\I‘ I" VI" " ' 'I ”III JV" “3?. ‘ H'H‘L‘Z‘ ' I“ | EI‘II‘ I I II”? .I‘n"Q \I‘d'VIII‘ ‘1! IT; :.7-' I u n1 :1? IMII. ‘ “I“ I “‘31 I" 1.1; ”a .11: ..‘ .I.’Q III-{42" ‘ 3i II, , w-U‘ ”:‘I‘IIII' III . ‘* : :- 1.1"! I'l ‘I ' F iy‘nW193;:A ‘ ’11.” I III: '2 I VI 1' ‘1‘ (“II t! 13):. Ltid‘ yd "5.:1 [3% . “lfigl‘ittfi 2 b '4' ‘ [I .I it .AItIah'fi‘?.t. 11:"; j r . H13. .13; ., H 'I .11. :.u'.;. ,1" 3 III. . I-IIIIIIJ: . ““3192.” I’I ."HI I: .II‘ I: III-WI" IJIIII|:I"I|I"I'1I II," ' MsIII MRI-“I; ' {‘I‘II. ‘. '.’..:‘-;“- 3.3.31.” 31",339 K-I‘l’- ....I’ mi Pi"I. nghfilh {:33 III .1 my”? 3 IL. ‘u: . wink 3,151“ bfiztmafls‘é‘bfi 'l‘g‘i“ '4:th “:2! “III?“ 'ELIII ." g)! 1* “‘3‘“? 1L}? ‘1‘}, ”13.11 II"; I314: . .Ii'fi... ,1? 3-.31. ”I“ lu'I‘ .5 i3 .. . {I m Ip‘fi:§vfif|\n}to.. Sgfiéw I .31 . ‘ q . ' J1 . w; rt ‘7 ‘$ IIII'O ' \-’3ILI'II::‘7‘ 5' 1 “M" II?“ $2.3? I'm} :z-‘II 5). I 3'3 “MI '3.“ I ' t! ”11.1.122m-21i.’ . E‘II' :‘u. I'LI’IIQIIIII‘I'I. «J “: III L'éu’: \«L'IIIII‘I. . LL“ ‘1 “I ~ I: 3‘ '0" v ‘I I ‘3' ‘ 1‘. h'o'.IHII,."' I‘I ‘ ‘ I I Pl“? l."~1~ 3 :I} " ' .x. I‘IE.L‘._ Mam”, . L . . It"? ‘ . I all!“ :I.‘ . .l‘ II‘II‘iI . 3" I“ 'V .‘r {9. 1 .13 . 'Ift‘t III‘I‘: ajvi 0.. . xv do . "€5.50“ II}: " I ._..-:1! .- . It?“ 2’4 . “I?" I..." ‘ .fs‘ 3-5... .I‘3. .'.-”"£ “-' I . .. . ""- . I‘... m ‘ j - 3:,1I, . , .' ‘. ‘ . "63'1""; '.3~:.1I.1\‘.30. ;' ”fr: 9; 1': I1 inf." if? "- I‘ ' .I . 3 ~ ,:A f‘ we. . v: I 3 a . ' sir-j: . r‘33i13'1‘"'=‘5.%“.‘13‘:*:§ EizIII‘II #3333.“ WWW fi'r"“3§W“-‘; 'tz“ -' ~, I-: . “I .‘, .I-"' 1‘ hit . - '3; . " '— Ifii' XIIINI, 993319.“!53'2'1-H11"'"’1*}33,”‘.,"ha‘ ”7?: I:.. .. '_ .. '5- , 1 ‘ ' I. .-..: ‘ 3,15. II" 2 :- ‘ ‘. Ij" INS“ :? -. .uI *‘l‘I‘II‘IH‘Iv 34- ‘1‘!" IIIEIIII'IIIAIII “I WWII ‘~"*""“‘I.’. fl '1 432“” l “[6 . 'I , X'J . .1 .). )1" IIIIII KIWI}??? I‘ .4 3M " 5.3.3 I « 9‘. “‘; -w7 Tag, a a‘ x; av.."<. ‘,-.,.:.- _' .._. .— w..- .‘o “.3: .-... -._->v - 5 ‘;;4 _.- ~ 9....-. __, - 1...-.— .... ., I?" . r.» . “"‘.—u-—: 3; ..o __. _, II . -"-.~ 1.13 It I I I. -‘ ..; ’ c v .I \ . lg? I“ ‘3 ‘41:." U “ I 'I' I 3 " a ‘ I .IJ,.‘ I . ' uh: ' ,1‘. Ir '3.‘ . I 1.3...“ THESIS lilllllUlH||||H||H1|HllllllilllilllIDIIHHIHIIHIHHHI 31293 01716 344 This is to certify that the dissertation entitled EVALUATION OF PRUNUS CERASUS GERMPLASM FOR COLD RESISTANCE presented by Hannah M. Mathers has been accepted towards fulfillment of the requirements for Ph.D. degree in Wre [lanai £41.” ( a Mlafi/professor t // Date /%/’7l/q 7 MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MTE DUE MTE DUE DATE DUE 1/” W14 EVALUATION OF PRUNUS CERASUS GERMPLASM FOR COLD RESISTANCE BY Hannah Mary Mathers A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Horticulture ABSTRACT EVALUATION OF PRUNUS CERASUS GERMPLASM FOR COLD RESISTANCE BY Hannah M. Mathers Knowledge of the level of cold stress resistance and how this resistance is inherited in sour cherry is essential to cultivar development and future germplasm collection. Twig and flower bud samples of two sweet cherries (Prunus avium L.), 12 sour cherries (Prunus cerasus L.) and one ground cherry (Prunus fruticosa Pall.) were collected once a month from August 1990 to March 1991, preconditioned and subjected to freeze tests and differential thermal analysis. Low temperature exotherms (LTEs) were detected in all stems of P. cerasus investigated and were strongly correlated to xylem incipient temperatures aITs) in December to March. LTEs were strongly correlated to phloem~cambium ITs in November. Cortical tissue and leaf bud injury were not related to the stem LTEs. Xylem ITS were the selected method for evaluating sour cherry cold resistance in December to March. Phloem-cambium ITS were the selected method for November. The degree of hardiness of the phloem—cambium in late fall and early spring is a factor determining P. cerasus germplasm geographic distribution and commercial production. Flower bud LTw values were strongly correlated to the flower bud LTEs. Ifl3ovalues were the selected measure for sour cherry flower bud hardiness. Exotherms were not detected in the flower buds of all selections on all evaluation dates but revealed better separation of selections in November, December and March than LTw measures. The principal component (PC) analyses, of the phloem- cambium and cortical tissues, depict gradations between minimum survival temperatures of the two presumed progenitor species of sour cherry, sweet cherry and ground cherry suggesting that cold is a major selective force, contributing to sour cherry population variation. The PC analysis of the flower bud Hmovalues evaluated for five months, indicate selective forces have played a more significant role in the hardiness range of sour cherry flower buds than simply geographic origin. This conclusion is further supported by our LTE’s occurring in midwinter within 3%Zcfi the reported average annual minimum temperature for the northern range of Prunus commercial production (Zone 6). Dedicated to Jim, Nona and Emma iv ACKNOWLEDGEMENT The author wishes to acknowledge, Dr. Amy Iezzoni for the provision of plant material, for her patience and perseverance with this dissertation and for this opportunity to have study under her direction. A special thanks to Dr. Stan Howell for the provision of facilities and equipment, for his proficiency in woody plant cold hardiness and willingness to teach. Special thanks are due to Dr. J.A. Flore, Department of Horticulture, Michigan State University for his constant support and scholarly guidance. I also gratefully acknowledge the help of Dr. K. Poff, Department of Botany, Michigan State University, Dr. R.E.C. Layne, Research Station, Harrow, ON and Dr. C. Chong, University of Guelph, Vineland Station, ON for serving on my guidance committee. The author also gives acknowledgement to the help of J. Beaver and C. Mulinix, in collecting samples, Mr. J. Kozub for statistical guidance and to the grace of God. TABLE OF CONTENTS LIST OF TABLES .............................................................................................. x-xiv LIST OF FIGURES ......................................................................................... xv-xvii LIST OF ABBREVIATIONS ................................................................ . .............. Xfifi INTRODUCTION ......................................................................................................... 1 LIST OF REFERENCES ......................................................................................... 19 Chapter 1: Supercooling and Cold Hardiness in Sour Cherry Germplasm: Part 1. Vegetative Tissue ....................................................................... 24 ABSTRACT ..................................................................................................... 24 INTRODUCTION ........................................................................................... 25 MATERIALS AND METHODS ....................................................................... 30 Plant Material ....................................................................................... 30 Acclimation and Freezing Treatments ...................................... 33 Visual Hardiness Determinations ................................................ 33 Differential Thermal Analysis ........................ . ........................... 35 Comparison of DTA and Visual Browning .................................. 36 PC Analysis .............................................................................................. 37 RESULTS ....................................................................................................... 38 Comparison of DTA and Visual Browning .................................. 38 Response Differences in Tissues ............................... . ................ 57 Hardiness of Prunus cerasus Open-pollinated and Hybrid Seedlings .................................................................................. 63 Xylem PC Analysis ................................................................................ 66 Phloem—cambium PC Analysis ........................................................... 69 Vegetative Bud PC Analysis ........................................................... 72 Cortical Tissue PC Analysis ......................................................... 75 DISCUSSION ................................................................................................ 78 REFERENCES ................................................................................................ 86 Chapter 2: Supercooling and Cold Hardiness in Sour Cherry Germplasm: Part 2. Flower Buds ..................................................................................... 94 vi ABSTRACT ..................................................................................................... 94 INTRODUCTION .................................................................................. . ........ 95 MATERIALS AND METHODS ....................................................................... 99 Plant Material ....................................................................................... 99 Acclimation and Freezing Treatments ...................................... 99 Flower Bud Hardiness Determinations ..................................... 99 Flower Bud Volume Determinations ........................................... 101 LTE Determinations for Flower Buds ....................................... 101 PC Analysis ............................................................................................ 102 Orchard Air Temperatures .............................................................. 102 RESULTS ..................................................................................................... 103 Correlation of Supercooling to Tissue Injury ................ 103 Response Differences in Tissues .............................................. 116 Hardiness of Prunus cerasus Open-pollinated and Hybrid Seedlings ................................................................................ 120 Relationship of Ambient Temperatures .................................. 123 Flower Bud PC Analysis .................................................................. 123 DISCUSSION .............................................................................................. 129 REFERENCES .............................................................................................. 137 Summary .......................................................................................................... 143 APPENDICES A. Mean square and F values for GLM analysis of LTE temperatures for the cherry selections for November to March .......................................................................................................... 147 B. Mean square and F values for GLM analysis of Xylem IT values for the 15 cherry cultivar/selections for August to March ................................................................................................... 148 C. Mean square and F values for GLM analysis of phloem- cambium IT values for the 15 cherry cultivar/seedlings for August to March ....................................................................... 149 D. Mean square and F values for GLM analysis of cortical tissue IT values for the 15 cherry cultivar/selections for August to March ......................................................................... 150 vii APPENDICES (cont.) Mean square and F values for GLM analysis of vegetative bud IT values for the 15 cherry cultivar/selections for November 1990 to March 1991 ....................................................... 151 Mean square and F values for GLM analysis of IT values for all factors ................................................................................... 152 Mean square and F values for GLM analysis of twig LTE values for all factors .................................................................. 153 Mean square and F values for GLM analysis of vegetative bud IT values from November 1990 to March 1991 ........... 154 Mean square and F values for GLM analysis of cortical tissue IT values over eight months ....................................... 155 Mean square and F values for GLM analysis of phloem tissue IT values over eight months ....................................... 156 Mean square and F values for GLM analysis of xylem IT values over eight months .............................................................. 157 Mean square and F values for GLM analysis of flower bud Ifl3ovalues for cherry selections for November 1990 to March 1991 .............................................................................................. 158 Mean square and F values for GLM analysis of flower bud LTE’s for the cherry selections for November 1990 to March 1991 .............................................................................................. 159 Mean square and F values for GLM analysis of flower bud Iflflvalues for all factors .......................................................... 160 Mean square and F values for GLM analysis of flower bud LTE values for November to March ............................................ 161 Mean square and F values for GLM analysis of flower bud LTg,values over five months ....................................................... 162 viii APPENDICES (cont.) Mean square and F values for GLM analysis of one cultivar (Pandy 114) IT and LT” values from August 1990 to March 1991, showing the relative importance of the various non-genotype sources of variation. The variable tree is like replication. ....................................... 163 Mean square and F values for GLM analysis of effects of preconditioning IT values for the cherry selections investigated for August, September and December. The preconditioning temperature was -3%3. ................................. 164 Mean square and F values for GLM analysis of the comparison of regression slopes of xylem (Fig. 2) and phloem (Fig.3). ................................................................................... 165 Mean square and F values for GLM analysis of the comparison of regression slopes of xylem (Fig. 2) and phloem (Fig.3) and flower bud (Fig IV). ........................... 166 Mean square and F values for GLM analysis of tissue types, IT values for 15 cherry selections for August to March. ....................................................................................................... 167 Comparison of five tissue types, phloem, xylem, cortex, vegetative buds and flower buds, of 15 cherry cultivar/seedlings evaluated by visual injury for August 1990 to March 1991 and over all months. ........... 168 ix LIST OF TABLES Table Page Chapter 1 1. Names, abbreviations, presumed origins, type designations and graph symbol of the cultivars and seedlings evaluated ........................................................................... 31 2. Comparison of an August collection of tissue types of 15 cherry cultivar/seedlings evaluated by visual injury of phloem, xylem and cortical shoot sections frozen to —35°C at -5°C per hour ............................................... 39 3. Comparison of a September collection of tissue types of 15 cherry cultivar/seedlings evaluated by visual injury of phloem, xylem and cortical shoot sections frozen to -35°C at -5°C per hour ............................................... 4O 4. Comparison of a October collection of tissue types of 15 cherry cultivar/seedlings evaluated by visual injury of phloem, xylem and cortical shoot sections frozen to -40°C at —5°C per hour ............................................... 41 5. Comparison of a November collection of tissue types of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -5°C per hour and by visual injury of phloem, xylem and cortical shoot sections and vegetative bud cross-sections frozen to -4@T2at -5%3 per hour ..................................................................................................... 42 ON Comparison of a December collection of tissue types of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -5°C per hour and by visual injury of phloem, xylem and cortical shoot sections and vegetative bud cross-sections frozen to -50WC at -5%3 per hour ..................................................................................................... 43 Table 10. ll. 12. LIST OF TABLES (cont.) Page Comparison of a January collection of tissue types of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -5°C per hour and by visual injury of phloem, xylem and cortical shoot sections and vegetative bud cross-sections frozen to -50W: at ~59: per hour ..................................................................................................... 44 Comparison of a February collection of tissue types of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -5°C per hour and by visual injury of phloem, xylem and cortical shoot sections and vegetative bud cross—sections frozen to -45WC at —5%3 per hour ..................................................................................................... 45 Comparison of a March collection of tissue types of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -5°C per hour and by visual injury of phloem, xylem and cortical shoot sections and vegetative bud cross-sections frozen to -40W3 at —5%3 per hour ..................................................................................................... 46 Summary of F values for twig LTE’s and incipient temperatures of four vegetative tissues by month, summed over sweet, sour and ground cherry selections. Analyses were conducted on the calculated IT means of three sub-samples per cultivar and seedling. The SAS procedure GLM was used to perform the analyses ............. 48 Correlation coefficient values and significance for xylem incipient injury and twig LTE’s for different time periods ........................................................................................... 54 Comparison of tissue types of 15 cherry cultivar /seedlings evaluated over 8 months by exotherm analysis conducted at -5°C per hour and by visual injury of phloem, xylem and cortical shoot sections and vegetative bud cross—sections frozen to -50W3 at -5°C per hour ..................................................................................................... 60 xi Table 13a. 13b. 14a. 14b. 15a. 15b. 16a. 16b. II. LIST OF TABLES (cont.) Page Eigenvalues of xylem tissue IT values correlation matrix ......................................................................................................... 67 Eigenvectors of xylem IT values PC analysis .................... 67 Eigenvalues of the phloem-cambium IT values correlation matrix ......................................................................................................... 7O Eigenvectors of the phloem-cambium IT values PC analysis ..................................................................................................... 70 Eigenvalues of the vegetative bud IT values correlation matrix ......................................................................................................... 73 Eigenvectors of the vegetative bud IT values PC analysis ..................................................................................................... 73 Eigenvalues of the cortical tissue IT values correlation matrix .............................................................................. 76 Eigenvectors of the cortical tissue IT values PC analysis ..................................................................................................... 76 Chapter 2 Names, abbreviations, presumed origins, and type designations of the cultivars and seedlings evaluated ...................................................................................... ."munmuumlOO Comparison of a November collection of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -EPC per hour and by visual injury of floral bud cross-sections frozen to -3SWC at -5°C per hour .......................................................................................................... 106 xii LIST OF TABLES (cont.) Table Page III. Comparison of a December collection of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at.-£PC per hour and by visual injury of floral bud cross-sections frozen to -50WC at —5°C per hour .......................................................................................................... 107 IV. Comparison of a January collection of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -£PC per hour and by visual injury of floral bud cross-sections frozen to -5OWC at -5°C per hour .......................................................................................................... 108 V. Comparison of a February collection of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -SW$ per hour and by visual injury of floral bud cross-sections frozen to -459C at -5°C per hour .......................................................................................................... 109 VI. Comparison of a March collection of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -59C per hour and by visual injury of floral bud cross-sections frozen to -4OWC at -5°C per hour .......................................................................................................... 110 VII. Names, type designations, bud volumes, number of scales per bud and number of flowers per bud for the cultivars and seedlings evaluated .............................................................. 113 VIII.Correlation coefficient values and significance for flower buds LTw values and LTE’s calculated for different time periods ................................................................ 115 IX. Comparison of 15 cherry cultivar/seedlings evaluated over 5 months by exotherm analysis conducted at -5%3 per hour and by visual injury of floral bud cross- sections frozen to a maximum of-4MVC at -5°C per hour .......................................................................................................... 121 xiii LIST OF TABLES (cont.) Table Page Xa. Eigenvalues of the floral bud tissue LTm3correlation matrix ....................................................................................................... 126 Xb. Eigenvectors of the flower bud LTw PC analysis .......... 126 xiv LIST OF FIGURES Figure Page Chapter 1 1. A comparison of the differential thermal response (a typical profile), represented by the dashed line (---) with xylem injury scores, represented by *. The average injury score for four twigs is given: 0 = no injury, 1 = a trace of injury; 2 = slight injury, 3 moderate injury and 4 = severe injury. The solid line represents the regression line fitted by PLOTITm Non-Linear Regression function Y = B(1)/(l.0+ B(2)* EXP (-B(3)*X)) (PLOTIT‘D, Scientific Programming Enterprises .................................................................. 47 2. Linear regression of xylem incipient injury and twig LTE’s for November 1990 to March 1991 summed over sweet, sour and ground cherry selections ........................... 50 3. Linear regression of phloem—cambium incipient injury and twig LTE’s for November 1990 to March 1991 summed over sweet, sour and ground cherry selections ............... 51 4. Linear regression of phloem-cambium incipient injury and twig LTE's for November 1990 summed over sweet, sour and ground cherry selections ........................................... 52 5. Linear regression of xylem incipient injury and twig LTE's for December 1990 to February 1991 summed over sweet, sour and ground cherry selections ........................... 55 6. Linear regression of xylem incipient injury and twig LTE’s for March 1991 summed over sweet, sour and ground cherry selections ................................................................................ 56 7. Cultivar/seedling X month interaction of incipient temperature data of the phloem-cambium tissue, sampled from August to March ......................................................................... 61 XV Figure 8. 10. 11. 12. 13. LIST OF FIGURES (cont.) Page Cultivar/seedling X month interaction of the low temperature exotherms, sampled from November 1990 to March 1991 ................................................................................................ 64 Comparison of phloem-cambium tissue injury in P. cerasus ‘Montmorency' and in P. cerasus seedlings over all evaluation dates, August 1990 to March 1991 ........... 65 Positions of PC scores for xylem tissue of two sweet cherries, 12 sour cherries and one ground cherry on the first three PC axes. Abbreviations as in Table 1 ...... 68 Positions of PC scores for phloem-cambium tissue of two sweet cherries, 12 sour cherries and one ground cherry on the first three PC axes. Abbreviations as in Table 1 ....................................................................................................... 71 Positions of PC scores for vegetative bud tissue of two sweet cherries, 12 sour cherries and one ground cherry on the first three PC axes. Abbreviations as in Table 1 ....................................................................................................... 74 Positions of PC scores for cortical tissue of two sweet cherries, 12 sour cherries and one ground cherry on the first three PC axes. Abbreviations as in Table l ........ 77 Chapter 2 DTA profile for March 1991 of Cigany Meggy flower buds showing three peaks at -20, -21 and -22C ......................... 104 xvi Figure II. III. IV. VI. VII. VIII. IX. LIST OF FIGURES (cont.) Page DTA profile for January 1991 of Montmorency flower buds showing three peaks at —l7, —19 and -22.5C .................... 105 Linear regression of flower bud LTE's and bud volumes summed over November to March for the four seedlings showing LTE's in January with large bud volumes ......... 114 Linear regression of flower bud LTw and LTE mean values for November 1990 to March 1991 summed over sweet, sour and ground cherry selections ......................... 117 Linear regression of flower bud LTw and LTE means for November 1990 to February 1991 summed over sweet, sour and ground cherry selections ..................................................... 118 Linear regression of flower bud LTw and LTE means for March 1991 summed over sweet, sour .and ground cherry selections .............................................................................................. 119 Cultivar and seedling X month interaction of the LTW data, sampled from November 1990 to March 1991 ........... 122 Comparison of floral bud LTw values in P. cerasus ‘Montmorency’ and P. cerasus seedlings over all evaluation dates, November 1990 to March 1991.. ........... 124 Ambient temperatures and flower bud LTw values from November 1990 to March 1991 at Clarksville, MI ........... 125 Position of PC scores for flower buds of 15 cherry and seedlings on the first two PC axes. Abbreviations as in Table 1 .............................................................................................. 128 xvii LIST OF ABBREVIATIONS Term Abreviation Differential Thermal Analysis DTA General Linear Model GLM High Temperature Exotherm HTE Low Temperature Exotherm LTE Incipient Temperature IT Lethal Temperature to 50% of sample LTw Principle Component PC 2, 3, 5-Triphenyltetrazolium Chloride TTC xviii INTRODUCTION Statement of Problem Horticultural practices are limited by the plants' genetic potential to acclimate to naturally imposed environmental stresses. In Michigan, spring frosts during blossoming kill sour cherry gynoecia and subsequently reduce cherry yield. Lack of cold hardiness of sour cherry woody tissue is associated with tissue death resulting in production losses and increased susceptibility to diseases and boring insects. Potential exists to increase the cold tolerance of sour cherry through genetic selection, thereby increasing yields and extending its cultural range. vegetative Tissue Evaluation Aim Quamme (1983) conducted hardiness determinations on Prunus persica over the winter months, from mid—October through to early March. Quamme observed that large variances in the ability of the tree to resist freezing over this period. He noted that the loss of hardiness correlated strongly with air temperature fluctuations and water content changes in the tissues. Sour cherry trees in Michigan experience considerable temperature fluctuations in their dormant period. The 2 ability of various individuals in a sour cherry collection, and various tissue types within the tree, to successfully resist these fluctuations would be of considerable interest. For this reason, a study was conducted in 1990—91 to investigate cold hardiness of the xylem, phloem—cambium, cortex, vegetative buds and flower buds related to monthly changes in the trees, from August to March. Relationships have been found between geographic distribution and twig supercooling (Burke and Stushnoff 1979). There have been no obvious relationships found between geographic distribution and flower bud supercooling (Burke and Stushnoff 1979) or commercial range and twig supercooling. Limits in commercial range are usually attributed to lack of flower bud, or bloom hardiness. Geographical range is usually limited by inadequate vegetative tissue hardiness. However, limits in commercial range may also be associated directly or indirectly with lack of vegetative tissue hardiness. For this reason the cold hardiness determinations of the vegetative tissues, conducted in 1990-91, were analyzed using principle component (PC) analysis. PC analysis was used to determine the extent of influence of cold as a major selective force, contributing to population variation in sour cherry. Floral Tissue Evaluation Aim Burke and Stushnoff (1978) indicated that the occurrence or absence of flower bud deep supercooling was not related to geographic distribution but to flower bud morphology. Burke and Stushnoff (1978) classified the Prunus species into five groups dependent on presence of deep supercooling and rate of cold acclimation. The flower buds of hardy Prunus species such as Prunus besseyi are more resistant to heterogeneous nucleation (Quamme et a1. 1982). In these species, the vascular tissue of the flower bud is more fully differentiated. Thus, water migrates from the supercooled regions to the flower bud scales and pith more readily because the xylem is more fully differentiated (Ashworth, 1984). With regard to the evolutionary specialization, the vascular system appears to be slower to change (more conservative) than the organs that it serves. Consequently, investigations of vascular systems may reveal former boundaries, numbers, and categories of organs in flowers, other types of investigations could not (Esau 1977). In summary, Prunus species that have flower buds more resistant to heterogeneous nucleation are more specialized or advanced. P. avium and P. fruticosa are the two proposed progenitor species of P. cerasus (Olden and Nybom 1968, Beaver and Iezzoni 1993). In an examination of 4 flower buds from P. cerasus, one would expect to find northern types more like P.fruticosa, and southern types more like P. avium. The northern types would be more tolerant of heterogeneous nucleation than the southern types. For this reason the cold hardiness determination of the flower buds conducted in 1990-91 were analyzed using principle component analysis to investigate gradations between the northern and southern types. Acclimation Cold acclimation is the series of phenomena that enable a tree to increase its resistance to cold. Cold acclimation in Prunus and most woody species normally occurs in two stages (Hong and Sucoff 1982, Howell and Weiser 1970a, Siminovitch 1982, Weiser 1970). The first stage of acclimation in hardy species is normally induced by short days and warm temperatures. The first evidence of acclimation appears at vegetative maturity. Ketchie reports that until vegetative maturity (VM) is reached the tree will not acclimate regardless of the temperature of exposure. Short day induction of hardiness involves a translocatable factor that promotes hardiness by causing the cessation of vegetative growth (Howell and Weiser 1970). Acclimation can be imposed in woody plants without exposure 5 to short days. The rate, however, is much faster under the influence of a short day photoperiod (Howell and Weiser 1970, Siminovitch 1982, Tumanov et al. 1972). Growth cessation is important to the accumulation of reserves that will be utilized in the hardiness period: an actively growing woody plant does not acclimate (Howell and Weiser 1970b). Analysis has shown that actively growing trees have inadequate carbohydrate reserves and cannot be hardened (Tumanov et al. 1972). Because the effects of this first stage are translocatable it is also reversible until the second stage is initiated (Howell and Weiser 1970b). The second stage begins at vegetative maturity and is induced by exposure to low temperatures (Fuchigami et al. 1982). As the first hardening stage advances, plants become increasingly responsive to temperatures near or just below the freezing point of water, which cause the initiation of the second stage. The effects of this second stage in the acclimation process are non-translocatable and non- reversible (Howell and Weiser 1970b). Description of Supercooling and DTA One analytical procedure frequently used in cold hardiness and freezing injury studies is exotherm or differential thermal analysis (DTA). In DTA, heat of fusion 6 is detected by recording the temperature difference between a dry reference sample and a wet sample during freezing. The differential temperature is plotted against the reference temperature (Andrews et al. 1983, Barney 1989). The type of freezing observed by DTAFs occurs in plants that resist freezing by supercooling. Most Prunus species have the ability to maintain supercooled cellular water in their xylem ray parenchyma and flower primordia (Andrews and Proebsting 1983, Burke and Stushnoff 1979). Supercooling in tissue can only happen when effective barriers are created to isolate the internal cell solution from external nucleation (heterogeneous nucleation). The internal cell solution then remains stable, to the homogeneous nucleation point, the temperature at which water freezes in the absence of nucleators (Gusta et al. 1983, Quamme 1985). Some plant tissues supercool to as low as -47°C before homogeneous nucleation takes place. The depth and stability of this supercooling (called deep supercooling) distinguishes it from transitory supercooling that occurs in whole plants above —5TZ(Burke et al. 1976). The DTA profile in species which deep supercool indicates two exotherms. The first is associated with extracellular freezing that results from heterogeneous nucleation and is called the high temperature exotherm 7 (HTE). The second exotherm is associated with the tissues that supercool to the homogeneous nucleation temperature and is termed the low temperature exotherm (LTE) (Barney 1989, Ketchie and Kammereck 1987, Quamme 1985). Xylem Supercooling Gusta et al. (1983) stated that deep supercooling in the xylem of plants was the result of two conditions: 1) the cells were dispersed into independent freezing units made possible by the cell walls; and, 2) the plasma membrane remained intact, preventing the flow of extracellular water into the cell, so that the solutes within the cell were concentrated by dehydration, thus lowering the homogeneous nucleation temperature. For the first condition to be met, secondary thickening must be present in the cell walls (Wisniewski 1995). To summarize, when vegetative tissue freezes, ice forms either inside the cell (intracellular) or outside the cell walls (extracellular) (Wisniewski 1995). Intracellular freezing disrupts the integrity of the cell and is invariably lethal. Flower Bud Supercooling In several Prunus species, the LTE in the flower buds appears to be a discrete event and the flower primordium or 8 floret appears to be uniformly injured after the freezing event producing the LTE takes place (Quamme 1985). Flower buds, such as peach, which has a single flower produce a single LTE, whereas flower buds, such as sour cherry, which has several florets, produce multiple exotherms (Quamme 1974). In many Prunus flower buds, the location of ice formation is dispersed between different tissue types and is referred to as "extraorgan" freezing (Ashworth et al. 1983, Quamme et al. 1982). Two barriers prevent external nucleation of the supercooled flower primordium: a) the cuticle or epidermis that prevents nucleation of the primordium surface; and b) a "dry region" at the base that prevents an ice boundary from spreading into the scales during initial stages of freezing. The formation of the "dry region" at the base of the primordium results when water is withdrawn from the bud axis to freeze in the preferred sites, in the scales: water is not withdrawn from the flower primordium (Quamme 1978). Water freezes at higher temperatures in the bud scales and lower bud axis because these external tissues of the flower bud are lower in solute concentrations. Layne and Ward (1978) reported that concentrations of reducing sugars and other carbohydrates significantly rose in flower buds, 9 corresponding with cold acclimation periods. Gusta et al. (1983) stated that for water to supercool to low temperatures there must be a reduction in intracellular water and a concentration of solute within the cell. Initial ice formation will occur in the external tissues because of the raised solute concentration within flower primordia. Subsequent nucleation of the primordium is prevented by the dry region and the cuticle. As the vapor pressure of the external tissue ice-water mixture is less than that of the internal flower primordium water, a vapour pressure gradient is produced (Steponkus 1978). Further external freezing produces a greater gradient buildup. Eventually, if freezing continues, the ice front in the lower axis will advance. Once it advances to a point of contacting with the primordium, nucleation of the supercooled primordium cells is very rapid and produces sudden death. Seasonal changes in deep supercooling points of Prunus were found to be correlated with changes in water content of the flower bud and its parts, the flower primordium and vascular traces. Both deep supercooling point and water content of the flower bud and its parts correlated with air temperature preceding collection (Quamme 1983). If the water content of the flower decreased, then the LTE occur at 10 a colder temperature (Quamme 1995). Lower water contents were associated with colder minimum air temperatures preceding collection (Quamme 1983). Water content of the flower primordium and vascular traces, but not the whole flower bud, was critical to the deep supercooling point of the flower primordium. Water was lost from the flower primordium at a slower rate than the vascular traces and this loss was associated with decline of the deep supercooling point (Quamme 1995). Much of the fluctuation in flower bud hardiness during the mid winter season, therefore, appears to result from the redistribution of water during freezing and thawing and its effect on deep supercooling (Quamme 1983). During deacclimation in the spring, irreversible changes take place in the flower bud, eg. vascular differentiation, which prevent deep supercooling. The finding that deep supercooling occurs in flower buds has implications to breeding programs in which the objective is to improve cold hardiness. Exotherm analysis can be used to measure varietal differences in flower bud hardiness (Quamme 1985). Knowledge of the expression of deep supercooling mechanisms in different species and cultivars should allow better identification of the germplasm required for making the improvements. 11 Cold Hardiness Evaluations and DTA Too often it is inferred that different types of freezing stress are separate and distinct events that are mutually exclusive in a given plant. It is more appropriate, however, to conclude that the stresses that arise during freezing are a sequential series of events (Steponkus 1981). The plant‘s ability to survive depends upon its successful sequential avoidance or tolerance of each event. In 1972 Levitt stated that cold stress resistance in plants could be broken down into two major components: woody plants have mechanisms to either avoid or tolerate ice within their tissues, but not both. Since 1972, however, studies conducted during supercooling analyses have indicated that avoidance and tolerance are not mutually exclusive. When samples were cooled slowly or held at non-lethal subfreezing temperatures LTEs could be reduced in size or eliminated altogether, while the killing temperatures could be decreased due to protective dehydration (Biermann et al. 1979, Gusta et al. 1983). In many cases, researchers have discovered significant correlations between the temperatures at which injury began and those at which LTEs occurred, when conducted in midwinter, during controlled freezing experiments (Biermann et al. 1979). However, to conduct plant hardiness 12 evaluations based solely on midwinter resistance or the use of DTAs alone would not be appropriate (Ketchie and Kammereck 1987, Burke and Stushnoff 1979). Concerning DTA analyses, several problems are associated with the technique used. The technique chosen when conducting DTA's can substantially alter the exothermic patterns of the tissue. For this reason DTA twig evaluations need to be correlated with some other measure of tissue viability, and technique characterizations are required. There are essentially four areas requiring characterization: i) cooling rates and holding temperatures (Quamme et al. 1973, Biermann et al. 1979, Gusta et al. 1983); ii) sample collection, preconditioning and equilibration (Ashworth 1982); iii) sample size (Ashworth and Davis 1984); iv) freeze—nucleation and thawing protocol. Twig DTA Methodology Cooling rates and holding temperatures Cooling rates have varied widely in published studies from 120°C/hr to SOC/day (Sakai 1982) . Several studies have indicated that the occurrence of both the LT and HT exotherms were not independent of cooling rate (Quamme 1983, Quamme et a1. 1972, Ketchie and Kammereck 1987). The cooling rate can substantially influence the temperature at 13 which exotherms occur and the "width" of the exotherms (Quamme 1983). The cooling rate also affects the temperature at which tissues are injured and the way the injury occurs. Biermann et al. (1979) found that mean survival temperature from DTA analyses, derived at slow cooling rates were most closely related to lowest survival temperatures. Biermann et al. (1979) had conducted rate dependency studies at 65, 33, 8 and 6.5°C/hr. For DTA to be an accurate predictor of cold hardiness you must assume that controlled freezing tests in the lab impose the same freezing stress as the outside environment would. Changes as rapid.as 1200 per hour or even BOWC/hr do not occur naturally even in the most drastic conditions. Considering that a LTE does not represent a single temperature but is a picture of an event spanning several degrees it makes sense to utilize a slower freezing rate. Most of the current DTA work on stem tissues has been conducted at freezing rates ranging from.l.EPC/hr (Ashworth and Davis 1984) and TTVhr (Warmund and Slater 1988, Warmund et al. 1989) to EPC/hr (Montano et al. 1987). Some of the most classical studies relating supercooling and cold hardiness to geographic distribution of woody plants have been conducted at rates of.3TVhr (Quamme et al. 1982, Gusta et al. 1983). 14 Perhaps the most important factor to consider when determining the rate of freeze is the knowledge that DTA evaluations can not stand alone. The necessity to correlate the DTA data with a qualitative viability test sets a precedent for the freezing rate used. Viability tests such as 2, 3, 5-Triphenyltetrazolium Chloride (TTC) and visual browning are more rate dependent than DTA evaluations. Sample collection, preconditioning and equilibration The way in which samples are collected and prepared affects both their exothermic patterns and their hardiness, as measured in the lab. Several collection methodologies are outlined in the current literature (Andrews and Proebsting 1986, Andrews and Proebsting 1987, Ashworth and Davis 1984). The purpose of these methodologies is tissue equilibration. Stabilization, however, is achieved by ice formation and dehydration not by collection methodology, per se (Olien and Lester 1985). The use of an artificial hardening regime such as described by Sakai (1982) and Quamme et al. (1982), ensures maximum freeze resistance by equilibration. The artificial hardening regime is also particularly useful in heritability studies of cold resistance in order to limit environmental variation. Heritability of cold 15 resistance is greatly influenced by the environment: the greater the effects of the environment, the lower the heritability of the trait (Hansche 1983, Cummins and Aldwinckle 1983). Thus, considerable motivation is provided to reduce the effects of environmental variation. Sample size A.minimum of three samples per treatment is reported in the literature. Twig section sizes range from 3cm to 10 cm long, with a common diameter of 6 mm. All sampling is done from previous season's growth, at approximately the same height around the tree, 1 metre. Freeze-nucleation and thawing protocol. The wrapping of samples in moist cheese cloth is the standard procedure for ultra low freezes. This procedure ensures sites of nucleation for prevention of excessive dehydration. Thawing at UTiin the air is also currently ascribed in the literature (Sakai 1982, Quamme et al. 1982). After thawing, it is recommended to stand the samples in water (Sakai 1982), in a water saturated atmosphere for 5—8 days, at room temperature (Quamme et al. 1982). 16 Flower Bud DTA.Methodology In the flower buds as in woody tissue, researchers have discovered significant correlations between the temperatures at which injury began and those at which LTEs occurred during controlled freezing experiments (Quamme 1974, Quamme et a1. 1982). However, as with wood tissue evaluations, to base hardiness determinations on DTAs alone is not appropriate (Ketchie and Kammereck 1987). There are several problems identified with floral bud DT analyses, distinct from those problems encountered in stem tissue DTAFs. As in wood, the technique chosen can substantially alter the exothermic patterns of the tissue. There are essentially four areas of concern in floral bud DTA methodology: i) cooling rates and holding temperatures (Quamme 1974); ii) sample collection, preconditioning and equilibration (Ashworth 1982); iii) sample size (Ashworth and Davis 1984); iv) freeze-nucleation and thawing protocol. Cooling rates and holding temperatures Floral bud hardiness determinations are more freeze rate dependent than woody tissues studies. Studies should never be conducted at more than a SW: drop per hour (Quamme 1995). Anywhere between 1 and 5%: is reported in the literature (Quamme et. al. 1982, Ashworth 1984, Ashworth and 17 Davis 1984). Due to the critical rate dependency of floral bud hardiness determinations, a linear freeze regime is essential. Wolf and Pool (1987) showed that prolonged periods of subfreezing temperatures (-10 to -15W:) do increase bud hardiness. However, the ability to detect exotherms decreased. Fifty percent of the buds were killed suggesting that water migrated from bud primordia, and that this dehydration ultimately was associated with bud injury. Sample collection, preconditioning and equilibration The concerns of note when using DTA's in floral bud work are relatively the same as those expressed for woody tissue. However, hydration is more critical. Wolf and Pool (1987) found that a hydrated substrate was essential to ensuring consistent occurrence of HTE's at temperatures distinct from LTE's. Sample size Wolf and Pool (1987) also noted that the integrity of the primordia-nodal interface is a critical component of the barrier to ice nucleation of supercooled primordia. Exclusion of subjacent nodal tissue resulted in an ability of the bud primordia to supercool. This result emphasized 18 that bud excision must not negate structural barriers contributing to primordia supercooling and supported the findings of similar experiments (Ashworth 1982, Quamme 1978). Samples must include at least 2-3 cm of stem tissue (Quamme 1983). Freeze-nucleation and thawing protocol. The concerns of note when using DTA's in floral bud work are relatively the same as those expressed for woody tissue studies. LIST OF REFERENCES l9 Andrews, P.K. and Proebsting, E.L. 1983. Differential thermal analysis and freezing injury of deacclimating peach and sweet cherry reproductive organs. J. Amer. Soc. Hort. Sci. 108: 755-759. Andrews, P.K., Proebsting, E.L. and Campbell, 6.8. 1983. An exotherm sensor for measuring the cold hardiness of deep- supercooled flower buds by differential thermal analysis. HortScience. 18(1): 77-78. Andrews, P.K. and Proebsting, E.L. 1986. Development of deep supercooling in acclimating sweet cherry and peach flower buds. Hort Sci. 21(1): 99-100. Andrews, P.K. and Proebsting, E.L. 1987. Effects of temperature on the deep supercooling characteristics of dormant and deacclimating sweet cherry flower buds. J. Amer. Soc. Hort. Sci. 112(2): 334-340. .Ashworth, E.N. 1982. Properties of peach flower buds which facilitate supercooling. Pl. Physiol. 70: 1475- 1479. Ashworth, E.N. 1984. Xylem development in Prunus flower buds and the relationship to deep supercooling. Pl. Phys. 74: 862-865. .Ashworth, E.N. and Davis, GLA. 1984. Ice nucleation within peach trees. J. Amer. Soc. Hort. Sci. 109: 198- 201. .Ashworth, E.N., D.J. Rowse, and LMA. Billmyer. 1983. The freezing of water in woody tissues of apricot and peach and the relationship to freezing Injury. J. Amer. Soc.Hort. Sci. 108(2): 299-303. Barney, D.L. 1989. Differential thermal analysis. Am. Nurserymen. 169(11): 47-57. Biermann, J., Stushnoff, C. and Burke, MWJ. 1979. Differential thermal analysis and freezing injury in cold hardy blueberry flower buds. J.Amer. Soc. Hort. Sci. 104(4):444-449. 20 Burke, M.J., and C. Stushnoff. 1979. Frost Hardiness: A discussion of possible molecular causes of injury with particular reference to deep supercooling of water, p.197-225. In H. Mussell and R.C. Staples (eds.) Stress Physiology in Crop Plants. New York: John Wiley & Sons. Burke, M.J., Gusta, L.‘V;, Quamme, H.Au,‘Weiser, C. J. and Li, P. H. 1976. Freezing and injury in plants. Ann. Rev. Plant Physiol. 27: 507—528. Cummins, JgN. and Aldwinckle, H.S. 1983. Apple rootstock breeding, pp. 294-394. In J. Janick (ed.) Plant breeding reviews. Avi Publishing Co., Westport, Conn. Dayton, D.F., Bell, R.L. and Williams, E.B. 1983. Disease resistance. pp. 189-215. In J.N. Moore and J. Janick (eds.) Methods in fruit breeding. Purdue University Press, West Lafayette, Indiana. Esau, K. 1977. Anatomy of seed plants. John Wiley & Sons, New York, NY. Fuchigamd, L.H., C.J.‘Weiser, K.K0bayashi, R.Timmds, and L;V. Gusta. 1982. A degree growth stage (GS) model and cold acclimation in temperate woody plants, p. 93-116. In P.H. Li and A. Sakai (eds.) Plant Cold Hardiness and Freezing Stress. Vol. 2. New York: Academic Press. Gusta, L.V., Tyler, N.J. and Chen, T.H. 1983. Deep undercooling in woody taxa growing north of the 40%: isotherm. Plant Physiol. 72: 122-128. Hansche, P.H. 1983. Response to selection. pp. 154-171. In J.N. Moore and J. Janick (eds.) Methods in fruit breeding. Purdue University Press, West Lafayette, Indiana. Hong, 8., and E. Sucoff. 1982. Temperature effects on acclimation and deacclimation of supercooling in apple xylem. p.341-356. In P.H. Li and A. Sakai (eds.). Plant Cold Hardiness and Freezing Stress. Vol. 2. New York: Academic Press. Howell, 6.8., and C.J.‘Weiser. 1970a. The environmental control of cold acclimation in apple. Plant Physiol. 45: 390—394. 21 Howell, 6.8., and C.J.‘Weiser. 1970b. Similarities between the control of flower initiation and cold acclimation in plants. HortScience 5(1): 18—20. JUnttila, 0. 1980. Effect of photoperiod and temperature on apical growth cessation in two ecotypes of Salix and Betula. Physiol. Plant. 48: 347-352. Ketchie, D.O. and Kammereck, R4 1987. Seasonal variation of cold resistance in Ablus woody tissue as determined by differential thermal analysis and viability tests. Can J. Bot. Vol. 65: 2640-2645. Layne, R.C. and‘ward, G;M. 1978. Rootstock and seasonal influences on carbohydrate levels and cold hardiness of ‘Redhaven' Peach. J. Amer. Soc. Hort. Sci. 103: 408-413. Levitt, J. 1972. Responses of plants to environmental stresses. lst ed. Academic Press, New York, NY. Mbntano, J.N., Rebhuhn, M., Hummer, K. and Lagerstedt, H.B. 1987. Differential thermal analysis for large- scale evaluation of pear cold hardiness. HortScience. 22(6): 1335-1336. Olien, C.R. and Lester, 6.3. 1985. Freeze-induced changes in soluble carbohydrates of rye. Crop Sci. 25: 288-290. Quamme, H.An 1974. An exothermic process involved in the freezing injury to flower buds of several Prunus species. J.Amer. Soc. Hort. Sci. 99(4): 315—318. Quamme, H.S. 1978. Mechanism of supercooling in overwintering peach flower buds. J. Amer. Soc. Hort. Sci. 103(1): 57—61. Quamme, HMA. 1983. Relationship of air temperature to water content and supercooling of overwintering peach flower buds. J. Amer. Soc. Hort Sci. 108: 697-701. Quamme, H.A~ 1985. Avoidance of freezing injury in woody plants by deep supercooling. Acta Horticulturae 168: 11-30. 22 Quanme, ILA. 1995. Deep supercooling in buds of woody plants. Pages 183-199 in R. Lee, G.J. Warren and L.V. Gusta, eds. Biological ice nucleation and its applications. APS Press, St. Paul, Minnesota. Quamme, H.A~, Stushnoff, C. and‘weiser, C.J. 1972. The relationship of exotherms to cold injury in apple stem tissues. J. Amer. Soc. Hort. Sci. 97(5): 608—613. Quamme, H.Au,‘Weiser, G.J. and Stushnoff, C. 1973. The mechanism of freezing injury in xylem of winter apple twigs. Pl. Phys. 51: 273—277. Quamme, H.Au, Layne, R.E.C. and Ronald,‘W;G. 1982. Relationship of supercooling to cold hardiness and the northern distribution of several cultivated and native Prunus species and hybrids. Can. J.Plant Sci. 62: 137-148. Sakai, A. 1982. Freezing resistance of ornamental trees and shrubs. J. Amer. Soc. Hort. Sci. 107(4): 572-581. Siminovitch, D. 1982. Major acclimation of living bark of Sept. 16 black locust tree trunk sections after 5 weeks at 10 C in the dark -- Evidence for endogenous rhythms in winter hardening. p. 117-128. In P.H. Li and A. Sakai (eds.). Plant Cold Hardiness and Freezing Stress. Vol. 2. New York: Academic Press. Steponkus, P.L. 1981. Responses to extreme temperatures. Cellular and sub-cellular bases. Cornell Agronomy Papers pp. 372-402. Steponkus, P.L. 1978. Cold hardiness and freezing injury of agronomic crops. Advances in Agronomy 30: 51- 98. Tumanov, I.I., G.V. Kuzina, and L.D. Karnikova. 1972. Effect of vegetation time in woody plants on accumulation of reserve carbohydrates and nature of the photoperiodic response (in Russian). Fiziologiya Rastenii 19: 1122-1131. (Academy of Sciences of the USSR, Moscow, translation). ‘Warmund, M.R. and Slater, J;V. 1988. Hardiness of apple and peach trees in the NC-140 rootstock trials. Fruit Var. J. 42: 20-24. 23 Warmund, M.R., George, M.F., Ellersieck, M.R. and Slater, J;V. 1989. Susceptibility of blackberry tissues to freezing injury after exposure to IETL J; Am. Soc. Hort. Sci. 114(5): 795-800. 'weiser, C.J. 1970. Cold Resistance and Acclimation in Woody Plants. HortScience 5: 3-17. wolf, T.K. and Pool RqM. 1987. Factors affecting exotherm detection in the differential thermal analysis of grapevine dormant buds. J. Amer. Soc. Hort. Sci. 112(3): 520-525. Supercooling and Cold Hardiness in Sour Cherry Germplasm: Part 1. vegetative Tissue Abstract Knowledge of the level of cold stress resistance and how this resistance is inherited in sour cherry is essential to cultivar development and future germplasm collection. Twig samples of two sweet cherries (Prunus avium L.), 12 sour cherries (Prunus cerasus L.) and one ground cherry (Prunus fruticosa Pall.) were collected once a month from August 1990 to March 1991, preconditioned to induce maximum cold resistance and subjected to freeze tests and differential thermal analysis. LTEs detected in all stems of P. cerasus investigated and were strongly correlated to xylem ITs in December to March. LTEs were strongly correlated to phloem-cambium ITS in November, representing the acclimation period. Xylem ITS were the selected method for evaluating the cold resistance of sour cherry in December to March and phloem-cambium ITs were selected for November. The degree of supercooling and hardiness of the phloem—cambium in late fall and early spring appears significant in determining P. cerasus commercial range. 24 25 The principle component (PC) analyses, of the phloem- cambium and cortical tissues, depict gradations between minimum survival temperatures of the two presumed progenitor species of sour cherry, sweet cherry and ground cherry. Cultivars and progeny of crosses of northern origin parents showed hardiness values more comparable to ground cherry than did slections of less-cold-hardy parents suggesting that cold is a major selective force, contributing to sour cherry population variation. INTRODUCTION Prunus varieties are characteristically heterozygous (Fogle 1975) and their seedlings vary highly in hardiness. Not only are there differences within and among seedling populations, but also within the same tree, differences occur between phloem, cambium, xylem, bark, vegetative and floral buds (Quamme et al. 1982, Burke and Stushnoff 1979). Stushnoff et al. (1985) discusses the presence of unilateral maternal inheritance in Prunus. Nearly twice as many seedlings which were resistant to cold stress were obtained when the cold hardy species Prunus americana was used as the female parent in contrast to the tender Prunus salicina. There have also been reports of a maternal 26 parental effect on the resistance of germinating apple seed to low temperature damage (Stushnoff et al. 1985). The freezing response of certain tissues can be monitored using differential thermal analysis (DTA) as outlined by Quamme et al. (1972) and since modified to take advantage of recent advances in digital data-acquisition equipment and microcomputers (Wisniewski et al. 1990). The type of freezing observed by DTA's occurs in plants which resist freezing by supercooling. The DTA profile in species which deep supercool indicates two exotherms. The first is associated with bulk water freezing within lumens of tracheary elements and extracellular spaces and is called the “high temperature exotherm” (Wisniewski 1995). The second exotherm is associated with tissues that supercool to the homogeneous nucleation temperature within living cells and is termed the “low temperature exotherm” (LTE) (Ketchie and Kammereck 1987, Quamme 1985, Wisniewski 1995). The shape of the LTE and the temperature at which it is initiated is subject to seasonal changes (Quamme et al. 1972, Ketchie and Kammereck 1987, Wisniewski and Ashworth 1986, Arora et al. 1992) and in general follows a pattern similar to those of other processes involved in cold 27 acclimation: increasing in the fall, reaching a maximum in midwinter, and then decreasing in the spring (Wisniewski 1995). The basis for the complex shape of the exotherms, which is often accentuated during periods of acclimation and deacclimation (Ashworth 1993), is not understood (Wisniewski 1995). Xylem tissue in some species, rather than freezing homogeneously, may respond as a heterogeneous population of cells that freeze over a wide temperature range (Wisniewski 1995) as indicated by the complex shape of the LTE. In addition to the shape variability of LTEs, the number of LTEs can also vary (Ketchie and Kammereck 1987). Ketchie and Kammereck (1987) observed at least three exotherms in apple, and indicated that they represent the freezing of different tissues of the stem: pith, primary xylem and secondary xylem. In many cases, researchers have established a correlation between stem tissue injury and the temperature range of the LTE, at the stage of maximum cold resistance (Quamme et al. 1972, George et al. 1974, Ketchie and Kammereck 1987). Hong and Sucoff (1982), in a study of eight woody plant species, observed a linear relationship between the number of dead xylem ray cells and the amount of supercooled water that had frozen. Association of the 28 entire freezing curve with tissue injury is unclear in studies where the LTE is quite broad or exhibits several peaks. Ketchie and Kammereck (1987) observed that only during midwinter was the peak near the homogeneous nucleation point of water associated with xylem tissue injury. In some species, LTEs observable in early winter completely disappeared during prolonged exposure of the tissue to subzero temperatures approaching —38W: (Gusta et al. 1983). Wisniewski et al. (1993) noted that contact of the plasma membrane with the cell wall was essential for maximum supercooling to occur. Current research on cell wall structure and deep supercooling indicates that the type, amount and degree of cross-linking of pectin within the pit membrane may determine the size of the pores and/or their permeability to water (Wisniewski 1995). Pectin-mediated regulation of porosity/ permeability of the pit membrane is an attractive hypothesis because it provides a plausible basis to explain seasonal shifts that occur in the extent of deep supercooling (Wisniewski 1995). Most Prunus species have the ability to maintain supercooled cellular water in their xylem ray parenchyma 29 (Wisniewski et al. 1995, Rajashekar and Burke 1978). Xylem can supercool to as low as -47©C before homogeneous nucleation takes place. To date, no studies have examined the general presence, occurrence and influence of the environment on LTE's in sour cherry (Prunus cerasus L.). A diverse sour cherry germplasm collection was established at Michigan State University, Clarksville Horticultural Experiment Station, through Dr. Iezzoni's collection trips to Yugoslavia, Romania, Hungary, Bulgaria and Poland. The collection contains a range of plant material from P. avium types to and P. fruticosa types the two proposed progenitor species of P. cerasus (Olden and Nybom 1968, Beaver and Iezzoni 1993). Because a lengthy quarantine period is required for imported clonal material, the collection consists primarily of hybrid and open- pollinated seedlings of elite selections from eastern European breeding programs (Hillig and Iezzoni 1988). Because of the scope and diversity of this germplasm collection, an excellent opportunity to investigate supercooling and cold hardiness in sour cherry existed. An understanding of cold resistance heritability and environmental influence in sour cherry is critical to cultivar development and future germplasm collection. 30 At the whole plant level, resistance to cold is a complex quantitatively inherited character (Cummins and Aldwinkle 1983, Artlip et al. 1997). Principal Component (PC) analysis has been used successfully to understand the response of complex traits to imposed treatments or evolutionary pressures (Iezzoni and Pritts 1991). The main objective of this study was to compare DTA and visual browning as methods to determine minimum hardiness levels for cherry xylem, phloem—cambium, cortex and vegetative buds. The second objective was to determine if these methods detect differences in these tissues in the cherry germplasm from August to March. The third objective was to determine if the cherry selections differed in their cold resistance. The fourth objective was to use PC analyses to examine the possible association of minimum survival temperatures to geographic origin. MATERIAL AND METHODS Plant Material The plant material included two Prunus avium selections, a P. fruticosa selection, and 12 P. cerasus selections. The material was chosen to represent different geographical origins (Table 1). Four sour cherry selections Names, abbreviations, presumed origins and type designations of the cultivars and seedlings evaluated. 31 Table 1 Abbr. Presumed Origin Type Prunus avian Emperor Francis F Unknown Southern Schmidt S Unknown Southern Prunus cerasus Cigany Meggy o.p. G Hungary Southern Csengodi Csokras o.p. K Hungary Southern English Morello E Germany Northern English Morello X U Germany X Yugoslavia Northern Sumadinka Fructbare von Michurin V Moldavia Northern Meteor R Montmorency X Northern Russian Seedling Montmorency M France Northern Oblacinska o.p. O Yugoslavia Southern Pandy 114 o.p. A Hungary Southern Pitic de Iasi o.p. I Moldavia Northern Spaniole X Crisana o.p. C Spain X Romania Southern Wolynska X Sumadinka W Poland X Yugoslavia Northern Prunus firuticosa 586-1 P Unknown Northern 32 originated in Hungary (Cigany Meggy o.p., Csengodi Csokras o.p., and Pandy 114 o.p., Spaniole X Crisana o.p.) which has a milder climate than Germany and Moldavia which is represented by three selections (‘English Morello', Pitic de Iasi o.p., and ‘Fructbare von Michurin'). ‘Montmorency' is originally from France and Oblacinska and Sumadinka are from Serbia. ‘Meteor' is a hybrid seedling selected in Minnesota from the cross ‘Montmorency’ by a Russian seedling. The two other hybrid seedlings are from crosses made at MSU between parents differing in area of origin: ‘English Morello' X Sumadinka and Wolynska (originally from Poland) X Sumadinka. All studies were conducted with detached shoots of the previous season's growth collected randomly across the crown of trees growing outdoors at the Michigan State University, Clarksville Horticultural Experiment Station, Clarksville, MI. A three-way treatment structure was used in a completely randomized design with three sub-samples per treatment. The study was conducted over the period of August 1990 to March 1991. All twigs were cut in the field and placed in polyethylene bags for transit and storage. When collections were done in below freezing temperatures, the polyethylene bagged samples were placed in ice-packed coolers to prevent tissue thawing. 33 .Acclimation and Freezing Treatments All twigs were subjected to one of two artificial hardening procedures, depending on the hardiness of the material at the time of collection (Sakai 1982). For the December through March collections, the twigs were hardened at -3°C for 3 wk followed by -5°C for 1 wk and -10°C for 5 days and for August through November, 0 to —19C for a maximum of 20 days (Sakai 1982). These artificial hardening treatments were used to induce maximum hardiness in order to minimize environmental variation (Sakai and Weiser 1973, Gusta et al. 1997). After the artificial hardening treatment, the twigs were frozen utilizing a linear programmed freeze regime, dropping by 5°Ch"1 to a series of test temperatures at five-degree intervals from -5 to -50%3. During the freeze, the samples were wrapped in moist cheese cloth and aluminium foil to ensure ice nucleation and prevent excessive dehydration (McKenzie and Weiser 1975). At each test temperature, four twigs were removed and allowed to thaw in a cold storage at O”: for 16 hrs. ‘Visual Hardiness Determinations After thawing, the twigs were placed in a water- saturated atmosphere for five to eight days at room 34 temperature approximately ZOTL Cross sections of the twigs were examined under a dissecting microscope (40X) and were scored numerically (0-4) for injury to the cortical tissue, the phloem-cambium region and the xylem on the basis of the extent and intensity of oxidative browning. A rating of 0 represented no injury, 1 a trace of injury, 2 slight injury, 3 moderate injury and 4 represented severe injury (Quamme et al. 1982). The vegetative buds were also cut and scored numerically for browning (0-4) (Quamme et al. 1982). Quamme et al. (1982) defined the temperature of incipient injury as the temperature at which the average injury score did not exceed 1 using hand-fitted graphs of the injury scores. Here incipient injury was defined as the temperature at which injury not exceeding 1.0 using non-linear regression of the 0-4 injury scores (IT). The IT values were determined graphically using PLOT-IT Non-Linear Regression function Y = B(1)/(1.0+B(2)*EXP(-B(3)*X)) (PLOT-IT), (Scientific Programming Enterprises, Haslett, MI, 1991). This non-linear regression equation fits an exponential curve, where Y is a function of the exponent of X and B(i) is a coefficient, * means multiple and / means divide. Y is the dependent variable and X is the independent variable. 35 The general equation would be Y = bex. B(i) describe the rate of growth of the curve. The coefficients of B were determined by PLOT-IT and were different for each curve calculated. In the case of Figure 1, ‘Meteor’, xylem injury scores in February, B(1) is 18.09942, B(2) is 11052.55 and B(3) is -.1748091. For this curve the equation would read Y = 18.09942 / (1 + 11052.55 * art-”“809“ ‘X ), so that if x was -40, Y would be 18.09942 / 11.155876 or 1.622. Analyses were conducted on the calculated IT means of three sub-samples per selection and referred to as the variable plant in the General Linear Model (GLM). The SAS© procedure GLM was used to perform the analyses. Fisher's least significance difference test was used to compare means (SAS© Institute Inc. 1989). IT values were rounded to the nearest O.5k2(Cain and Andersen 1976). Differential Thermal Analysis The DTA utilized an EXP 16 multiplexer and amplifier, a 37 pin D male connector, a DAS 8 output board (MetraByte Corp., Taunton, MA) and thermoelectric modules (15 X 15 mm) (CPI 1.0-17—O6L, Melcor Materials Electronic Products Corp., Trenton, NJ). The EXP 16 was set for a gain of 1000 fold. The samples were affixed to each module with the aid of a 36 thermally conductive paste (Omegatherm 201, Omega Engineering, Inc., Stamford, CT). The modules were wrapped with parafilm and aluminium foil. A dried sample of comparable size was attached to the opposite side of the module. The modules were then suspended in a structural foam box within a converted (Mathers et al. 1991) ultralow freezer (Revco, Model 179OD, Rheem Manufacturing Co., Asheville, NC). Specimens and sensors were held for 16 hours at -3%:. The temperature was then lowered at a constant rate of 59C per hour with the aid of a programmable temperature controller (Micron 82300, Reserach Inc., Minneapolis, MN). The temperature was monitored with a 36— gauge copper-constantan thermocouple connected to the reference channel of the EXP 16. Four samples per cultivar and seedling were evaluated at each collection date. Analyses were conducted on the calculated LTE means of three plants per selection. The SAS© procedure GLM was used to perform the analyses. Fisher's least significant difference test was used to compare means (SAS© Institute Inc. 1989). Comparison of DTA.and Visual Browning Linear regression analyses were performed, where the 37 means of the LTE were regressed on the means of IT. The coefficient of determination (r2) values were determined using Microsoft Excel©, ChartWizard function (Microsoft Excel©, Version 5.0c, Microsoft Corporation, 1994, Seattle, WA). The square root of r2, the correlation coefficient R values were calculated for the xylem, phloem-cambium, cortex and vegetative buds from November to March. The statistically significant values of r were determined from significance tables (Steel and Torrie, 1980). PC Analysis Principal component (PC) analysis was performed using the PRINCOMP procedure in SAS, 1989. In each of the four analyses, the cultivar and seedling means were used to create a correlation matrix from which standardized PC scores were extracted. To determine which of the first three PCs in each analysis accounted for the greatest portion of the variance for each month, the eigenvectors of the three PCs were compared (Hillig 1988). The most variation attributed by each month was determined by identifying the eigenvector with the highest absolute value for that month. See Table 1 for abbreviations of cultivars and seedlings. 38 RESULTS Comparison of DTA and Visual Browning Low temperature exotherms were not present in any of the cherry selections from collections in August (Table 2), September (Table 3) and October (Table 4). However, most of the cherry selections exhibited LTEs in November through March (Tables 5 - 9). The notable exception was P. fruticosa which only exhibited a LTE in February. A typical DTA profile is represented by ‘Meteor' in February, and presented in Fig. 1. The first exotherm, which occurred from -7 to -12°C, represented the freezing of bulk water, the high temperature exotherm. The LTEs were present as single deflections (Fig. 1). The size of the LTE varied among the selections but tended to be smaller in the more hardy selections. When the LTE temperatures were compared among the cherry selections within each sampling month, significant differences were identified among the cherry selections (Appendix A). At the 0.05 level, the differences for twig LTE's were only significant in January and March (Table 10). Similarly significant xylem hardiness differences measured as oxidative browning were identified among the selections when comparisons were done for each month (Appendix B). At 3 9 Table 2 Comparison of an August collection of tissue types of 15 cherry cultivar/seedlings evaluated by visual injury of phloem, xylem and cortical shoot sections frozen to -35°C at -5°C per hour. Cultivar/Seedling Incipient Temperature Xylem Phloem Cortical Prunus avium Emperor Francis -21.5 az - 9.0 a -11.0 a Schmidt -14.0 a - 7.0 a -10.0 a Prunus cerasus CsengodiCsokas -10.0 a -8.5 a -10.0 a (SpanioleXCrlsana) -10.3 a -9.0 a -12.0 a Pandy114 -11.0 a -9.0 a -12.0 a Piticdelasi -11.5 a -10.0 a -12.0 a WolynskaXSumadinka -18.0 a -12.0 a -17.0 a EMXSumadinka -14.0 a -15.0 a -13.0 a Oblacinska - 9.5 a -9.5 a - 9.5 a Cigany Meggy 42.0 a —9.0 a -11.o a Montmorency -19.0 a -15.0 a -15.0 a Fructbare von Michurin -26.5 a -10.0 a -10.0 a English Morello -18.0 a -11.0 a -15.0 a Meteor -20.0 a -15.0 a -17.0 a P.frutioosa586-1 -16.0 a -19.0 a -18.0 a 1 Means in columns, based on three sub-samples, per four replicates, followed by different letters are significantly different (P=0.05) according to Fisher’s least significant difference test. 4 0 Table 3 Comparison of a September collection of tissue types of 15 cheny cultivar/seedlings evaluated by visual injury of phloem. xylem and cortical shoot sections frozen to -35°C at -5°C per hour. Cultivar/Seedling Incipient Temperature Xylem Phloem Cortical Prunus avium Emperor Francis -16.0 b2 - 9.0 a -16.0 a Schmidt -20.0 c -18.0 b -16.0 a Prunus cerasus CsengodiCsokas -12.5 a -10.0 a -15.0 a (SpanioleXCrisana) -20.0 c -18.0 b -15.0 a Pandy114 -17.0 b -11.5 a -16.0 a Piticdelasi -17.0 b -13.5 a -15.5 a WolynskaXSumadinka -17.0 b -14.0 a -17.0 a EMXSumadinka -16.5 b -14.0 a -16.0 a Oblacinska -11.5 a -10.0 a -14.5 a Cigany Meggy -18.0 c -14.0 a -16.5 a Montmorency -19.0 c -19.0 b -15.0 a Fructbarevon Michurin -19.0 c -19.0 b -18.0 a English Morello -23.0 c -20.0 b -20.0 a Meteor -18.0 c -22.0 b -18.0 a P. frutioosa 586-1 -23.0 c -19.0 b -20.0 a TMeans in columns, based on three sub-samples, per four replicates, followed by different letters are significantly different (P=0.05) according to Fisher’s least significant difference test. 4 1 Table 4 Comparison of an October collection of tissue types of 15 cherry cultivar/wedlings evaluated by visual injury of phloem, xylem and cortical shoot sections frozen to -40°C at -5°C per hour. Cultivar/Seedling Incipient Temperature Xylem Phloem Cortical Prunus avium Emperor Francis -25.0 2 - 9.0 -19.0 Schmidt -28.0 -10.0 -21.0 Prunus cerasus Csengodi Csokas -29.0 -19.0 -18.5 (Spaniole X Cn‘sana) -28.5 -24.0 -25.5 Pandy 114 -21.0 -22.0 -24.0 Pitic de Iasi -27.5 -23.5 -24.0 Wolynska X Sumadinka -28.5 -24.5 -26.0 EM X Sumadinka -27.0 -25.5 -27.0 Oblacinska -27.0 -24.5 -25.0 Cigany Meggy -26.0 -22.0 -25.0 Montmorency -28.0 -24.0 -25.0 Fructbare von Michurin -29.0 -26.0 -27.0 English Morello -18.0 -24.0 -28.0 Meteor -29.0 -25.0 -26.0 P. fruticosa 586-1 -29.0 a -27.0 c -27.0 a ' Means in columns, based on three sub-samples, per four replicates. followed by different letters are significantly different (P=0.05) according to Fisher's least significant difference test. 4 2 Table 5 Comparison of a November collection of tissue types of 15 cheny cultivar/mdlings evaluated by differential thermal analysis conducted at - °C per hour and by visual injury of phloem- cambium, xylem and cortical shoot sections and vegetative bud (VB) cross-sections frozen to - 35°C at -5°C per hour. Cultivar/Seedling LTE (0°) Incipient Temp. (C°) Twig Xylem Phloem Cortical VB Prunus avium Emperor Francis -22.0 a2 -23.0 a -19.0 a -26.0 a -15.0 a Schmidt -26.0 a -26.0 a -21.0 b -24.0 a -17.0 a Prunus cerasus CsengodiCsokas ~22.0 a ~25.5 a ~17.0 a ~24.0 a 48.5 a (SpanioleXCrisana) -22.0 a -18.0 a -20.3 a -25.5 a -24.0 b Pandy114 -23.0 a -23.0 a -22.5 b -26.0 a -22.5 b Piticdelasi -25.0 a -24.0 a -25.0 b -25.0 a -22.0 b WolynskaXSumadinka -28.0 a -28.5 a -26.0 b -24.0 a -24.5 b EMXSumadinka -25.5 a -27.5 a -27.5 b -19.5 a -23.0 b Oblacinska -27.5 a -27.0 a -27.0 b -25.0 a -25.5 b CiganyMeggy -26.0 a -26.5 a -22.5 b -25.0 a -20.5 a Montmorency -23.0 a -27.0 a -19.0 a -26.0 a -15.0 a chtbare von Michurin -30.0 a -27.0 a -27.0 b -26.0 a -27.0 b English Morello -23.0 a -11.0 a -28.0 b -27.0 a -25.0 b Meteor -27.0 a -29.0 a -28.0 b -28.0 a -26.0 b P. fruticosa 586-1 -— -20.0 a -29.0 b -25.0 a -22.0 b zMeans in columns, based on three sub-samples, per four replicates, follmrved by different letters are significantly different (P=0.05) according to Fisher’s least significant difference test. 4 3 Table 6 Comparison of a December collection of tissue types of 15 cheny cultivar/seedlings evaluated by differential thermal analysis conducted at -5°C per hour and by visual injury of phloem-cambium, xylem and cortical shoot sections and vegetative bud (VB) cross-sections frozen to -50°C at - 5°C per hour. Cultivar/Seedling LTE (C°) Incipient Temp. (C°) Twig Xylem Phloem Cortical VB Prunus avium Emperor Francis -15.0 a2 -29.0 a -15.0 a -21.0 a -14.0 a Schmidt -15.0 a -15.0 a -15.0 a -17.0 a -15.0 a Prunus cerasus Csengodi Csokas -19.5 a -26.5 a -19.5 a -18.0 a -17.5 a (Spaniole X Crisana) -14.0 a -22.0 a -14.0 a -16.0 a -18.0 a Pandy 114 44.5 a -19.0 a -14.5 a -17.0 a -15.5 a Pitic de Iasi -20.0 a -32.0 a -20.0 a -25.0 a -19.5 a Wolynska X Sumadinka -23.0 a -30.0 a -23.0 a -28.5 a -19.0 a EM X Sumadinka -23.0 a -35.0 a -26.0 a -25.5 a -16.5 a Oblacinska -21.0 a -26.5 a -21.0 a -20.5 a -21.5 a Cigany Meggy -17.5 a -24.0 a -17.5 a -19.0 a -19.5 a Montmorency -28.0 a -30.0 a -28.0 a -30.0 a -18.0 a Fructbare von Michurin -25.0 a -33.0 a -25.0 a -25.0 a -18.0 a English Morello -20.0 a -24.0 a -20.0 a -15.0 a -16.0 a Meteor -28.0 a -41.0 a -28.0 a -32.0 a -23.0 a P. fruticosa 586-1 — -26.0 a -26.0 a -24.0 a -15.0 a ' Means in columns, based on three sub-samples, per four replicates, followed by different letters are significantly different (P=0.05) according to Fisher’s least significant difference test. 4 4 Table 7 Comparison of a January collection of tissue types of 15 cheny cultivar/seedlings evaluated by exotherm analysis conducted at - °C per hour and by visual injury of phloem-cambium, xylem and cortical shoot sections and vegetative bud (VB) cross-sections frozen to -50°C at -5°C per hour. Cultivar/Seedling LTE(C°) Incipient Temperature (C°) Twig Xylem Phloem Cortical VB Prunus avium Emperor Francis -22.0 a2 -25.0 a -16.0 a -23.0 a -24.0a Schmidt -25.0 a -31.0 a -15.0 a -27.0 a -22.0a Prunus cerasus CsengodiCsokas -34.0 b -34.0 a -27.5 b -30.5 a -25.0a (SpanioleXCrisana) -36.0 c 43.0 c -31.5 b -36.0 b -26.5a Pandy114 -40.0 c -42.0 c -31.0 b -35.5 b -28.5a Piticde Iasi -38.5 c -44.5 c -31.0 b -37.5 c -24.0a WolynskaXSumadinka -37.5 c -44.0 c -31.0 b -40.0 c -30.0a EMXSumadinka -21.0 a -40.0 c -29.5 b -37.0 c -31.0a Oblacinska -37.0 c -43.5 c -31.5 c -38.0 c -34.0a CiganyMeggy — -38.5 b —27.0 a -34.5 b -27.0a Montmorency -38.0 c -39.0 b -32.0 c -34.0 b -28.0a Fructbare von Michurin -37.0 c -43.0 c -33.0 c -40.0 c -31.0a English Morello -41.0 c -37.0 b -29.0 b -32.0 b -25.0a Meteor -42.0 c -43.0 c -33.0 c -40.0 c -30.0a P. fruticosa 586-1 — -45.0 c -37.0 c -41.0 c -34.0a 1 Means in columns, based on three sub-samples, per four replicates. followed by different letters are significantly different (P=0.05) according to Fisher’s least significant difference test. 4 5 Table 8 Comparison of a February collection of tissue types of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -5°C per hour and by visual injury of phloem-cambium, xylem and cortical shoot sections and vegetative bud (VB) cross-actions frozen to 45°C at -5°C per hour. Cultivar/Seedling LTE(C°) Incipient Temp. (C°) Twig Xylem Phloem Cortical VB Prunus avium Emperor Francis -25.0 a2 -28.0 a -25.0 a ~26.0 a -28.0 a Schmidt -27.0 a -34.0 a -30.0 a -32.0 a -29.0 a Prunus cerasus Csengodi Csokas -27.5 a -32.0 a -20.5 a -27.0 a -32.5 a (Spaniole X Crisana) -24.5 a -24.0 a -26.0 a -26.5 a -31.0 a Pandy 114 -28.0 a -31.0 a -29.0 a -31.5 a -32.0 a Pitic de Iasi -30.5 a 3 -35.0 a -34.5 a -33.0 a -32.5 a Wolynska X Sumadinka -34.0 a -38.0 a -33.0 a -37.0 a -31.0 a EM X Sumadinka -34.5 a -36.0 a -31.0 a -33.5 a -32.0 a Oblacinska -27.5 a -30.0 a -26.5 a -30.5 a -31.0 a Cigany Meggy -32.5 a -35.5 a -32.0 a -32.0 a -32.0 a Montmorency -33.0 a -37.0 a -33.0 a -35.0 a -31.0 a Fructbare von Michurin -30.0 a -37.0 a -32.0 a -35.0 a -31.0 a English Morello -25.0 a -33.0 a -24.0 a -33.0 a -30.0 a Meteor -33.0 a -37.0 a -34.0 a -37.0 a -32.0 a P. fwticosa 586-1 -32.0 a -32.0 a -30.0 a -33.0 a -31.0 a ‘ Means in columns, based on three sub-samples, per four replicates, followed by different letters are significantly different (#005) according to Fisher's least significant difference test. 4 6 Table 9 Comparison of a March collection of tissue types of 15 cheny cultivar/seedlings evaluated by exotherm analysis conducted at - °C per hour and by visual injury of phloem-cambium, xylem and cortical shoot sections and vegetative bud (VB) cross-sections frozen to -40°C at - °C per hour. Cultivar/Seedling LTE (C°) Incipient Temperature (C°) Twig Xylem Phloem Cortical VB Prunus avium Emperor Francis -10.0 a2 -24.0a -10.0 a -19.0 a -15.0a Schmidt -19.0a -27.0a -16.0 b -27.0 b -19.0a Prunus cerasus Csengodi Csokas -33.0 b -34.5 b -25.0 c -30.5 b -27.0 c Spaniole X Crisana — -36.0 b -23.0 b -32.0 c -25.0 b Pandy114 -31.0 b -34.0 b -22.0 b -30.3 b -25.0 b Piticdelasi -32.5 b -34.0 b -29.0 c -30.0 b -29.0c WolynskaXSumadinka -31.0b -35.0 b -29.0 c -32.0 c -30.0c EMXSumadinka -32.0 b -35.0 b -30.5 c -32.0 c -26.5c Oblacinska -32.0 b -36.0 b -28.5 c -32.0 c -29.5c Cigany Meggy -27.5 b -35.0 b -26.0 c -3o.o b .2505 Montmorency -32.0 b -34.0 b -23.0 b -30.0 b -24.0b Fructbare von Michurin -33.0 b -37.0 b -32.0 c -32.0 c -23.0 b English Morello -31.0 b -35.0 b -26.0 c -29.0 b -27.0c Meteor -33.0 b -35.0 b -29.0 c -31.0 c -31.0c P. fmticosa 586-1 — -23.0 a -34.0 c -33.0 c -27.0c ’Means in columns, based on three sub-samples, per four replicates, followed by different letters are significantly different (P=0.05) according to Fisher's least significant difference test. 47 .Aamma momaumuoucm manEoumoum oauaucoaom .BHBoqcv “Ax¢.mvm -vcxmsxmcm +o. HV\AHVm u » noduocsu scammonmom Moocaq :oz c.8HHqu >n oouuam coda Godmuuumou was mucomoumou soda oaHOm one .xuanH ono>om u v oco >mshca ououoooe I m .>ushca unmaHm u a .xusncA no woman a u H .xuanca on u o .co>am a“ amasu use“ Mam swoon xusflca omouo>m 0&8 8V menu/immerse can 2.: or... on: on: 81.8: B: 07 an o fifth C . . _ . . . Trikfiwrwvflwmbi “Lt _ . . . o __ . : __ C i _ : . __ __ l ,N 5843,.) n GE , no mm“: H Amvm mvmmpflwi u Em in “ sow. u we 9.93.2 H mm . . 2x .. amicaxm . gm + or \ :E .s an ooucomoumou .mououm >nsnca anxx nua3 Ar--. coda oonooo onu xn oouoououmou .Aoaauoun Hmuamhu my uncommon fineness Hoaucououuao on» no couwuoaeoo 4 .H .mE m JD 3 13 NH hi3 flN 8U AV .l S as we 3d 0 mm S 3 4 8 Table 10 Summary of F values for twig LTE's and incipient temperatures of four vegetative tissue by month. summed over sweet, sour and ground cheny selections. Analyses were conducted on the calculated IT means of three sub-samples per cultivar and seedling. The SAS procedure GLM was uwd to perform the analyses. Month Pr>F for Twig LTE’s and Tissue lT values Twig LTE Xylem IT Phloem IT Cortical IT Vegetative Bud IT August 0.3060W 0.3687"5 0.372.?“5 September 0.0097" 0.0060” 0.2039” October 0.3120“ 0.0145" 0.0820“ November 0.2762“ 0.1395“ 0.0356* 0.4332W 00346" December 0.0804” 0.3236“ 0.0763NS 0.2915“ 0.5697“ January 0.0043" 0.0001m 0.0004*** o.ooo2*** 0.6340” February 0.5377“ 0.1812W 0.1206“ 0.0559“ 0.2027"5 March 0.0219* 0.0006“ 0.0022" 0.0001m 0.0063“ "sf, **, *** Nonsignificant and significant at the 0.05, 0.01 and respectively. .001 levels of probability, 49 the 0.05 level, significant differences for xylem browning were only found for September, January and March (Table 10). The most significant difference, at the 0.05 level, occurred for phloem-cambium browning, appearing in September, October, November, January and March (Table 10). Significant differences for the cortical browning, at the 0.05 level, were only found in January and March and for the vegetative buds only in November and March (Table 10). The months which showed the most significant differences between twig LTE’s and browning scores for the vegetative tissues evaluated were March, January and September (Table 10). Only March had significant differences in all vegetative tissues and the twig LTE's. There was a significant linear relationship between oxidative browning, measured as IT, and the LTE for the xylem and phloem—cambium tissues from November to March (Fig. 2 and 3). The (Group * X) factor is non-significant indicating the slopes of the two tissues are not significantly different (Appendix S). Cortical tissue and vegetative bud injury were not correlated to the stem LTEs. In the acclimation period, represented by the November collection, the best correlation of the LTE was to the phloem-cambium (Fig. 4). The correlation coefficient (r) 50 32628. .920 case» 28 39. .395 58 .8883 32 €82 2 83 59562 e8 MBA mes“ can be? .5385 82.? Co 368.38 .883 .n .wE av. ov. an. on. Gov ah.— wu. on. 2. 2. n- c O ”dr- .._..1- 1)— _... --. - W4; 3- '3 ~ . o ‘1' I'lrl' .Ilrl r n ‘3‘ (30) annual mi midi-m1 o "P mm- -MlL- , -.__.__..(___ _.___-._.L._._ ._ M-———-—-———- -——-——4 n9. 51 accuse—8. abuse 23% one .58 48.5 86 BEES 32 see—2 2 33 Lon—83.52 he 9mg..— EE 2: was be? .5565 EEnEflrEQOEQ Co 3.82%: 58:3 .n .uE 6055 as e... an. on. a. o". 2- 2. n. o _ an ear." _ £23321. 1 ,0 m . l c c 4 _ in o l c l o l e .. a r . 7 ll 0 0 w A c 1 \ or- 52- oz- 91- or- (Do)mwdmi WP“! SC" 52 6.. m5 22828 Eons .58 new L826 56 ooEEam 82 89562 to. ME... 92 new be? 8222.. EaaEuoEooEQ Co 53352 .35.. .v .9“. w- Sl' x summon: 8.75.5.1 lo.) mammal wardraur 53 for the phloem-cambium IT's and the twig LTE's during this period is 0.6777, significant at pS0.0l (Fig. 4). The phloem-cambium browning occurred within 5%: of the LTE, and for 11 of the 14 cultivars and seedlings studied browning occurred before the LTE (Table 5). In November, only the phloem-cambium and vegetative buds revealed significant differences among cultivars and seedlings (Table 10). In midwinter, represented by the December to February collections, the best correlation of the LTE was to the xylem browning (Table 11). The correlation coefficient (r) for the xylem during this period is 0.8347, significant at pS0.0l (Fig. 5). The xylem browning occurred within 5W3, on average, of the LTE, and in almost every case the LTE occurred at a higher temperature than the xylem browning (Tables 6, 7 and 8). In deacclimation, represented by the March collection, the best correlation of the LTE was to the xylem browning (Table 11). The correlation coefficient (r) for the xylem and the stem LTE’s during this period is 0.9574, significant at pS0.0l (Fig. 6). The xylem browning began within 4°C, on average, of the LTE, and in every case the LTE preceded browning (Table 9). 54 Table 11 Correlation coefficient values and significance for xylem incipient injury and twig LTE’s for different time periods. Time Sample size r} r Nov-Mar. 69 0.6092 0.7805“7 Nov. 14 0.3234 0.5685' Dec.-Feb. 42 0.6968 0.8347" Mar. 13 0.9166 0.9574" 55 22.8.8 .920 958» new 58 48% 36 388832 beacon 3 82 .BEooOQ .8.“ MP: we»: use be? .5565 E2? .6 22.4832 30:3 .n .uE 60cm: 2.. o... 2. ea. 3. on. 2. 2. a. o . . e _ W “a _ ii 2. Same n u _ . 5.: saloon» _ r 17 2- m. p . m 8- u x. m c c _ O \ n- m c \ u o _ . a n E .lillrrlll rill. i- r ,- , Ii :1??? , 1-. mm. c c c \ $1 9.. c O ‘ O . irrrl llirlrlrirl 1 ill (trillilirllrpe illllrllllrrn'l 56 mm on 3 so. P: an n _ o. n llrilll ‘llllll..lllllllll.1.,l.ll lili'lrlr‘ \ 15o a u M £3: .. .322. .. a 52.8.3. bee—.6 .58 55 .85 8.6 3553 .8. 55.2 5.. mm: 33. o5. be“... .556... 52b. .8 cosmonaut 52.5 .e .mE cu nu on .3 CV (.30) "Wuhan warden] 57 Response Differences Among Tissues The comparison of xylem injury means for September (Table 3) (Appendix B), January (Table 7) (Appendix B) and March (Table 9) (Appendix B) illustrated significant differences among cultivars and seedlings. The comparison of xylem injury means for August (Table 2) (Appendix B), October (Table 4) (Appendix B), November (Table 5) (Appendix B), December (Table 6) (Appendix B) and February (Table 8) (Appendix B) did not illustrate significant differences among cultivars and seedlings. The comparison of phloem—cambium injury means for September (Table 3) (Appendix C), October (Table 4) (Appendix C), November (Table 5) (Appendix C), January (Table 7) (Appendix C) and March (Table 9) (Appendix C) illustrated significant differences among cultivars and seedlings. The comparison of phloem-cambium injury means for August (Table 2) (Appendix C), December (Table 6) (Appendix C) and February (Table 8) (Appendix C) did not illustrate significant differences among cultivars and seedlings. The comparison of cortical injury means for January (Table 7) (Appendix D) and March (Table 9) (Appendix D) illustrated significant differences among cultivars and 58 seedlings. The comparison of cortical injury means for August (Table 2) (Appendix D), September (Table 3) (Appendix D) October (Table 4) (Appendix D), November (Table 5) (Appendix D), December (Table 6) (Appendix D) and February (Table 8) (Appendix D) did not illustrate significant differences among cultivars and seedlings. The comparison of vegetative bud injury means for November (Table 5) (Appendix E) and March (Table 9) (Appendix E) illustrated significant differences among cultivars and seedlings. The comparison of vegetative bud injury means for December (Table 6) (Appendix E), January (Table 7) (Appendix E) and February (Table 8) (Appendix E) did not illustrate significant differences among cultivars and seedlings. The GLM of the IT data showed significant differences due to cultivars and seedlings, tissue and month (Appendix F). The interaction of cultivars and seedlings X month, cultivars and seedlings X tissue, tissue X month and cultivars and seedlings X plant were also significant (Appendix F). The interaction of tissue X plant and cultivars and seedlings X tissue X month was non-significant (Appendix F). 59 The comparative hardiness and range of temperatures over which injury occurred to vegetative tissues, evaluated over all months, are presented in Table 12. Xylem was on average 5.4°C hardier than the phloem-cambium in 14 of 15 selections evaluated. Xylem was on average 2.8°C hardier than vegetative buds in 13 of 15 selections evaluated. Xylem was significantly hardier than all tissues evaluated in all months except November and February. In November, the xylem, phloem and cortex had similar mean hardiness values and in February, the xylem and cortex had similar mean hardiness values (Appendix V). The most susceptible vegetative tissue, evaluated over all months, in 14 of 15 selections evaluated, was the phloem-cambium (Table 12)(Appendix V). Consistently in the least hardy taxon, P. avium and cultivars and seedlings of P. cerasus, Csengodi Csokas, Spaniole X Crisana, Pandy 114, Oblacinska, and Cigany Meggy the phloem-cambium was the most susceptible vegetative tissue. For P. avium the lowest survival temperatures for the phloem-Cambium tissue occurred in February (Fig. 7) and the most susceptible time for midwinter injury was December and January (Fig. 7). In December and January the phloem was significantly less hardy than the xylem (Appendix V). The phloem-cambium was strongly 60 Table 12 CornparlsonoftissuctypssoftSchenycrossea/cultivarsevaluatedover8monthsbyexothelmanalysisoonductedst5°0per hoursnd byvisusl injury of phloem-cambium, xylem and cortical shoot sections and vegetative bud (V8) cross-sections frozen to ~ail“Cat-5"Cpcrhour. Cultivar/Cross Low Temp. Exotherm (C°) Incipient Temp. (C°) ng' Xylem Phloem Cortical vs Prunus avium Emperor Francis ~190 s ~24.0 a ~14.0 3 ~50 a ~19.5a Schmidt ~22.5 a -24.5 s ~16.5 s ~220 a ~20.5a Prunus cerasus Csengodi Csokss ~27.0 b ~55 a ~190 a ~220 s ~240 b Spanich Crisana ~50 3 ~50 a ~220 b ~55 b ~50 b Pandy114 ~27.5 b ~24.5 a ~200 b ~24.0 b ~24.5 b Piticdelasi ~290 b ~50 b ~230 c ~50 b ~50 b Wolynska X Sumadinka ~290 b (1)0 b ~24.0 c ~50 c -27.0 b EMXSumadinka ~27.5 b ~29.0 b ~24.5 d ~55 c ~26.0 b Oblacinska 50.0 b ~26.5 s ~22.5 c ~24.5 b ~28.5 b Cigany Meggy -- ~27.0 b ~21.0 b ~24.0 b ~50 b Montmorency 31.0 5 ~50 b ~240 d ~26.5 c ~230 b FructbarevonMichurin 31.0 b 31.5 b ~55 d ~27.0c -26.0 b English Morello ~280 b ~50 a ~50 b ~50 b ~50 b Meteor ~330 b 81.5 b ~270 d ~29.0 c -28.5 b P.fludcasa§86~1 - ~27.0 b ~28.0 d ~280 c -26.0 b zMeans in columns, based on three sub-samples, per four replicates, followed by different letters are significantly different (P=0.C5) according to Fisher's least significant difference test. . 61 .35. 5.5:. 8 so... .253. :5... 3.953 .25.. 5535358....— o... .5 3.... 8382.50. 59.52.. .5 5.595.... 5.55 on «5598.53.50 H .5... (3°) urn-urinal 62 influenced by the warm temperatures experienced in late November before the December collection (ambient temperature date not shown). In most cases the xylem continued to increase in hardiness between November and December, however, the phloem-cambium deacclimated (Tables 5 and Table 6)(Appendix V). In March also, the phloem-cambium deacclimated much more rapidly than the xylem showing a loss in hardiness of 3.8°C averaged over the 15 selections versus O.5°C for the xylem (Tables 8 and 9). In March the phloem was significantly less hardy than the xylem or cortex. As in December, the loss of hardiness in March, in the phloem tissue was most pronounced in the P.avium cultivars (Table 9). The cortical tissue also decreased in hardiness between November and December by 3.?K2averaged over the 15 selections (Tables 5 and 6)(Appendix V). However, the deacclimation of the cortical tissue in March was not as pronounced as the phloem-cambium (Tables 8 and 9). In November and January, the vegetative buds, of almost all taxa and cultivars and seedlings, were the most susceptible vegetative tissue (Tables 5 and 7). In November, the vegetative buds were significantly the least hardy vegetative tissue (Appendix V). In the March 63 collection, the phloem-cambium, of 9 of the 15 selections, was the most susceptible vegetative tissue (Table 9)(Appendix U and V). In November, the hardiness of vegetative buds and phloem-cambium of some cultivars and seedlings of P. cerasus, were similar (Table 5). In the phloem-cambium significant differences occurred between taxa and cultivars and seedlings from northern and southern areas (Fig. 7). The relationship appeared to be dependent on area of origin. The LTEs of the month X cultivars and seedlings interaction also revealed clear differences between taxa and cultivars and seedlings from northern and southern areas. (Fig. 8). Hardiness of Prunus cerasus Open-pollinated and Hybrid Seedlings The hardiness of the phloem-cambium of P. cerasus ‘Montmorency’ and open-pollinated and hybrid seedlings are presented in Fig. 9. Some of the P. cerasus seedlings, of northern origin, exhibit superior hardiness in the phloem- cambium when compared to ‘Montmorency’ in late fall and early spring. 64 .32 5.82 8 :3— uBEQSz 82.. 33:3. £552.85 2382.83 .32 a... .8 538.53 5.8:. x «5893.52.30 .a .uE 5:62 3.533 x 2m . \ Sam on 3.2563 \\ s33 3§m§m .. (3°) umuadml \ \ mas—80 60990 x . 65 Fig. 9. Comparison of phloem-cambium tissue injury in P. cerasus 'Montmorcncy' and P. cerasus seedlings over all evaluation dates, August 1990 to March 1991. 66 Xylem PC Analysis The first three PCs of the xylem IT data account for 74% of the total variance among cultivar and seedling means; i.e., for 35%, 22%, and 17% of the variance, respectively (Table 13a). See Table l for the abbreviations representing the cultivars and seedlings on the PC analysis figures. The northern selections are at the negative end of PCl (Fig 10). The only exceptions being ‘English Morello' (E) and P. fruticosa (P). The P. fruticosa used in this study is an open-pollinated seedling of unkown origin. Proceeding from negative to positive values of PCI, the cultivars and seedlings means decrease in hardiness. The heaviest loading of PCl is represented by November, December and February (Table 13b). The northern origin seedlings ‘English Morello’ X Sumadinka (U), Wolynska X Sumadinka (W) and Pitic de Iasi (I) cluster with the northern cultivars ‘Montmorency’ (M), ‘Fructbare von Michurin’ (V), and ‘Meteor’ (R)(Fig. 10). The northern cultivars and seedlings are also at the negative end of PC2. The only exception is Pitic de Iasi(I) an open—pollinated seedling from Moldavia. From negative to positive values of PC2 the cultivars and seedling means decrease in hardiness during acclimation, represented by 67 Table 13a Eigenvalues of the xylem tissue IT values correlation matrix. Eigenvalue Difference Proportion Cumulative PRINl 2.76844 1.03543 0.346055 0.34605 PRIN2 1.73301 0.32153 0.216626 0.56268 PRIN3 1.41148 0.46122 0.176435 0.73912 PRIN4 0.95026 0.36746 0.118783 0.85790 PRIN5 0.58281 0.28810 0.072851 0.93075 PRIN6 0.29470 0.10699 0.036838 0.96759 PRIN7 0.18772 0.11614 0.023465 0.99105 PRIN8 0.07158 0.008948 1.00000 Table 13b Eigenvectors of the xylem tissue IT values PC analysis. PRINl PRIN2 PRIN3 PRIN4 Aug. 0.204706 0.650038 -.023604 -.188433 Sept. -.215990 0.520304 0.328070 0.444998 Oct. 0.368043 -.109636 -.326072 0.628663 Nov. 0.491617 -.124363 -.37l605 -.033063 Dec. 0.476468 0.156679 0.091936 -.073994 Jan. 0.250407 -.210518 0.578369 0.456122 Feb. 0.428517 (.324264 0.134535 -.l64607 Mar. 0.246877 -.324399 0.535105 ".359919 PRIN5 PRIN6 PRIN7 PRIN8 Aug. -.223147 0.270922 0.583887 -.188055 Sept. 0.192011 0.216655 -.213547 0.497563 Oct. -.059027 0.466940 -.160253 -.322159 Nov. 0.326222 -.048615 0.293539 0.639307 Dec. -.662669 -.247182 -.426490 0.228864 Jan. -.024187 -.372089 0.449605 -.096127 Feb. 0.600849 -.238805 -.336665 -.366934 Mar. 0.051944 0.634577 -.065159 0.091140 68 .H magma CH mm mHonE>m pom mcofiumfibwubnm .mmxm um omucu umuaw on» so huuozo ocsouo oco can mofiuuono usom NH .moflnumno umozm 03» mo oommflu Ema>x now mouoom um mo mcofluflmom .OH .on mm_._u mm..ml . F7__m& , I I '1 a 4:: :5: 69 August and September and increase in hardiness during deacclimation represented by March. All the southern origin hybrid seedlings Spaniole X Crisana (C), Cigany Meggy (G), Pandy 114 (A), Csengodi Csokras (K), and Oblacinska (0) cluster together at the positive end of PC2. The southern cultivars, however, ‘Schmidt’ (S) and ‘Emperor Francis' (F) are at the negative end of PC2. ‘English Morello’ (E) a seedling of Germany origin represents the negative extreme of PC2. Phloem-cambium PC analysis In an analysis of the phloem-cambium tissue, the first two PCs of the phloem data account for 74% of the total Variance among cultivar and seedling means; i.e., for 61% and 13% of the variance, respectively (Table 14a). P. fruticosa (P) the presumed northern progenitor of sour Cherry, and P. avium ‘Emperor Francis’ (F), the southern progenitor, are situated at the extremes of PC1, representing the extremes in their hardiness (Fig. 11. \_l The P. cerasus cultivars and seedlings evaluated are situated between these two presumed progenitors. The only other P. avium cultivar investigated ‘Schmidt’ (5) to is located close ‘Emperor Francis’. 70 Table 14a Eigenvalues of the phloem-cambium IT values correlation matrix. Eigenvalue Difference Proportion Cumulative PRINl 4.88770 3.83441 0.610962 0.61096 PRIN2 1.05329 0.31837 0.131661 0.74262 PRIN3 0.73492 0.19032 0.091865 0.83449 PRIN4 0.54460 0.05012 0.068075 0.90256 PRIN5 0.49448 0.31125 0.061811 0.96437 PRIN6 0.18324 0.10831 0.022904 0.98728 PRIN7 0.07493 0.04808 0.009366 0.99664 PRIN8 0.02685 0.003357 1.00000 Table 14b Eigenvectors of the phloem-cambium IT values PC analysis. PRINl PRIN2 PRIN3 PRIN4 Aug. 0.357647 0.230392 0.168475 -.613937 Sept. 0.253067 0.494536 0.601222 0.314689 Oct. 0.406569 -.344539 -.045936 0.018165 Nov. 0.336693 -.l69991 0.321099 0.454405 Dec. 0.381768 0.303301 -.084414 -.353177 Jan. 0.394411 -.341102 -.104953 -.145618 Feb. 0.245992 0.513427 -.681882 0.360683 Mar. 0.409349 -.287752 -.147749 0.202994 PRIN5 PRIN6 PRIN7 PRIN8 Aug. -.271287 0.535594 -.222668 -.060559 Sept. 0.469825 -.003449 -.088424 0.048247 Oct. 0.305682 0.062341 0.195960 -.760149 NOV. -.665658 0.067383 0.310015 0.066346 Dec. -.146546 -.752495 0.206547 -.004019 Jan. 0.382587 0.163233 0.358355 0.628039 Feb. -.021369 0.268138 0.090891 -.014488 _y@r. -.028639 -.199773 -.792937 0.130832 71 ._ 638. E 8 82.3.6.5”? .838 um 93 3.5 2: so .526 95on 28 98 856:... .58 2 .8526 .83 a»: .8 can: EsBanéooEa .5.— 888 on .3 98:59. .: .uE ~25: 1P Ito—I No 1! db up w. .. 2. m 9 3nd. M. H... 0. <0 ZNRIJ 72 Proceeding from negative to positive values of PCI, the cultivar and seedling means decrease in hardiness. The majority of separation of PCI was due to differences among the selections for minimum hardiness values in October, December, January and March (Table 14b). The sweet cherry and Hungarian selections are at the positive end of PCI with positive values above 0, while the selections which are derived from the colder regions of Germany and Moldavia have values less than 0. PC2 is difficult to interpret because there are no clusters. The variation of PC2 is due to differences in September and February (Table 14b). 'Vegetative Bud PC Analysis The first two PCs of the vegetative bud analysis account for 63% and 16% of the variance, respectively (Table 15a). As with the flower buds, ‘Meteor’ (R) and ‘Emperor Francis’ (F) are situated at the extremes of PCI and all five months contributed to the variation along PCl (Fig. 12). Selections at the positive end of PCl had vegetative buds that were more cold susceptible than selections at the negative end of PC1. However, unlike the flower bud analysis, but more like the other vegetative tissue Table 153 Eigenvalues of the vegetative bud IT values correlation matrix. Eigenvalue Difference Proportion Cumulative PRINl .15622 .36358 0.631244 .63124 PRIN2 .79264 .30303 0.158529 .78977 PRIN3 .48962 .08788 0.097923 .88770 PRIN4 .40173 .24195 0.080347 .96804 PRIN5 .15978 0.031957 .00000 Table 15b Eigenvectors of the vegetative bud IT values PC analysis. PRINl PRIN2 PRIN3 PRIN4 PRIN5 Nov. Dec. Jan. Feb. Mar. .449211 .431583 .370841 .455488 .516675 .351369 .321474 .749036 .406841 .215915 .129022 .777072 .094832 .581230 .180806 .792452 .269092 .538268 .098942 .009360 .173559 .184767 .051929 .528576 .808486 74 ._ 033. E 8 22652.3(. .838 0a 0...: 6E 22 so .926 258» 25 23 8.2.66 59. 2 .8E26 383... 95 Co 25m: 25 2656»? 8.. 888. Um .6 2828a .m— .9..— ~25: q» 0 v '- db in «I- (‘l I a o :2- an. D. Avn.°n m. C. .0 ZNRId 7S analyses, the heaviest loading of PC2 is due to differences in selections in January (Table 15b). The vegetative tissues of most cultivars and seedlings evaluated were at their maximum cold resistance in January (Table 7). Cortical Tissue PC Analysis The first three PCs of the cortical tissue analysis account for 50%, 19% and 13% of the variance, respectively (Table 16a). The analysis of the cortical tissue illustrates clear division between northern origin cultivars and seedlings and southern origin cultivars and seedlings (Fig. 13). Proceeding from negative to positive values of PCI the means of the selections decrease in hardiness. The majority of separation of PCl was due to differences among selections for minimum hardiness in October, January and February (Table 16b). Northern origin cultivars and seedlings are at the negative end of PCI and southern origin cultivars and seedlings are at the positive end. November and December are seperated out in eigenvector two and three (Table 16b). The separation can be accounted for by the dramatic deacclimation of the cortical tissue between November and December (Tables 5 and 6). Table 16a Eigenvalues of the cortical tissue IT values correlation matrix. Eigenvalue Difference Proportion Cumulative PRINl 4.01861 2.51877 0.502326 0.50233 PRIN2 1.49984 0.48705 0.187479 0.68981 PRIN3 1.01279 0.31687 0.126599 0.81640 PRIN4 0.69592 0.34059 0.086990 0.90339 PRIN5 0.35534 0.15153 0.044417 0.94781 PRIN6 0.20381 0.02704 0.025476 0.97329 PRIN7 0.17677 0.13984 0.022096 0.99538 PRIN8 0.03693 0.004616 1.00000 Table 16b Eigenvectors of the cortical tissue IT values PC analysis. PRINl PRIN2 PRIN3 PRIN4 Aug. 0.376559 0.310595 -.143701 -.312102 Sept. 0.277992 0.486114 0.377021 -.407402 Oct. 0.430758 -.053072 0.318482 -.058458 Nov. 0.045575 0.613724 0.119136 0.767503 Dec. 0.319648 0.011549 -.730245 0.102153 Jan. 0.431802 -.280402 0.127365 0.297891 Feb. 0.425501 0.047142 -.284489 -.032803 Mar. 0.351590 -.454719 0.301941 0.209493 PRIN5 PRIN6 PRIN7 PRIN8 Aug. 0.686121 -.203020 -.340815 -.124670 Sept. -.150982 0.482357 0.332899 0.109577 Oct. -.268166 -.743505 0.163901 0.234960 Nov. 0.028324 -.032410 -.097099 0.082042 Dec. 0.026161 0.074536 0.444599 0.387416 Jan. 0.158946 0.119431 0.386771 -.663959 Feb. -.600514 0.143439 -.541392 -.245806 Mar. 0.217168 0.363139 -.310430 0.509079 77 .H magma cw mm maone>m cam mooHumH>ounn¢ .moxm om mourn umueu may no >uuono oasoum oco cam mofluuozo noon NH .wofluuono pomzm 03» mo osmmeu Hmoepuoo How mouoom om uo maceuflmom .MH .owh 2.2-2:- 0:. Es. m.) . e t. .: 2E 78 DISCUSS ION The observation that LTEs occur in midwinter in the xylem of P. avium and P. cerasus is similar to that reported previously (Burke and Stushnoff 1979, Rajashekar and Burke 1978, Quamme et a1. 1982) . In these studies the twig LTE's in January were reported to occur between -37 and. -—41°C for P. cerasus. Quamme et al. (1982) indicated lower temperatures for LTE's in P. avium then were found in our investigations. They reported LTE’s occurrences, on average at -40.5°C for the taxon. Our values were up to 18.50C higher, but were more representative of the average minimum temperature isotherm at the northern geographic range of P. avi um, -29OC (Quamme et al. 1982) . The pattern of LTE occurrence was similar to that indicated by Wisniewski (1995) increasing in the fall, reaching a maximum in midwinter, and decreasing in the spring, LTE’s absent in Allgust, September and October. Cells need to become thickened and rigid before they show supercooling Chalfacteristics. Non—rigid cells do not supercool (Raj ashekar and Burke 1996) . In acclimation, cell walls typically become thicker, increase their tensile strength and decrease their pore sizes (Rajashekar and Lafta 1996) 79 In August, September and October acclimation was not far enough advanced to affect these changes. The absence of LTEs in August, September and October is also similar to the findings of Ketchie and Kammereck (1987) in their work with Malus. Perhaps exotherms were present but our equipment was unable to measure them during these three months. Current research indicates the cell wall and. plasma membrane are critical to the occurrence of deep supercooling (Wisniewski 1995). Wisniewski et a1. (1993) noted contact of the plasma membrane with the cell wall was essential for maximum supercooling to occur. In the early part of the season perhaps this contact is not far enough progressed. The type, amount and degree of cross-linking of pectin within the pit membrane has also been proven important in the occurrence of deep supercooling and offers a plausible explanation for seasonal shifts in occurrence. Perhaps there is not enough cross-linking established in Auglist, September and October for supercooling to occur. Gerletic variability in sour cherry, in this pectin-mediated reghllation of the pit membrane, is very likely since there was variability in the size and occurrence of the LTE within the germplasm. 80 The presence of a linear relationship between xylem browning and supercooling corresponds with earlier work by Hong and Sucoff (1982) . The complexity of the LTE, in terms of tissue correlation during acclimation, is similar to that indicated by Wisniewski (1995) . The complexity of shape (Ashworth 1993) and number of exotherms (Ketchie and Kammerick 1987) is accentuated during acclimation. The correlation of the LTE with the phloem-cambium, in November, conforms with the findings of Ketchie and Kammerick (1987) where correlation of the LTE and xylem injury did not occur until maximum cold resistance had been obtained. Wisniewski (1995) reported that the xylem in some species does not freeze in a homogeneous manner but instead responds as a heterogeneous population of cells that freeze over a wide temperature range. From our observations, this 18 true for P. cerasus. During acclimation, represented by November, browning injury preceded the LTE. During November, the DTA equipment was measuring some amount of extracellular freezing or non-homogeneous freezing which correlated best to the phloem-cambium. During maximum aLCClimation, December to February, browning occurred after the occurrence of the LTE and generally within 50C, this is ) (3. (I) '(1 ’41 V 81 in agreement with earlier papers (Quamme et al. 1972, George et a1. 1974 and Quamme 1976). Although average annual minimum temperature isotherms are often used to establish northern limits of distribution, the frequency of low temperature extremes and their distribution with respect to the acclimation and deacclimation cycles may be more critical in determining plant distribution. Significant differences in hardiness of the phloem-cambium existed between P. cerasus seedlings of northern origin compared to, the industry standard, ‘Montmorency’, in late fall and early spring. These decreases in hardiness during acclimation and deacclimation may also affect plant distribution indirectly by increasing For example, cold damaged tissue sus ceptibility to disease . predisposes the tree to infection from various pathogens and lIlsects. Diseases such as Pseudomonas sp. are presumed related to lack of hardiness in the fall. Increased phloem—— can:IkDium hardiness during this critical period might result in reduced susceptibility. The phloem-cambium is also the tissue which expresses the most rapid deacclimation response and thus maybe the most critical tissue determining geographic distribution and c . OmIIIe rc1a1 range . 82 The observation that supercooling occurs in the :midwinter xylem of the majority of sour cherries studied and ‘that significant differences in LTE occurrence could be found between southern and northern types implies that supercooling plays a critical part in survival and distribution of sour cherry. However, our studies were only investigated temperature responses and our factors besides temperature such as day length would influence distribution. The observation that supercooling limits P. cerasus germplasm distribution is similar to that made for other woody species in which supercooling also exists (Quamme 1976, George et al. 1974, Sakai 1978). DTA is useful for measuring the hardiness of the xylem in December to March in sour cherry, but it does not provide a measure of vegetative bud or phloem—cambium hardiness, bOth of which are considered to be more critical to survival than xylem hardiness (Quamme 1991) . For this reason and because of the strong correlation of xylem ITs to the LTEs, and the consistency of. browning score occurrence versus LTE occurrence, xylem ITs were found to be the preferred method for evaluating the cold resistance of sour cherry in Dec—ember to March. Because of the strong correlation of phlCDem-cambium ITs to the LTE in November and again the 83 consistency of occurrence of the phloem browning scores versus the LTE’s, phloem-cambium ITs were preferred for November. Cortical IT’S and vegetative bud IT's were also of interest depiciting times of susceptible to winter injury that could not be correlated to the stem LTE’s. The survival curves developed using the PLOT-IT Non— Linear regression equation can be used to express hardiness in absolute terms. It is possible to relate absolute hardiness indices to environmental survival (Quamme 1978). Future studies in this area should focus on this characterization. In the development of hardier P. cerasus cultivars, our study suggests that it is possible to increase hardiness of the xylem and phloem-cambium through hybridization. Significant gains in hardiness over the industry standard ‘Montmorency’, could be made in late fall and early spring. The significance of phloem-cambium susceptibility over the entire dormant period and the opportunity to improve the resistance of this tissue has not been fully investigated. Direct gains in hardiness and indirect gains by way of disease resistance may be forthcoming. 84 One conjecture regarding P. cerasus origin is that there were two presumed progenitor species, P. avium and P. fruticosa. Following on this conjecture, we hypothesize that in an analysis of a P. cerasus germplasm one would expect to find a full range of sour cherries with cold resistance levels similar to the cold tender, P. avium through to the cold tolerant, P. fruticosa. Those sour cherries exhibiting cold resistance similar to P. avium would predominantly have their origin in southern regions, whereas those similar to P. fruticosa, northern regions. Also, since ‘Montmorency’ is the standard cultivar in the P. cerasus industry there would also be selections hardier than ‘Montmorency' and selections less hardy than ‘Montmorency.’ This separation is especially clear in the vegetative bud PC analyses where ‘Montmorency’ is positioned in the middle of PCl and PC2 and there are less hardy and hardier selections on each side. The PC analysis of the phloem—cambium, cortical tissue and xylem supports the two presumed progenitor species and southern, northern type hypothesis by the loading of minimum survival temperatures during acclimation and midwinter. During these periods there are clear separations between northern and southern types along the first PC axis. The PC 85 analysis of the vegetative buds supports this hypothesis by the loading of minimum survival temperatures during midwinter. Midwinter or maximum acclimation contributed the most to the clear separation of the vegetative buds between seedlings more hardy than ‘Montmorency’ and seedlings less hardy than ‘Montmorency,’ along the second PC axis. The results of the phloem-cambium PC analysis are especially clear in support of the two presumed progenitor species hypothesis and are of particular significance in view of the phloem—cambium being the most susceptible tissue to freezing injury in the acclimation period and its correlation to the LTE’s during this period. 86 REFERENCES: Artlip, S., Callahan, Ang, Bassett, C.L., and‘Wisniewski, M.E. 1997. Seasonal expression of a dehydrin gene in sibling deciduous and evergreen genotypes of peach (Prunus persica [L.] Batsch). Plant Molecular Biology 33: 61-70. .Arora, R.,'Wisniewski, M., Scorza, Rm 1992. Cold acclimation in genetically related (sibling) deciduous and evergreen peach. Plant Physiol. 99: 1562-1568. Ashworth, E.N. 1993. Deep supercooling in woody plant tissues. Pages 204-213 in Advances in cold hardiness. P.H. Li and L. Christersson, eds. CRS Press, Boca Raton, FL. Beaver, JMA. and Iezzoni, AiF. 1993. Allozyme inheritance in tetraploid sour cherry (Prunus cerasus L.). J. Amer. Soc. Hort. Sci. 118: 873-877. lBurke, MWJ., and C. Stushnoff. 1979. Frost Hardiness: A (iiscussion of possible molecular causes of injury with Pairticular reference to deep supercooling of water. Pages 1597-225 In H. Mussell and R.C. Staples, eds. Stress PhY‘siology in Crop Plants. John Wiley & Sons, New York, NY. 87 Cain, DJW. and Andersen, R.L. 1976. Sampling procedures for minimizing non-genetic wood hardiness variation in Peach. J. Amer. Soc. Hort. Sci. 101(6): 668-671. Cummins, J.N. and.A1dwinkle, H.S. 1983. Apple rootstock breeding, pp. 294-394. In J.Janick (ed.) Plant breeding reviews. Avi Publishing Co., Westport, Conn. Fbgle, wa. 1975. Cherries. Pages 348-366 In J. Janick and J.N. Moore, eds. Advances in Fruit Breeding. Purdue University Press, West Lafayette, Indiana. George, M;F., Burke, M.J., Pellet, H.M. and Johnson,.A.G. 1974. Low temperature exotherms and woody plant distribution. HortScience 9: 519—522. Gusta, L;V., Tyler, M.J., and Chen, T.H. 1983. Deep undercooling in woody taxa growing near the -4U%Iisotherm. Plant Physiol. 72: 122—128. (hista, LJV., O’Conner, B.J. and MacHutcheon, M.G. 1997. the Serlection of superior winter—hardy genotypes using a Prkblonged freeze test. Can. J. plant Sci. 77: 15-21. 88 Hilligy KL 1988. A multivariate analysis of a sour cherry germplasm collection. Ms Thesis, Michigan state university, East Lansing. Hillig, KLWZ and Iezzoni, AHF. 1988. Multivariate analysis of a sour cherry germplasm collection. J. Amer. Soc. Hort. Sci . 113: 928-934 . Kong, 8., and Sucoff, E. 1982. Rapid increase in deep supercooling of xylem parenchyma. Physiol. 69:697—700. Iezzoni, AWF. and Pritts, M.P. 1991. Applications of principal component analysis to horticultural research. HortScience 26(4): 334-338. Ketchie, D.O. and Kammereck, R. 1987. Seasonal variation of colci resistance in Malus woody tissue as determined by - 3,. ‘F .1}- «>— «>- O .1 OCw 104 # coo— ENOAHBPQ : con— ? coon -1s;:ezl- con" 1‘ 1.!" it". 51.03.15.111]. .3!‘.} . lt.|lln)ul.0\v' it l~ 105 U— N- can .3- .2- an 868 82: metro—6 83 6.6: 656552 .6 32 .9252. Se 2.65 «ab .= .wE ovgflofioh — . . . . . . . . . . . . . .l . . GI WfrrWWCCCCCCCCCZ-GIINVIIO.... cc z I. 6 8 L L .L 9 9 9 .6 L f. 6 L C. 7. r) C. r. C. bhhhphlh» nth phhhbrhbypphbrhhrbkktbLnth»b»~thbP->bhbb~bbP.bLup.phhh.wnthFrbrtPghFlplf_blrh_>»p._»”_—>.b.—-.PF_LL»»-H_»>h ~— 4<4M444<4Jj4—d~q<4u444-.{Mdqud4-144—444-q~qqq_«a~11614ij‘44<4<<4414~a4111114<44444~6—Afiijei d34<1<1144l41d44fi41 44<<1 o :8“ $002 :82 yooon it p l: .- i -3 -f§;.,i.s.li: .-!..-Lr can" rtli tall.) 106 Table I] Comparison of a November collection of 15 cherry mitivar/sccdlings evaluated by exothenn analysis conducted at -5°C per hour and by visual injury of floral bud cross-sections frozen to -3 5°C at -5 °C per hour. Cultivar! Seedling Flower Bud Low Temperature Exotiienn Flower Bud LT“ Prim avium Empemr Francis - 41.0 a Schmidt - 4 2.0 a Prunus cerasus Csengodi Csokas 42.5 b ' 4 6.0 a (Spaniole X Crisana) —- 47.0 a Pandy114 41.0 b 43.03 Pitic de Iasi 4 6.5 c 48.5 a Wolynska X Sumadinka 42.5 b 46.0 a EM X Sumadinka - 8.0 a 47.5 a Oblacinska 47.5 c -20.0 a Cigany Meggy 41.5 b 43.3 a Montmorency 46.0 c 48.0 a Fructbare von Michurin 47.5 c 48.0 a English Morello -- 47.0 a Meteor -— -23.0 a P. fruticosa 6864 48.5 c 45.0 a 'Means in columns. based on three sub-samples. be: four replicates, followed by different letters are significantly different (P=O.w) according to Fisher's least significant difference tea. 107 Table III Comparison of a December collection of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -5°C per hour and by visual injury of floral bud cross-sections frozen to -50°C at -5 °C per hour. Cultivar! Seedling Flower Bud Low Turpentine Flower Bud LT... Exotherrn anrrs Mum Emperor Francis — 44.0 a Schmidt _ 4 5.0 a Prurrus census Csengodi Csokas -125 b' -155 b (Spaniole x Crisana) —- 48.5 b Pandy114 42.5 b 45.0 a Pitic de Iasi -- 4 8.5 b Wolynska X Sumadinka 47.5 c 49.5 b EM X Sumadinka 44.0 b 46.5 a Oblacinska 47.5 c -20.5 b Cigany Meggy 41.0 a 44.0 a Montmorency -17.o c .200 b Fructbarevon Michurin 45.0 c 48.0 b English Morello - -1e.o b m _. -23.o b P. Most see-1 4 4.0 b 47.0 b ‘Means in columns, based on three sub-samples, per four replicates, followed by different letters are significantly different (P=0.t5) according to F isher's least significant difference test. 108 Table IV Comparison of a January collection of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -5°C per hour and by visual injury of floral bud cross-sections frozen to -50°C at -5 °C per hour. Cultivar! Seedling Flower Bud Low Temperature Flower Bud LT“ Exotherrn Pumas avium Emperor Francis — 48.0 a Schmidt -- 4 7.0 a Pnrnm cerasus Csengodi cm 48.0 a‘ -1 7.5 a (Spaniole X Cflsam) - 49.0 a Pandy114 48.0 a 49.0 a Pitic de Iasi — 48.5 a Wolynska X Sumadinka — 48.0 a EM X Sumadinka 47.0 a 49.5 a Oblacinska — -20.0 a Cigany Meggy 45.5 a 46.0 a Montmorency 49.0 a -22.0 a Fructbare von Michurin - 48.0 a English Morello —— -20.0 a Meteor — -24.0 a P. fruflcosa 688-1 -- 48.0 a 'Means in columns. based on three sub-sampbs. per four replicates, followed bytdifferent letters are significantly different (P=0.w) awarding to Fisher's least significant difference test. 109 Table V Comparison of a Febniary collection of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -5°C per hour and by visual injury of floral bud cross-sections frozen to 45°C at -5 °C per hour. Cultivar! Seedling Flower Bud LowTerrperature Flower Bud LT“ Exotherrn Prim avium Emperor Francis 4 7.0 a‘ -20.0 a Schmidt _. -20.0 a Prm cerasus Csengodi Csokas -22.0 a -23.5 b (SpanioleXCrisana) 4 7.0 a 49.0 a Pandy114 -zo.o a -23.o b Piticdelasi 49.0 a -22.0 b Wolynska x Sumadinka 48.0 a -205 a EM X Sumadinka 47.5 a -20.5 a Oblacinska -20.0 a -22.0 b CiganyMeggy 48.5 a -20.5 a Montmorency -22.0 a -25.0 c Fructbarevon Michurin -20.0 a -24.0 b English Morello -22.0 a -25.0 c Meteor .. -26.0 c P. turbos: 6864 -- ~27.0 c 'Means in columns, based on three sub-samples. per four replicates. followed by different letters are significantly different (P=0.05) according to F isher's hast significant difference test. 110 Table VI Comparison of a March collection of 15 cherry cultivar/seedlings evaluated by exotherm analysis conducted at -5°C per hour and by visual injury of floral bud cross-sections frozen to 40°C at -5 °C per hour. Cultivar! Seedling Flower Bud Low Temperature Flower Bud LT“ Exotherrn Prurmsevm Emperor Francis 43.0 a' 40.0 a Schmidt — 48.0 a Prunus census CsengodiCsokas 48.0 b 48.0 a (Spaniolex Crisana) -20.0 c 45.0 a Pandy114 45.5 b 46.5 a Piticde Iasi -22.5 c -20.5 a Wolynska X Sumadinka -2f .0 c ~20.5 e EM X Sumadinka -22.0 c 49.5 a Oblacinska -20.0 c ~20.0 a Cigany Meggy -21.0 c 45.5 a Montmorency 49.0 c .200 a Fructbare von Michurin 48.0 b 48.0 a English Morello -22.0 c 48.0 a Meteor — -27.0 a P. Masons-1 —- 49.0 a ’Means in columns. based on three sub-samples. per four replicates, followed by different letters are significantly different (P=0.(5) according to Fisher's least significant difference test. lll cultivars and seedlings (Table IV), in February 12 cultivars and seedlings (Table V) and in March 12 cultivars and seedlings (Table VI). In November there were significant differences in the flower bud LTE’s but not in the LTsovalues (Table II). In December there were significant differences in both the flower bud LTE's and the LTm,values (Table III). However, the LTE’s and the LTygvalues gave different results. With the LTm3values the majority of the northern types were significantly different from the southern types except for ‘English Morello’ X Sumadinka. In contrast, LTE values separated the northern types into two significantly different hardiness classes (Table III). The flower bud LTE’s expressed more differences between cultivars and seedlings than the LTmyvalues. The LTE's were significant at p50.01 in December (Appendix M), whereas, the LTmyvalues were signicant at p50.05 (Appendix L). In January there were no significant differences found in either the flower bud LTE’s or the LTM3values (Table IV). In February, only the Lngvalues indicated differences between cultivars and seedlings (Table V). In March, however, only the LTE's showed cultivar and seedling differences (Table VI). 112 There is a relationship between bud volume and geographic origin (Table VII). The majority of cultivars and seedlings of northern type designations had significantly smaller buds (d’s), except for ‘English Morrello' X Sumadinka, Pitic de Iasi and Spaniole X Crisana. The majority of significantly larger buds were from southern origin cultivars and seedlings (a, b, c's), with Csengodi Csokras o.p. having significantly larger buds than any other cultivar or seedling evaluated (Table VII). Southern types ranged in size from 0.76 mm? to 0.32 nmP with an average size of 0.43 HmP. Northern types ranged from 0.53 nmP to 0.19 nm3 with an average size of 0.34 nmN. No relationship existed between geographic origin and number of bud scales or number of floret per bud (Table VII). There was a significant correlation between exotherm occurrence and bud volume (Fig III). The correlation coefficient (r) was 0.950, significant at p50.05. In January, when exotherm occurrence was the poorest, exotherms were generally present only in cultivars and seedlings with larger bud volumes (Tables IV and VII). On all dates evaluated, the correlation of the flower buds LTayvalues to the LTE's were significant at p50.05 (Table VIII). There was a significant linear relationship 113 Table VII Names, type designations, bud volumes, number of scales per bud and number of flowers per bud for the cultivars and seedlings evaluated. Name Type Vblume of thmflmur Number Designation Bud (m3) of Bud of Scales Flowere/ Bud Prunus avium Emperor Francis Southern. -—— - «-- Schmidt Southern .36 c‘ 6 3 Prunus cerasus Cigany Meggy Southern .37 c 6 3 o.p. Csengodi Csokras Southern .76 a 7 3 o.p. English Morello Northern —-- - --- English Morello Northern .53 b 6 4 X Sumadinka Fructbare von Northern .27 d 7 3 Michurin Meteor Northern .28 d 6 3 Montmorency Northern. .23 d 6 3 Oblacinska o.p. Southern .37 c 6 3 Pandy 114 o.p. Southern .43 b 6 3 Pitic de Iasi Northern .46 b 6 3 o.p. Spaniole X Southern .32 c 6 3 Crisana o.p. Wolynska X Northern .40 c 7 3 Sumadinka .Prunus fruticosa.INorthern .19 d 6 3 586-1 :Means in columns, based on five replicates, followed by different letters are significantly different (P=0.05) according to Fisher’s least significant difference test. 114 on. 26:28 .520 2:2» 28 58 48.3 8.6 BEE—a 82 £982 2 82 380.62 a8 mos—a.» :85 PS 28 ca...— 83 330: cc cosmewoa 80:3 .5 .uE Gov 8.82 ME...— 2. OT n- o 3.33 r. . sass . 5.83 n » 0 oz- 91- or- s- 00) "madam MINI SZ' 115 Table VIII Correlation coefficient values and significance for flower buds LTu,values and LTE's calculated for different time periods. Time Sample size :- r Overall 48 0.4684 0.6844" Nov.- Feb. 36 0.7254 0.8517" Mar. 12 0.4706 0.6860' NS"""MNonsignificant or significant at P=0.05 or 0.01, respectively. 116 between oxidative browning, measured as LTW, injury to 50% of all flower, and the flower bud LTE for all dates evaluated. The lower the appearance of the LTw, the lower the LTE. The correlation coefficient (r) for the flower bud LTM,values and LTE's, over all months was 0.6844, significant at p30.01 (Fig. IV). For the acclimation and midwinter period, represented by November to February collections, the correlation of the LTE’s to the Lngvalues was the strongest, with a correlation coefficient (r) for this period of 0.8517, significant at p50.01 (Fig. V). The average LTw occurred before the LTE and within 2.EPC (Tables II, III, IV, V). In deacclimation, represented by the March collection, (r) was .6860, significant at p50.05 (Fig. VI). The LTw began within 2.0T» on average, of the LTE, and in 11 of 12 cultivars and seedlings preceded the LTE (Table VI). Vegetative bud injury was not correlated to the flower bud LTEs (data not shown). Response Differences in Tissue The flower buds were more susceptible to low temperatures than all vegetative tissues, in almost all taxa, on all evaluation dates (Mathers et al. l99xa). The scone—om Page 258» can 38 483 .26 BEER SE 5.82 9 32 398962 5c 82? :86 mph 98 op: 33 .538 cc 5523.. 30:3 .>— .uE Gov 3302 man— 117 nu- Till I. .Il‘ .Irte. lllrro ( SI’ Ol' S" (De) mammalian-1 OZ' $Z‘ sconce—om Pogo 959% can 38 .383 3.5 3588 3% bacon 8 82 aBEoSZ he £82: P: 93 can .03 .35: co commas.— Socfi .> .3..— 118 Ge 282 a: a. 8. n _. 2. n. o o “I :3 o n a $33 . x Rag u a 2. 2. on. r- l.lrfnlllrirll.lloi lllrrll 2..llr.. it? “NI (Do) modular Misti] 119 sconce—8 Ease v52» 98 38 48.5 3.6 3588 32 €32 he «:8:— mhq 98 3.5 can assoc .8 eBay—woe 30:5 ._> .uE an a 2 S Gov n50: ”NF—.4 2 n— 2 c— $ «— \. v— w— «— on . -. r i-l!l-ill...l.l-l-.lr.:-..... .r N" be) madam :03:de 120 LTm of all flower buds occurred well before the detection of the LTE in the stem (Mathers et al. 199xa). The comparative hardiness and extent of temperatures over which injury occurred, evaluated over all months, are presented in Table IX. Significant differences occurred, in the LTa3values, among the 15 cherry selections (Table IX). A representative array of selections is presented in (Fig. VII). The relationship appeared to be dependent on area of origin. The GLM of the flower bud LTE showed significant differences due to cultivars and seedlings and month (Appendix N). The interaction of cultivars and seedlings X month was also significant (Appendix N). The GLM of the LTm data showed significant differences due to cultivars and seedlings and month. The interaction of cultivars and seedlings X month was also significant (Appendix 0). The comparison of means of LTygvalues for December (Table III) (Appendix L) and March (Table VI) (Appendix L) indicated significant differences among cultivars and seedlings. Hardiness of Prunus cerasus Open-pollinated and Hybrid Seedlings The floral LTygvalues of P. cerasus ‘Montmorency’ and open-pollinated and hybrid seedlings, summed over all months l 2 1 Table [X Comparison of 15 cherry cultivar/seedlings evaluated over 5 months by exotherm analysis conducted at - 5°C per hour and by visual injury of floral bud cross-sections frozen to a maximum of -50°C at -5 °C per hour. Cultivar! Seedling Flower Bud Low Temperature Flower Bud LT” Exotherrn Prunus avium Emperor Francis — 45.0 a Schmidt — 46.0 a Prunus cerasus Csengodi Csokas 48.0 b‘ 48.1 b (Spaniole X Crisana) — 47.5 b Pandy114 47.0a 47.5 b Pitic de Iasi - "9'5 b Wolynska X Sumadinka - 4 9.0 b EM X Sumadinka 4 7.0 a 4 9.0 b Oblacinska - -20.5 c Cigany Meggy 4 7.0 a 46.0 a Montmorency -20.0 b -21.0 c Fructbare von Michurin — 49.0 b English Morello —- -20.0 c Meteor — -25.0 c P. Mose m4 — 49.0 b ' Means in columns. based on three sub-samples. per four replicates. followed by different letters are significantly different (P=0.C5) according to Fisher‘s least significant difference test. 122 as as: s 82 85082 as. Bass .58 8.5 as s 88225 585 x 8603.225 .5 es :2 5.32 as >02 l mu. \Ivn. l. "NI \ - . :8. £533 55 x \ a: 80:: .._\ x a 1:. \\ \ \ 0 ee 0 .000 \ O 0000000000 0000000000 x .......... T2. 0 ”.5 \ .. \ eeeee \x 1' t \ \ 0 . eeee _ e \ \ O . O... s \ u 0 .\ 00000 t \ ... 0 x ..... .\ eeeee O eeee ) .0 ace 0 . O I 00 C .0 IV~I 0 e... O 000. O 0000 O 0000 O .000 O 0 see. . .‘O. . eeee . -~.. 0... eeeee .ll ...... INA—1 eeeeee I! Ale... ...... eeeeeee , .‘nv. eeeeeee a. eeeeeee -. Oral-n. eeeeee eeeee :4... 00 e no a e 0000000000000 I it .0. 0000000000000000 r 0 I t. . ......... \ e— le!!! eeeeeeeee ... , I eeeeee eeee _ . ..r.. e 0 00000000000 00 . . .1 i. anOHeuonononoooo I!!! l. 0.... [IL Denial? h!!! . I'll-Ill V..- I) 1. (Do) 091.1 123 evaluated are presented in Fig. VIII and Table IX. The flower buds of the industry standard ‘Montmorency’ were hardier than all P. cerasus seedlings examined, summed over all evaluation dates (Table IX and Fig VIII). P. cerasus ‘Meteor’ was similar in floral hardiness to ‘Montmorency’, ‘English Morello’ and Oblacinska, summed over all evaluation dates (Table IX). Relationship of Ambient Temperatures The ambient temperatures to which floral tissue was exposed are presented in (Fig. IX). Warm temperatures in the deacclimation period reduced the capacity of flower buds to avoid freezing. This was particularly evident in the southern types (Fig. IX). P. cerasus ‘Meteor’, Pitic de Iasi, Wolynska X Sumadinka, Oblacinska and ‘Montmorency’ all showed significantly later deacclimation in March (Table VI) than the P. avium cultivar ‘Emperor Francis’. Flower Bud PC Analysis The first two PCs of the analysis of the flower bud LTw data accounts for 77% and 12% of the variance, respectively (Table Xa). 124 .33 £032 9 32 eBEoSZ £83 nouns—«>0 =5 .66 @5808 .3488 .m 96 6:065:62. afieaecfi E was? or: 25 3.6: .6 cam—3800 .E> E..— 5.5: new 658.552 . x. \ >02 1 n". \ \ £5.58 x is? \ \ \\ 363% x 2m \ i ”No .I ”NI 3 ages \ (Do) 05.1."! rafil Ike—I r. ._s_ 6553.20 5 32. £222 2 camp 5nEo>oZ E9: mo:_w> 8.5 can 526: pew macaw—8E3 EmEE< .x. .94.. 125 Cog awnmmwmmmmmzwmmmmwmwmmmmmmmmmmw 332?... a ems ammmm mm m wmm m wmw w rtlxwtxxixnxwxfi.xx$4£ttxiEx“xxxxxxvxxilxrlxxxrxfixfirxtfiifi:xrixxlxxrx“infiltrate:Ext. 8 82021 I l H .mm- §§+ x H... n .. 9.5323 x 5m IOI m m m m n -. 8- e a isscflxl m n r n. ma. EscoxsgcamIXI « l. . #9. m 5558+ . m £88m§onEm .- 0:38:35 . .22anth 2.5 E2555. l0! . 126 Table Xa Eigenvalues of the floral tissue LTw correlation matrix. Eigenvalue Difference Proportion Cumulative PRINl 3.85425 3.23637 0.770851 0.77085 PRIN2 0.61789 0.28150 0.123577 0.89443 PRIN3 0.33638 0.22502 0.067277 0.96170 PRIN4 0.11137 0.03126 0.022273 0.98398 PRIN5 0.08011 0.016022 1.00000 Table Xb Eigenvectors of the flower bud LTw PC analysis. PRINl PRIN2 PRIN3 PRIN4 PRIN5 Nov 0.472171 .341497 .235063 .225191 .744626 Dec 0.473321 .342588 .087115 .545325 .594669 Jan 0.445773 .033004 .801746 .383521 .101549 Feb 0.365197 .874308 .101699 .299484 .046724 Mar 0.470041 .022567 .532933 .644306 .281789 127 Data from all five months, November, 1990 to March, 1991, contributed to the variation along PCl (Table Xb). Selections at the positive end of PCl had flower buds that were more cold susceptible than selections at the negative end of PCl (Fig. X). ‘Meteor’ (R) and ‘Emperor Francis’ (F) are situated at the extremes of PC1, representing their divergence in floral hardiness as in Table IV. ‘Meteor’ (R) is an outlier, with the closest cultivar being its maternal parent ‘Montmorency’ (M). Proceeding from negative to positive values of PC2 the cultivars and seedlings decrease in hardiness in February. February as indicated Table V represents the time of maximum floral cold resistance for all cultivars and seedlings evaluated. P. fruticosa (P) and Spaniole X Crisana (C) are situated at the extremes of PC2, representing their differences in magnitude of floral midwinter hardiness. Seven of the 12 P. cerasus cultivars and seedlings evaluated, and all of the P. avium, are situated between P. fruticosa and Spaniole X Crisana; i.e., between 0 and —1.5 on the PC2 axis. Three of the five P. cerasus seedling exceptions are either from Yugoslavia or are of Yugoslavian parentage, the southern origin seedlings Oblacinksa (O), 128 A Bomb 5 mm Ecua$2£< .85“ um 93 “mac «.5 .6 $5333.36 .523 2 .6 mean assoc 6.“ 668 Um .6 628m .x .3..— ‘F l I? d! (b {D 0 II «p up 0.. a o :2. m 0 find. "*0 >9 :3 5 e _e >2. 3. :— 0. mm 129 from Yugoslavia and Spaniole X Crisana (C), the northern origin seedlings Wolynska X Sumadinka (W) and ‘English Morello’ X Sumadinka (U), with Yugoslavian parentage and Pitic de Iasi (I) from Moldavia. DISCUSSION The observation that floral tissue browning occurred after the LTE is in agreement with earlier papers regarding Prunus flower bud supercooling (Andrews and Proebsting 1987, Quamme et al. 1982). Andrews and Proebsting (1987) reported that the LTw’s occurred anywhere from 0.Tkito 2.8%Zafter the LTE’s. Quamme et al. (1982) observed LTw occurrences EPC after the LTE’s, averaged over seven species and cultivars. Exotherm size and occurrence were similar to that reported by Andrews and Proebsting (1987) for P.avium in February. They reported P. avium, February, LTE’s at -17.39C in 1983 and -19.9W: in 1984, compared to our LTE occurrence at -l7°C, in February. In March their P. avium LTE’s where again similar occurring at -14.5W: in 1983 compared to our -IBWC. Our P. cerasus ‘Montmorency’ LTE occurrences were 129C warmer than the P. cerasus ‘Montmorency’ LTE’s reported by Quamme et al. (1982). They reported ‘Montmorency’ LTE's at -31°C for January versus our 130 -19%3. However our P. cerasus cultivar LTE occurrences in February, representing maximum acclimation, were within 3%: of -25%3. The temperature of -25W: is reported to be the normal extent of supercooling for common commercial Prunus cultivars (Quamme 1978). Minus 259C approximately coincides with the average annual minimum temperature at the northern range of Prunus commercial production (Quamme 1978). Our P.cerasus LTE’s were also within 2.70C, on average, of those reported by Rajashekar and Burke (1978). As with our findings, other researchers have discovered significant correlations between LTw values and LTE’s (Biermann et al. 1979, Quamme 1976, Andrews and Proebsting 1987, Quamme 1995). It has been demonstrated in P. besseyi, P. pennsylvanica (Rajashekar and Burke 1978), P. persica (Quamme 1983) and Vitis riparia (Pierquet et al. 1977) that pre-freezing exposure to temperatures below Wklreduces the size and temperature of occurrence of the LTE in flower buds. As discussed, the temperatures at which the January LTEs occurred in the flower buds of our selections were slightly higher than those found by Quamme et al. (1982), using a similar pre-freezing protocol. Quamme (1983), working with flower buds of P. persica, and Andrews and 131 Proebsting (1987), working with flower buds of P. avium, showed that floral tissue hardiness is particularly influenced by air temperature and water-content at time of collection. It appears that a similar effect occurs with P. cerasus and P. fruticosa. There were some warm temperatures in December that seemed to decrease the temperature of January LTE occurrences in these species. The relationship between ambient temperatures to floral tissue hardiness indicates two critical times for flower bud injury, November and March. November injuries would occur in years when sudden cold temperatures occur without sufficient pre—exposure to freezing temperatures. March injuries would occur in years when warm days were followed by sudden freezing temperatures. This type of injury would be most pronounced in southern types. However, because our findings are based on one year’s observations a subsequent year of testing would be necessary to verify this assumption. The relationship between small bud volumes and lack of LTE occurrence in January is similar to results reported by Kadir and Proebsting (1994a). They found that Prunus species with small buds had low water contents and lacked exotherms in December and January after two days at -7%2. 132 Spring freeze injury could be significantly reduced by the selection of cultivars and seedlings that have delayed deacclimation. P. cerasus ‘Meteor’, Pitic de Iasi, Wolynska X Sumadinka, Oblacinska and ‘Montmorency’ all showed significantly later deacclimation in March (Table 7) than southern types and P. avium cultivars. This delayed deacclimation trait, of these five P. cerasus cultivars and seedlings, makes them advantageous choices in a sour cherry breeding program. Flower buds were more susceptible to low temperatures than all vegetative tissues, in almost all taxa, on all evaluation dates (Appendix V). The floral buds being the most susceptible tissue agrees with the observations of many researchers in various woody taxa (Sakai 1982, Rajashekar and Burke 1978, Quamme 1991). The finding also supports the observation of Quamme et al. (1982) that flower bud hardiness is probably more important to distribution of commercially grown fruits than those grown in home gardens or native fruit grown in the wild. The lack of significant differences found between selections with either LTE evaluations or LTw evaluations, in January, corresponds to findings presented by Quamme (1986) for grapes. Quamme (1986) indicated that cultivars 133 responded differently to preconditioning temperatures and that preconditioning may be used to improve cultivar separation. Exposure of buds to low temperatures for 3 days improved the separation of three grape cultivars in autumn when they were similar in hardiness. In midwinter, however, after they had already been exposed to low environmental temperatures there was no separation improvement (Quamme 1986). We found a similar effect of preconditioning in sour and sweet cherry (Appendix R). In midwinter there was no cultivar separation (Quamme 1986). In months when flower bud LTE evaluations determined more difference between selections than LTw evaluations perhaps the freezing rate of'-EPC per hour was too rapid. Viability tests such as browning scores are more rate dependent than LTE evaluations (Quamme 1991). The freezing rate may have reduced the ability of the browning evaluations to distinguish small differences between selections. Kadir and Proebsting (1994) reported that DTA analysis of P. avium floral buds, separated selections that clearly differed in floral bud hardiness, between December to March. Our findings would be similar for P.cerasus except that our preconditioning minimized cultivar and seedling differences 134 in midwinter (January-February) as previously discussed (Quamme 1991). Because of the strong correlation of flower bud (LTw) values to the flower bud LTEs, and the LTE ability to separate selections in November, December and March, LTE evaluations were the selected method for flower bud hardiness determinations in a sour cherry germplasm. PC analysis of the flower buds LTw values evaluated for eight months depict some gradations between the northern and southern types. However, selective forces seem to have played a more significant role in the hardiness range of sour cherry flower buds than simply geographic origin as illustrated by the cultivar ‘Meteor’ representing the extreme in midwinter cold resistance. This conclusion is supported by our LTE’s in midwinter occurring within 3%:«of the reported average annual minimum temperature for the northern range of Prunus commercial production (Zone 6). In February the floral tissue reached its maximum cold resistance. Similar to reports by Kadir and Proebsting (1994) for P. avium, floral hardiness decreased rapidly after early February. Kadir and Proebsting (1994) showed that in early February, LTE occurrence in P. avium ‘7147-13’ was at -19.8%3. The LTE occurrence then decreased rapidly 135 during late February through March to -l3.9kh Our February evaluations were conducted early, on February 9 and indicated P. avium ‘Emperor Francis’ LTE occurences at -17”C. By the time we conducted our March collections, March 8, the P. avium ‘Emperor Francis’ flower buds had deacclimated to -13%L There was a dramatic increase in hardiness from January to February (Tables IV and V). P. fruticosa increased in hardiness by 9% in this period and some of the P. cerasus selection by 533 and 6WZ. This large increase in hardiness caused the high positive value on PC2 for February (Table Xb). If floral hardiness was related to geographic distribution you would expect to see a clear separation along PC2 for southern and northern types (Fig. X). This clear separation does not occur, indicating as others have reported that flower bud hardiness is related to commercial range versus geographic distribution (Burke and Stushnoff 1979). One obvious requirement in any breeding program for hardiness is to be able to quickly develop hardy parental material with desirable commercial attributes. Our data indicates we have the capability to rapidly carry out artifical freezing and objective flower bud evaluations of 136 viability. DTA evaluations can be used to accelerate breeding for cold tolerance in P. cerasus, but in order for progress to be made, careful attention to procedure is necessary. 137 REFERENCES: Anisko, T. and Lindstrom, O.M; 1996. Survival of water— stressed Rodoodendron subjected to freezing at fast or slow cooling rates. HortScience. 31(3): 357—360. Andrews, P.K., Proebsting, E.L. and Campbell, G.S. 1983. An exotherm sensor for measuring the cold hardiness of deep- supercooled flower buds by differential thermal analysis. HortScience. 18: 77—78. Andrews, P.K. and Proebsting, E.L. 1987. Effects of temperature on the deep supercooling characteristics of dormant and deacclimating sweet cherry flower buds. J. Amer. SOC. Hort. Sci. 112(2): 334-340. Biermann, J., Stushnoff, C, and Burke, M.J. 1979. Differential thermal analysis and freezing injury in cold hardy blueberry flower buds. J. Amer. Soc. Hort. Sci. 104: 444-449. Bittenbender, H.C. and G.S. Howell. 1974. Adaptation of the Spearman-Karber method for estimating the LTw of cold stressed flower buds. J. Amer. Soc. Hort. Sci. 99: 187~190. 138 Burke, M.J., and C. Stushnoff. 1979. Frost Hardiness: A discussion of possible molecular causes of injury with particular reference to deep supercooling of water. Pages 197-225 In H. Mussell and R.C. Staples, eds. Stress Physiology in Crop Plants. John Wiley & Sons, New York, NY. Esau, K. 1977. Anatomy of seed plants. John Wiley & Sons, New York, NY. Flore, J.An and Howell, G.S. 1987. Environmental and physiological factors that influence cold hardiness and frost resistance in perennial crops. Int. Conf. on Agrometeorology, Cesena. Hillig, K. 1988. A multivariate analysis of a sour cherry germplasm collection. Ms Thesis, Michigan state university, East Lansing. Iezzoni, AHF. and Hamilton, R.L. 1985. Differences in spring floral bud development among sour cherry cultivars. HortScience. 20(5):915-916. 139 Kadir, S.A~ and Proebsting, E.L. 1992. Freezing behaviour of Prunus, subgenus Padus, flower buds. J. Amer. Soc. Hort. Sci. 117: 955-960. Kadir, S.A~ and Proebsting, E.L. 1993. Dead Prunus flower- bud primordia retain deep-supercooling properties. HortScience 28(8): 831-832. Kadir, SMA. and Proebsting, E.L. 1994a. Screening sweet cherry selections for dormant floral bud hardiness. HortScience 29(2): 104—106. Kadir, SMA. and Proebsting, E.L. 1994b. Various freezing strategies of flower—bud hardiness in Prunus. J. Amer. Soc. Hort. Sci. 119(3): 584-588. Mathers, H.M., Iezzoni,.A.F. and Howell G.S. 199xa. Supercooling and cold hardiness in sour cherry germplasm: Part 1. Vegetative tissue. Pierquet, P., Stusnoff, C. and Burke, M.J. 1977. Low temperature exotherms in stem and bud tissues of Vitis riparia Michx. J. Amer. Soc. Hort. Sci. 97: 608-613. 140 Quamme, H.Au 1974. An exothermic process involved in freezing injury to flower buds of several Prunus species. J. Amer. Soc. Hort. Sci. 99: 315-317. Quamme, H.Au 1976. Relationship of the low temperature exotherm to apple and pear production in North America. Can. J. Plant Sci. 56: 493-500. Quamme, H.Au 1978. Mechanism of supercooling in overwintering peach flower buds. J. Amer. Soc. Hort. Sci. 103: 57-61. Quamme, H.Au, Layne, R.E.C. and Ronald,‘W.G. 1982. Relationship of supercooling to cold hardiness and northern distribution of several cultivated and native Prunus species and hybrids. Can. J. Plant Sci. 62: 137-148. Quamme, H.Aa 1983. Relationship of air temperature to water content and supercooling of overwintering peach flower buds. J. Amer. Soc. Hort. Sci. 108: 697-701. Quamme, H.An 1985. Avoidance of freezing injury in woody plants by deep supercooling. Acta Horticulturae 168: 11—30. 141 Quamme, H.Aa 1986. Use of thermal analyses to measure freezing resistance of grape buds. Can. J. Plant Sci. 66: 945-952. Quamme, H.Aa 1991. Application of thermal analysis to breeding fruit crops for increased cold hardiness. HortScience 26: 513-517. Quamme, HaA. 1995. Deep supercooling in buds of woody plants. Pages 183—199 in R. Lee, G.J. Warren and L.V. Gusta, eds. Biological ice nucleation and its applications. APS Press, St. Paul, Minnesota. Quamme, H.A., Su, weiaA. and veto, L.J. 1995. Anatomical features facilitating supercooling of the flower within the dormant peach flower bud. J. Amer. Soc. Hort. Sci. 120(5): 814-822. Rajashekar, C.B. and Burke, M.J. 1978. The occurrence of deep undercooling in the genera Pyrus, Prunus and Rosa: A preliminary report. Pages 213-225 in P.H. Li and A. Sakai, eds. Plant cold hardiness and freezing stress. Academic Press, New York, NY. 142 Sakai,.A. 1982. Freezing resistance of ornamental trees and shrubs. J. Amer. Soc. Hort. Sci. 107: 572-581. Sakai,.A. 1979. Deep supercooling of winter flower buds of Cornus florida L. HortScience 14:69-70. £W_W '. SUMMARY Most cold hardiness investigations that have been conducted, by other researchers, have examined the xylem only and been evaluated in midwinter, spanning one or two months. These studies have found a significant correlation between the xylem and LTE’s and have related LTE's to geographic distribution. Our study was conducted over eight months and spanned all four seasons, late summer, fall, winter and summer. We observed a significant correlation between the phloem and LTE’s in acclimation, and no significant correlation to the xylem until midwinter. The only other study that includes the same time frame as ours was conducted by Ketchie and Kammereck (1987) in Malus. Ketchie and Kammereck (1987) had similar observations to ours. The phloem was found to be significantly less hardy than xylem in every month except November and summed over all months was significantly less hardy than the other two woody tissues xylem and cortex. We suggest that phloem is the critical tissue limiting commercial distribution of sour cherry, due to its lack of hardiness during acclimation (September to October) and deacclimation (March). The finding that phloem is important in commercial distribution 143 144 is supported in the PC analysis conducted on the phloem. The lack of phloem hardiness during acclimation and deacclimation may have influenced commercial distribution directly and/or indirectly by increasing susceptibility to diseases such as Pseudomonas sp. Phloem is also critical to plant survival because it reacts the most rapidly to deacclimation following warm temperatures and thus is limiting in areas that experience midwinter temperature fluctuations. The work presented is relevant in a sour cherry breeding program and indicates that the future focus for direct gains in cold resistance and indirectly gains in disease resistance should focus on the phloem in acclimation and deacclimation. The survival curves developed using the PLOT-IT Non- Linear regression equation can be used to express hardiness in absolute terms. We are the first to have done this. It is possible to relate absolute hardiness indices to environmental survival (Quamme 1978). Some further investigations building on the results presented in this dissertation would be appropriate. The research findings presented for sour cherry floral tissue indicates the flower buds reach their maximum cold 145 resistance level in February and rapidly deaclimate after early February. This is similar to reports by Kadir and Proebsting (1994) for P. avium. Another critical injury time for floral injury is November. The floral tissue of P. cerasus ‘Montmorency’ was superior in hardiness compared to all the P. cerasus seedlings evaluated, summed over all evaluation dates. However, ‘Montmorency’ was similar in hardier to the P. cerasus cultivars ‘English Morello’ and ‘Meteor’, summed over all months. Significant differences in hardiness occurred within flower buds between taxa and cultivars and seedlings from different geographic origins. Generally, flower buds from plants of northern areas were hardier. The relationship appeared to be dependent on area of origin. Gains in floral hardiness over the industry standard ‘Montmorency’ will be difficult but attention should focus on mid-February to late March, and November, finding selections that deacclimate slowly and acclimate quickly with little pre-exposure to freezing temperatures. For future investigations to improve cold resistance in sour cherry flower buds, DTA should be used as the evaluation method. The LTE data gave better separation of the cultivars and seedlings than did visual browning. 146 Preconditioning should be used in acclimation stage investigation for both floral and vegetative tissue to further improve cultivar and seedling sepearation. As indicated above this work is relevant in a sour cherry breeding program but similar cold hardiness evaluations could be conducted in any woody plant with broad geographic distribution. Determination of what tissue is critical to commercial range is essential to grower and breeders. Earlier studies have focused on the xylem and its role in determining geographic distribution. We are the first to implicate phloem as the tissue critical to commercial range. APPENDICES Appendix E Mean square and F values for GLM analysis of vegetative bud IT values for the 15 cherry cultivar/selections for November 1990 to March 1991. Month Source of variation df Mean Pr > F square November Cultivar and seedlings (C) 14 18.64 0.0346 Error 13 6.59 Total 27 December Cultivar and seedlings (C) 14 8.98 0.5697 Error 13 9.86 Total 27 January Cultivar and seedlings (C) 14 20.82 0.6340 Error 13 25.08 Total 27 February Cultivar and seedlings (C) 14 2.11 0.2027 Error 13 1.32 Total 27 March Cultivar and seedlings (C) 14 22.12 0.0063 4 Error 13 5.15 firtal 27 ' Appendix F Mean squares and F values for GLM analysis of IT values for all factors. Source of variation df Mean square Pr > F Cultivar and seedlings (C) 14 276.13 0.0001 Month (M) 7 4277.78 0.0001 Tissue(T) 4 3540.97 0.0001 M X C 98 28.57 0.0001 M X T 22 146.10 0.0001 T X C 56 19.43 0.0002 T X Plant 8 8.42 0.5723 c x Plant 13 34.67 0.0001 5 M X T X C 308 10.62 0.3125 Error 421 10.09 Total 951 152 Appendix G Mean square and F values for GLM analysis of Twig LTE values for all factors. Source of variation df Mean square Pr > F Cultivar and seedlings (C) 14 79.02 0.0001 Month (M) 4 802.01 0.0001 C X M 52 24.02 0.0976 C X Plant 13 34.53 0.0336 Error 34 15.78 Total 117 153 Appendix H Mean square and F values for GLM analysis of vegetative buds IT values from November 1990 to March 1991. Source of variation df Mean Sqaure Pr>F Cultivars and seedlings (C) 14 38.31 .0001 Month (M) 4 744.60 .0001 C X M 56 8.59 .3254 C X Plant 13 17.68 .0157 Error 52 7.58 Total 139 154 Appendix I Mean square and F values for GLM analysis of cortical tissue IT values over eight months. Source of variation df Mean Sqaure Pr>F Cultivars and seedlings (C) 14 55.18 .0001 Month (M) 7 1279.08 .0001 C X M 98 12.23 .0498 C X Plant 13 12.85 .1406 Error 91 8.69 Total 223 Appendix L Mean square and F values for GLM analysis of flower bud LTw values for the cherry selections for November 1990 to March 1991. Month Source of variation df Mean Pr > F square November Cultivar and seedlings (C) 14 14.49 0.2544 F” Error 13 9.99 Total 27 December Cultivar and seedlings (C) 14 11.40 0.0294 Error 13 3.86 Total 27 ”_ January Cultivar and seedlings (C) 14 5.85 0.1763 Error 13 3.46 Total 27 February Cultivar and seedlings (C) 14 10.49 0.0009 Error 13 1.62 Total 27 March Cultivar and seedlings (C) 14 19.57 0.0576 Error 13 7.99 Total 27 l58 Appendix M Mean square and F values for GLM analysis of flower bud LTE’s for the cherry selections for November to March. Month Source of variation df Mean Pr > F square November Cultivar and seedlings (C) 9 20.05 0.0026 Error 9 2.56 Total 18 December Cultivar and seedlings (C) 8 13.15 0.0013 Error 7 0.98 Total 15 January Cultivar and seedlings (C) 4 4.24 0.3895 Error 6 3.44 Total 10 February Cultivar and seedlings (C) 11 5.45 0.2581 Error 10 3.58 Total 21 March Cultivar and seedlings (C) 11 13.40 0.0121 Error 7 2.19 Total 18 159 Irr- Appendix N Mean square and F values for the GLM analysis of flower bud LTE values for all factors. Source of variation df Mean square Pr > F Cultivar and seedlings (C) 14 22.23 0.0001 Month (M) 3 245.91 0.0001 g C X M 24 8.73 0.0014 3 C X Plant 13 6.33 0.0194 E Error 22 2.34 Total 76 160 Appendix 0 Mean squares and F values of the GLM analysis of flower bud LTE values for November to March. Source of variation df Mean SqaureL Pr>F Cultivars and seedlings (C) 14 79.02 .0001 Month (M) 4 802.01 .0001 C X M 52 24.03 .0976 C X Plant 13 34.53 .0336 Error 34 15.78 Total 117 161 ;._,._,."_',-:-o."_~‘ '4‘- Appendix P Mean square and F values for GLM analysis of flower bud T50 values over five months. Source of variation df' Mean Sqaure Pr>F Cultivars and seedlings (C) 14 41.61 .0001 Month (M) 4 131.04 .0001 C X M 56 5.05 .0047 C X Plant 13 17.10 .0001 Error 52 2.45 Total 139 162 Appendix Q Mean square and F values for GLM analysis of one cultivar (Pandy 114) IT and LTw values from August 1990 to March 1991, showing the relative importance of the various non- genotype sources of variation. The variable tree is like replication. Source of variation df Mean Square Pr>F Month (M) 7 623.56 0.0001 M X TREE 16 19.95 0.0350 Tissue (T) 4 341.02 0.0001 M X T 22 27.64 0.0016 Error 52 10.18 Total 101 163 Appendix R Mean square and F values for GLM analysis of effects of preconditioning IT values for the cherry selections investigated for August, September and December. The preconditioning temperature was -3%L Date Source of variation df Mean Pr > F square August 14 Cultivar and seedlings (C) 6 1.71 0.0831 (Day 1) Error 14 0.71 Total 20 August 16 Cultivar and seedlings (C) 6 6.86 0.0003 (Day 3) Error 14 0.71 Total 20 August 28 Cultivar and seedlings (C) 2 1.71 0.1357 (Day 1) Error 4 0.50 Total 6 August 30 Cultivar and seedlings (C) 2 7.00 0.0110 (Day 3) Error 4 0.67 Total 6 Sept. 5 Cultivar and seedlings (C) 2 27.00 0.0001 (Day 9) Error 4 0.33 Total 6 Dec. 6 Cultivar and seedlings (C) 2 79.00 0.0001 (Day 1) Error 4 0.67 Total 6 Dec. 8 Cultivar and seedlings (C) 2 57.00 0.0001 (Day 3) Error 4 0.33 Total 6 Dec. 11 Cultivar and seedlings (C) 2 43.00 0.0001 (Day 6) Error 4 0.67 Total 6 164 Appendix 3 Mean square and F values for GLM analysis of the comparison of regression slopes of xylem (Fig. 2) and phloem (Fig. 3). Source of variation df Mean Pr > F square Group 1 1286.51 0.0001 X (LTE) 1 251.48 0.0001 X*Group 1 10.64 0.4030 Error 135 15.11 Total 138 .165 Appendix U Mean square and F values for GLM analysis of tissue types, IT values for the 15 cherry selections for August to March. Month Source of variation df Mean ' Pr > F square August Cultivar and seedlings (C) 14 58.52 0.0001 Tissue 2 74.61 0.0010 C * Tree 20 30.14 0.0011 C * Tissue 28 11.25 0.2101 Error 26 8.20 Total 90 September Cultivar and seedlings (C) 14 39.17 0.0001 Tissue 2 53.61 0.0001 C * Tree 20 5.17 0.0649 C * Tissue 28 5.96 0.0252 Error 26 2.75 Total 90 October Cultivar and seedlings (C) 14 32.01 0.0001 Tissue 2 109.19 0.0001 C * Tree 20 16.18 0.0063 C * Tissue 28 14.47 0.0091 Error 26 5.64 Total 90 November Cultivar and seedlings (C) 14 35.68 0.0003 Tissue 4 343.21 0.0001 C * Tree 20 15.02 0.1073 C * Tissue 56 16.05 0.0357 Error 52 9.75 Total 146 December Cultivar and seedlings (C) 14 144.27 0.0001 Tissue 4 418.38 0.0001 C * Tree 20 49.75 0.0007 C * Tissue 56 15.11 0.6264 Error 52 16.48 Total 146 January Cultivar and seedlings (C) 14 110.38 0.0001 Tissue 4 1931.61 0.0001 C * Tree 20 31.25 0.0001 C * Tissue 56 10.91 0.0279 Error 52 6.43 Total 146 February Cultivar and seedlings (C) 14 58.10 0.0001 Tissue 4 538.82 0.0001 C * Tree 20 17.03 0.0019 C * Tissue 56 9.91 0.0460 Error 52 6.23 Total 146 March Cultivar and seedlings (C) 14 43.16 0.0001 Tissue 4 702.42 0.0001 C * Tree 20 4.06 0.6831 C * Tissue 56 10.02 0.0059 Error 52 4.97 Total 146 167 Appendix V Comparison of 5 tissue types, phloem, xylem, cortex, vegetative buds and flower buds, of 15 cherry cultivar/seedlings evaluated by visual injury for August 1990 to March 1991 and over all months. Month Tissue '1pr Grand Means August Xylem -15.3az Phloem -1l.1b Cortex -12.6b September Xylem -17.7a Phloem -15.3c Cortex -16.5b October Xylem -26.4a Phloem -21.9c Cortex -24.5b November Xylem -24.2a Phloem -23.9a Cortex -25.1a Vegetative Buds -21.8b Flower Buds -16.4c December Xylem -27.4a Phloem -20.5b Cortex -22.1b Vegetative Buds -l7.6c Flower Buds -17.4c January Xylem -39.5a Phloem -29.0c Cortex -35.0b Vegetative Buds -28.0c Flower Buds -18.9d February Xylem -33.1a Phloem -29.6b Cortex -32.la Vegetative Buds ~30.9b Flower Buds -22.6c March Xylem -32.9a Phloem -25.6c Cortex -30.0b Vegetative Buds -25.5c Flower Buds ~18.1d Over all Months Xylem -27.7a Phloem -22.1c Cortex -24.7b Vegetative Buds -l9.8d Flower Buds -18.7e zMeans in columns, based on three sub-samples, per four replicates, averaged over 15 selections, followed by different letters are significantly different (P=0.05) according to Fisher's least significant difference test. 168