THS THESIS -—-w V" LIBRARY Michigan Stats University ' * "a. This is to certify that the thesis entitled THE USE OF CONTROLLED FREEZING TO EVALUATE FACTORS INFLUENCING CRITICAL TEMPERATURES FOR FREEZE INJURY IN DEVELOPING GRAPEVINE BUDS presented by David Elwood J onhson has been accepted towards fulfillment of the requirements for M. S . Horticulture degree in MM . Major professor Date Aug. 11, 1980 0-7 639 OVERDUE FINES: 25¢ per day per item RETURNING LIBRARY MATERIALS: Place in book return to remove charge from circulation records THE USE OF CONTROLLED FREEZING TO EVALUATE FACTORS INFLUENCING CRITICAL TEMPERATURES FOR FREEZE INJURI IN DEVELOPING GRAPEVINE BUDS By David Elwood Johnson A DISSERTATION Submitted to Euchigan State University in partial fulfillment of the requirements for the degree of MMSTER OF SCIENCE Department of Horticulture 1980 ABSTRACT THE USE OF CONTROLLED FREEZING TO EVALUATE FACTORS INFLUENCING CRITICAL TEMPERATURES FOR FREEZE INJURY IN DEVELOPING GRAPEVINE BUDS By David Elwood Johnson A controlled freezing technique was developed and used to determine critical temperatures for developing grape- vine buds. Critical temperatures were estimated for ’Concord' grapevine buds at defined stages of development under both wet and dry surface conditions. Tissue surface moisture strongly influenced freeze resistance. Hardiness was similar for developing 'Concord' buds whether on green- house forced cuttings, on cuttings taken directly from the field during spring development, or on outdoor grown whole potted vines. Pre-freeze temperature had no apparent effect on hardiness of developing buds. ‘While freeze resistance always decreased with advancing phenological deve10pment, morphological characteristics prior to bud development had no effect on bud hardiness or rate of development. Cultivar differences affected both the rate of bud development, and the hardiness at a given stage, of buds forced from stored cuttings. ACKNOWLEDGMENTS I wish to thank Dr. Frank Dennis and Dr. Jerry Cash for their guidance and constructive criticisms of my research and writing: Jim Wolpert and Tim Hansfield for their invaluable friendship and help; my major professor, Dr. Stan Howell, for his guidance, confidence, and inspira- tion: and.my wife, Connie, for her love and support, and many hours of help in the production of this manu- script. ii TABLE OF CONTENTS LIST OF TABIES O O I O O O O O O O O C 0 LITERATURE REVIEW . . . . . . . . . . . Introduction . . . . . . . . . . . Critical Temperatures. . . . . . . Freezing in Cold Tender Plant Tissues. Controlled Freezing. . . . . . . . . . . . Use of 7(2 and the Rx2 Contingency Table in the Analysis of Hardiness Data . . . . . PURPOSES. I O O O O O O O O O O O O O 0 MATERIALS AND METHODS . . . . . . . . . PlantMaterial . . . . . . . . . . Hardiness Evaluations. . . . . . . DevelOpmental Studies. . . . . . . Bud Weight andIMOisture Content. . RESULTS AND DISCUSSION. . . . . . . . . Controlled Freezing Technique. . . . . . . Factors Determining Critical Temperatures in Developing Grapevine Buds . . . . . . Estimation of Critical Temperatures. Summary and Conclusions. . . . . . LITERATURECITED. . . . . . . . . . . . APPENDIX. 0 O O O O O O O I O O O O O 0 iii Page iv w 11 14 18 19 19 22 25 26 26 30 37 39 5h 60 Table 1. 10. LIST OF TABLES Effect of surface moisture during controlled freezin on primary bud hardiness in 'Concor ' grapevines at several stages of phenological development . . . . . . . . . Effects of cane excision and pre-freeze con- ditions on 'Concord' primary bud hardiness during controlled freezing . . . . . . . . Effect of stage of phenological development on primary bud hardiness of 'Concord grapevine cuttings exposed to controlled fmeZing O O O O O I I O I O C O O O O I 0 Effect of node position on development of primary buds on 6-node 'Concord' grape- vine cuttings. . . . . . . . . . . . . . . Effect of pre-freeze treatments on hardiness of full-swell 'Concord' grapevine buds subjected to controlled freezing . . . . . Effects of cane characteristics and bud po- sition on hardiness of deve10ping 'Concord' grapevine primary buds . . . . . . . . . . Effect of eXposure status on rate of primary bud development in l-bud 'Concord' grape- vine cuttings. . . . . . . . . . . . . . . Effect of cutting origin on rate of primary bud development in l-bud 'Concord' grape- vine cuttings. . . . . . . . . . . . . . . Effect of cane diameter and original nodal osition on rate of primary bud deve10pment n l-bud 'Concord' cuttings. . . . . . . . . Effect of cultivar on primary bud develOpment rate in l-bud grapevine cuttings . . . . . . iv Page #1 #2 #3 AA #5 A6 A7 #8 A9 50 Table Page 11. Effect of cultivar on hardiness and per cent mortality of deve10ping grapevine primary buds subjected to controlled freezing. . . . 51 12. Effect of cultivar on tissue weight and moisture content of developing grapevine buds forced from.cuttings in a greenhouse. . 52 13. Estimated critical temperatures for devel- Oping grapevine buds forced from cuttings in a greenhouse. . . . . . . . . . . . . . . 53 Al. Raw mortality data for all freezing experi- ments on 'Concord' buds frozen on l-node cuttings O O O O O I O O O I O O O O O O O O 60 A2. Raw mortality data for freezing experiments on whole potted 'Concord' vines. . . . . . . 67 A3. Raw mortality data for freezing experiments on buds from 3 different cultivars . . . . . 68 LITERATURE REVIEW Introduction Grapes are the leading horticultural crop in terms of‘worldewide production (Harlan, 1976). Among the United States, Michigan is fourth in production, with 15,800 acres of vines (Lovett and Collins, 1978), and a 1979 crop val- ued at over $10 million (W.R. Grevelding, personal come munication). Freezing injury represents a major economic loss to United States agriculture and 63% of this loss is to fruit crops (Mayland and Cary, 1970), with an estimated 10% of the grape crop being lost annually (Reingold, 1960). Spring freeze damage can be devastating to grapes, and no viticultural area in the United States is completely free of this hazard. In.Michigan, substantial crop reduction from spring freeze has occurred in 11 of the 21 years from 1957 to 1977 (Howell and Wolpert, 1978). In two recent years the Michigan grape industry has been officially de- clared a ”disaster area” as a result of spring freeze damage (Howell, 1976). Spring freeze damage can be reduced by site selection, site modification, cultural practices, and microclimate modification (Rogers and Swift, 1970; Ballard and Proebsting, 1972). Certain aspects of these techniques, as they relate 2 specifically to grapes, are presented by Dethier and Shaulis (196A). methods of frost protection not currently used on a large scale, but under investigation, include modification of pruning techniques in grapes to retard spring bud development (Howell and Wolpert, 1978), application of insulating stable foam (Bartholic g§_§1,, 1970) and the use of growth regulators and systemic cryo- protectants (Burns, 197A: Ketchie and Murren, 1976). Frost protection may be divided into two categories; passive or preventative measures such as site selection and cultural manipulation, and active or palliative measures such as heating or sprinkling. The former are taken well in advance of possible frosts, while the latter require a critical decision to begin protective measures. The consequences of error in the decision to start protec- tive measures are excessively high operation costs on the one hand, and crop loss on the other. The decreasing availability and increasing cost of fuel, has stimulated interest in reducing energy consumption for freeze protec- tion. Increasing the accuracy and availability of hour to hour weather forecasts with infrared thermal imagery, and developing protective systems with rapid start-up capa- bilities, may signigicantly aid this effort (Proebsting, 1975). The final decision to start protective measures muSt be based on an understanding of the temperature likely to cause injury or the "critical temperature" (Proebsting and.Mills, 1978). Critical Temperatures Critical temperature, defined as the lowest temp- erature endured for 30 minutes or less without injury, ‘was apparently introduced by Young (1920). LT5O, or the temperature estimated to cause 50% kill, (Timmis, 1977; Dennis 9331., 1975) as well as LT10 and LT90 (Proebsting‘g§_gl,, 1978) have also been used as indi- cators of the critical temperature. Critical tempera- tures for several fruit species, including grapes, are given by Young (19A?) and Rogers and Swift (1970). Both publications point out the deficiencies in the data used to arrive at these temperatures.‘ Critical temp- eratures for tree fruits and small fruits in Michigan during spring deve10pment were listed by Eichmeier gt_§l, (1965), but the source of the data and methods used to determine these temperatures were not given. .Mbre detailed information on critical temperatures has been given for tart cherries in Michigan (Dennis and Howell, 197A), and for 6 deciduous tree fruit species in.Wash- ington (Ballard et al., 1971). Critical temperatures for developing grape buds have not been as extensively studied. Clore gt_§l, (197A) suggest that grape buds in full swell can tolerate -5.5°C. Preliminary estimates of critical temperatures for grape buds at several stages of spring deveIOpment were reported by Proeb- sting et a1. (1978). The concept of critical temper- ature was extended from fruit species to cone buds in A coniferous seed orchards by Timmis (1976). Critical temperatures are affected by several fac- tors. The most important factor is the stage of phen- ological development, (Ballard and Proebsting, 1972). Early studies on the relationship of developmental stage and frost hardiness in peach buds were presented by Blake and Steelman (19AA), and Proebsting and Mills (1961). Generally, the more advanced the stage of devel- opment, the higher is the critical temperature, although this is not always strictly true (Rogers and Swift, 1970). For example, developmental stage had no effect on freeze resistance in the frost tender reproductive tissues of deveIOping wheat ears after emergence from the leaf sheath (Single and Narcellos, 197A). Suitable defini- tion of the stages of development can be a problem. Young (19A7) used 3 descriptive stages for all fruit spe- cies. Most recently, Proebsting and Mills (1978) used 7 to 9 stages separately defined for each of 6 tree fruit species studied. Stages of grape bud develop- ment were described and given letter designations by Baggiolini (1952). Proebsting gt_§;, (1978) used de- scriptive terms with definitions, and an effort has been made to coordinate these definitions with other investi— gators (E.L. Proebsting, personal communication). Species and cultivar are also important when con- sidering critical temperatures. Early studies indi- cated that apple blossoms were more tender than peach, 5 which were more tender than cherry (Chandler, 1913). Varietal differences may be related to developmental rate. Early blooming cultivars of apricot showed 3 times the injury from a naturally occurring frost than late blooming cultivars (Layne, 1966). At one site in Huchigan, during the severe frost of April 27, 1976, trace primary bud kill occurred in the grape cultivar 'Vidal-256"whose buds did not reach full swell until may a, 1976. However, primaries of 'Foch"were 5-7cm long when frosted and suffered 95% kill (Howell, 1976). Variability in bud development within a cultivar can also be a factor. Flowers in late blooming lateral buds of 'Golden Delicious' apple may survive a frost that kills buds on the spurs (Ballard and Proebsting, 1972). Some cultivar differences appear to be inherent rather than being related to deve10pment rate. Buds of several peach cultivars showed varying degrees of injury at the same temperature and stage of develOpment (Blake and Steelman, 19AA). 'Earliril' apricot and 'Chinook' cherry are early blooming cultivars which appear to be hardier than late blooming cultivars (Ballard and Proebsting, 1972). In addition to cultivar differences, characteristics of grape vines such as sunlight exposure of the leaf at a given node, periderm color, cane diameter, presence of persistant laterals, and node position on the cane rel- ative to base and apex have been related to winter hard- iness of dormant grapevine buds (Howell and Shaulis, 1980). 6 Preliminary observations by N.J. Shaulis (personal commu- nication) have suggested that poor sunlight exposure dur- ing the previous growing season may increase spring frost injury in 'Concord' grapevines. The duration of the low temperature may be important. Generally, the longer the temperature stays below the critical level, the greater the damage (Rogers and Swift, 1970). Ballard and Proebsting (1972) suggest that dura- tion is important as a factor in tissue-air temperature equilibrium but that the minimum temperature is rela- tively more important than the duration. Humidity, usually measured as dew point, can be a factor in spring freeze damage. Rogers and Swift (1970) suggest that in deciduous fruits, more damage results from a given temperature at lower humidities. Ellison and Close (1927) concluded that when the dew point is low, severe injury occurs to apple blossoms at temperatures causing only slight injury when the dew point is high. Ballard and Proebsting (1972) state that dew point prob- ably has no effect on actual tissue hardiness, but that low dew points favor radiation to the sky and evaporative cooling that may result in tissue temperatures 3°to AOF lower than the air temperature on calm nights, while 'winds of 2 miles per hour or more tend to keep tissue temperatures close to air temperatures. The latent heat released.when dew or frost forms as the temperature falls to the dew point may considerably retard the rate of 7 fall (Rogers and Swift, 1970). This, coupled with the fact that no visible frost forms under conditions of sufficiently low dew point, may have led early investi- gators to conclude that low dew points result in greater injury at a given temperature. Recent results on citrus (Young, 1969) and mulberry leaves (Kitaura, 1967) suggest that the reverse is true. External ice such as hoar frost can readily nucleate plants via entry sites such as sto- mates, lenticels, and wounds (Burke,gt_al,, 1976). Conditions prior to the frost may also affect the temperature necessary to cause injury. Ballard and Proebsting (1972) state that temperatures during the 24 hours preceding the freeze can have profound effects on hardiness of tree fruit buds prior to bloom, although they do not describe the nature of these effects. Hewett gt. 3;, (1978) were not able to find any clear relationship between damage to blossom buds at a given temperature following warm weather, and damage following cool weather. Storage up to 6 days at 1°C did not increase the freeze resistance of develOping wheat ears (Single and Narcellos, 197A). Longer term effects may be important, as Ballard and Proebsting (1972) state that apple blossoms deve10ped in cool weather have been shown to be more frost resistant than those develOped in warm weather. Under conditions of severe soil moisture shortage, water content of apple blossoms was reduced and damage from frost was candid- erably less than that to blossoms of well watered trees (Mbdlibowska, 1961). Hewett et a1. (1978) noted severe 8 Spring freeze damage to developing grapevine buds following sprinkler irrigation while adjacent unsprinkled vines were unaffected. These authors also found that buds from sprin- kled peach and apricot trees had higher moisture content, and suffered more damage from artificial freezing, than unsprinkled buds. They suggest that if buds have been sub- jected to recent sprinkler irrigation or rain, critical temperatures must be revised and that care should be taken to avoid sprinkling before predicted freezes. Freezing in Cold Tender Plant Tissue New shoots and leaves of non—acclimated woody peren- nials are injured by temperatures only slightly below 0°C. Supercooling and freezing point depression are means by which some resistance is achieved, but when ice nucleation occurs, rapid intracellular freezing takes place, resulting in destruction of membrane continuity (Burke‘§§_gl,, 1976). Injured foliage appears flaccid and water soaked, as cell membranes have lost their semipermeability and intracellu- lar compartmentalization has been destroyed. Intracellu- lar freezing is nearly always fatal (Scarth, 19Ah; Evert, 1967). In herbaceous tissues that tolerate ice formation, (e.g., hardened'winter wheat, cabbage, turf grass, etc.), ice appears to form in extracellular spaces. During freez- ing in hardened barley crown tissue, water readily diffused from the protoplast to extracellular nucleation sites, while freezing in cold.tender tissuesnwas characterterized 9 by explosive ice formation resulting in the rupture of plasma membranes and release of cell contents into extracellular spaces (Olien, 196A). In mature moss shoots ‘water moved to sites of extracellular ice formation allow- ing survival, while the water in actively growing shoots supercooled, then froze intracellularly; in addition, cells of the mature shoots lost water faster than cells of young shoots when plasmolyzed (Hudson and Brustkern, 1965). These findings indicate that the plasma membrane of cold tender tissues is less permeable and will not readily allow move- ment of water to external nucleation sites. Increases in cell permeability have been associated with the hard- ening process (Scarth, 194A; Evert,1967). In some frost susceptible plants such as Solanum tuberosum L., the quantity of fluid lost by cells, or its rate of loss may be insufficient, causing ice to form at many points through- out the tissue (Hudson and Idle, 1962), rather than at sites which accomodate ice with little damage to the plant, as is the case with hardy‘woody tissue (Burke g§_gl,, 1976). Exotherm studies have shown that azalea flower primor— dia (George §t_g;,, 197A), peach flower primordia (Quamme, 1978) and blueberry buds (Bittenbender, 197A) are killed at the onset of ice formation, and rely on supercooling for survival even during midwinter. Buds of the wild grapevine liti§_riparia Michx. apparently behave similarly (Pierquet, g§_§;,, 1977). This is not the case for apple buds during midwinter (Burke gt_§;,, 1976). The extent of supercooling lO apparently is related to the number and quality of ice nucleating sites. The freezing temperature of conifer needles rises with increasing needle length and the quan- tity of ice nucleators, rather than the quantity of water, may be critical (Kaku and Salt, 1968). A similar rela- tionship exists in mature‘ngus leaves, although in imma- ture leaves, age was the important factor regardless of size (Kaku, 1971). Changes in hardiness within a tissue that relies on supercooling may be related to the abun- dance of nucleating centers within cells and or the pres- ence of effective barriers to nucleation (Burke M” 1976). Supercooling apparently provides some spring frost resistance to the develOping buds of deciduous fruit species. Mbdlibowska (1962) observed supercooling of apple blossoms in the orchard during natural spring frosts. Ballard and Proebsting (1972) report trace injury vs. 100% kill in adjacent peach trees and attribute this to super- cooling that did not persist in the damaged tree. Hewett g§_gl, (1978) state that the increased spring freeze injury to sprinkled grape buds may have resulted from lower bud temperatures due to evaporative cooling or from reduced supercooling, but they state elsewhere that the wet bulb temperatures were not low enough to account for the injury. It therefore appears that supercooling was a factor in the superior resistance of the unsprinkled buds. 11 Controlled Freezing There has been widespread use of controlled freezing techniques in studies on cold hardiness. Controlled freezing of fruit buds to determine critical temperatures has been used for decades (Proebsting and Mills, 1971). All controlled freezing studies assume a close relation- ship between the behavior of artificially frozen material and material frozen under natural conditions. This assump- tion has proven accurate enough for these tests to be very useful (Levitt, 1951; Lapins, 1961; Proebsting and.Mills, 1978). Injury to WMontmorency' cherry from naturally occurring spring freezes tended to be greater in early stages of bud deveIOpment and less in advanced stages than predicted by controlled freezing (Dennis gt_gl,, 1975). .Most freezing studies have made use of a standard, or some- what modified, chest freezer. Scott and Spangelo (196A) described the design and construction of a portable cold stress unit for‘inugitgbfreezing of whole fruit trees. Quamme §§_gl. (1972) described a technique utilizing the vaporization of liquid nitrogen for freezing biological materials. A control system and solenoid valve are used to either regulate release of nitrogen vapor or activate an electric heater. Advantages are accurate temperature control, rapid response, and a wide range of possible temp- eratures, (+5000 to -100°C), although capacity is rather limited. Whatever the system employed there are several 12 important factors that should be considered in any con— trolled freezing experiment: 1. Iggperaturegstratification'within the freezing chamber can cause some material to be subjected to temp- eratures different from other material or recorded temp- eratures. A fan within the freezer improves temperature distribution but also adds heat (Proebsting and Mills, 1971) . 2. Eguctuationg in freezer tem erature, caused by thermostatically controlled on-cff cycling, can be sig- nificant. True temperatures are alternately above, then below intended temperatures. This problem has been cir- cumvented by setting the freezer to a constant low temp- erature and insulating the plant material within the freezer, resulting in a gradual decline in tissue temp— erature as equilibrium is established (Howell and Weiser, 1970). Temperature within a conventional freezer can be lowered by controlling the coolant flow rather than by on- off cycling of the entire system. This can be accom— plished using an expansion pressure regulator (EPR) valve (Lumis 2131., 1972). 3. Rate of temperature change, particularly rate of fall, is generally considered important (Lapins,196l; Proebsting and Mills, 1971; Daniel and Crosby, 1971). How- ever, the rate of cooling did not alter the freeze resist- ance of guayule (Mitchell, 19th), and early studies have shown that cold tender tissues are not affected by the 13 rate of freezing (Levitt, 1951). Perhaps this is re— lated to the lack of significant movement of intracellu- lar water to sites of extracellular ice nucleation. This may be the reason that Proebsting and Mills (1971) state that, for fruit buds during dormancy, temperatures should be lowered at 2°F per hour, while during the prebloom period, rate of fall may be increased to 5°F per hour. The same authors point out that if the tissue mass is so large that its heat capacity and release of heat of fusion causes tissue temperature to lag measurably behind air, (box), temperature, very slow rates of fall may be nec- essary. Cold tender beet root tissue frozen to ~h°C suffered less damage from ultraslow cooling (0.2°C/hr.) than from moderately slow cooling (3.3°C/hr.), while ultra- fast rates produced no more damage than the moderate rate (Finkle g§_§l,, 197A). This may indicate some movement of water out of the cells to extracellular nucleation sites when rates of fall are slow enough. Rates of fall of 0.200 per hour probably seldom occur in natural freezes. Rates of tissue thawing may be important (Siminovitch and Briggs, 1953) but Ballard and Proebsting (1972) suggest that for developing fruit buds, this is a very minor fac— tor if it has any influence at all. A. Supercooling was prolonged, and damage from con- trolled freezing was reduced, in apple blossoms when spurs or single flowers were cut from trees, while blossoms on 'whole potted trees seldom supercooled throughout a freezing 1h experiment (Modlibowska, 1962). Mulberry leaves super- cooled.more when detached from the stem (Kitaura, 1967). If excised.plant samples tend to supercool more than whole plants, erroneous conclusions may be made about crit— ical temperatures for natural freezes. To help prevent this, Proebsting and Mills (1971) suggest wetting the surface of all samples before placing them in the freezer. This reduces apparent cold resistance somewhat and Hewett g§_gl, (1978) state that under these conditions, test injury correlates well with field injury. Because the excising of tissues prolongs supercooling, it would seem particularly important to reduce this by‘wetting the tissues if the test temperatures are not maintained for long periods. HE? of‘X2 and the Rx2 Contingency Table in the Analzsis of Hardiness Data Horticulturists often misuse the analysis of vari- ance (Evert and Howell, 1979). This and other para- metric tests require random sampling from a population ‘with normal distribution and homogeneous variance for all treatments. Mbst parametric statistics also require an interval scale of measurement, that is, a scale that re- flects the size of differences between measurements (Conover, 1971). Discrete data, such as living vs. dead buds, are not interval measurements, but rather, they are nominal measurements. Mbst nonparametric tests assume either a 15 nominal scale or an ordinal (ranked) scale to be appro- priate, and that the distribution function of the ran- dom variable producing the data is unspecified (Conover, 1971). There are several nonparametric tests that may be appropriate in the analysis of hardiness data. The modified Friedman test as described by Evert and Howell (19790 appears well suited to certain hardiness studies ‘where the amount of injury among the various treatments can be ranked. The Rx2 contingency table as described by Cramer (19th), Steel and Torrie (1960), Conover (1971) and Meddis (1975) provides another technique of evaluating hardiness data. A contingency table is an array of natural numbers in matrix form in which the numbers represent counts or frequencies. The Rx2 table consists of R rows of categories or treatments divided into 2 classes (e.g., alive vs. dead). The question to be answered is "Do the treatments or categories significantly alter the proportion of objects or results in each of the 2 classes?” The null hypothesis (Ho) would then be: P1=P2=P3=...PR, or that the proportion of buds in a given class is the same for, or independent of, the various categories or treatments. The alternate hypothesis (H1) 'would be that in at least 2 of the categories or treat- ments, the proportions are not the same. The calculation of the test statistic used to examine these hypotheses is given in different but equivalent forms by Conover 16 (1971, page 152), and Steel and Torrie (1960, page 371). The test statistic is distributed approximately as X2 with (Rel) degrees of freedom (Cramer, 19h6). Large values affix? indicate that the deviation from the null hypothesis is signigicant. Cramer (19A6) and Steel and Torrie (1960) discuss methods of determining the degree of dependence indicated by the sample. Because the asymptotic distribution, (12), is used, the approxima- tion of significance levels may be poor if the expected values of a given contingency table are small. The expected value for any cell in a contingency table is the product of the row and column totals for that cell, divided by the grand.total. The approximation is consid~ ered poor if 20% or more of the expected values are less than 5 (Neddis, 1975: Conover, 1971). If some of the expected values are too small, several categories may be combined.provided they are similar in some respects such that the hypotheses retain their meaning (Cramer, 19h6; Conover, 1971). Steele and Torrie (1960) indicate that data such as the numbers alive and dead at several test temperatures may be pooled into one contingency table. A 2x2 contingency table can be used to determine significant differences between any 2 of the various categories in an Rx2 table. If 5 cultivars are sub- jected to freezing stress, the 5x2 contingency table can be used to determine whether the frequencies of living and 17 dead buds are dependent on cultivar. A signigicantly high value of the test statistic indicates only that the frequencies for at least 2 of the 5 cultivars are different. To determine whether any given 2 of the 5 cultivars differ significantly from each other, the appropriate 2x2 table must be used. In discussions of 2x2 contingency tables, a correction for continuity, ("Yates correction”) is sometimes recommended (Steel and Torrie, 1960) to compensate for the use of a contin- uous distribution function (X?) to approximate the discrete distribution function of the test statistic. Conover (1971) cites several authors with whom he is in agree- ment in recommending against the use of "Yates correc- tion” as being overly conservative. The use of a one-tailed test, with its increased sensitivity, is apprOpriate for the 2x2 contingency table if one category is expected to have a greater proportion in a given class (Conover, 1971). Therefore, when com- paring 2 cultivars, or 2 developmental stages, where one cultivar or stage is expected to be less hardy, the one tailed comparison should be used. PURPOSES With the ever increasing cost of fuel, more emphasis will probably be placed on frost protection methods other than high energy consuming microclimate modification. This study was initiated for 3 primary purposes. The first was to deve10p a suitable controlled freezing technique for developing grapevine buds that would facil- itate the investigation of new frost control measures by repeated and readily controlled evaluations of their effects. The deve10pment of this technique would also make possible the accomplishment of the second purpose, that of determining actual critical temperatures for deve10ping grapevine buds. This would allow more effi- cient use of existing microclimate modification tech- niques. The third purpose was to investigate various factors that might affect critical temperatures and there- fore influence further investigations into frost protec- tion as well as the use of current frost control methods. 18 MMTERIALS AND METHODS ‘Plgp§,Materials. During midwinter of 1977 and.1978, cane cuttings were collected from a vineyard of 6-year—old 'Concord' (132i; labruscana Bailey) grapevines at the Mich- igan State University Horticultural Reasearch Center. In preparation for the 11 April 1978 experiment, cuttings were stratified in two groups at the time of collection. One group consisted of cuttings with light periderm color taken only from the interior of the vine canopy. The other, with dark periderm color, was from the vine canopy exterior. Cuttings for the 13 April 1978 experiment were stratified as to collection from either main canes or laterals and only cuttings of medium diameter (5.0-7.5mm) were used for forcing and hardiness evaluations. Each of the cuttings collected for the 17 April 1978 experiment included nodes 1 through 16. This allowed stratification based on nodal position as well as cane diameter. In November 1978, additional cuttings of the cultivars 'Baco-l' and 'Vidal—256' (interspecific hybrids of‘Eigig)‘were also collected from ”Tabor Hill” vineyard in Berrien county, Michigan. Cuttings were stored in moist peat at 1—2°C until used. Storage time was suf- ficient to satisfy rest and the length of time depended upon material needed for forcing in the greenhouse. Buds ‘were forced on a greenhouse mist bench unless otherwise 19 20 stated. .Mist cycle varied from 1 to A three second mistings every half-hour depending on outside conditions as this affected greenhouse temperature and evaporation rate. For the 11 December 1977 freezing experiment, buds were also forced in the same greenhouse without misting (cane bases in water), and in a growth chamber on a 1A hr—13°C/10 hr—8°C day/night cycle. These con- ditions approximate the mean max./min. temperature and photoperiod occurring during bud expansion in the spring of 1977 at the Michigan State University Horticultural Reasearch Center. Controlled freezing studies were also carried out on naturally developing buds taken directly as cuttings from a block of 12—year-old 'Concord' grape- vines and on non-excised buds on 2-year-old potted 'Con- cord' vines, grown at the MSU Horticultural Reasearch Cen- ter. Grapevine buds were classified into 5 deve10pmental stages based on those described by Proebsting,g§_gl. (1978). Stages are as follows: Sgglg_§£§g§, This is the first visible indication that growth has begun. A small crack occurs between the hard outer most bud scales as the bud begins to swell. If these scales have been removed or damaged, as often happens during handling of stored cuttings, this stage can not be accurately assessed. This stage would be inter- mediate between A and s as defined by Baggiolini (1952). 21 1§i3§§:§!211. This stage is approximately the equiv- alent of stage B or ”cotton tip" swell of Baggiolini and of "first swell" of Proebsting,gt_gl. At this stage the bud has swollen out of the hard outer bud scale and is globular, doe colored, and fuzzy in appearance. No green or pink color is visible. Egllggggll. This stage corresponds to stage C or "green tip" swell of Baggiolini and "full swell“ of Proebsting,g§_gl, The bud has elongated, being roughly 1.5 to 2 times as long as wide. One or more bulges of leaf tissue are visible and appear green or pink. The bud remains closed around the growing point. ,Bgdgbgggt, This stage is roughly equivalent to stage D or "first-leaf" of Baggiolini, (1952) and to "burst" of Proebsting g§_gl, (1978). Here the leaves have sepa- rated at the tip, usually exposing the growing point. No leaf has, as yet, made a right angle with the stem. Egpandedfighggp, Here the young shoot is h—6cm in length with 1-3 small leaves at right angles to the stem. This stage most closely approximates stage E or "leaf expansion” of Baggiolini.g§_gl, (1952), and "2nd leaf” of Proebsting g§_gl, (1978A). While bud develOpment is a continuum, only buds judged to typify a given defined stage were used for hardiness evaluations. In all deveIOpmental studies, buds‘were recorded as being at the stage they most closely fit. 22 Hardiness Evaluations. Freezing technique was a modification of that used by Howell and Weiser (1970). Buds were cooled, and at each of several test tempera- tures a portion of the buds was removed from the freezer and allowed to thaw at 2°C. Test temperatures were selec- ted such that the warmest temperature produced no injury and the coldest was lethal for all buds. The tempera— ture interval was 1.500. Freezing was conducted in a special walk-in freezing unit at the MSU Horticultural Research Center. This unit is equipped with several large blowers to minimize temp perature stratification within the freezer box. Temper- ature fluctuation from freezer on-off cycling does not occur as this unit is equipped with an expansion pressure regulator valve which regulates coolant flow. Box teme perature is lowered by the gradual cpening of this valve controlled by a time clock. Rate of temperature fall was at 3°C/hour. Box temperature was monitored via several 2h guage copper-constantan thermocouples distributed around and among the plant material. Bud tissue temperature 'was monitored by thermocouples inserted into several extra buds not being used for hardiness evaluation. Preliminary evaluation showed only minor (50.2%) dif- ferences between tissue and air temperature. Air temp perature therefore served as recorded test temperatures. Because developing grape buds are very susceptible to mechanical damage, they could not be bundled together for 23 freezing as suggested by Proebsting and.Mills (1971). The bases of one node cuttings were instead stuck into pieces of moist floral foam to provide both mechanical support and a continuing moisture supply. Depending on availability of buds of appropriate stage and or treatment, buds were randomly assigned to 1 group of 8 to 15 buds, (early experiments), or to 3 groups of 6 to 10 buds, (later experiments), per test temperature. A group from every stage or treatment was placed at random on a piece of moist floral foam. Therefore, either 1 foam piece, (early experiments), or 3 foam pieces, (later exper- iments), were removed at each test temperature. The number of foam pieces per temperature and the number of buds per treatment per foam piece are recorded in Table A1. In whole vine studies the entire potted vines were placed in the freezer with 3 vines selected at random being removed at each test temperature. To reduce supercooling, buds in the 27 April 1977 freezing experiment were moistened prior to being placed in the freezer, as suggested by Proebsting and Mills (1977). Results suggested this to be insufficient moistening and in the 13 May 1977 experiment buds were moistened in the freezer just prior to the beginning of freezing, (approx- imately 0°C). To reduce variability resulting from moist- ening, and to further reduce supercooling, buds were reg- ularly moistened throughout freezing experiments unless stated otherwise. In all cases moistening was via a fine 2h H20 mist from a hand held spray bottle. Bud viability was evaluated after 1 week of regrowth on a greenhouse mist bench. Buds were sectioned with a razor blade and lack of growth and tissue browning were used as criteria for determining the live-dead status. Buds were recorded as dead if the growing point or vascu- lar tissues had browned but not if portions of leaves had browned while the main axis was green and growing. All results were recorded as numbers of dead buds out of the total number for each treatment or developmental stage at each test temperature, (see Table A1). Significance of the differences in the frequencies of dead buds of the various treatments was determined using the?!2 test for Rx2 contingency tables, (see Literature Review p. 1A). Temperatures resulting in either 0% or 100% kill in all treatments in any one comparison were not used in the calculation ofXZ. LTSO calculations were made via the Spearman-Karber equation (Bittenbender and Howell, 197A). DeveloEmental Studies. Data on developmental rates of grapevine buds when excised from the parent vine were obtained by forcing stored dormant cuttings. Prior to the 3 August 1977 freezing experiment, 20 six-node 'Concord' cuttings were selected at random from the stored material and placed on the mist bench. The stage of develOpment of the buds at each node on each cutting was recorded at 2-day intervals. Prior to the 11, 1A, and 17 April 1978 freezing experiments A groups of 32 one-node cuttings were 25 selected at random from each category of 'Concord' cut- tings and were randomly arranged on the mist bench. The stage of development of each bud was recorded at 2 day intervals for all categories of cuttings. Prior to the 29 June 1978 freezing experiment, 3 groups of 2A one-node cuttings were selected at random from each of the 'Concord',_'Baco-l', and 'Vidal-256' cuttings, and bud deve10pment was recorded at 2 day inter- vals. Buds remaining dormant at the end of all deveIOp- mental studies were sectioned.to determine live—dead status and data were recorded based on the number of live buds. Bud Weight and Mbisture Content. Prior to the 29 June 1978 freezing experiment, 3 replicates of 5 buds each were randomly selected from 'Baco-l' and 'Vidal-256' cuttings at the full-swell stage of bud develOpment, and from the 'Concord' cuttings at first-swell, full-swell, and bud-burst stages of bud development. Primary buds 'were excised and each group of five placed quickly into separate air—tight glass vials (25x50mm) with ground glass tops. Vials were weighed on a Mettler H31 single pan balance, opened, and tissues dried at 70°C for 72 hours. Vials were closed and weighed again, emptied and reweighed. Fresh weight, dry weight, and water loss were calculated by difference. Moisture content was expressed as grams of water per gram of tissue dry'weight. RESULTS AND DICUSSION Controlled Freezing Technigue Preliminary experiments showed that the use of a ”Part- low", cam programmable temperature control device was unsat- isfactory for planned experiments. True freezing chamber temperature varied by greater than 2°C alternately above and below the set temperature as a result of on-off cycling of the entire unit. The use of an expansion pressure regu— lator valve (EPRV) remedied this problem, resulting in smoothly declining chamber temperatures. Because coil tem- perature was very near box temperature with EPRV control, and because of substantial air circulation via 2 large blow- ers,'within-chamber temperature stratification was very much reduced. Temperature seldom varied more than 0.25°C between thermocouples distributed in and around the plant material. Because the cuttings were not bundled together, air circulation kept tissue temperatures very near air tem- peratures. When grapevine buds were moistened in the freez- ing chamber before freezing commenced, tissue temperatures ‘were oftern 0.50-1.000 below air temperature as a result of evaporative cooling. When the applied surface moisture froze, tissue temperatures rose briefly above air tempera— tureture as a result of the release of the latent heat of fusion. Critical temperatures should be based on air 26 27 temperature because the grower generally has no accurate means of measuring tissue temperatures (Rogers and Swift, 1970). For this reason, and because of the modest vari- ation between tissue and air temperature under these experimental conditions, test temperatures were based on air temperature. Regrowth of the cold stressed grapevine buds on a greenhouse mist bench proved satisfactory for evaluating freezing injury. Because actively growing, cold tender, tissues tend to supercool, (see Literature Review, p.8), and because freezing under these conditions is very injurious (Burke g§_gl,, 1976), cold damaged buds were easily separated from uninjured buds. ‘Well advanced buds suffering freeze injury appeared wilted and water soaked within several hours. Nearly all injured buds, at all stages of development, became desiccated and brown within 3 to 6 days, while buds that escaped.injury remained green and continued deve10pment. On rare occasions a bud showed no visible injury, but did not continue development. 'When sectioned, these buds showed browning of the vascular cylinder while peripheral tissues remained green. Because ice grows most rapidly through vascular tissue (Burke et al., 1976), these buds had probably begun to freeze just prior to removal from the freezer. These budS'were considered functionally dead. ,Iissue surface moi§§E£§_during freezing had a drama- tic effect on survival (Table 1). At full—swell, the LT50 28 of grapevine buds that were misted regularly during freezing (treatment C) was 3.700 higher than similar buds that were not misted (treatment A). A similar relationship was found for buds at the first-swell and bud-burst stages as well. The LTSOB of buds of a given stage and.moistening treatment were very similar in the 27 April 1977 and 1A May 1978 experiments, while buds on 13 May 1977 received an intermediate moistening treat- ment (treatment B) and were intermediate in hardiness. Mbisture on the surface of buds may reduce apparent hardiness several ways. First, evaporative cooling from a wet surface may lower tissue temperatures. Hewett g§_gl. (1978) obtained results very similar to those in Table 1 for developing peach and apricot buds. Because dormant peach buds were not rendered.more cold sensitive by a wet surface unless they remained wet for extended periods prior to freezing, Hewett g§_gl, (1978) concluded that evaporative cooling within the freezing chamber was not responsible for the increased injury. The previously discussed deviations of bud tissue temperatures from champ ber air temperatures after misting can not explain the freezing injury differences found here. Secondly, a ‘wet surface could increase bud tissue moisture content, resulting in lowered freeze resistance by lowering cell sap concentration (Modlibowska, 1962) or by eliminating barriers to ice nucleation (Quamme, 1978). Hewett gt_g;, (1978) were, in fact, able to demonstrate increased bud 29 tissue moisture levels after extended periods of wetting. However, buds in the 1h May 1978 experiment (Table 1) were all forced.under mist and differed only in the within- freezer treatment. However, during the period in the freezer, the relative water content of the unmisted buds may have decreased and contributed to their greater freeze resistance. Finally, in this study, the most important effect of a wet surface on increased injury was probably the reduction of supercooling via ice crystal inoculation. Kitaura (1967) found that a supercooled.mmlberry leaf ‘would freeze when touched by an ice crystal. Results of the 10 June 1978 freezing experiment (Table 1) do not support the view (Mbdlibowska, 1962) that supercooling is greater in small excised pieces of tissue than in whole plants. However, because the chance of ice nucleation increases with the time at a given subfreezing temperature,and because test temperatures were not maintained for extended periods in this study;super- cooling should be minimized if injury at a given tempera- ture is to reflect injury likely to occur in the field at that temperature. Therefore the "standard” freezing technique should include regular misting of the buds during freezing to promote ice nucleation. The use of artificially forced grapevine buds, rather than naturally develOping buds from the field, for con- trolled freezing experiments, would greatly facilitate development of non-microclimate modifying frost control 30 techniques. For this reason, two experiments were con- ducted in which buds forced from.dormant cuttings on a greenhouse mist bench were compared, in the same freezing experiment, to buds developed naturally in the field ‘ during the spring. No significant differences in hardi- ness were found at 3 different stages of develOpment (Table 2: 1A May 1978 and 20 May 1978 experiments). Hardiness in these 2 experiments was similar to that of buds of the same stage on whole potted vines in separate experiments (Tabel 2; 10 June 1978 and 13 June 1978 exper— iments). These data indicate that the hardiness of grape- vine buds forced from cuttings in a greenhouse is similar to that of buds developing in the field. Factors Determining Critical Temperatures in DevelOping Grapevine Buds Stage of phenological development is generally con— sidered the most important factor determining the tempera- ture that a developing fruit bud will tolerate (Ballard pp_pl., 1972). Hardiness of developing grapevine buds decreases with advancing stage of development (Table 3) whether the buds deve10p naturally in the field or are artificially forced on a greenhouse mist bench. Hardi- ness differences between stages of develOpment may be related to quantitative factors. A.simple volume effect may be important in that the large size of more advanced buds would.make them, by chance, more likely to contain 31 more and better ice nucleators. Kaku and Salt (1968) found that the freezing temperature of conifer needles increased with increasing needle length and attributed this to increased numbers of ice nucleators. Because of both size and shape, the area in contact with surface ice'will tend to be larger in more advanced buds. Mbdlibowska (1962) found that apple blossoms supercooled more when closed than when open. Differences in freeze resistance between developmental stages may also be re- lated to qualitative differences in tissue hardiness. Kaku (1971) found that hardiness in immature,§p§p§ leaves ‘was related to age, independent of leaf size. Changes in maturity, or stage of deveIOpment, may involve structural changes that affect the efficiency of nucleating sites. It is critical to any frost control technique designed to retard development, that the relationship between stage of development and freeze resistance be independent of the rate of development. One such tech- nique (Howell and Wolpert, 1978) makes use of apical dominance to retard develOpment of buds at nodes that will be kept for fruiting. If however, the slowly developing basal buds are no hardier than the more rapidly devel- oping apical buds, the technique loses its effectiveness. Apical buds suppress the development of buds basal to them even in excised cane sections (Table A). There- fore, by selecting buds from 2 groups of 6-node cuttings placed on the greenhouse mist bench several days apart, 32 it was possible, in one experiment, to cold stress full- swell buds that had developed slowly (i.e., basal buds from the first group) vs. rapidly(i.e., apical buds from the second group). The buds whose development had been retarded by buds apical to them, and therefore exposed longer to dehardening conditions, were no less hardy than rapidly developing buds at the same stage (Table 5: 3 August 1977 experiment). Supporting this finding is the fact that the hardiness of buds developing naturally in the field in the spring of 1978 did not differ appreciably at a given stage over 3 sampling dates (17, 20, and 23 May 1978: Table 3). Pro-freeze conditions, particularly the temperature during the 2A hours prior to freezing and during the entire period of spring bud development, reportedly influence critical temperatures (see Literature Review, p.7). The 3 August 1977 and 5 may 1978 freezing experiments (Table 5) both indicate that exposure of full-swell buds to tempera- tures of 1°C for 3 days proir to freezing did not result in rehardening. The hardiness of full-swell buds allowed to develop entirely at cool temperatures did not differ significantly-from that of full-swell buds developed at warm temperatures (Table 5, 12 October 1977 experiment). Data for field vs. greenhouse grown buds support this conclusion in that mean max./min. tempera- tures from.May 1-20 were 170/806 in the field and BOO/17°C in the greenhouse (Table 2: 1A and 20 May 1978 experiments). 33 These observations agree with Hewett gp_gl. (1978) and indicate that for developing grapevine buds, pre-freeze temperatures do not significantly affect resistance to freezing stress at a given stage of development. This contrasts with a report on apple blossoms (Ballard and Proebsting, 1972) indicating that blossoms developed in cool weather were hardier than those develOped in warm weather. Misting vs. no misting during development did not affect hardiness of full-swell buds forced simultane- ously in the greenhouse (Table 5; 12 October 1977 experi- ment). This conflicts with the report of Hewett gp_gl, (1978) who found reduced hardiness and increased moisture content in developing peach buds exposed to prolonged misting. The misting both groups of grapevine buds received during controlled freezing may, however, have obscured any differences due to prior treatment. Characteristics of grapevine buds and canes prior ‘pp_§pring development were investigated as to their effect on the hardiness of deveIOping buds. Light periderm color and large cane diameter have been associated with reduced ‘winter hardiness of dormant grape buds (Howell and Shaulis, 1980). Preliminary observations by Nelson Shaulis (per- sonal communication) have suggested that poor leaf ex- posure to light during the previous summer’may result in increased spring frost injury to 'Concord' buds. Pre- deve10pment characteristics could affect spring frost 34 injury in two general ways. First, they may alter the rate of phenological development (Anderson.gp_gl,, 1980) and second, they may affect the actual hardiness of buds at a given stage of develOpment. No differences were found in the hardiness of either first-swell or full-swell buds based on pro-development sunlight exposure status (Table 6). In addition, no differences were found in the days to first-swell of one- node cuttings from either group under greenhouse forcing conditions (Table 7). These data indicate that exposure status during the previous summer does not affect freezing damage to deve10ping buds the following spring. Partridge (1925) warned against the retention of laterals for fruiting canes because of their generally smaller cane diameter, suggesting that buds on small canes develop more rapidly in the spring than buds on large canes, resulting in more damage from spring frost. With cane diameter restricted to medium size, (5.0-7.5mm), no differences were found in the rate of development (Table 8) or in first-swell and full-swell bud hardiness (Table 6) between buds arising from laterals and those arising from main canes, when forced from cuttings. In a separate experiment, buds were stratified as to the diameter of the cane section from which they arose and to their original nodal position to prevent confounding these characteristics, as small cane sections tend to occur at more apical portions of the original cane. ,This 35 allowed comparisons of developing buds arising from large vs. small canes, stratified as to node position: and of buds from apical vs. basal positions, stratified as to cane diameter. Although insufficient numbers of buds on large diameter, apical cane sections prevented their evaluation, no differences were found in the hardiness of full-swell buds (Table 6) or in deveIOpmental rates of the buds from the other 3 categories (Table 9). These data indicate that, for buds forced from one-node cuttings, main cane vs. lateral origin, cane diameter, and orig- inal nodal position do not affect the hardiness of devel- 0ping buds or the rate of develOpment. Cutting canes into one-node cuttings may, however, prevent the influence of cane diameter on bud development rate in a manner analagous to the loss of apical dominance. It is inter- esting to note that Antcliff and.May (1961) observed that development of buds on one-node cuttings did show the pattern of original apical dominance if the cuttings were taken one month or loss before normal spring bud-burst. The importance of the data in the 3 experiments pre- sented in Table 6 and Tables 7-9 lies in sampling consid- erations for future evaluations of frost protection measures. If 'Concord' buds are forced from cuttings for controlled freezing, the cuttings need not be stratified as to exposure status (periderm color), lateral vs. main cane origin, cane diameter, or original nodal position. 36 Cultiyar differences could affect spring freeze resistance through variations in developmental rate, and through actual differences in tissue hardiness at a given phenologic stage of develOpment. Buds of different culti- vars are known to develop at different rates under field conditions (Anderson gp_pl,, 1980). Because of this, an experiment was conducted to determine relative development rates for cuttings forced under greenhouse conditions. The reported order of field deve10pment was 'Baco-l', 'Concord', 'Vidal—256' (Howell, 1976: Anderson M” 1980). When buds were greenhouse forced, the order of development was 'Baco-l', 'Vidal-256', 'Concord' (Table 10). This apparent reversal of the normal relative development rates of 'Con- cord' and 'Vidal-256' buds may have resulted from differ- ences in either threshold temperature to induce bud devel- Opment, or in whole vine vs. forced cutting response. Further investigation of this may lead to an understanding of, and possibly to the manipulation of, variability in bud develoPment rates in the spring. Knowledge of relative greenhouse forcing rates per— mitted programming so that buds of these 3 cultivars could be subjected to controlled freezing while at the same stages of deve10pment. For first—swell and full-swell buds, freeze resistance is not the same in these 3 cultivars (Table 11). Fifty-six per cent of the 'Baco-l' buds, A0$ of the 'Vidal- 256' buds and only 22% of the 'Concord' buds were killed by the same test temperatures. A possible explanation 37 of these findings lies in the data presented in Table 12. While moisture content does not differ significantly, the larger size (fresh or dry weight) of the less cold resistant buds may simply provide more opportunities for fatal ice nucleation. Alternatively, hardiness differences may actually reflect innate differences in tissue hardiness. One possible explanation for the superior resistance of 'Concord' buds is the tomentose nature of young leaf tissue surfaces, while those of 'Vidal-256' and 'Baco-l' are more glabrous. The felt-like surface of developing 'Concord' buds may reduce the probability of surface ice contacting the supercooled tissue water, causing flash crystallization of cellular water. These hardiness differences can be important, particularly in determining critical temperatures for existing cultivars and in breeding new grape cultivars for spring freeze resistance. However, differences in natural development rates are apparently more important, as Howell (1976) attributed the 60-65% primary bud kill in 'Concord' vs. only trace kill in 'Vidal-256' at the same site after a natural spring freeze, to differences in deve10pmental rates. Estimation of Critical Temperatures One of the primary aims of this investigation was to develop a reliable technique for the estimation of practical critical temperatures for develOping grapevine buds at various stages of development. Results indicate 38 that these temperatures may depend on conditions other than developmental stage. While cultivar differences appear to be of some importance, the presence or absence of surface moisture or ice is crucial. Because the presence of surface moisture or hoar frost during a spring freeze may vary from one growing region to another or from one natural freeze to another, critical tempera- tures may have to be adjusted accordingly. Estimated critical temperatures of developing 'Concord' grape buds for wet and dry tissue surface conditions are presented in Table 13. Temperatures for wet conditions were obtained by averaging, for each stage of development, LTSO values for 5 freezing experiments (Table 3). In all of these experiments, the buds were regularly misted throughout controlled freezing. Critical temperatures for dry conditions were obtained by averaging LT50 values for buds not misted during freezing, (Treatment A, Table 1). Critical temperatures (Table 13) for buds‘with surface moisture or frost may well be relevant to Michigan or Eastern viticultural conditions while the lower values listed for dry buds, or the likewise lower values given by Proebsting and.Mills (1978), may be more relevant to the more arid regions such as Washington's Yakima Valley. Extensive comparisons with field injury from natural spring frosts over several years is the only method of assessing their accuracy and the natural conditions under which they apply. 39 Summapy and Conclusions The results of this study indicate that the con- trolled freezing technique employed is capable of de— tecting small hardiness differences between the stages of develOping grapevine buds or between deve10ping buds of different cultivars. The surface moisture levels must be controlled during freezing, as this markedly affects hardiness. Buds on forced cuttings, on cuttings taken directly from the field, and on whole vines behave similarly‘when exposed to controlled freezing. Therefore, the use of stored cuttings forced in a greenhouse should provide a continuous source of developing grapevine buds for the evaluation on nonemicroclimate modifying techniques of frost protection, such as use of surface and systemic cryoprotectants and breeding of frost resistant cultivars. These cuttings do not have to be stratified for previous exposure status, lateral vs. main cane origin, cane diameter, or original nodal position. All of these morphological conditions will have an effect on whether the primary bud survives that dormant season (Howell and Shaulis, 1980), but once growth begins, a living primary bud's cold resist- ance is not further influenced. There appears to be cultivar difference in cold resist— ance at a common stage of phenological development. This should be further studied to assess the range of resistance extant and to assess the potential value for genetic hO improvement of bud resistance to low temperature stress during spring freezes. The stage of phenological deveIOpment appears to be the most important endogenous factor affecting bud hardiness. This relationship is independent of the rate of development, temperature during develOpment, and tem- perature immediately prior to controlled freezing. It ‘would appear that developing grapevine buds can not be rehardened by exposure to temperatures just above freezing. Knowledge of critical temperatures can be helpful to growers in efforts to use current frost control measures more economically. These critical temperatures may have to be based on prevailing weather conditions such as dew point and precipitation probabilities, as well as the overall bud development status. L1 .mo.ona .opoo oopam o no mach odnpaz.oo:oofimanw«o opooaood oooaop :oaon otoppoa coo oqasaoo canvas ooaooHMHnwao opooaood oonao> woazoaaom oaomwoq .oohopohoaaop poop Hooapdho no oooapthOHQ oooolobaa mo oaohaodo Nywhn soapohoaoo 9A .mnduoohm ooHHthnoo on oopoonnno oho3.oo:H> ooppoa tacihoohim oHonsx .oosonooohw on» ea poaa goons owowppoo do ooonom oosmh .pooaahogxo woauoohm mo boo on» oHoam scam noxop owofipaaou n o . mu eoosnuoooaoaun n o o or? a on- a 3.. o oo. senseless n o n o m.mu «.5: n e.m| p H.s| o o.e| on H.>I Haozmiaass n o a .3. a E- a do. desolate so 4.0I xoohoIoHoom o e o a m < wooed. Sons soonossoono Born agar“ ms ease S we so: i: R so: 3” E dame R moon mo oohsoo coo oopoo moduoohm .AabmH.HHozom coo homnonoopvamv ooavoswo honhomlooahooam on» an oopoHSoHoo oo Aoovo 9A oho mosaop .moauoohm poonws0hap oHophopna poadmoh po oooopodoa on coo » uoovwdauoohm mo moaqoawon on» go vooopmfioa Amv «haqo modsooam op Moan oooopoaoa Aaw o so mean manna: oooooamacwfio opooaoqd oooao> radon opopvoa coo osasaoo nanpaz_oooooamaowao opoofiooa oosHo> woazoaaom onoppoq .ogpohomaop poop Hooaflfio uo odoaphonHOMQ oooolosa mo oaohaodo «Rho. ooapoaoaoo Omaqh .oooooha moauoohm on» p:onw:otnp oHoSNopna hoasmou no pods osam o sub nonopoaoa moon npaz.oooaonoop wmauoohm onoooopmu in: o T? o 0.? annfinesm m w 2.. n RT o H? a in: Serenades m w d a a on- a are. as air. all nastiness o o m < m < ensem— pooapooha mm onselmla mm some 3 on as: ou 2. Us: .2 psoeahoqwo‘MdHaoohm no ouoa .Aabma.aaosox coo noodondoppamv soapoovo nonuoulothooam on» hp oopoflaoaoo oo Oman ono oooaob .onooolmolpso macaw oooab vaoinoohia_ooppoa oHogz on coo ”scoop pods oooonooohm o no ooohom owodppso Amv .oaoam scam hapoohfic noxop omnappao A cocononoa noon nnnr.ooononoon monsoonm chooconmo o o.~l noonolvoonogwm n m.ml o m.ml o o.ml o o.ml nonsnloom o m.ml n m.ml o m.ml n 4.ml o H.ml Haozolaash o m4: a a4: a 4.4.. n 0.4.. Haofiannnnm o ©.ml so m.ml MoohoIoHoom owes: do 3.. was on E. as: S mines in E. Hash. ow. oosmm woaodh hoooonooohw onoo.mo«oooam coo moon no oohoom .AdhoH.HHo3om coo noonopoonnamv noanosvo ponaoMIooauoomm onn hp oonoanoaoo oo AooVOmBA oho oosaob o.mnaooonm ooHHonnsoo on ooooaxo omnnnnSo oofinomopm .ohooooo. mo oooowonon ooh hhoeanm do nooemoao>oo Hooawoaooosm mo owono mo noommm .m oHnoa Table A. Effect of node position on deve10pment of primary buds on 6—node 'Concord' grapevine cuttings.. greenhouse mist of live buds at a given stage. Twent nch. cuttings were forced on a Values are per cent Nodegposition Bud Basal Apical stage 1 2 3 A 5 6 7-27-77 Dormant 65 #0 44 LL 26 43 Scale-crack 35 6O 50 50 63 50 First-swell O O 6 6 11 7 Full-swell 0 O O O O 0 7-29-77 Dormant A5 35 25 31 11 29 Scale-crack #5 65 38 19 32 21 First-swell 10 0 31 50 58 50 Full-swell 0 O 6 O 0 0 7-31-77 Dormant 50 35 13 25 11 1h Scale-crack 3O 30 25 19 0 1A First-swell 10 35 25 13 26 7 Full-swell 10 0 38 AA 63 6A A5 Table 5. Effect of pre-freeze treatments on hardiness of full-swell 'Concord' grapevine primary buds subjected to controlled freezing.z Values are LT 0 as calculated by the Spearman—Karber eqaation (Bittenbender and Howell, 197A). Pre-freeze treatment I"T50 .3 August 7? Experiment Days in Days held greenhouse at 1°C N.S. 12 November 77 Experiment 13 days in greenhouse (27°/19°C) on mist bench -h.0 no mist -3.7 17 days in growth chamber (130/80) no mist -h.3 N O S C 23 Mayg78 Experiment Cuttin 8 taken from field directly frozen -3.8 held 3 days at 1°C -A.O N.S. zStandard freezing technique with buds moistened via a fine mist at regular intervals throughout the freezing process. yN.S.- Not significant. Significance determined by 22 analysis of live-dead proportions at critical test temper- atures. Comparisons within one freezing date only, p=0.05. L6 Table 6. Effects of cane characteristics and bud position on hardiness of develOping 'Concord' grapevine primary buds. Buds were forced as 1-noda cuttings on a greenhouse mist bench for all controlled freezingz experiments. calculated by the (Bittenbender and Howell, 197A). Values are LT5o(°C) as Spearman-Karber equation 11 April 78 Experiment First-swell Full-swell Cane exposure status: Well exposedy -A.3 -3.7 Poorly exposedx -h.3 -3.8 N.S.W N.S. 13 April 78 Experiment Buds arising from: Main canes(5.0-7.5mm diam)v -A.9 -3.3 Laterals(5.0-7.5mm diam) -A.8 -3.A N.S. N.S. 17 April 78 Egperiment Cane diam(mm) Node position Large(6.7-10.0) Basal(2-6) -3.2 Small(h.0—6.5) Basal(2-6) -3.0 Small(A.0-6.5) Apical(11-15) -3.o N.S. zStandard freezing technique with buds moistened via a fine mist at regular intervals throughout the freezing process. yOnlg cuttings collected from exterior of vine canOpy and wit dark periderm color were considered."well exposed." Onlg cuttings collected from interior of vine canopy and wit light periderm color were considered ”poorly exposed.” wN.S.— Not significant. Significance determined bylxz analysis of live—dead proportions at critical test te era— tures. Comparisons within one freezing date only, p=0. 5. vCane diameter measured at midepoint of internode below bud. #7 Table 7. Effect of exposure status on rate of primary bud deveIOpment in 1-bud 'Concord' grapevine cuttings. One hundred twentyheight cuttings of each type were forced on a greenhouse mist bench. Values are per cent of live buds at or past first-swell. Days on mist bench Exposure No. of status live buds 12 1h 16 18 20 Well exposed? 121 0 17 67 9O 98 Poorly exposedy 109 0 17 72 95 99 zOnly cuttings collected from the exterior of the vine canopy and with dark periderm color were considered "well exposed.” yOnly cuttings collected from the interior of the vine canopy and with light periderm color were considered ”poorly exposed." h8 Table 8. Effect of cutting origin on rate of primary bud development in 1-bud 'Concord' grapevine cuttings. One hundred twentyheight cuttings of each type were forced on a greenhouse mist bench. Values are per cent of live buds at or past first-swell. Days on mist bench No. of Origin of cuttings live buds 8 10 12 1h 16 Persistent laterals 122 0 5 72 93 97 Main canes 122 O 6 66 93 97 A9 Table 9. Effect of cane diameter and original nodal position on rate of primary bud development in 1-bud 'Concord' cuttings. twentybeight cuttings of each type were forced on a greenhouse mist bench. One hundred Values are per cent of live buds at or past first-swell. Cane diameter Nodal position Large(6.7-10.0mm) From nodes 2-6 Small(A.0-6.5mm) From nodes 2-6 Small(A.O-6.5mm) From nodes 11-15 No. of Days on mist bench live buds 8 10 82 o 56 125 o 65 122 o 42 12 1A 16 9A 98 100 91 99 100 8A 98 99 50 osdnoaohm tabla on onoh noosaoaonoo to: hhosaha so honanaso mo noommm mm mm mm mm mm o o o no .ouoosoo. 00H 00a 00H am mm o o o no .ommtaoonb. OCH OCH OCH OCH OCH no om o no .Hlooom. Haozo noasn HHoso nonsn Haoso nonsn Adorn nohsn noon o>wm, hobanfloo nouns com nouns 35 no.4; pom nohnh can mo 62 0H NH w 4 meson none :0 whoa .oonooaofi ooonm one noon no no noon o>nH mo nsoo non oao mosaob .noson nods oosonsoohm o :0 coupon onos.ho>«naso sooo mo owsannno oznlhnsobom .omswnnso .oa oHnoe 51 Table 11. Effect of cultivar on hardinessz and per cent mortalityy of deveIOping grapevine primary buds subjected to controlled freezing.x 'Baco-l' 'Vidal-256' 'Concord' LTSO (first-swell) -2.9 a" -3.6 a -A.1 a a b b LT50 (full-swell) -2.3 b -2.7 b -3.A b a a b % Mortality 56.3 39.6 21.9 a b c zHardiness given as LT 0(°C) calculated by the Spearman— Karber equation (Bittenbender and Howell, 197A). yPer cent mortality is per cent of first-swell and full- swell buds killed at -2.00 and -3.5°C. xStandard freezing technique with buds moistened via a fine mist at regular intervals throughout freezing process. WSeparation by 12 analysis of live-dead proportions at critical test temperatures. The same letter within a column or row indicates that reapective values are not significantly different, p=0.05. 52 .mo.ona .nsohommao hansooamasmao no: oho nonnoa osoo onn hp oosoaaom osooz .noon owsom oamanasz o.soo:sn hp osESHoo on soanohomoo soozh .nnmao:.hho osooan sohM\honoz.osohoo o mH.o o Has.o a mHH.o o omm.o Haosnnaasa .wleHdfidb. o no.o o 4mo.a o omH.o o 4s~.a Haosouflass .Hloowm. n om.m o mso.a o oo~.o o ms~.H nonahiosm o mm.m o mam.o a Hmo.o o mao.o Haosmnaaae a ma.4 a n4a.o a 4mo.o as ssa.o anomalooaan .vhoosoo. apaooaoo Aaosnlmwwv Anson m\wo Anson axov omaooneam goono: omm Heron phones b5 names: moons .oosonsoohm o on omsannso scam poohom moon osa>omonw modaoaoboo mo noonsoo onsnoaoa woo nnwaoz.o5ooan so hobanflno mo noommm .NH oHnos 53 Table 13. Estimated critical temperatures for developing 'Concord' grapevine primary buds. Values are LT 50 ( 0C) as calculated by the Spearman-Karber equation (Bittenbender and Howell, 197A). Surface moisture status Stage of develOpment Wet2 Dry Scale-crack -5.7 -9.A First-swell -A.A -7.9 Full-swell -3.5 -7.1 Bud-burst -3.1 -6.2 Expanded-shoot -2.6 2Indicates the presence of frost, dew, ice, or water from precipitation or irrigation. LITERATURE CITED LITERATURE CITED Anderson, H.A., G.S. Howell, and.J.A. Wolpert. 1980. Phenological development of different Vitis cultivars. Fruit Varieties Journal 3A:5-7. Antcliff, A.J. and P. May. 1961. Dormancy and bud burst in Sultana vines. Vitis 2:1-1A. Baggiolini, M; 1952. Les stades reperes dans le develOppement annual de la vigne et leur utilization pratique. Rev. Rom. Agric. Vitic. Arboric. 8:A—6. Ballard, J.K. and E.L. Proebsting. 1972. Frost and frost control in Washington orchards. 'Wash. State Univ. Ext. Bul. 634. , and R.B. Tukey. 1971. Critical temperatures for blossom buds. ‘Wash. State Univ. Ext. Cir. 369-37A. Bartholic, J.F., MQD. Heilman, and.B.M. Farris. 1970. Large volume generator of stable foam.for freeze protection. HortScience 52A86-A88. Bittenbender, H.C. and G.S. Howell. 197A. Adaptation of the SpearmaneKarber method for estimating T50 of cold stressed flower buds. J. Amer. Soc. Hort. Sci. 99:187-190. Blake, MLA. and C.H. Steelman. 19AA. Preliminary in- vestigation of the cold resistance of peach fruit buds at the pink bud stages of develOp- ment. Proc. Amer. Soc. Hort. Sci. A5:37-A1. Burke, M.J., L.V. Gusts, H.A. Quamme, C.J. Weiser, and P.H. Li. 1976. Freezing and injurg in plants. Annu. Rev. Plant Physiol. 27:507-52 . Burns, R.M; 197A. Effects of spraying chemicals on young citrus trees for frost protection. Calif. Agric. 28:13-1A. Chandler, W.H. 1913. The killing of plant tissue by low temperature. Missouri Agric. Expt. Sta. Res. Bul. 8. (cited in Proebsting and Mills, 1978) . 5A 55 Clore, WiJ., MgA.'Wallace, V.P. Brummand, and R.D. Fay. 197A. Cold hardiness studies. {1p Proc. Wash. State Grape Soc. p20-28. Conover, W.J. 1971. Practical nonparametric statistics. John Wiley & Sons Inc., New York. A62p. Cramer, H. 19A6. Imathematical methods of statistics. Princeton Univ. Press, Princeton, N.J. 575p. Daniell, J.W. and F.L. Crosby. 1971. The relation of physiological stage, preconditioning, and rate of fall of temperature to cold injury and decline of peach trees. J. Amer. Soc. Hort. Sci. 96:50-53. Dennis, F.G. Jr.,'W.S. Carpenter, and.W.J. MacLean. 1975. Cold hardiness of 'Mbntmorency' sour cherry flower buds during spring development. HortScigncg 10:529—531. and G.S. Howell. 197A. Cold Hardiness of tart cherrg bark and flower buds. Mich. State Univ. Res. eport 220. Dethier, B.E. and N. Shaulis. 196A. Minimizing the hazard of cold in New York vineyards. Cornell Ext. Bul. no. 1127. Ellison, E.S. and W.L. Close. 1927. Missouri Weather Rev. 55:11-18. (cited in Gardner et aI., I952). Evert, D.R. 1967. The physiology of cold hardiness in trees. Ip_A3rd Proc. Int. Shade Tree Conf. pAO-SO. and G.S. Howell. 1979. The modified Friedman test- a simple alternative to the F-test for the randome ized complete-block design. HortScience 1A:19-20. Finkle, B.J., E.S.B. Pereira, and.Mgs. Brown. 197A. Freezing of nonwoody plant tissues. I. Effect of rate of cooling on damage to frozen beet root sections. Plant Physiol. 53:705—708. Fuchigami, L.H., M; Hotze, and C.J. Weiser. 1977. The relationship of vegetative maturity to rest, development, and s rin budpbreak. J. Amer. §QCQ Hort. 301. 10g645 -l|'520 Gardner, V.R., F.G. Bradford, and.H.D. Hooker. 1952. The fundamentals of fruit production. 3rd. ed. MCGraw- Hill Book Company Inc., New York. 739p. 56 George, M.F., MQJ. Burke, and C.J. Weiser. 197A. Super— cooling in overwintering peach flower buds. J. Amer. Soc. Hort. Sci. 103:57-61. Harlan, J.R. 1976. The lants and animals that nourish mane §£io mere 2 5:88.97. Hewett, E.W., K. Young, E.L. Proebsting, and H.H. Mills. 1978. Madification of critical freezing tempera- tures in fruit buds by elevated tissue water content. HortScience 13:247-2A9. Howell, G.S. 1976. Frost damage in Michigan - What our 1976 experience tells us. Great Lakes Fruit Grow- ers News. Nov. 1976. and.N. Shaulis. 1980. Factors influencing within-vine variation in the cold resistance of cane and prima bud tissues. Amer. J. Enol. Viticult. 31:15 -161. and C.J. Weiser. 1970. Fluctuations in the cold resistance of apple twigs during spring dehardening. J. Amer. Soc. Hort. Sci. 95:190-192. and.J.A.'Wolpert. 1978. Nodes per cane, primary bud phenology, and spring freeze damage to 'Concord' grapevines. A preliminary note. Amer. J. Enolp, Viticult. 29:229-232. Hudson, MIA. and P. Brustkern. 1965. Resistance of young and mature leaves of Mnium undulatum (L.) to frost. Planta 66:135-155. and D.B. Idle. 1962. The formation of ice on plants. Planta 57:718-730. Kaku, S. 1971. Changes in supercooling and freezing processes accompanying leaf maturation in Buxus. Plant & Cell Physiol. 12:1A7-l55. and RfW. Salt. 1968. Relation between freezing temperature and length of conifer needles. Can. J. Bot. A6:1211-1213. Ketchie, D.O. and C. Murren. 1976. Use of cryoprotec- tants on apple and pear trees. J. Amer. Soc. Hort. Sci. 101:57-59. Kitaura, K. 1967. Supercooling and ice formation in mulberry trees. Int. Conf. Low Temp. Sci. Sapporo, Japan. 2:1A3-156. 57 Lapins, K. 1961. Artificial freezing of 1-year—old shoots of apple varieties. Can. J. Plant Sci. Layne, R.E.C. 1966. Relation of bloom date and blossom temperature to frost in ury and fruit set in apricots. Fruit Variet es and Horticultural Digest vol.21, no. 2. Levitt, J. 1951. Frost, drought, and heat resistance. Apnu. Rev. Plant thsiol. 2:2A5-268. Lovett, H.W. and J.K. Collins. 1978. .Michigan Agri- cultural Statistics. .Michigan Department of Agriculture, Marketing Division. Lumis, G.P., H.A. Mecklenburg, and K.C. Sink. 1972. Factors influencing winter hardiness of flower buds and stems of evergreen azaleas. J. Amer. Soc. Hort. Sci. 97:12A-127. Mayland, H.F. and.J.W. Cary. 1970. Frost and chilling injury to growing plants. Adv. Agron. 22:203-23A. Maddie R. 1975. Statistical handbook for non-statis- ticians. McCraw-Hill Co. Inc., London. 162p. Mitchell, J.W. 19AA. Winter hardiness in guayule. Bot. Mbdlibowska, I. 1961. Effect of soil moisture on frost resistance of apple blossoms, including some observations on "ghost" and "parachute" blossoms. Jo Home $010 36:186-1%o Olien, C.H. 196A. Freezing processes in the crown of "Hudson" barley, Hordeum vul are (L. amend. Lam.) Hudson. Crop Sci. I:9I-95. Partridge, W.L. 1925. The fruiting habits and pruning of the 'Concord' grape. ZMich. State Univ. Tech. Bul. 69. 39p. Pierquet, P.C., C. Stushnoff, and.MLJ. Burke. 1977. Low temperature exotherms in stem and bud.tissues of Vitis riparia Michx. J. Amer. Soc. Hort. Sci. 102:55-55 Proebsting, E.L. Jr. 1975. Reducing energ consumption in cold protection. HortScience 10: 63-A65. , V.P. Brummund and.W.J. Clore. 1978. Critical emperatures for 'Concord' grapes. Wash. State Univ. Ext. Mimeo. A330. 58 and H.H Mills. 1961. Loss of hardiness by peach fruit buds as related to their morphologi- Ecal development during the pre-bloom and bloom period. Proc. Amer._§oc. Hort. Sci. 78:10A-110. and . 1971. The use of controlled Treezingtechniques for estimating critical? temperatures for fruit buds. ‘Wash. State Univ. Agric. Expt. Sta. Cir. 5A8. and . 1978. Low temperature resistance of developifig flower buds of six deciduous fruit species. J. Amer. Soc. Hort. Sci. 103:192-198. Quamme, H.A. 1978. Mechanism of supercooling in over- 'wintering each flower buds. J. Amer. Soc. Hort. Sei. 103:5 -61. Reingold, N. 1960. Cold Weather Agriculture: Senate Doc. 91. 0.3. Gov't. Printing Office, washington, D.C. Rogers, W.J. and H.L. Swift. 1970. Frost and the -prevention of frost damage. U.S. Dept. of Com- merce, NOAA, Silver Spring, Md. Scarth, G;W. 19AA. Cell physiological studies of frost resistance: A review. New Phypol. A3:1-12. Scott, K.R. and L.P.S. Spangelo. Portable low tempera- ture chamber for winter hardiness testing of frgit trees. Proc. Amer. Soc. Hort. Sci. 8A:131- 13 . Siminovitch, D. and D.R. Briggs. 1953. Studies on the chemistry of the living bark of the black locust in relation to its frost hardiness. III. The validity of plasmolysis and desiccation tests for determining frost hardiness of bark tissue. Plant Physiol. 28:15-3A. Single, W.V. and H. Marcellos. 197A. Studies on frost injury to wheat. IV. Freezing of ears after emergence from the leaf sheath. Aust. J. Agric. Res. 25:679-686. Steel, R.G.D. and J.H. Torrie. 1960. Principles and procedures of statistics. MbGraw-Hill Co. Inc., ew York. A81p. Timmis, R. 1977. Critical temperatures for Douglas- fir cone buds. Can. J. For. Res. 7:19-22. 59 Young, F.D. 1920. Frost and the Bravention of damage by it. USDA Farmers Bul. 1 96. . 19A7. Frost and the prevention of frost damage. .8. Dept. of Commerce W. . 1A8A. Young, R. 1969. Cold hardening in 'Redblush' grapefruit as related to sugars and water soluble proteins. APPENDIX 6O «H «H «H NH NH «H «H «H 0.0: «H «H «H «H NH NH 0H HH m.s: «H «H 0H NH 0H NH 4 n 0.o: a HH 4 «H m 0H m o m.4: ANHV m 0 m m 0 0 0 0 0.m: 0 0 0 0 0 0 0 0 Honesoo 0 Aoososaoonov 55:0H:s m m m m m m o.o: m m m m o w m.o: m m o m e m o.m: 0 o 0 H 0 H “.m: Hmv 0 H 0 0 0 0 0.~: 0 0 0 0 0 0 Honoaoo m AoHoHsv as:HH:m o m b b m m o H 0.0: 4 o 4 o m m o H m.b: m m 0 H H H 0 0 0. 0 0 0 0 0 0 0 0 m. : Amv 0 0 0 0 0 0 0 0 0.m: 0 0 0 0 0 0 0 0 Honosoo < AeHoHsv ss:s~:4 lullmjiluij sea... as: as as... as as noono nouns Haoso Haoso xooho nooa .nan oooohm onnono ouoohh .gxm tom Hash nouns oHoom oooohMIohm aeosaoHoroo esp oH oo moron .msHooosm noonmsOAnn hanson :anuonsH osonoHoE on oso "Au 0v msdooohm mo moassamon onn no oosonofioa Amy «haso odooohm on house ooswnoaoe A .omannno ocosla so sooohm our: .ohoosoo. so onoosasoaxo monsoohm sow onoo hnaaonhoa 3mm .H< oHnoa 61 NH NH 0.0: A0 H no ohoo m NH NH 0.0: osHm oosonsoohw 0 NH n.4I :H ohoo HHV 4 HH 0.m: ANHV H H m.HI cocoohonoh 0 0 Hounsoo o stHaoHonoo hHhom NH NH NH NH NH NH m.bl NH NH NH NH NH NH 0.0: HH NH NH NH 5 w 0.4: Aoosomsoohm m NH 0 b N N 0.m| ANHV sH o o0 JHV 0 H 0 0 0 0 m.H: 0 0 0 0 0 0 Hoansoo 0 manoHoboo onoA NH NH NH NH NH NH m.h: NH NH NH NH 0H 0H 0.0: HH NH 5 0 m m m.4: Aoosomsoono H HH 4 m 0 H 0.m: HNHV sH neon HHV 0 0 0 0 0 0 m.H: 0 0 0 0 0 0 Honpsoo 0 maHnoHo>o0 hHose ne:m:m ON 0H oN 0H 0N OH oN oH 0N oH Aomummson aHonon 05mg Aoohsoo 05mg omwww nooso nohsn HHozo HHozo goose n a .nan oooohm osnono h .aww com HHSh nohHh oHoom ouoopmlohm nnosaomoroo esp oH «0 soono 8 . poo oql. H__...<_ ._ _o_ Home 62 aH HH aH mH m.o: Ho m\HH m 0 mH 4H 0.m: possess amsonov 0 0 H m «.m: AmHv 0 0 0 0 0.N: nan noon: 0 0 0 0 Hoopsoo 0 ooonoo noz mH mH mH mH m.o: NH mH HH 4H 0.“: 10o0H\NN oosomaoonov 0 o H H m.m: AmHv 0 0 0 0 0.N: Scoop noHa so 0 0 0 0 Honpaoo 0 season ooz as:NH:HH NH NH 0.N: HH NH m.ml Aoosonsoohm o 0 0.4: sH mane 5H0 o 0 m.N: HNHV 0 0 0.H: wsHQOHoboo 0 0 Hosanoo 0 onsH some NH NH NH NH 0.N: HH NH NH NH m.m: 4 NH N 0H 0.4: Hooaoasoonm 0 o 0 4 m.N: HNHV aH than 4H0 0 0 0 0 0.H: 0 0 0 0 H9580 0 058350 33 this oN oH oN oH oN oH oN oH oN oH Howwwaaon HHooon some Hoonsoo some owwmw noono nohsn HHozo HHoso xoono a .nan oNoonh osnono h .oxm osm HHss noeHm oHoom nooEQOHoboo cop oH mo amonm ooooHMIonm H0.eaooe.H< oHnee 63 .hHoo oosn HHozolHHsa pom ohsnohomson noon nod oven 0m» .thO oops HHozoInoHHm pom onnnohogaon noon hon oosp mHu 0m 0H 0H 0H 0.0: 0m OH «H 0H m.0: sAomv 5N 0m HH 0H 0.0: H HH 0 N m.m: sAoHv Hoosoesoosuv 0 0 0 0 0.N: 0 0 0 0 Honnnoo 0 ooooaxo HHos 0m 0m 0H 0H 0.0: 0N 0H 0H sH «.0: alone 0H 0m 0 NH 0.m: H 0 0 H m.m: aAoHV Hoosomaoonov 0 0 0 0 0.N: 0 0 0 0 Honnsoo 0 oomomxo thooi os:HH:4 mH mH “.0: H000H\NN 4H 0H 0.m: oosonsoouov N m m.m: AmHv 0 0 0.N: Hanan noHs H0.nnoov 0 0 Honnaoo 0 so ooonos se:NH:HH ON OH ON OH ON OH ON OH ON OH Aqu.oson AHonon osmv Aoohsoo osmv “not #0003 pmhfin H.Hozm 4..ng xomho Puma. .389. mumah mfipmvm 0 02h .axo 0am HHss nonHm oHaoo oaoono:onm nsoamOHowoo tsp OH HO omonm Ho.paoo0.H< oHooa 6A .hHsO oosn HHozolHHsm Mom ohsnohogson noon non oosn 0mh .hHso oops HHosoInohHm how ohsnohogaon noon hon oosn wHu 0m 0m 0m 0m 0m 0H 0.0: HN 0N 0H 0N 0H 0H 0.0: HmH:HH ooooa HooHeono N 5N 0 sN H NN m.m:: Homv Hasn.o:0. VHHaao 0 0 0 0 0 0 0.N: 0 0 0 0 0 0 Honnsoo 0 o:N noooewHooonam N H N H N H ass.0:0.4 HHnsn O O O o O o 0 o H H0:N noooawMMomn. lull: o I. 0 9H HHoso:HHso so none oH HHH Haso 0H s 0 H on:NH:4 0N 0H 0H 0H «.0: NHONV HN 0n m 0H 0.0: 0 ON 0 0 m.m: aHoHv Honsomsoonov 0 0 0 0 0.N: 0 0 0 0 Hounsoo o thO oosoo 0m 0H 0H 0H 0.0: AOHV H.H 0H m a 0.“: a 0 0H 0 N m.m: oAmHV Aoosonsoonuv 0 0 0 0 0.N: 0 0 0 0 Honeaoo 0 .38 oHanoeaH 05:4H:4 ON OH ON OH ON OH ON OH ON OH auovwmson aHonon tong ~oohsoo comm] omwww nooso nohsn HHozo HHoro xoono n a .nan oooohh odnono m .nxm osm HHas onnHa oHaom osoono:onm nsoEnOHoboU orb OH HO owonm A0.nnoog.H¢ oHnoa 65 On On On On 0N 0N 0.0: On On on On 0H HN 0.0: 0N 0m «N 0N NH 4H 0.0: N 4H N o 0 0 0.0: H000 0 0 0 0 0 0 0.N: 0 0 0 0 0 0 Honnaoo 0 AoHoHsv 0n:sH:m 4N 4N 0.0: 0H 0H 0.0: N 0 0.0: H4Nv 0 0 0.N: 0 0 Honnsoo 0 HoHoHav 4N 4N 4N 4N 4N 4N 0.0: 4N 4N 0N 4N NH 0H 0.“: 0 ON N 4H 0 m 0.0: A4NV 0 0 0 0 0 0 0.N: 0 0 0 0 0 0 Honoaoo 0 Housemaoonov 4N 4N HN HN 0H NH 0.0: HH HH 4 4 m 4 0.0: 0 NH H H 0 0 0.0: 0 H 0 H 0 0 0.N: 3N0 0 0 0 0 0 0 0.N: 0 0 0 0 0 0 H9580 H Hoonoasoonov 0s:4H:m oN oH oN oH oN oH oN oH oN oH Homemaooo Ammaoo some Hoonsoo 0000 oooo noono nohsn HHozo HHozo xooho nooa .nan oooohh osnono oooohh .axo 000 HHsm eonHN oHooo oaoono:oni oaosooHoooo esp oH 0o omoao Ad.nnOOV.H< oHnoa 66 on on 0.0: 0H 4N 0.0: HeHonv 0 HH 0.N: H000 0 H 0.N: A0 H as nose 0 0 0 Honoaoo 0 0Homvooeoonoeom 00 on on on on on 0.0: 0m 00 NN 0N NH 0N 0.0: 0 0N m 0H N 0H 0.0: Home H w 0 H 0 H 0.N: 0 0 0 0 0 0 Honnsoo 0 HoHonv 05:0N:m on on On On 0m 00 0.0: 0H Om 0H 0N 0 0H 0.m: 0 0H 0 HH 0 N 0.0: Acmv 0 0 0 0 0 0 0.N: 0 0 0 0 0 0 Honoaoo 0 AoHoHov on On On 00 0.0: 0H 0N 0 NH 0.0: 4 4H H m 0.0: Homv 0 0 0 0 0.N: 0 0 0 0 Honssoo 0 Hoosomsoonov 0e:0N:m oN oH oN oH oN oH oN oH oN oH Hoowwmmmm::HHonoa 0:00 Hoonsoo 0000 omwuwm noono noasn HHoro HHoso goose .nan oooohh osnono .mxm osm HHfih poan oHooo nsoeamHopoo esp oH no omeno oooohulohm Ad.nGOOv .H< oHnoa ‘? 0.N: Honnsoo nonsp:0sm 00 0 0n:mH:0 mom 0 o 0 mm 4 m.0: m 0.0: 4N m.m: O 0.Nl o Honncoo HHozmlHHsh o: u 67 i"? H OIAO m 0 m.m: o Houncoo HHozolHHsm 5N < mbIOHI0 0oHHHx oven OH Aomv.qaon nmmw owono 05m Honon 00m nooEnoohn onoo ooooah oooonh .msHooohm noonwsOHnn thsonnHEHonsH coconoHOa A00 oso “wsHooopm wsHhso oosonoHOs nos “<0 «oho3_onsoanooan oooohm .owoaoo oooopm Mom oooooooo oho3.n:osmOHo>o0 mo owono OonoOHosH onn no on on oosdsponooloha oosn H hHso .ohdnohomson noon pod coon oso3.oosH> oohna .oosH> .0noosoo. tonnoa wHOns.sO onsoanoawo wsHuooam sou onoo hnHHonhoa sow .N< oHnos Table A3. 68 Raw mortality data for freezing experiment on buds from 3 different cultivars. the number of buds killed out of 2A buds per test temperature per stage of develo ment. were frozen on 1—node cuttings with misting throughout freezing. Values are Buds ntermdttent Buds were forced from cuttings under mist in the greenhouse. Stage of 10 bud development First Full swell swell Freeze date Cultivar temp.(°C) 1° 2° 1° 2° 6-29-78 'Baco-l' Control 0 0 O -6.5 24 2A 2A 2A 'Vidal-256' Control 0 O O -6.5 2A 2A 2A 2A 'Concord' Control 0 0 0 O -3.5 7 2 1A 3 -6.5 23 23 2A 2A lllllllllll H l "I mll Mll will 3 1293 03047 0581 l llHllWIl