killed by freezes in the winter or grow less vigorously due to lower temperatures (Moss, 1973a). Interestingly there are no comparative reports on the response of citrus trees pruned in late-f all or winter even though these are common times of pruning for commercial orchards in Florida.
17 CHAPTER 2 COMPARISON OF COLD AND DROUGHT FLOWER BUD INDUCTION IN ORANGE ( Citrus sinensis OSBECK) Introduction Predicting bloom intensity is an important component for assessing crop load in orange trees in a given season. A model for predicting bloom date in sweet orange trees has been designed and provides an accuracy of days, and estimates a general level of flowering intensity (Albrigo et al., 2006). Th is model is based on the effect of temperature on flower bud induction, bud growth initiation, a nd bud development using historic al weather data of Central Florida, a humid subtropical climate. Several m odels have been developed for predicting bloom date using weather data from regions with me diterranean climates (Bellows and Morse, 1986; Ben Mechlia and Carrol, 1989; Lovatt et al., 1984) These models may not be applicable under humid subtropical conditions due to marked differences in weat her conditions during the flower bud induction season. The presence of drought duri ng the citrus flower bud induction season is very common in humid subtropical climates, as opposed to mediterranean climates with relatively rainy winters. A recent study has shown evidence supporting some additive role between cold and drought flower bud induction in another humid subtropical climate: So Paulo, Brazil (Albrigo et al., 2004). However, it is not completely clear whether low temperatures and drought simultaneously modifies the flower ing response of sweet orange trees. Determining whether there is an interaction be tween temperature and drought in the flower bud induction would be useful in fine-tuning th e bloom prediction model for humid subtropical climates. This project tested the hypothesis that low temperatures and drought interact to induce flowering in sweet orange trees, thus modifying flowering intens ity. Therefore, the objective of this study is to determine whether low temperat ure and drought flower bud induction signals are additive, competitive or do not interact during the flower bud induction process in oranges.
18 Materials and Methods Growth Chamber/Greenhouse Experiment I Growth chamber/greenhouse experiments were conducted at the University of Floridas Citrus Research and Education Center (CREC) in Lake Alfre d, Florida. Young (4 year old) Valencia seedless orange trees grafted on Carriz o citrange growing in 45L pots filled with high organic peat growing media were used for the experiment. Before the beginning of the experiment all fruits were removed. The experiment was conducted using a complete ly randomized design with 4 replicates per treatment in a 4x2x2 factorial ar rangement. The factors evalua ted were: time under inductive conditions (3, 5, 7, and 9 weeks), temperature (15/12 and 23/18C, day/night; near-optimal inductive and marginally inductive respectively), and water deficit (drought and irrigated). Each replicate consisted of one tr ee and represented one indepe ndent experimental unit (64 experimental units total). At the beginning of the experiment, plants were transferred to 2 iden tical growth chambers with temperatures set according to the levels in dicated above. Illuminatio n was the same for both chambers averaging about 250 molm-2h-1 with a photoperiod of 12h/12h (day/night). The two levels of water deficit were ach ieved by applying polyet hylene glycol 8000 (4.44 gL-1; PEG 8000, Fisher Scientific) to the soil medi a via irrigation water. The desired level of depressed water status was later fine-tuned by modifying the freque ncy of irrigation and recycling the water that leached out of the pots. Midday stem water potential ( s) was used as an indicator of water stress and was measured weekly using the pressure chamber method (Kaufmann, 1968; Scholander et al., 1965) using covered leaves (Be gg and Turner, 1970). Average s readings in the wellirrigated plants were -0.79. 04 and -0.69.03MPa, in the cool and warm chamber, respectiv ely. Average reduction of s in drought treatments represented a
19 very mild water stress with average readings of -1.00.02 and -1.08.04MPa in the cold and warm chamber, respectively. After induction conditions were applied for the time indicated by the treatment, the plants were transferred to a greenhouse at >20C and PEG was washed out by applying excess water. Higher non-inductive temperatures and non-restricted irrigation were intended to stop the induction process and promote bud sprouting. The variables evaluated we re: number of available buds per shoot, number of new sprouts per shoot, number of inflorescences per shoot and total number of flowers per shoot. These variables were used to calculate the number of flowers per inflorescence and number of flowers per available bud. All the shoots from the prev ious summer and spring flushes were tagged on each tree, and at least 12 shoots pe r tree were present. The values of the variables were averaged for each experimental unit (tree) and these averages were used for analysis. Data were analyzed using ANOVA (MINITAB 14.20, Minitab Inc.). Th e variables, number of inflorescences per shoot, number of flowers per shoot and number of flowers per available bud were transformed to their square root values in order to achieve normality of their distributions. Growth Chamber/Greenhouse Experiment II A second growth chamber experiment was cond ucted six months afte r the conclusion of experiment I using the same group of plants. This experiment (II) was conducted using the same methodology as Experiment I but higher temperatur es and more severe water stress was used (102 gL-1 PEG). The levels of temperature used in this experiment were: 20/10C (inductive) and 27/21C (presumably non-inductive). Drought wa s more severe in this second experiment and caused severe water st ress in the plants ( s = -3.2MPa and -3.6MPa in drought treatments of the cool and warm chamber respectively).
20 Field Experiments Field experiments were conducted over two inductive season s (2005 and 2006) at CREC in Lake Alfred, Florida (28 N, 81 W). Each season three adult sweet orange [cv. Valencia (2 blocks) and Hamlin (1 block)] blocks and one grapefruit (cv. Marsh) block were used for the experiment. Sweet orange tr ees were growing in Candler sand (hyperthermic, uncoated Typic Quartzipsamments) while the grap efruit trees were growing in Apopka fine sand (loamy, siliceous, hyperthermic Grossarenic Paleudults). Experiments were conducted under a completely randomized block design with 5 replicates1 and two treatments (d rought and irrigation). Levels of the block factor were 2 sweet orange [Valencia (2 blocks) and Hamlin] and one grapefruit (Marsh) cultivar. Each replicate consisted of one tree, and measurements were taken on 20 shoots distributed on each side of the hedgerow. Irrigated treatments consisted of plants receiving adequate i rrigation throughout the induction season, while drought treatments consisted of plants that were deprived of irrigation just before the induction season started (m id-November) and remained un-irrigated for 75 days. In both seasons, one sweet orange plot (cv. Valencia) had an a dditional treatment which consisted of extra drought. Extra drought was ach ieved by covering the soil under the trees with a sheet of Tyvek (E. I. DuPont de Nem ours and Company, Rich mond, Virginia, US) and withholding irrigation as before. Depression of s was used as an indicator of water stress in the trees, and was monitored throughout the experime nt using the pressure chamber method using covered leaves (Kaufmann, 1968; Scholander et al., 1965). Maxima, minima and mean ambient temperature, as well as rainfall data were obt ained from the IFAS-FAWN weather station at 1 In the first year of data collection, 2 replicates from one of the blocks were lost, therefore, for that year the experimental design consisted of an unbalanced completely randomized block design.
21 Lake Alfred. The total number of hours unde r optimum inductive conditions (10CC), and total inductive conditions (<20C) were calculated from these data. Water balances of nonirrigated plots were calculated as follows: volumetric availabl e soil water holding capacity for Candler sand was calculated to a depth of 0.90 m by subtracting the volumetric soil water content at permanent wilting point (0.01%; soil matric potential -1.5MPa) from the volumetric soil water content at field capacity (0.08%; soil matric potential = -5-3MPa; Obreza et al., 1989). Volumetric soil water content at field ca pacity represented 100 % available soil water content (ASWC) and the volumetric soil water co ntent at permanent wilting point represented 0% ASWC. Available soil wate r content at the beginning of the experiment was 0.08% equivalent to field capacity (100% ASWC). Afte rwards, water use due to crop evapotranspiration was subtracted from the available soil water c ontent of the previous day and water input from precipitation (rainfall only) was added on daily basi s. Thirty-three percent and 50% depletion of the available ASWC were considered as referenc e values for calculating total number of days under moderate and severe drought (Obreza et al., 1997; Smajstrla et al., 1987). Bud differentiation date, after which additional accumulation of induction units was not effective, was obtained from the predictions of the Citrus Flowering Monitor (CFM) of the Decision Information System for Citrus ( University of Florida, http://www.minuetto.net/bloom ). The variables evaluated were the same as those described for the growth chamber/greenhouse experiment. Sampled shoots we re spring and summer shoots formed during the previous season. For each experimental unit, data was averaged and analyzed as indicated before, but in this case data did not need to be transformed.
22 Results Growth Chamber/Greenhouse Experiments The main effect of temperature significantly (p 0.05) increased flowering with lower temperatures, for all the variables evaluate d. Time under inductive conditions was also significant for increasing all the variables except for the number of flowers per inflorescence, whereas water deficit by itself did not affect any of the variables. In general, low temperatures and increased time of exposure to inductive conditi ons produced more intense flowering (Figure 2, Table 2). These observations are in agreement with previous reports so will not be further discussed here (Moss, 1969; Southwick and Davenport, 1986). The interaction between temperature and water deficit was significant (p 0.05) for all the variables except for the number of flowers per inflores cence. Water stress improved the flowering response as measured by all the variables at higher te mperatures whereas the opposite effect was observed at colder temperatures (Figure 2). Results of the growth chamber experiment II partially supported the results observed in experiment I. Unfortunately, many of the plants did not flower after exposure to the inductive treatments (Table 2.2). Therefore, it was not possibl e to evaluate the variab les statistically as in experiment I. While temperatures in this expe riment were expected to provide less flower induction to trees, we believe that uneven vegetative flushing before the start of the experiment and low carbohydrate levels in some plants mi ght have contributed to the extremely low flowering response observed. Nonetheless, most of th e plants that flowered were from the cooler temperature (20/10C), primarily the drought treatments. Furthermore, at 27/21C only trees subjected to drought flowered, and similarly to experiment I, th e peak in the number of trees flowering at both temperatures occurred at about 7 weeks. Again, these data should be interpreted with caution since they only represen t numerical differences. Also, the level of water
23 stress associated with the drought treatments was apparently exce ssive since severe leaf drop and drying of young twigs was observed, potentially another limitation in this experiment. The tendency towards more flowering w ith drought in the cooler chamber is contradictory to the first experiment but this may be because temperature in the cooler chamber in the second experiment was also slightly higher than that used in the first experiment. Field Experiments Inductive season 2005. Drought significantly increased (p 0.075) the response of most of the variables evaluate d (Figure 2). Drought increased by almost one unit the average number of sprouts and the aver age number of inflorescences per shoot in the sweet orange cultivars, but not in Marsh grapefruit. Trees fr om the drought treatments in all the cultivars produced about 2 more flowers per shoot than trees from the irriga ted treatments. The number of flowers per inflorescence was apparently less affected by the drought treatment (p=0.095). The main effect of blocks (cultivars) was always significant (p<0.05) but did not interact with drought. In general sweet orange cultivars (Valencia and Ham lin) had higher values for all the variables than Marsh grapefruit. Extra drought significantly (p <0.075) increased the average number of sprouts and inflorescences per shoot, as well as the aver age number of flowers per available bud. Trees subjected to extra drought (Val encia only) produced about on e more sprout and one more inflorescence per shoot than trees that were onl y deprived of irrigati on (Figure 2). Again, the variable number of flowers per inflorescence was less affected by the additional water stress treatment. Table 2 and Figure 25 summarize weather co nditions relevant to flower bud induction in the two seasons of the e xperiment. Season 2005 accumulated 7 days more of combined induction than season 2006. In season 2005, 33% depletion of the ASWC occurred six
24 days before the bud differentiati on date predicted by the CFM. Tw o days after the predicted bud differentiation date, depletion of the ASWC was 50%, and drought continued afterwards for 24 days more until irrigation resumed. Total rainfall accumulation during this experiment was approximately 48mm. Inductive season 2006. No significant differences were found between drought and irrigated treatments in the experiment, however some significant differences were found in extra drought treatments applied to Valencia trees (Figure 2). Again, the ef fect of cultivars was significant but did not interact with water defi cit. Extra drought signi ficantly increased the average number of sprouts and the average numbe r of inflorescences per shoot, but significantly reduced (P<0.075) the number of flowers per infl orescence. The number of flowers per shoot and number of flowers per available bud were not affected by water st ress in all the blocks. Weather conditions for season 2006 were al ready presented in Table 2. In this season, days of simultaneous cold and drought in ductive conditions started accumulating prior to the predicted date of bud differentiation, but occu rred in two intervals of no more than 7 days interrupted by days with rain and low cold induction. Total rainfall accumulation during this experiment was approximately 169mm. Discussion Results from the first growth chamber experime nt indicate that both factors, temperature and water deficit, can interact during the flower bud induction pr ocess in citrus. Due to this interaction, trees subjected to drought responde d differently when exposed to optimal or marginal inductive temperatures (Figure 2). Based on the main effect of the factors, the values of the variables were inversely proportional to te mperature and water deficit. However, since the effect of the interaction was significant, simple additivity of the effects of temperature and water deficit alone can not explain the response obser ved. Apparently, drought e nhances flowering at
25 warmer temperatures, whereas it decreases floweri ng at cold temperatures It is of practical interest that water deficit and temperature may inte ract to enhance flowering at temperatures that are marginally inductive. In s eason 2005 trees exposed to drought in the field flowered more intensely than those that were irrigated (Figure 2) possibly confirming the positive effect of drought detected at marginal inductive te mperatures in the growth chamber. Severe drought (ASWC<50%) is apparently not required to enhance flowering intensity in field grown trees exposed to natural induc tive conditions. In season 2005, severe drought occurred only after the predicted bud differentia tion date and therefore it may not have been necessary to trigger the enhanc ing effect observed in drought treatments since buds were already determined to flower. Days of simulta neous cold induction and moderate drought (50%
26 prior to bud differentiation support the hypothesis that 7 days of combined drought and low temperatures is not enough time to produce a respons e and that flower induc tive drought starts at levels of ASWC higher than our reference value fo r moderate drought (66%). It is interesting to note that in this season trees pr oduced more flowers and inflorescen ces than in the previous year, despite equal or reduced accumulation of cool or drought induction. Since the shift was synchronized in all the blocks, this is might have been due to year-to-ye ar climatic variations affecting tree vigour, but more specifically previous-season crop load and carbohydrate levels might have also been a factor. Further evidence supporting a ro le of drought in enhancing flowering above the cool temperature level comes from the experiments in which trees were subj ected to extra drought by covering the soil with an impermeab le sheet of Tyvek to exclude rain (Figures 2 and 2). In this treatment more days in which water stre ss and cool induction oc curred simultaneously should have occurred as well as earlie r accumulation of induction units (before bud differentiation date) since soil water was not replenished by s poradic rains. The number of sprouts and inflorescences per s hoot in the extra drought treatmen t were clearly increased when the soil was covered. This response was consiste nt between both years for the number of sprouts and inflorescences per shoot. However, fewer flow ers per inflorescence were produced in extra drought treatments during season 2006 a nd this numerically reduced (though not significantly) the overall le vel of flowering intensity, as meas ured by the number of flowers per available bud, when compared with irrigated trees. The effect of water stress enhancing flower ing at warmer inductive temperatures may be inherently related with the fl ower bud induction process and its internal mechanisms (Lovatt et al., 1988). Nevertheless, what is not clear is why drought reduce d flowering at optimal cold
27 inductive temperatures in the growth chamber. One hypothesis is that trees became overstressed when both stimuli were present at the same ti me. However, water stress applied in the first growth chamber experiment was very mild as s was never less than -1.1MPa, no more than 0.3MPa below that of irrigated trees, and no visi ble symptoms of wilting were observed. It is interesting to note that this level of s was almost the same as that registered in field trees on which irrigation was withhold through the seas on. Since ambient temp eratures during the induction season in humid subtropical climates usually are not within the optimal inductive temperature range, this level of water stress was apparently adequa te to enhance flowering in the field under Florida conditions. It is known that greater accumulation of co ld or drought induction hours leads to an increased number of total sprouts and inflores cences (Moss, 1969; Southwick and Davenport, 1986). Therefore, the increase in number of sprouts and inflores cences reported for these field experiments suggests that buds of drought treat ments perceived more induction units than irrigated trees (Figures 2, 2 and 2). A simila r increase occurred in the growth chamber experiment providing evidence supporting the hypothesis of increased perception of the inductive stimuli. On the other hand, the number of flowers per infloresce nce was the variable that was least sensitive to combinations of cold and water stress (the only exception is in Figure 2B). This indicated that the number of spr outs and inflorescences produced may be more important than the number of flowers per inflor escence in determining the level of flowering intensity in a given season. Nevertheless, the number of flowers per inflores cence is an imperfect indicator of the type of inflorescence formed, and infl orescence leafiness is another indicator of the level of induction perceived by the buds (Moss, 1969). Leafy and l eafless inflorescences may often bear the same
28 number of flowers, and a reduction in the number of flowers per inflorescence may more likely imply a transition toward leafier inflorescences and thus slightly lowe r perceived level of induction. The study reported here did not evaluate inflorescence leafiness, so small differences between irrigated and drought treatments in flower induction may not have been detected in this aspect. Future studies sh ould evaluate inflorescence leafiness to assess more precisely the level of induction perceived by the buds. From a practical standpoint, l eafier infloresce nces (probably related with lower levels of induction) are more desirable since they set more fruit than leafless inflorescences. Growth chamber experiments indicated that wate r stress and temperature can interact in the flower induction process in citrus This interaction had a positive effect at warmer marginal inductive temperatures and a ne gative effect at colder optimal inductive temperatures. When compared to irrigated trees, water stress increas ed the flowering respons e variables between 10% and 130% at warmer temperatures, but reduced flowering by 30% under lower near-optimal inductive temperatures. Field e xperiments support a role for wa ter stress enhancing flowering intensity at warmer inductive temperatures. The average number of sprouts and inflorescences per shoot had the greatest gain (about 40%) in flowering due to drought for all the variables evaluated in the field experiments. At least unde r our experimental conditi ons, covering the soil with an impermeable sheet to exclude rain fu rther increases floweri ng by 20% when compared with trees for which only irrigation was withhel d. Based on the results of the growth chamber experiment, this increase was probably caused by greater accumulation of days in which cold and drought occurred at the same time. Field trials using imperm eable sheets to cover the soil and different irrigation program s would be useful to elucid ate whether this hypothesis is supported. The negative effect of drought reported at lower optimal induction temperatures
29 indicates that drought conditions during the inductive season may not be as useful to stimulate flowering in climates where high accumulation of cool temperatures occurs (i.e. mediterranean climates). However, for humid subtropical c limates, where temperatur es during winter are relatively warm; withholding irrigati on in winter may be useful to increase flowering intensity in commercial groves. Winter precipitation may pose a practical limitation for application of water stress in the field, as noted in the results of season 2006. Evidence reported here indicates that the reported increase in flowering intensity caused by drought may be due to increased sprouting and number of infl orescences rather than incr eased number of flowers per inflorescence. Experiments with more levels of inductive and non-inducti ve temperatures would be helpful to better characterize the interacti on of cold and drought in flower bud induction.
30 Mean Sprouts/shoot 9 7 5 3 2.8 2.4 2.0 1.6 23C / 18C 15C / 12C Drought Irrigated Time under Induction Temperature Water AvailabilityWeeks Inflorescences/shoot 9 7 5 3 1.6 1.2 0.8 0.4 23C / 18C 15C / 12C Drought Irrigated Time under Induction Temperature Water AvailabilityWeeks Mean Flowers/shoot 9 7 5 3 5 4 3 2 1 23C / 18C 15C / 12C Drought Irrigated Time under Induction Temperature Water AvailabilityWeeks Flowers/Inflorescence 9 7 5 3 2.6 2.4 2.2 2.0 23C / 18C 15C / 12C Drought Irrigated Time under Induction Temperature Water AvailabilityWeeks Flowers/Available Bud 9 7 5 3 0.8 0.6 0.4 0.2 23C / 18C 15C / 12C Drought Irrigated Time under Induction Temperature Water Availability WeeksFigure 2. Main effect plots of time under induction, temperatur e and water deficit on five flowering variables of potted 4 year old V alencia orange trees. A. Average number of sprouts per shoot. B. Average number of inflorescences per shoot. C. Average number of flowers per shoot. D. Average number of flowers per inflorescence E. Average number of flowers per available. Main effects of time under induction and temperature were always significant (P 0.05) except for the effect of in time of induction on the number of flowers per inflor escence (P=0.16). Main effect of water deficit was never signif icant according to ANOVA. A B C D E
31 Table 2. Analysis of variance table of five flowering variables of 4-year-old Valencia orange trees after exposure to different flower inductive conditionsy. Sequential mean squaresP-value Source df Sprouts per shoot Inflorescences per shoot Flowers per shootz Flowers per inflorescencez Flowers per available budz Time of induction (A) 3 1.830.05 0.88<0.01 2.780.01 1.210.16 0.420.01 Ambient temperature (B) 1 25.75<0.01 9.05<0.01 26.99<0.01 3.610.02 4.44<0.01 Water deficit (C) 1 0.330.49 0.010.80 0.010.89 0.160.63 0.000.89 A B 3 1.200.16 0.370.10 0.810.29 0.090.93 0.160.18 A C 3 0.060.96 0.060.76 0.290.71 0.370.64 0.040.73 B C 1 3.270.03 1.080.01 2.830.03 0.010.91 0.340.06 A B C 3 0.490.53 0.160.41 0.860.26 0.230.78 0.090.42 Error 40 0.66 0.17 0.63 0.66 0.09 Total 55 yTime of induction (3, 5, 7, and 9 weeks), ambient temperature (15/12C and 23/18C, day/night), water deficit (irrigated and drought). zANOVA with square root transformed values due to non-normality of residuals of the original data.
32 A Average Sprouts/Shoot Drought Irrigated 3 2 1 15C / 12C 23C / 18C B Average Inflorescences/Shoot Drought Irrigated 2 1 0 15C / 12C 23C / 18CC Average Flowers/Shoot Drought Irrigated 8 7 6 5 4 3 2 1 0 15C / 12C 23C / 18CD Average Flowers/Inflorescence Drought Irrigated 3.5 3.0 2.5 2.0 1.5 15C / 12C 23C / 18C E Average Flowers/Available Bud Drought Irrigated 1.25 1.00 0.75 0.50 0.25 0.00 15C / 12C 23C / 18C Figure 2. Growth Chamber Experiment I. Inte raction plots Temperature x Water Deficit for five flowering variables of 4-year-old potted Valencia orange trees. A. Average number of sprouts per shoot. B. Average number of inflorescences per shoot. C. Average number of flowers per shoot. D. Average number of flowers per inflorescence E. Average number of flower s per available bud. Values are means and bars are SE (n=8). Reference dashed line in the y axis is the grand mean of the variable (n=64). All interactions presente d were statistically significant (P<0.05) according to ANOVA, except for the number of flowers per inflorescence (P=0.91)
33 Table 2. Growth Chamber Experiment II. Numb er of 4-year-old potte d Valencia orange trees that flowered after e xposure to inductive conditionsy. Time of Exposure to In ductive Conditions (weeks) Temperature, Plant Water Status 3 5 7 9 20/10C, Droughtz 2 3 4 3 20/10C, Irrigated 2 2 27/21C, Drought 2 1 27/21C, Irrigated y Numbers are counts out of a total of 4 trees fo r each treatment. Number of flowers ranged from 1 95 flowers per tree. Hyphens indicate that all the trees in that treatment produced only vegetative shoots afte r floral induction. z Plants of the drought treatments had midda y stem water potential readings 2.7.0MPa below those of irrigated plants. Stem water potential of irrigated plants ranged between -0.46 and 0.64MPa.
34 Season 2005 Sprouts or In f lorescences/Shoot Inflorescences Total Sprouts Marsh Valencia1 Hamlin Marsh Valencia1 Hamlin 5 4 3 2 1 0 Irrigated Drought P < 0.01 P < 0.01 A Season 2006 Sprouts or Inflorescences/Shoot Inflorescences Total Sprouts Marsh Valencia2 Valencia1 Hamlin Marsh Valencia2 Valencia1 Hamlin 5 4 3 2 1 0 Irrigated Drought P = 0.95P = 0.84Season 2005 Flowers Flowers/Available bud Flowers/Inflorescence Flowers/Shoot Marsh Valencia1 Hamlin Marsh Valencia1 Hamlin Marsh Valencia1 Hamlin 20 15 10 5 0 Irrigated DroughtP < 0.01 P < 0.01 P = 0.10 Season 2006 Flowers Flowers/Available bud Flowers/Inflorescence Flowers/Shoot Marsh Valencia2 Valencia1 Hamlin Marsh Valencia2 Valencia1 Hamlin Marsh Valencia2 Valencia1 Hamlin 20 15 10 5 0 Irrigated DroughtP = 0.47 P = 0.68 P = 0.59 B Figure 2. Effect of winter drough t on five flowering variables of field-grown citrus trees. A. Total number of spring sprouts and inflores cences. B. Average number of flowers per shoot, average number of flowers per inflor escence, and average number of flowers per available bud. Error bars are SE, n=5 fo r all cultivars except for the control of Valencia1 in season 2005 where n=3. P-values provided to assess significance of the difference between treatments (irrigate d vs. drought) for each variable according to ANOVA. Differences among blocks (cultiv ars) were always significant (P<0.05).
35 Sprouts or Inflorescences/Shoot Inflorescences Total Sprouts 4 3 2 1 0 Drought Extra DroughtP = 0.05 P = 0.06 A. Flowers Flowers/Available bud Flowers/Inflorescence Flowers/Shoot 12 10 8 6 4 2 0 Drought Extra DroughtP = 0.12 P = 0.96P = 0.05 B. Figure 2. Effect of extra-drought on five flowering variables of field-grown Valencia orange trees (Season 2005). A. Average number of total spring sprouts and inflorescences per shoot. B. Average number of flowers pe r shoot, average number of flowers per inflorescence, and average number of flower s per available bud. Error bars are SE, n=3 and n=4 for drought and extra drought tr eatments respectivel y. P-values provided to assess significance of the difference be tween treatments (drought vs. extra drought) for each variable according to ANOVA.
36 Table 2. Accumulation of days under cool and drought flower bud induction conditions for citrus in seasons 2005 and 2006 in Lake Alfred, Floridaw. <15Cx 15CC >20C Sum 2005 <50%ASWCy 11(0)z 9(0) 6(0) 26(0) 50%%ASWC 3(2) 4(3) 2(2) 9(7) > 66%ASWC 17(17) 17(17) 10(10) 44(44) Sum 2005 31(19) 30(20) 18(12) 79(51) 2006 <50%ASWC 0(0) 1(0) 0(0) 1(0) 50%%ASWC 3(2) 16(9) 3(1) 22(12) > 66%ASWC 20(11) 17(12) 24(22) 61(45) Sum 2006 23(13) 34(21) 27(23) 84(57) w Season 2005 from 11/15/05 to 02/01/06 ( 78 days). Season 2006 from 11/14/06 to 02/05/07 (83 days). x Days with at least 12h at th e indicated temperature range. y ASWC = Volumetric available soil water content. z Numbers in parenthesis indicate days accumulate d from the beginning of the experiment until the predicted date of bud differentiation for each category.
37 Season 2005 %ASWC 1/29/2006 1/14/2006 12/30/2005 12/15/2005 11/30/2005 11/15/2005 100 75 50 25 0 66 01/04/06 Season 2006 %ASWC 1/28/2007 1/13/2007 12/29/2006 12/14/2006 11/29/2006 11/14/2006 100 75 50 25 0 66 01/09/07 Figure 2. Distribution of days under natural cool induction for citrus and available soil water content in non-irrigate d plots in seasons 2005 and 2006. Patterned bars indicate days with temperatures <20C fo r at least 12 hours (days of cool induction). Reference lines in the y axis represent m oderate (66%) and severe drought (50%). Arrowheads indicate the predicted date of flower bud differentiation for the strongest wave of flowering of the season.
38 Sprouts or Inflorescences/Shoot Inflorescences Total Sprouts 8 7 6 5 4 3 2 1 0 Irrigated Drought Extra Droughtab b b b aA. Flowers Flowers/Available bud Flowers/Inflorescence Flowers/Shoot 20 15 10 5 0 Irrigated Drought Extra Drought a ab b a a a a aaB. Figure 2. Effect of two levels of drought on five flowering variab les of field-grown Valencia sweet orange trees (Season 2006). A. Aver age number of total spring sprouts and inflorescences per shoot. B. Average number of flowers per shoot, average number of flowers per inflorescence, and average num ber of flowers per available bud. Error bars are SE, n=5. Different letters above the bars indicates st atistical difference according to Fishers protected LSD procedure ( =0.075).
39 CHAPTER 3 EFFECT OF LATE FALL TIPPING OF SU MMER FLUSH ON SUBSEQUENT FLOWER BUD INDUCTION IN FLORIDA ORANGES Introduction Hedging and topping in citrus may induce regrowth flushes composed of only vigorous vegetative shoots without flower s if performed too intensely or followed by warm weather (Bevington, 1980; Bacon, 1981). In addition, more frequent and intense vegetative regrowth increases the susceptibility of citrus trees to pest and diseases, including citrus canker ( Xanthomonas axonopodis pv. citri ) and greening ( Candidatus Liberibacter, Huanglongbing, HLB) (Graham et al., 2004; Halbert and Manj unath, 2004). Limiting fl ushing frequency and coordinating its timing and intensity within a region was proposed to help improve pesticide coverage and thereby decrease susc eptibility of the trees to thes e diseases (Albrigo et al., 2007). Light pruning of the outer 0.2m of the canopy in spring just prior to flowering is regarded as the most effective pruning time and intensity for commercial citrus orchards in the mediterranean climate of Australia (Bacon a nd Bevington, 1978; Bevingt on, 1980). At this time of the year, the risk of freezes killing the new flus h is low, as well as the potential for flower and fruit losses (Davies and Albrigo, 1994), at least for early and midseason varieties. Nonetheless, early spring pruning may remove a number of indu ced flower buds, thus re ducing that seasons flowering intensity and increasing the producti on of vegetative shoots (Davies and Albrigo, 1994). In mediterranean climates, fall hedging produ ces less vigorous regrowth and reportedly reduced fruitfulness of the regrowth than spring pruning (Bacon and Bevington, 1978; Bacon, 1981). Reduced vigour of this flush has been rela ted with the onset of low temperatures during winter that limits vegetative growth (Moss, 1973a) On the other hand, reduced fruitfulness may be a consequence of immature fall flush meristems from the 1st regrowth not being able to
40 respond to floral-inductive winter temperatures (incompetence to fl ower). Citrus shoots require 6 weeks of development in order for their buds to become fully responsive to floral induction (Albrigo, unpublished data). If he dging occurred close enough to the onset of floral-inductive winter temperatures, the occurr ence of a fall flush would be a voided and only mature buds would be subjected to floral inducti on. Precise timing of this program may be more important in humid subtropical climates, such as Fl orida (U.S.A.) or So Paulo (Br azil) in which accumulation of hours below the floral-inductive threshold (20 C, Moss, 1969; Valiente and Albrigo, 2000) is less than in mediterranean climates. Also, Valie nte and Albrigo (2004) noted that bud positions 1 (counting from the apex) produce most of th e flowers in a given season (>80%), supporting previous views of apical domi nance influencing flower bud induc tion (Lovatt et al., 1984). Since more basal bud positions are still able to flower but at a much lesser intensity, these buds are competent to flower but may be kept from doing so by apical dominance effects of distal buds during the floral-induction process. In this study, we tested th e hypothesis that removal of the four most apical buds of moderately vigorous summer shoots (>8 nodes long) just prior to the onset of floral-inductive temperatures (mimicking a light late-fall hedging) changes flowering intensity in reference to intact shoots but does not precludes flowering If true, then tipping tran sfers flowering to the new terminal buds (formerly less inducible basa l buds) during next springs flush. The objective of this study was therefore to determine if late fall tipping modifies the flowering intensity of the next spring flush. Results of this study shoul d provide useful information for the further exploration of hedging timing for horticultural and pest-management purposes in citrus. Materials and Methods A field experiment was conducted using adult Valencia and Hamlin sweet orange trees ( Citrus sisnesis Osbeck) growing in Candler sa nd (hyperthermic, uncoated typic
41 Quartzipsamments). The two adjacent experimental plots with similar characteristics were located at the University of Fl oridas Citrus Research and Edu cation Center in Lake Alfred, Florida (28 N, 81 W). The experiment was conducted in two successive seasons at the same location (2005 and 2006). Experiments were conducted using a randomi zed block split-plot design with 10 and 6 replicates in season 2005 and 2006, respectively. Individual trees were used as blocks and were nested within the main plot. Each experiment al unit consisted of half of all the vigorous (>8 nodes long) summer shoot pairs present on the trees. On each tree there were at least 4 subreplicates per treatment. The factors evaluated were Cultivar (main plot) and Tipping (subplot). Tipping consisted of two treatments: 4 most api cal buds (nodes) removed, and in tact shoots which served as controls. Cultivar consisted of two levels: Valencia and Hamlin sweet orange trees. Treatments were applied to shoots >8 nodes lo ng and similar in vigour. Tip removal was done manually using pruning clippers. Inductive c ool hours (<20C) accumulated by the date the treatments were applied were 322h and 155h for the 2005 and 2006 inductive seasons, respectively. Subsequent hours of induction af ter tipping for these seas ons were 1216 (01/04/06) and 1156 (01/09/07) as of the pred icted date of bud differentiation responsible for the strongest flowering wave (DISCs Citrus Flowering Monitor, University of Florida). Trees received proper irrigation throughout the experiment. The number of available buds, number of sprouts, number of inflores cences (all types: leaf-abundant, leaf-deficient, and single flowers), total number of flowers per shoot, sprout position (1, 5, >8), and type of sprout were evaluated at bl oom. The proportion of inflorescences forming and proportions of each type of inflorescences were calculated from the
42 variables listed above. For the sake of simplification of the analys is, inflorescences were divided in three types: leaf-abun dant inflorescences (La; inflorescen ces with a leaf to flower ratio 1), leaf deficient inflorescences (Ld; inflorescences w ith a leaf to flower ratio < 1) (Lovatt et al., 1984) and single flowers (one flower no leaves). Data were analy zed using analys is of variance (SAS, SAS Institute Inc.). In season 2006, an additional comparison was performed between whole tipped shoots and bud positions >4 of intact shoots with the in tention to indirectly compare the flowering response of shoots tipped in the fall (tipped shoots) and shoots tipped after bud differentiation (positions >4 in int act shoots; bud position #5 in intact shoots corresponds to bud position #1 in late-fall tipped shoots). For th is last comparison, only the number of sprouts, number of inflorescence s, and type of sprout were analyzed. Results Tipping the four most apical meristems remove d about 27% of the total number of buds available for induction taking the number of availa ble buds in intact shoots as reference. Tipping reduced the number of sprouts, inflorescences, and flowers per shoot when compared with intact shoots but did not affect the number of flowers per inflorescence (Figure 3). However, in percentages, tipped shoots were not different than intact shoots in the proportion of available buds sprouting (Table 3). Tipped shoots produ ced either higher or equal number of leafabundant inflorescences than int act shoots but had less leaf-deficient inflor escences than intact shoots, and had more vegetative sprouts per shoo t than intact shoots (Fig ure 3). Results for the number of single flowers were in consistent in that differences between tipped and intact shoots were significant only in the season 2006 but not in 2005, however, tipping numerically reduced the number of single flow ers formed when compared with intact shoots in both seasons (Table 3, Figure 3). The effect of cultivars a nd interactions between cultivars and tipping were not consistent between the two seasons eval uated for any of the variable above mentioned.
43 Roughly, 1 sprouts and inflorescences were lost on each shoot due to tipping using intact shoots as reference. All the treatments flowered at the same time and none of the tagged shoots showed late-fall/early winter sprouting. Sprouting at different bud positions along the shoo t was less in tipped shoots than in intact shoots (Figure 3). However, at bud positions 1, differences in the number of sprouts between tipped and intact shoots were numerically smaller and sometimes not signifi cant (season 2005) when compared to bud positions 5 and >8. The percentage of leaf-deficient inflorescences and single flower inflorescences formed in the 4 most apical buds of tipped shoots was lower than in the 4 most apical buds of inta ct shoots, but the percen t of vegetative sprouts was higher in these bud positions (1) in the ti pped shoots than of intact shoots (Figure 3). The proportion of leaf-abundant inflorescences formed in the 4 most apical buds was not affected by tipping, but for Valencia it wa s numerically higher. Again, neith er cultivars nor interactions between cultivars and tipping significantly affected sprouting and the type of inflorescence/sprout formed at different bud positions. In a theoretical comparison of buds tipped in late fall and after the bud differentiation date (buds >4 in intact shoots), ther e were significantly more sprout s, inflorescences, leaf-abundant inflorescences and vegetative sprouts in tipped sh oots than in intact shoots, whereas the number of leaf-deficient inflorescences and single fl ower inflorescences we re the same for both treatments (Table 3). Discussion Late fall tipping reduced flower ing intensity in the next sp ring, but stimulated sprouting and flowering in buds that ot herwise would not have flow ered. The overall reduction in flowering intensity may be due to two factors: a) reduction of the number of buds available for flower-induction and/or b) loss of units of flower induction (h ours <20C) in the buds removed.
44 Less flowering in tipped shoots (versus intact shoots) due to reduction in the number of buds available for induction was verified by: 1) significant differences in the number of sprouts per shoot but non-significant diffe rences in the percen t of buds sprouting in each shoot (Figure 3 and Table 3), and 2) the number of sp routs per shoot in tipped shoots was 75% that of intact shoots, the same proportional reduction th at tipping caused in the total number of buds available for flower induction ( 73%). On the other hand, the difference in the numbe r of sprouts per shoot between tipped and intact shoots roughly parallels the difference in the number of infloresce nces per shoot between tipped and intact shoots (Figure 3 ), suggesting that those sprouts that were lost were mainly inflorescences. In addition, the difference in th e number of leaf-abundant inflorescences and vegetative shoots between tipped an d intact shoots was fractional whereas the difference in the number of leaf-deficient inflor escences was often greater or e qual to 1 (Figure 3), indicating that less flowering in tipped shoots, as compared with intact shoots, was due to less leaf-deficient inflorescences forming and an increase in th e number vegetative spr outs. The proportion of leafier inflorescences (based on th e leaf to flower ratio) and ve getative sprouts increases when citrus trees are exposed to less intense induc tive temperatures (Moss, 1969); therefore, considering that buds in tipped s hoots were competent to flower, this may indica te that some induction units already accumulated by the date of tipping were lost in those buds that were removed. However, it is interesting to note that the number of flowers per inflorescence was not different between tipped and intact shoots, s uggesting that leaf-abundant and leaf-deficient inflorescences had approximately the same number of flowers, and therefore the main difference between the types of inflorescences must have been the number of leaves accompanying the flowers.
45 Higher sprouting and flowering at the four most apical buds of tipped shoots (formerly buds 5 before tipping) when compared with mo re basal positions indicates that apical dominance was reset in the new most apical buds (positions 1 in tip ped shoots) (Figure 3). Sprouts in new positions 1 of tipped shoots were mostly inflorescences, but there was a higher percentage of vegetative sprouts in tipped shoots than in intact s hoots. This supports our previous claim of lower level of induction of the buds in tipped shoots due to loss of partially induced buds after tipping. Future experiments including earlier dates of tipping and different ti pping intensities could be useful to maximize the exposure of the buds in the tipped shoots to floral inductive stimuli possibly increasing the intensity of flowering. Also, these studies would help determine how early in the fall tipping can be done without stimulating a late fall flush. However, even though the number of flowers produced in a given season is an important factor determining yield (Moss, 1973b), it has been shown also that inflores cences with leaf:flower ratio >1 set more fruit than inflorescences with leaf:flower ratio <1 (Sauer, 1954; Mo ss, 1970). In this experiment small or no differences were found for the number of leaf-abundant inflorescenc es produced in tipped and intact shoots. Therefore it is not clear whether the reduction in flowering intensity observed here will have a parallel effect on fruit-set and yield since most of the flowers that were lost due to tipping came (hypothetically) from inflorescen ces that had fewer number of leaves than flowers (less fruitful; Sauer, 1954). Future experi ments should include fru it-set data to clarify this point. The comparison of late fall tipped shoots and bud positions >4 of intact shoots provided insights on the difference of the flowering re sponses of shoots tippe d early during flower induction and after the bud differe ntiation date, when floral induc tion is over. Late fall tipped
46 shoots produced more sprouts and inflorescences than bud positions >4 of intact shoots, indicating that late fall tippi ng may minimize reductions in flow ering intensity when compared with tipping times later than the bud differentia tion date (Table 3). Moreover, a tendency toward leafier inflorescences in late fall tipped sh oots implies that fruit se t could be better after late fall tipping than tipping after bud differentiati on. More vegetative sprouts in late fall tipped shoots at the cost of a reduction in leaf-deficient and single flower inflorescences may be useful to provide young bearing-wood for ne xt seasons flower induction. The results of these experiments provide evid ence supporting the hypothesis that late fall hedging changes the flowering intensity observed in the following spring flush. The observed reduction in flowering intensity af ter tipping in late fall may have been caused by two factors: the reduction in the tota l number of buds available for flow er induction, and the loss of induction units already accumulated in the removed buds. Wh ile the first factor was unavoidable, tipping earlier in the fall may reduce the loss of induc tion units and increase flowering intensity. However, it is not clear if increasing flowering intensity per se in our experiments would have been beneficial from a commercial standpoint. The loss in flowering intensity observed was related to a reduction in the numbe r of inflorescences with less nu mber of leaves than flowers, which generally set less fruit than inflorescences with more leaves than flowers and an increase in vegetative sprouts. However, the number of leafier inflorescences, which tend to set more fruit, remained almost unaffected after tipping. Late fall hedging may not avoid yield losses but it may recover much of the flowering bud potentia l otherwise lost from hedging after induction. It is not clear how this effect compares with yields derived from other timings of tipping. Further experiments including more tipping times and fru it-set/yield data, as well as tipping intensity (number of buds removed) during maintenance he dging in commercial groves will help elucidate
47 whether late fall hedging is a viable option for tree size control and maintaining adequate new vegetative flushes while having less impact in flow ering and yields of sweet orange trees grown in humid subtropical climates.
48 A Valencia Hamlin 7 6 5 4 3 2 1 0 Valencia Hamlin 6 5 4 3 2 1 0 Valencia Hamlin 35 30 25 20 15 10 5 0 Valencia Hamlin 5 4 3 2 1 0 Total Sprouts Inflorescences Flowers Flowers/Inflorescence Intact Tippeda b a b a b a b a b a b a a a aB Valencia Hamlin 8 7 6 5 4 3 2 1 0 Valencia Hamlin 7 6 5 4 3 2 1 0 Valencia Hamlin 30 25 20 15 10 5 0 Valencia Hamlin 5 4 3 2 1 0 Total Sprouts Inflorescences Flowers/Shoot Flowers/Inflorescence Intact Tippeda b a b a b a b a b a b a a a aFigure 3. Primary flowering intensity paramete rs of late-fall tipped vigorous summer shoots of Hamlin and Valencia sweet orange trees under field conditions on two successive seasons. A. Season 200607. B. Season 2006. Values are means of 10 and 6 trees for season 2005 and 2006 respectively ( 4 subreplicates/tree/treatment). Bars are 1SE of the mean (pooled). Different letters within each variety indicate significant differences be tween treatments (tipped vs. intact) for that variable accord ing to Fishers protected LSD ( =0.05).
49 Table 3. Types of sprouts formed in late-f all tipped vigorous summer shoots of Hamlin and Valencia sweet orange trees under fi eld conditions in two successive seasonsx. % Inflorescencesy Total available buds % of Buds sprouting Total Leafabundant Leafdeficient Single Flowers % Vegetative Sproutsy 2005 Hamlin Intact 1158 41.93.1 72.27.9 41.04.8 28.26.8 3.01.3 27.87.9 Tipped 835 44.82.7 51.69.7 37.15.1 11.74.4 2.81.2 48.49.7 Valencia Intact 951 51.22.7 88.08.3 40.45.1 44.96.7 2.71.5 12.08.3 Tipped 690 52.32.5 82.55.6 58.25.7 22.94.8 1.40.6 17.56.0 P-value Treatment n.a. 0.15 <0.01 0.03 <0.01 0.35 <0.01 Cultivar n.a. 0.03 0.05 0.14 0.08 0.56 0.04 Interaction n.a. 0.51 0.02 0.02 0.39 0.44 0.02 2006 Hamlin Intact n.a. n.a. 89.44.3 53.19.2 17.52.0 18.87. 7 10.64.3 Tipped n.a. n.a. 66.111.0 48.110.6 9.22.0 8.86.7 33.910.9 Valencia Intact n.a. n.a. 88.54.1 42.99.5 40.38.5 5.52.7 11.54.1 Tipped n.a. n.a. 76.58.0 46.78.0 28.28.3 1.71.7 23.58.4 P-value Treatment n.a. n.a. <0.01 0.98 0.02 0.04 <0.01 Cultivar n.a. n.a. 0.61 0.62 0.03 0.11 0.61 Interaction n.a. n.a. 0.29 0.49 0.66 0.30 0.29 x Values are means SE of 10 and 6 observations (tre es) for seasons 2005 and 2006, respectively ( 4 subreplicates/tree/treatment). P-values provided to indicate significance of the main effects and interactions according to ANOVA. y As percent of the total number of sprouts. L eaf-abundant inflorescences = leaf to flower ratio 1. Leaf-deficient inflorescences = leaf to flower ratio <1. z n.a. = Not available.
50 Average Number of Sprouts / Shoot Vegetative Single Flower Leaf-deficient Leaf-abundant Valencia Hamlin Valencia Hamlin Valencia Hamlin Valencia Hamlin 3 2 1 0 Intact TippedP = 0.23 P < 0.01P = 0.16 P = 0.01A Average Number of Sprouts / Shoot Vegetative Single Flower Leaf-deficient Leaf-abundant Valencia Hamlin Valencia Hamlin Valencia Hamlin Valencia Hamlin 5 4 3 2 1 0 Intact TippedP = 0.02 P < 0.01 P = 0.01 P = 0.01B Figure 3. Distribution of different types of sprouts on la te-fall tipped and intact summer shoots of Hamlin and Valencia sweet orange trees, under field conditions. A. Season 2005, B. Season 2006. Leaf-abundant in florescences = leaf to flower ratio 1. Leaf-deficient inflorescenc es = leaf to flower ratio <1. Error bars are SE of the mean (n=10 and n=6 for A and B, respectively, with > 4 sub replicates/treatment/tree). P-values provided to assess significance between treatments (tipped vs. intact).
51 Average Number of Sprouts / Shoot Buds >8 Buds 5-8 Buds 1-4 Valencia Hamlin Valencia Hamlin Valencia Hamlin 4 3 2 1 0 Intact TippedP = 0.16P < 0.01 P < 0.01 A Average Number of Sprouts / Shoot Buds >8 Buds 5-8 Buds 1-4 Valencia Hamlin Valencia Hamlin Valencia Hamlin 4 3 2 1 0 Intact TippedP = 0.02 P < 0.01P < 0.01 B Figure 3. Distribution of sprouts of late-fall-tipped and in tact summer shoots of Hamlin and Valencia sweet orange trees on different bud positions (counting from the apex after tipping), under field conditions. A. Season 2005, B. Season 2006. Error bars are SE of the mean (n=10 and n=6 for A and B, respectively, with at least 4 sub replicates/treatment/tree). P-values provided to assess significance of the differences between treatments (tipped vs. intact).
52 Valencia Hamlin 100 80 60 40 20 0 Valencia Hamlin 100 80 60 40 20 0 Valencia Hamlin 100 80 60 40 20 0 Valencia Hamlin 100 80 60 40 20 0 %Leaf-abundant %Leaf-deficient %Single Flower %Vegetative Full Tipped a a a a a b a b a b a b b a a b Figure 3. Percent distribution of different types of spring spr outs on the first 4 nodes of latefall tipped and intact vigorous summer s hoots of Hamlin and Valencia sweet orange trees under field conditions (seas on 200607). Leaf-abundant inflorescences = leaf to flower ratio 1. Leaf-deficient inflorescences = leaf to flower ratio <1. Values are the averages of 6 observations (trees) with at least 4 sub replicates per tree. Different letters indicate significant differences (P<0.05) between tipped and intact shoots according to ANOVA.
53 Table 3. Comparison of five fl owering parameters between la te-fall summer shoots tipped at bud position 4 (from the apex) and bud positions >4 of intact summer shoots Hamlin and Valencia sweet orange tr ees, under field conditions (season 2006 07)y. Inflorescences Sprouts/Shoot Total Leafabundantz Leafdeficientz Single Flowers Vegetative Sprouts Hamlin Intact 3.74.68 2.96.69 2.12.75 0.41.11 0.43.17 0.64.28 Tipped 4.74.77 3.10.76 2.34.66 0.46.16 0.30.20 1.64.43 Valencia Intact 2.61.61 2.04.39 1.20.41 0.66.16 0.19.09 0.32.16 Tipped 3.71.38 2.80.43 1.81.42 0.95.25 0.04.04 0.91.32 P-value 0.01 0.04 0.06 0.38 0.18 <0.01 y Values are means SE of 6 observations (trees ) with at least 4 sub replicates per tree per treatment. P-values provided to assess significa nce of the differences between means of intact and tipped shoots according to ANOVA (split-plot with nested randomized blocks). No difference for cultivars or interaction was detected. z Leaf-abundant inflorescences = leaf to flower ratio 1. Leaf-deficient inflorescences = leaf to flower ratio <1
54 CHAPTER 4 CONCLUSIONS We studied some factors that affect flower bud induction in sweet oranges under humid subtropical conditions and show some potential to enhance flowering in commercial orchards in Florida. Several conclusions follow: Drought enhances flower bud induction in sw eet orange trees at marginal inductive temperatures but reduces induction at lo wer near-optimal inductive temperatures, apparently due to an interaction between ambient temperatures and water deficit. Under humid subtropical conditions of Florida, reducing water availability to trees during by only withholding irrigation du ring the winter can increas e flowering intensity if rainfall is scarce. Late fall tipping of summer shoots can reduce subsequent flowering in these shoots when compared to intact shoots, but some flower ing was transferred to more basal buds that otherwise would neither sprout nor flower. Shoot tipping in late fall may provide more intense flowering in the spring flush than tipping in winter or tipping in spring just before bloom. Late fall tipping can cause an increase in the number and proportion of inflorescence leaves, inflorescences and ve getative shoots formed in the spring compared to tipping after induction. Future growth chamber experiments includi ng more levels of temperature and drought would be useful to confirm the interactive eff ects observed in our growth chamber experiment. On the other hand, confirmation of in teractive effects in th e field is more elusiv e, but analysis of accumulation of days of cool induction, drought induction and both simultaneously on more seasons and more plots could be useful to co mplement findings of growth chamber studies dealing the with potential interact ive effects of ambient temperatur es and water deficit in citrus flower bud induction. For our tipping experiment, if the response obs erved in summer shoots tipped in the fall represents the response of trees hedged using commercial equipment, late fall hedging may represent a viable time for routine tree si ze control hedging. However, to determine the
55 effectiveness of late-f all hedging under commercial conditions further studies are needed in which fruit-set and yield variables should be evaluated.
56 LIST OF REFERENCES Abbot, C.E. 1935. Blossom-bud differentiati on in citrus trees. Am. J. Bot. 22:476. Albrigo, L. G. and V. Galen Sa uco. 2004. Flower bud induction, flow ering and fruit-set of some tropical and subtropical fruit tree crops w ith special reference to citrus. Acta Horticulturae 632:81. Albrigo, L. G., J. I. Valiente and C. Van Parys de Wit. 2004. Influence of winter and spring weather on year-to-year citrus fruit set and yield variation in Sao Paulo, Brazil. Proc. Int. Soc. Citriculture 2004. 1:263. Albrigo, L.G., H.W. Beck, and J. I. Valiente. 2006. Testing a flow ering expert system for the decision information system for citrus. Acta Hort. 707:17. Albrigo, L.G., J.P. Syvertsen, T.M. Spann. 2007. Canopy flush control for management of canker and greening. Citr us Industry 88(3):12. Araki, T. 2001. Transition from vegetative to re productive phase. Curr. Opin. Plant Biol. 4:63 68. Bacon, P.E. 1981. The effect of hedging time on re growth and flowering of mature Valencia orange trees. Aust. J. Agric. Res. 32:61. Bacon, P.E., K.B. Bevington. 1978. Effect of time of hedging on shoot growth and flowering in citrus. Proc. Int. Soc. Citriculture 1974. 314. Begg, J. E., N.C. Turner. 1970. Water potential grad ients in field tobacco. Plant Physiol. 46:343 346. Bellow, T.S., J.G. Morse. 1986. Modeling flower development in Navel oranges. Scientia Horticulturae. 30:117. Ben Mechlia, N., J.J. Carroll. 1989. Agroclim atic modeling for the simulation of phenology, yield and quality of crop production. I. Citrus res ponse formulation. Intl. J. Biometeorology. 33:36. Bevington, K.B. 1980. Response of Valencia orange trees in Australia to hedging and topping. Proc. Fla. State Hort. Soc. 93:65. Borroto, C.G., A.M. Rodrguez. 1977. The effect of water stress on flowering and fruit set of Valencia oranges in Cuba. Proc. Int. Soc. Citriculture 1977. 3:1069. Cassin, J., J. Bourdeaut, A. Fougue, V. Furon, J. P. Gaillard, J. LeBourdelles, G. Montagut, C. Moreuil. 1969. The influence of climate upon the blooming of citrus in tropical areas. Proceedings 1st International Citr us Symposium 1:315.
57 Cooper, W.C., A. Peynado. 1958. Effect of giberell ic acid on growth and dormancy in citrus. Proc. Am. Soc. Hort. Sci. 72:284. Davenport, T.L. 1990. Citrus flowering. Hort. Rev. 12:349. Davies, F.S., Albrigo, L.G. 1994. Citrus, p.158 61. CAB International, Wallingford, UK. Edwards, G.R. 1986. Ammonia, arginine, polyamines and flowering in apple. Acta Horticulturae 179:363. Erickson, L.C. 1968. The general physiology of citr us, p. 97. In: Reuther, W., L.D. Batchelor and H.J. Webber (eds.). The Citrus Industry vol. 2 Chapter 2. Univ ersity of California Division of Agricultural Sciences, Riverside, California Garca-Luis A, V. Almela, C. Monerri, M A gust, and J.L. Guardiola. 1986. Inhibition of flowering in vivo by existing fruits and applied growth regulators in Citrus unshiu Physiol. Plant. 66:515 Garca-Luis, A., F. Fornes, A. Sanz, J.L. Guar diola. 1988. The regulation of flowering and fruit set in Citrus: relationship with carboh ydrate levels. Israel J. Bot. 37:189. Garca-Luis, A., M. Kanduser, P. Santamarina, J.P. Guardiola. 1992. Low temperature influence in citrus: The separation of inductive and bud dor mancy releasing effects. Physiol. Plant. 86:648. Garca-Luis, A., P. Santamarina, J.L. Guardiola. 1989. Flower formation from Citrus unshiu buds cultured In Vitro Annals of Botany 64:515. Goldschmidt, E.E., A. Golomb. 1982. The carbohydrat e balance of alternate bearing Citrus trees and the significance of reserves for flower ing and fruiting. J. Am. Soc, Hort. Sci. 107:206. Graham, J.H., T.R. Gottwald, J. Cubero, D.S. Achor. 2004. Xanthomonas axonopodis pv. citri : factors affecting successful eradication of citrus ca nker. Mol. Plant Path. 5:1. Halbert, S.E., K.L. Manjunath. 2004. Asian citr us psyllids (Sternorrhyncha: Psyllidae) and greening disease of citrus: a literature review and assessment of risk in Florida. Florida Entomologist 87:330. Kaufmann, M. R. 1968. Evaluation of the pressure chamber method for measurement of water stress in citrus. J. Am. Soc. Hort. Sci. 93:186. Koshita, Y., T. Takahara, T. Ogata, A. Go to. 1999. Involvement of endogenous plant hormones (IAA, ABA, GAs) in leaves and flower bud formation of Satsuma mandarin ( Citrus unshiu Marc.). Scientia Horticulturae 79:185. Koshita, Y., T. Takahara. 2004. Effect of wa ter stress on flower-bud formation and plant hormone content of Satsuma mandarin ( Citrus unshiu Marc.). Sci. Hort. 99:301.
58 Krajewsky, A.J., E. Rabe. 1995. Citrus flowering: a critical evaluation. J. Hort. Sci. 70:357. Lenz, F. 1969. Effect of day length and temperatur e on the vegetative and re productive growth of Washington Navel orange. Proceedings 1st International Citrus Symposium 1:333. Lord, E.M., K.J. Eckard. 1985. Shoot development in Citrus sinensis L. (Washington Navel Orange). I. Floral and inflorescen ce ontogeny. Botanical Gazette 146:320. Lovatt, C.J., S.M. Streeter, T.C. Minter, N.V. OConnell, D.L. Flaher ty, M.W. Freeman, P.B Goodell. 1984. Phenology of flowering in Citrus sinensis (L.) Osb., cv. Washington navel orange. Proc. Int. Soc. Citriculture 1984. 1:186. Lovatt, C.J., Y. Zheng, K.D. Hake. 1988. Demonstr ation of a change in nitrogen metabolism influencing flower initiation in citrus. Israel J. Bot. 37:181. McDaniel, C.N., S.R. Singer, S.M.E. Smith. 1992. Developmental states associated with the floral transition. Dev. Biol. 153:59. McKnight, T.L., Hess, D. 2000. Physical Geography: A Landscape Appreciation, p.221. 8th ed. Prentice Hall, Upper Saddle River, NJ. Monselise, S.P. 1985. Citrus and relate d genera. p.275. In: CRC Handbook of Flowering vol. II. CRC Press, Boca Ratn, Florida. Monselise, S.P., A.H. Halevy. 1964. Chemical in hibition and promotion of citrus flower bud induction. Proc. Am. So c. Hort. Sci. 84:141. Moss, G.I. 1969. Influence of temperatur e and photoperiod on flower induction and inflorescence development in sweet orange ( Citrus sinensis L. Osbeck). J. Hort. Sci. 44:311. Moss, G.I. 1970. Fruit-se t in sweet orange ( Citrus sinensis ): the influence of inflorescenceleaves. Phyton 21:141. Moss, G.I. 1971a. Promoting flowering in sw eet orange. Aust. J. Agric. Res. 22:625. Moss, G.I. 1971b. Effect of fruiting on flowering in relation to biennial bearing in sweet orange ( Citrus sinensis ). J. Hort. Sci. 46:177. Moss, G.I. 1973a. Regrowth and flowering of sweet orange after pruning. Aust. J. Agric. Res. 24:101. Moss, G.I. 1973b. Major factors influencing flower formation and subsequent fruit set of sweet orange. Proc. Int. So c. Citriculture 1973. 1:215. Moss, G.I. 1976. Temperature effects of flower initiation in sweet orange ( Citrus sinensis ). Aust. J. Agric. Res. 27:399.
59 Nir, I., R. Goren, and B. Leshem. 1972. Eff ects of water stress, gibberellic acid and 2 chloroethyltrimethylammoniumchloride (CCC) on flower differentiation in Eureka lemon trees. J. Am. Soc. Hort. Sci. 97:774. Obreza, T.A., D.J. Pitts, L.R. Parsons, T.A. Wheaton, and K.T. Morgan. 1997. Soil waterholding characteristic affects citrus irrigati on scheduling strategy. Proc. Fla. State Hort. Soc. 110:36. Putterill, J., R. Laurie, R. Mac knight. 2004. Its time to flower: the genetic control of flowering time. BioEssays 26:363. Reuther, W., D. Ros-Castao. 1969. Comparison of growth, maturation and composition of citrus fruits in subtropical Californi a and tropical Colombia. Proceedings 1st International Citrus Symposium 1:277. Sauer, M.R. 1951. Growth of orange shoots. Aust. J. Agric. Res. 2:105117. Sauer, M.R. 1954. Flowering in the sweet orange. Aust. J. Agric. Res. 5:649. Scholander, P. F., H.T. Hammel, A.D. Bradst reet, E.A. Hemmingsen. 1965. Sap pressure in vascular plants. Science 148:339. Smajstrla, A.G., D.S. Harrison, F.S. Zazueta, L.R. Parsons, K.C. Stone. 1987. Trickle irrigation scheduling for Florida citrus. Fla. Coop. Ext. Service Bulletin 208. University of Florida, Gainesville. Southwick, S. M. and T.L. Davenport. 1986. Characterization of water stress and low temperature effects on flower inducti on in citrus. Plant Physiol. 81:26. Syvertsen, J.M., C. Goi, and A. Otero. 2003. Fruit load and canopy shading affect leaf characteristics and net gas exchange of Spr ing navel orange trees. Tree Physiology 23:899. Valiente, J.I., L.G. Albrigo. 2000. Modeling flower ing date of sweet orange trees in central Florida based on historical weather. Proc. Int. Soc. Citriculture 296. Valiente, J.I., L.G. Albrigo. 2004. Flower bud induction of sweet orange trees [ Citrus sinensis (L.) Osbeck]: effect of te mperature, crop load and bud age. J. Am. Soc. Hort. Sci. 129:158.
60 BIOGRAPHICAL SKETCH Eduardo Jos Chica Martnez was born on June 9, 1982 in Guayaquil, Ecuador. In July 2005, he obtained the degree of Inge niero Agropecuario (a 4-yearsdegree in agriculture) from the Escuela Superior Politcnica del Litoral in Gu ayaquil. In the same year, he was admitted as a student in the Master of Science program in the Horticultural Science Department at the University of Florida. He finished his program in August 2007.