• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

A microscopic and phenological study of pollen development and bloom in selected cultivars of hazelnut (Corylus avellana)

Dissertation
Author: Chantalak Tiyayon
Abstract:
Pollen development is an important process in male flower development, the timing of which may be correlated with time of pollen shed in hazelnut (Corylus avellana L.). Early to very late blooming cultivars were identified and the relationship of microsporogenesis and microgametogenesis, and time of pollen shed were studied in nine hazelnut cultivars. Most advanced catkins from a single tree of each cultivar were collected each week from 4 Aug. to 6 Dec. 2002, and on 17 Jan. 2003, stained and analyzed by light microscopy. The phenology part of this dissertation studied the role of the chilling requirement as chill units (CU) and heat requirement as growing degree hours (GDH) in pollen shed. Hazelnut twigs of three cultivars; 'TGDL', 'Barcelona', and 'Hall's Giant' were collected at weekly intervals starting from early Fall 2006 through the time of anthesis in the field in winter 2007. Twigs were then held at a different constant temperature 0, 5, 10, 15, or 20 °C. Observing these twigs weekly, the time of anthesis (50% pollen shed) was recorded. A parallel study was conducted in more controlled conditions by collecting hazelnut twigs of the same three cultivars on 1 Nov. 2006 and holding them at 5 °C in a cold room. Five twigs of each genotype were brought out to room temperature at 5-day intervals in order to force them to bloom. Numbers of catkins that shed pollen were recorded every 5 days. From the results, we propose a model of hazelnut staminate flower development that begins with catkin differentiation concurrent with early stages of pollen development. Catkin length increased steadily and reached a lag phase at the end of microsporogenesis. While there is no external change, microspores continue microgametogenesis, the catkins are endodormant, unable to be induced to shed pollen, and accumulate chilling during this period of time. We propose that the chilling requirement is met when pollen reaches the mature stage and when catkins are in the latter stages of the lag phase of growth. At this point, GDH accumulate and catkins become ecodormant. As chilling continues to accumulate, the amount of GDH required for pollen to shed is reduced. A similar quantity of GDH was required for all genotypes from pollen grain maturity to catkin elongation.

TABLE OF CONTENTS

Page

CHAPTER 1 GENERAL INTRODUCTION....…...…………...…………….1 1.1. Biology of Hazelnut ..……………………..………………..……... 2 1.1.1. Taxonomy…………….……...………..……………………...2 1.1.2. Morphology…………………..…..………………………...… 3 1.1.3. Staminate Flower Development…………………….……... 5 1.2. Pollen Development………………………..……………..………. 7 1.2.1. Microsporogenesis………….………………………….........7 1.2.2. Microgametogenesis…………………………..………...…..9 1.2.3. Pollen Grains………………...……………………..……..… 11 1.3. Phenology………………….………………………...…………….. 12 1.3.1. Temperature Factor………………………………..….…..... 12 1.3.2. Chilling Requirement and Heat Accumulation………..….. 12 1.3.3. Modeling as Tool to Predict Bloom……………………...… 15 1.3.4. Comparison between Models…………………………….... 19 1.4. Research Objectives……………………..………………..……… 20 Literature Cited…………………………………..…………………….... 20 CHAPTER 2 MICROSCOPIC STUDY OF POLLEN DEVELOPMENT IN NINE HAZELNUT GENOTYPES………….………..…….26 Abstract…………………………………..…………………………...….. 27 Introduction………………………………………..…………………...… 28

TABLE OF CONTENTS (Continued)

Page

Materials and Methods……………………………………………..…...33 Plant Materials…………………………………………………….....33 Microscopy…………………………………………..…………........ 34 Pollen Development……………………………..…………..….......34 Chilling Unit and Growing Degree Hour Calculations…………... 35 Results and Discussion…………..………………………………..…... 35 Hazelnut Pollen Development………………………………...…... 35 Relationship of Catkin Growth and Pollen Development with Cumulative Chilling Units and Growing Degree Hours………..... 39 Conclusions………………………………………………………........... 41 Acknowledgements………………………………………………...…… 42 Literature Cited………………………………………………………….. 42 CHAPTER 3 STAMINATE BLOOM PHENOLOGY IN THREE CULTIVARS OF HAZELNUT…………………………..….... 52 Abstract……………………………………………...…………..………. 53 Introduction………………………………………..…………..………… 54 Materials and Methods…………………………………………..……...61 In Situ Chilling and Growing Degree Hour Accumulation until Anthesis……………………..………….………………………….... 61 Controlled Chilling and Growing Degree Hour Accumulation until Anthesis………..…..…………………………………….......... 62

TABLE OF CONTENTS (Continued)

Page

Chilling Unit and Growing Degree Hour Calculations………..… 62 Results and Discussion………………………………………………… 64 In Situ Chilling and Growing Degree Hour Accumulation until Anthesis…..…………………….……………………………….…... 64 Controlled Chilling and Growing Degree Hour Accumulation until Anthesis……..……..…………………………………………...68 Conclusions…………………………………………………………….. 72 Literature Cited…………………………..……………………...……….74 CHAPTER 4 GENERAL CONCLUSION……....…...…………...………… 89 BIBLIOGRAPHY…………….…………………..………….………………….96 APPENDICES………………..……………………………………………...... 103 Appendix A Stages of pollen development of collected cultivars from August 2002 to January 2003……..………..…… 104 Appendix B Chilling hour and growing degree hour to reach tetrad stage, mature pollen stage, and anthesis of nine hazelnut cultivars collected from August 2002 to January 2003…………………………………………. 105 Appendix C Comparison of cumulative chilling hour, chilling portion, chilling unit according to Utah model, chilling unit according to this study, and cumulative growing degree hour in 2006-2007 growing season....……….. 106

LIST OF FIGURES

Figure Page

2.1 Stages of hazelnut pollen development..………………........….......45 2.2 Stages of microsporogenesis and microgametogenesis during pollen development of nine hazelnut cultivars..…..…...……….….. 46 2.3 Stage of hazelnut pollen development and catkin length in relation to calendar date, cumulative chill hours, and growing degree hours…………………………………………………………… 48 2.4 The model describing the relationship of hazelnut catkin length and stages of pollen development to time and temperature …..… 50 3.1 Percent pollen shed of three hazelnut cultivars, A) ‘TGDL’, B) ‘Barcelona’, and C) ‘Hall’s Giant’, cut weekly from 2 Oct. 2006 to 6 Feb. 2007, placed at five different temperatures and observed for up to 9 weeks or until the twigs reached 50% anthesis...……………………………………………………………… 77 3.2 Number of days required for three hazelnut cultivars to reach anthesis after cutting at weekly interval and holding twigs at 5, 10, 15, and 20 o C relative to sample day interval from 1 Oct. 2006……………………………………………………………………. 78 3.3a Regression between chilling hours and cumulative growing degree hours for catkins to reach anthesis of three hazelnut cultivars that were held at 5, 10, 15, and 20 o C…………………..... 79 3.3b Regression between cumulative chilling portions and cumulative growing degree hours for catkins to reach anthesis of three hazelnut cultivars that were held at 5, 10, 15, and 20 o C..…...…… 80 3.3c Regression between cumulative chilling units, according to the Utah model, and cumulative growing degree hours for catkins to reach anthesis of three hazelnut cultivars that were held at 5, 10, 15, and 20 o C…………………………………………….………....…. 81

TABLE OF FIGURES (Continued)

Figure Page

3.3d Regression between cumulative chilling units, according to this study, and growing degree hours for catkins to reach anthesis of three hazelnut cultivars that were held at 5, 10, 15, and 20 o C………………………………………………………..… 82 3.4 Percent pollen shed of three hazelnut cultivars cut on 1 Nov. 2006 and held at 5 o C for up to 70 days, then forced at 20 o C for up to 30 days………………………...…………………………….. 83 3.5 Catkin survival percentage and regression between days at 5 o C and days to reach anthesis of three hazelnut cultivars cut on 1 Nov. 2006 and held at 5 o C for up to 70 days, then forced at 20 o C for up to 30 days ……...……………………..……... 84 3.6 Regression between cumulative chilling units, according to this study, and cumulative growing degree hours of three hazelnut cultivars cut on 1 Nov. 2006 and held at 5 o C for up to 70 days, then forced at 20 o C for up to 30 days.…………………………....… 85 4.1 The proposed model of catkin length and stages of pollen development in relative to chilling requirement and growing degree hour accumulation……….……………………………......…. 95

LIST OF TABLES

Table Page

2.1 Bloom time and chilling requirement of hazelnut cultivars categorized from Mehlenbacher, 1991……………………..……….. 51 3.1 Regression between number of days sampling from 1 Oct. 2006 and number of days to reach anthesis of three hazelnut cultivars after weekly cutting interval which combined four temperatures, date to reach anthesis estimated from the equations, and actual period that anthesis occurred in the field……………………………. 86 3.2 Regression between days at 5 o C and number of days to reach anthesis of three hazelnut cultivars, and estimated days at 5 o C and chilling units that anthesis may occur without forcing………… 87 3.3 Coefficients and their 95% confidence intervals of linear- equations transformed from power-equations in Figure 3.3d and 3.6…………………………………………………………………...88

1

CHAPTER 1

GENERAL INTRODUCTION

2 CHAPTER 1 GENERAL INTRODUCTION

1.1. Biology of Hazelnut 1.1.1. Taxonomy Hazelnut is categorized in the Genus Corylus, Family Betulaceae (the birch family) (Lagerstedt, 1975; Thompson et al., 1996), or previously Family Corylaceae (the hazel family) (Menninger, 1977; Rosengarten, 1984). The name Corylus has originated from korys in Greek meaning a helmet, hood (Rosengarten, 1984), or bonnet (Lagerstedt, 1975), to which the husk resembles. There are about 15 major species in this genus, all of which have edible nuts, spread across the temperate zones of North America, Europe, Northern Africa, and Asia (Menninger, 1977). In the United States, native species are C. americana Marsh. (American filbert), and C. cornuta Marsh. or C. rostrata (beaked filbert) (Rosengarten, 1984). Indigenous to Turkey and the Balkans is C. colurna L. (the Turkish tree hazel). The species grown in commercial orchards is European hazelnut, C. avellana (Thompson et al., 1996). Hazelnuts have many names around the world. For example, “Avellana” in Spanish, “Aveleira” in Portuguese, “Bunduq” in Arabic, “Findik” in Turkish, “Haselnuss” in German, “Hasselnöt” in Swedish, “Hazelaar” in Dutch, “Lesnoi Orekh” in Russian, “Nocciola” in Italian, “Noisette” in French,

3 “Hashibami” in Japanese, and “Chên Tzu” in Chinese (Rosengarten, 1984). However, the term “hazelnut” is used worldwide for many Corylus species. There are other common names such as hazel, cob, cobnut, lambert nut, Lombardy nut, Spanish nut (Lagerstedt, 1975), and other scientific names that were given based on appearance, shape, husk, and origin (Bunyard, 1920). In Oregon, hazelnuts have commonly been called filberts. The name “filbert” may have come from “full beard”, or Vollbart in German (Rosengarten, 1984), which refers to the long-leafy husk around the nut. It may also have evolved from St. Philibert’s Day that occurs on 22 Aug. which is coincident with the time the nuts start to ripen (Lyle, 2006). Also in the past, the “filbert” was often used for varieties where the husk is longer than the nut. The term “cob” referred to those nuts where the husk was about the same length as the nut. “Hazel” was the term for varieties where the husk was shorter than the nut (Menninger, 1977; Lyle, 2006). 1.1.2. Morphology Reed (1976), as cited in Handbook of Nuts (Duke, 1989), gave a description of C. avellana as a deciduous shrub or small tree. Trees are generally up to 6 m tall and often thicket-forming. A very old, untrained, C. avellana tree in the northwestern U.S. was reported to be 8 m in height and 10 m in canopy diameter (Lagerstedt, 1975). Twigs are dark brown, smooth, with glandular-hair. Leaves are 5 to 12 cm long, orbicular, long-pointed, and hairy on both surfaces with double serrated-margins. Catkins, staminate

4 inflorescences, are 2-8 cm long, pendulous, and appear in clusters of 1 to 4 (Reed, 1976). Staminate catkins are borne at nodes on one year wood (Germain, 1994). Catkins are composed of 130-290 flowers; each has one bract and two bracteoles (Germain, 1995). There are four filaments bearing eight anthers in each catkin bract [Trotter, 1947 (in Italian) cited by Lagerstedt, 1975]. Pistillate flowers are bud-like, erect, and approximately 5 mm long. Fruits occur in clusters of one to 4. Nuts are 1.5-2 cm in diameter with brown color, and are enclosed by deeply lobed and irregularly toothed bracts as of variable length. Flowers bloom from January to March while nuts mature in the fall (Reed, 1976). Hazelnuts are distinctive from other orchard crops because they are monoecious and anemophilous (wind pollinated), and bloom in midwinter (December to March in the northern hemisphere) (Germain, 1994). Each catkin produces 4 million (Kelley, 1980) to 40 million (Pisani et al., 1968) pollen grains. However, plants are generally self-incompatible. Most of the cultivars are also dichogamous. The staminate and pistillate flowering times may not overlap, and cultivars can be either protogynous or protandrous. Therefore cross-pollination with compatible pollinizers that shed pollen when female flowers are receptive is required for optimal nut production (Lagerstedt, 1975). Hazelnut pollen is triangular or ellipsoidal in shape. It has three germinative pores, each located over a small bump in the pollen wall, or over a

5 small vacuole in some cases. The approximate size of pollen is 25-40 μm in diameter (Trotter, 1947 cited by Lagerstedt, 1975; Germain, 1995). 1.1.3. Staminate Flower Development In France, the first sign of catkin differentiation occurs in mid-May, and a month later catkins are visible in the leaf axils. Free pollen grains are observed inside the anthers around mid-August (Germain, 1995). In Oregon, ‘Barcelona’ hazelnut catkins can be seen in the axils of basal leaves on the current season’s stem in late June. Catkins grow rapidly in the first three months and then stay in a lag phase for approximately 10 weeks. After that, growth resumes until anthesis (Lagerstedt, 1975). Catkin abscission in the fall prior to anthesis has been observed at high temperatures (~25°C or greater) when catkins are presumed to be dormant (Woodroof, 1979; Mehlenbacher- personal communication). Dimoulas (1979) did a comprehensive microscopic study on pistillate and staminate flower bud development of three hazelnut cultivars (‘Ronde du Piemont’ (TGDL), ‘Fertile de Coutard’ or ‘Barcelona’, and ‘Merveille de Bollwiller’ or ‘Hall’s Giant’) in Bordeaux, France, from May to September. He found that male inflorescence differentiation starts early in the season and the process takes less than three months. In his study, catkin development was categorized into eight stages:

6 I. No sign of catkin differentiation in the apical meristem during the first half of May; II. Initiation of the first catkin’s bract inside the axial bud which becomes convex and the outline of the catkin evolves between 15 and 20 of May; III. Catkin elongation inside the bud, open passage at the top between the scales of the bud, and the contour of the secondary catkin or future bud can be observed in mid-June; IV. Emergence of catkins (which have reddish color at the tip) from the bud as a result of pedicel elongation that occurs during the second half of June; V. Exiting of catkins out of the buds (approximately 10 mm long), initiation of the bracteoles inside the bracts, and gradual extension of the basal bracts to those at the tip of the catkin at the end of June and the beginning of July; VI. The initiation of anthers, outline of anthers begins at the base of the catkin during the first and second week of July; VII. Appearance of pollen mother cells (PMCs): anthers are completely formed and pollen sacs are seen with sporogenous cells in the middle happened on the third week of July; and VIII. Formation of tetrads and pollen grain: meiosis occurs approximately two weeks after the appearance of the PMCs (the end of July to mid-

7 August). At this time, the catkins are 2-3 cm long. The formation of the pollen grains starts from the basal bracts to the top bracts of the same catkin. There was roughly a week difference in the timing of these developmental processes among the three cultivars used in his study. The process of microsporogenesis in hazelnut has not yet been observed in detail nor do we know if the timing of the various stages of pollen grain maturation is different between early, mid, and late pollen shedding cultivars. 1.2. Pollen Development A review by Scott et al. (2004) stated that unlike most plants organs, which derive from meristem, the development of the anther is unusual in that the microsporangia (anther) arise from a single archesporial cell. Pollen development consists of two processes, microsporogenesis and microgametogenesis. The former is the formation of microspores and the latter is the development of microspores into pollen grains. 1.2.1. Microsporogenesis Microsporogenesis can be divided into two major phases, the formation of pollen mother cells (PMCs) which are also called microsporocytes or microspore mother cells, and the development of microspores from PMCs through meiosis.

8 Pollen mother cell formation: Davis (1966) described the general process of anther wall development and microsporogenesis in angiosperms. The initial periclinal division of archesporial cells forms the primary parietal cells on the outer portion and the sporogenous cells on the inner portion of the microsporangium. The primary parietal layer then divides periclinally, resulting in two secondary parietal layers, while the sporogenous cells enlarge into PMCs. The outer secondary parietal layer divides periclinally, forming an endothecial layer on the outside and a middle layer on the inside. The inner secondary parietal layer develops into the middle layer on the outside and the tapetal layer on the inside. However, anther wall layers in different plant species may develop differently, which would result in a different number of endothecial layers, middle layers, and tapetal layers. There are two major types of tapetum, the glandular or secretory and the amoeboid tapetum. The glandular tapetum builds up endoplasmic reticulum and dictyosome-derived vesicles while the PMCs are in prophase. The tapetum begins to break down after meiosis, releasing its lysed cell walls and disintegrating cytoplasm into the locule. The amoeboid tapetum protoplasts stay intact when tapetal walls are lysed and intrude among the developing pollen grains (Esau, 1977). Of the 231 plants families listed by Davis (1966), 181 families have glandular tapetum, including the genus Corylus. Clément et al. (1998) reported that the tapetal cytoplasm is rich in ribosomes and rough endoplasmic reticulum (RER) saccules, but the locular

9 fluid of the PMC, which consists of neutral polysaccharides, pectins, and protein, could be detected only in small amounts. Meiosis of PMCs to the release of microspores: Meiosis is one of the most complex events that occurs during gametogenesis because it involves the transition from a diploid to haploid state (Morohashi et al., 2003). Each of the PMC undergoes meiosis, resulting in a tetrad of haploid cells. These cells are held together by a callose (β-1,3-glucan) wall (Bewley et al., 2000). The synthesis of callose begins with initial deposition during the first meiotic prophase, and eventually spreading over the entire microsporocyte surface (Bhandari, 1984). The duration of meiosis can be as short as less than a day to as long as three months (Bennett et al., 1971). Sporopollenin is observed on the proUbisch bodies and on primexine matrix in the tetrad. Dictyosomes in the tapetum reach the maximal development at this stage (Clément et al., 1998). The tetrads are released by the action of callase (β-1,3-glucanase), an enzyme produced by the tapetal cells (Bewley et al., 2000). The middle layers cannot expand themselves as the sporogenous cells are developing into PMCs, because their cells are unable to divide anticlinally so they are crushed into “inexorable endothecium” (Davis, 1966). 1.2.2. Microgametogenesis After uninucleate microspores are released, they develop into pollen grains. The microspores undergo two mitotic divisions after release from the

10 callose wall. The first division is asymmetric and yields the vegetative cell, a larger cell, and the smaller generative cell. A second mitotic division of the generative cell yields two sperm cells (Bewley et al., 2000). Nuclear pore complex (NPC) density of the vegetative nucleus is twice as high as that of the sperm nuclei (Straatman et al., 2000). During pollen mitosis of the bicellular pollen grain stage, pollenkitt, composed of lipids and proteins, is located at the interface of the tapetal plasma membrane and loculus (Clément et al., 1998). The process when the generative nucleus undergoes mitosis to produce two sperm cells is called spermatogenesis. It can occur either in the pollen grain or in the pollen tube, depending on generative cell mitosis timing (Southworth and Russell, 2001). When division happens in the pollen grain, it becomes a tricellular pollen. If in the pollen tube, the pollen sheds as bicellular pollen (Rudall, 2007). Of 192 plant families, 137 families shed their pollen as bicellular and 55 families as tricellular (Davis, 1966). The pollen type of genus Corylus has not yet been reported. After the first mitotic division in the pollen, the generative cell wall is reduced or completely disappears in some species (Southworth and Russell, 2001). In higher plants, the generative cell migrates to the interior of the vegetative cell and remains wrapped by the vegetative cell (Russell et al., 1996).

11 1.2.3. Pollen Grains The mature pollen wall is composed of two layers, an inner, intine, and an outer, exine. The intine wall is largely pectocellulosic (Bedinger, 1992). The protein components of the intine are thought to be derived from gametophytic gene expression, whereas those of the exine are thought to be produced by the sporophytic tapetal layer (McCormick, 1991). The exine wall is composed of sporopollenin, a complex substance that is very resistant to degradation (Bedinger, 1992), which is currently thought to be a mixed polymer containing both phenolics and long-chain fatty acid derivatives (Bedinger et al., 1994). The exine is also reported to be synthesized of highly specialized extracellular matrix (ECM) which protects the pollen grain when it is released into the environment (Steiglitz, 1977). In maize pollen, there is a high accumulation of proteins, such as actin and tubulin, after microspore mitosis. These proteins will be used in the growth of the pollen tube (Mascarenhas, 1990). Tryphine on the pollen surface is important for pollen hydration in some species. It may be required for direct or indirect signalling to the stigma (Bedinger et al., 1994). The protein content of pollen from wind-pollinated plants tends to be lower than from insect-pollinated plants, and can range from 10 to 30% among plant species (Burgett et al., 1989). Although hazelnut is considered a wind- pollinated crop (Germain, 1994), it is one of the pollen sources for honey bees in the Pacific Northwest (Burgett et al., 1989).

12 Pollen viability of some hazelnut cultivars varies from 50% to 70% (Barbeau, 1972; Romisondo, 1977). Pollen grains of Corylus avellana require 90-95% RH for 2-3 hours for optimum germination. Immediately after dehiscence from anthers, pollen grains can have a high germination rate (87%). However, the germinability is significantly reduced when pollen grains are stored under desiccating conditions for 12 hours at 18-20 o C and 27 o C, and is eliminated at 40 o C (Heslop-Harrison and Heslop-Harrison, 1985). 1.3. Phenology 1.3.1. Temperature Factor One major factor that affects plant development and more specifically flowering is temperature (Ingram and McCloud, 1984). Faust (1989) reviewed research about time of bloom of temperate tree fruit and stated that most fruit trees enter a dormant period in late fall and resume growth in early spring. During this dormant or “rest” period plants cannot resume growth until the chilling requirement is met. Trees also have a heat requirement before bloom can occur. Besides temperature factors, Faust (1989) concluded that natural climate and cultural practices such as irrigation, sprays, and bud scale removal in the previous and current growing season can all have an impact on ‘time of bloom’ of fruit trees. Hampson (1995) found that catkin density was reduced by 64-74% in ‘Ennis’ and ‘Barcelona’ hazelnut trees which had been

13 under heavy shade. However, flowering was not as sensitive to shading as yield. 1.3.2. Chilling Requirement and Heat Accumulation Chilling Requirement The chilling requirement has been studied in various temperate fruit and nut species. In pistachio, Küden et al. (1995) found that in their studies of two male and five female cultivars, male cultivars had lower chilling requirements than female cultivars. In ‘Ahmadaghaei’, ‘Fandoghi-Ghafuri’, and ‘Chorouk’ pistachio, the optimum chilling hours were 1000, 1200, and 1400 hrs at 4-5 o C, respectively (Esmaeilizadeh et al., 2006). In 12 almond cultivars, Rattigan and Hill (1986) found that 220-320 CU was required for breaking dormancy in flower buds and subsequent floral development. Heat Accumulation Spring temperatures are very important with respect to the time of bloom in temperate fruit trees. The amount of heat required for each plant varies upon physiology of each species (Faust, 1989). The heat requirement is normally calculated as growth (growing) degree days (GDDs) or growth (growing) degree hours (GDHs). Richardson et al. (1975) defined one growing degree hour (GDH) as one hour at a temperature 1 o C above the base temperature of 4.5 o C. GDH is calculated by subtracting 4.5 o C from each hourly temperature between 4.5

14 and 25 o C. All temperatures above 25 o C are assumed equal to 25 o C, so the greatest accumulation for any one hour is 20.5 GDH. Ashcroft et al. (1977) used statistical methods to calculate the CU and GDH requirements of the deciduous fruit trees ‘Tilton’ apricot, ‘Italian’ prune, ‘Elberta’ peach, ‘Bing’ cherry, ‘Bartlet’ pear, and ‘Delicious apple’. Using temperature data and dates of full bloom from a period of six years, they were able to determine the CU required for rest and GDH required for full bloom in the above crops. They suggested that this method could be used to determine other phenological stages such as bud swell, but many years of data specific to that stage would have to be available. Spiegel-Roy and Alston (1979) suggested that heat requirement alone would be an adequate criterion for pear cultivar selection in Israel. They made this suggestion based upon a weak correlation between chilling requirement and bloom date, and a strong correlation between heat requirement after chilling and bloom date. Similar results from experiments on peach and western sand cherry by Werner et al. (1988) suggested that the basis for the difference in time of bloom was due to a difference in the base temperature of heat accumulation and not related to chilling requirements. In addition, Gianfagna and Mehlenbacher (1985) suggested that late flowering in apple is not a result of high chilling requirement, but of high heat and high minimum temperature requirements for bud growth.

15 1.3.3. Modeling as Tool to Predict Bloom Being able to estimate the time at which rest completion occurs would help growers determine whether specific cultivars of crops of interest can be grown in their area, when the growing degree hours accumulate enough to induce bud development, the time at which cultural practices such as irrigation should be applied, and the time at which the trees lose their cold hardiness and begin to grow with warm temperatures (Richardson et al., 1974). Cumulative chilling and heat unit models have been created for various tree fruit species. Two important examples are the Chill-Units model (also known as Utah model) (Richardson et al., 1974) and the dynamic model (Erez et al., 1990). Chill-Units Model Richardson et al. (1974) developed the Chill-Units model for ‘Redhaven’ and ‘Elberta’ peach, where an hour of exposure to 6 o C equals one chill-unit (CU), and the CU value is lower as the temperature increases or decreases. The value of the chill-unit is presented in following paragraph. This model was tested and works well in Washington, Georgia and Utah. < 1.4 o C = 0 CU 1.5-2.4 o C = 0.5 CU 2.5-9.1 o C = 1 CU 9.2-12.4 o C = 0.5 CU

16 12.5-15.9 o C = 0 CU 16-18 o C = -0.5 CU > 18 o C = -1 CU Aron (1975) could not duplicate the calculation successfully in California. Richardson et al. (1975) reported that this discrepancy is a result of the land-sea breeze effect along the California coast, where there is an air mass change twice each day. Such changes in air mass compromise the chill-unit method of synthesizing hourly temperatures. The Dynamic Model The dynamic model was developed in South Africa in a dormancy- breaking study of peach buds under controlled conditions (Erez et al., 1990). This model started as a two-step model (Fishman et al., 1987a) and evolved into the dynamic model over at least two decades. The Two-Step Model was developed by Fishman et al. (1987a). This model describes thermal dependence of the dormancy breaking phenomenon, assuming that the level of dormancy completion is proportional to the amount of a certain dormancy breaking factor which accumulates in plants by a two- step process. “The first step represents a reversible process of formation of a precursor for the dormancy breaking factor at low temperatures and its destruction at high temperatures. The rate constants of this process are assumed to be dependent upon the temperature according to the Arrhenius

Full document contains 122 pages
Abstract: Pollen development is an important process in male flower development, the timing of which may be correlated with time of pollen shed in hazelnut (Corylus avellana L.). Early to very late blooming cultivars were identified and the relationship of microsporogenesis and microgametogenesis, and time of pollen shed were studied in nine hazelnut cultivars. Most advanced catkins from a single tree of each cultivar were collected each week from 4 Aug. to 6 Dec. 2002, and on 17 Jan. 2003, stained and analyzed by light microscopy. The phenology part of this dissertation studied the role of the chilling requirement as chill units (CU) and heat requirement as growing degree hours (GDH) in pollen shed. Hazelnut twigs of three cultivars; 'TGDL', 'Barcelona', and 'Hall's Giant' were collected at weekly intervals starting from early Fall 2006 through the time of anthesis in the field in winter 2007. Twigs were then held at a different constant temperature 0, 5, 10, 15, or 20 °C. Observing these twigs weekly, the time of anthesis (50% pollen shed) was recorded. A parallel study was conducted in more controlled conditions by collecting hazelnut twigs of the same three cultivars on 1 Nov. 2006 and holding them at 5 °C in a cold room. Five twigs of each genotype were brought out to room temperature at 5-day intervals in order to force them to bloom. Numbers of catkins that shed pollen were recorded every 5 days. From the results, we propose a model of hazelnut staminate flower development that begins with catkin differentiation concurrent with early stages of pollen development. Catkin length increased steadily and reached a lag phase at the end of microsporogenesis. While there is no external change, microspores continue microgametogenesis, the catkins are endodormant, unable to be induced to shed pollen, and accumulate chilling during this period of time. We propose that the chilling requirement is met when pollen reaches the mature stage and when catkins are in the latter stages of the lag phase of growth. At this point, GDH accumulate and catkins become ecodormant. As chilling continues to accumulate, the amount of GDH required for pollen to shed is reduced. A similar quantity of GDH was required for all genotypes from pollen grain maturity to catkin elongation.