.. i . V. V . . . 4 V “V. mmwwi . VVVVVVVVmfiVwVV ~! I 14 V 4.3., V x k L4 a 32V“. «a? Vm... VV 3 , VV an, . human" :3, . V. Emmwwmw . .. .mmwz... .. . V . V. 7. i V . y a . V V. V V3.22: .u. 2:9 V . V. x. i‘ 73.1.. V 1 V . V. VV L.:.aA.ifl¥59ia.~flY V .V :V 3:_1V£)l:§.¢x.z . » 1.3 Vfiltiz V . . V V : . 5., tiafifi V . .. .V . 2&8332195. ifs: .. V. 3...; 1... E12,. :i?::.ah;3 fatal... V . ,V IV? V \. 1355.: 31:1... .9221: .5152... V. ziittflt§t7tgtsftgszag .V; q Vyiftgjlhqi 93) in?! 6415.59;V§3I§V§JXOKA41§5$§II9~J V V V . V1. V x. V was.» 1‘ évVNVVMVuhHWVifli , ~Q§ViVThzhflfi§Vu§jfigfiilg r; V V a V. I... V. 7:..kt!..§\gt\tz’§u§i§ :3 .33 V... .35..“3; 2L. $§Ezf3=h *3! $3:ah§vl.1§h? V. 55.3313 V V J1 - lit?! is, 5.1.1 5.73111. tigxial Vt!) IV . ti. 'Ist‘. a 1.3.1:. is. ll 1.311.... IL .151”?! alatq‘ih. .995! V .3 . . 5.3%... flunrxcfimflzy.§. V .Vbshxmflwmfiunhnuflmiaaw V. VIE-V15 3.5.52.2: 11:13:91.1! V V . 1.15 V . V... V tit... ::::: 1: .5: V.. a V. s. V V.- > V.‘ V .- .Vaahqlr: V1.2}... he... Vt 1 .91.... 1.1 1 . V1,, 4 , . v .i .V. 14mm... tfififlfihu: VV, xi V . 2. ‘0 V norfi V A ‘ p. v. r , . V)V:DV.!VV.V|;Z{)8VI.V.1V V1,?VLVJVV7V5V. (.3 V ::::::::: if; Vigtzi dsfiiigu. 3223:213qu ! tuft]; «Witt. , at $.39; : V! 2‘ V. {TA 22‘... gist... .. IQVMVlfiV-luhi 9%. gui.§.l.i.t.§;.zafl .v VI.“ mun “HUM... .) F: )\ .\L¥V€§t£§§t§vflg it!!! 38.1 C‘gfg‘ r it attilrtl. {OI-lg“. g!!! i ‘l‘ ’I‘. u Lyyiiipzaflsgi . V 7., Oil-2:28.113; gnu. £3 , naggi‘sg V V. #159213)?“ :Z‘murfihfi r11 . (18.1)} I! . is); . . V. I. V V V, . VS ~v9 V Q Y. . It 73.1 “fit. T , fit: Inn. , $.21] V. .5 .2 \ 7)}. t V. V Yo... .. 1 Vh’ V5 idirfi s 311.14... 1|: 1 . \>. . V é!“ . .. aka“ 0% V. Itu. V V. 1... I... v V... .fihnx 2 9. :fi V.» . . I o. V5.11 n. \ V 1: F. u Oflvflfi _ r 55%.!» . St :3 n ‘ Etigs‘g V... . .2. Nu ... V thaw? 24$. v V... .1 514.z....u._u....zdl . 2.1:! .naduzstlé.t3mmu.i . . V. 5??! T5199...“ 1.: 179. V. {515... 7. V. 2!. It” » V 15.} . h)? I a . ’5! 51‘th gfini-fluh .1 ix \‘tIIS. 3 "R . Mug”. gigg 3‘! “2 $3 , t .15.. . g L... ti}. mauilixtszni...‘ . .: V .1 .p .1: .ds: . vhf??? s... n; V “2.11.... $Ef§3§utl§atz V... V. I; , 3.2+ ..§I§ V. umm1§n3 1.4- 3) 72. til». I: f P gi‘?! V1 2 V. its; K , ii: V .1 V zyg’liz I; H. IVS V V L 5% “HM 1991...?! 1:1“: I}?! V V \t. . . all; it... I ha? u: . z. 8an .va .zfiait‘guzosug: V a J . fin“- !‘VV’ ; iiiga‘fi“ J V 210.“. 11! . . .2 V 154‘! Vttittlx 1 .tht b!« V.. \l. .P V... III. , r1 ,. 5:.“ i V . . it? wit-d. {{ .22 )1 3 V. . V: . V I . $3.1”? VV . "t: V 11 . Vt V .1 t QIIIIUAVHQSJ’UEIIIVANMH V V V. V inlttl(‘l§(2§§i I: V. V V I .. 11.2 52:? $41.5? 23.55.32-113? “if an? . 523331.!- 5): ’23))?! \p ,7‘52V3I.}£V§.z;§5§ .IV .V V «51(4 . 12.x V) .41 53),!!!) tillzslvggy ‘7 $19.52 1‘!!!" V V , .3. 1.3 V V... . 5%..» .i V xhflflfliv ~ . . . \ Vittxutf‘tx.kt)a\\.(.‘ SLVITKV 1 I: .V V : ya .PV. V .QIVV .34... V. .\ 5...: 1...VVV§V1.V:...V.3V..1.$:.IOIHL:VUJV .. \\5n 1 >»V. 7).. 9., tit; I» .V EV‘.‘§1 5? (till? 13...}: £13“? a. V 1. V11c|3§12§1tlsslltzttnn \qu§V£ttzl.lllt«§b\ttix. anemia V3.73. .VVV .at’uu. V _ I A1.V_\:§§.x..§$Vi£\§[x€2g§g \figg‘t 35! V V. (in: 55.. LI la 225~A£§~53~t L5» i?2§2\(2[512§i\\l gigfitzkfttizi§$szzzztlfi V 5 1233571.! 5... $3174.. le.t!1. : 4.).{3Viizzuzx =ng ‘ tl‘vzti, if§§\!$o§ Vtgfg‘git. , V . liigiilfix V YIR£CVV§VVI V V2 ..... ’V QVLvltVnViisztrg... {git}; \izz’VlrittlrrVnttl \4 s; ILLIV \9 V. .3 . . . V V. .Vrienzt ,n 34.....3Lnu‘ xStrAK ‘.§‘..¢§x§\it§t\tt§.¥:n .. . . .t 5H A: .6; Paul. I .Vx|5tt\\|(\tttxfiz(.k|§ ..... x. V .. $1! .57“? thtVIV.«\t\t ‘14::‘1 (VI-«5.) .V pix“)? A 1‘5." .. 19:52:13.1} .xtl “xix: . Vimflglt. 9.52 VV.V(V),910..i.OVV 21th.“... flit}? tilt... «. tit; {(4125: in. szxtsnflf s1. V25. ‘k‘u’k‘ 1. "YVlnfi. f‘v’zi‘c; ,K‘\.'z‘§.k.nlt ‘K.\ \\ \n‘K’lz ‘ l.§\\IVV ’0‘ i\\‘|‘.tklt|x(‘xfi\ £5: .2 .17: «(Lt L3V-YAV‘IVVLV-‘Ktt ( 253132.: 1.22133), ztv‘VgD-llgttz )3 i(\\‘\3¢( 2|!) s(\t(\t.\ x 120. ..... V524 2.1.?2L4sziz t§~¥t§$i§sxbt £..\. 2.3.3 1 k)! V )1“ 57 (S\VV.... is .(a At :51) «\t‘}? ‘5 :5 .3: 7?.1 aizV‘xz {55.29:}...er k t V(.VVV.Vt)O\t.V$L¥Ci (t.l\t61(l\)~\l§t9tt «Inf-t;- . .53: .tt\\. V Int-Atl azig K V Vi gift l5l}...,\§l.»\£t$s.ss ‘52»:ch V (1‘ (A)? {fit I...» fit ($1574! fig} z‘\:l\.~l Z$lliz I. V ‘fl V! ~|$I§h4|l~t2§ §\ \lkth :1 \ «£53.; .Q‘IK 59“ \‘l \V’. 5“ I:u.| 7 3 < ’73)): Silt};{7s:}llt(51tl~tfi.((<\tzt.l££ V4 fixtxkxl)‘ .{\s5\t~|s\ tin...“ V. V \kntx\\fi(q1|Iu> ,V L, 7V V. .«thvxlriliz‘VlrrrV: A 2.1L. . . Alnllulttfrvnbk I n V 1.9.7! . at. I 72, V V V . V V V V LI B RAR Y Michigan State University 1 U: ., ’r ' mymMini/Mimi);”MW This is to certify that the thesis entitled PHYSIOLOGIC AND GENETIC STUDIES ON WOOD HARDINESS _~ ‘ 0F PEACHES (Prunus Persica (L.) Batsch) GROWN NEAR THE NORTHERN LIMITS OF COMMERCIAL ADAPTATION presented by David Wayne Cain has been accepted towards fulfillment of the requirements for Ph.D. degree in Horticulture Major profe sor Date 5/3/78 0-7639 RETURNING MATERIALS: IV‘ESI_J Place in book drop to LIBRARIES remove this checkout from gage-g; your record. FINES will be charged if book is returned after the date stamped below. .. 4~ 5 fit/.m PHYSIOLOGIC AND GENETIC STUDIES ON WOOD HARDINESS OF PEACHES (PRUNUS PERSICA (L.) BATSCH) GROWN NEAR THE NORTHERN LIMITS OF COMMERCIAL ADAPTATION BY David Wayne Cain A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Horticulture 1978 ABSTRACT . PHYSIOLOGIC AND GENETIC STUDIES ON WOOD HARDINESS OF PEACHES (PRUNES PERSICA,(L.) BATSCH) GROWN NEAR THE NORTHERN LIMITS OF COMMERCIAL ADAPTATION BY David Wayne Cain Based on their orchard survival, 'Velvet', 'Redhaven' and 'Siberian C' were considered representative of tender, intermediate and hardy peach cultivars, respectively. Con- trolled freezing tests on 13 dates in 2 winters showed that 'Siberian C' always had less inner bark and xylem injury than 'Redhaven', while, for both tissues, 'Velvet' had slightly more injury than 'Redhaven'. In the fall, both inner bark and xylem injury increased rapidly as temperature decreased, but during midwinter, inner bark injury increased slowly with temperature decline, while xylem injury still increased rapidly with temperature decline. .This was indicative of changes in freezing processes. Fully acclimated xylem of all 3 cultivars was always killed near -37°C, indicating that it may have deep supercooled. Within dates, cultivar injury ratings paralleled moisture content. Increasing tissue mois- ture content prior to freezing did not affect xylem injury, but increased inner bark injury more in 'Redhaven' than in David Wayne Cain 'Siberian C'. High temperature pretreatment increased xylem injury in both cultivars, and increased inner bark injury more in 'Siberian C' than in 'Redhaven'. An electrophoretic mobility technique was used to eval- uate tissue freezing patterns. Even though bark moisture was twice that of the xylem, bark exhibited equilibrium freezing patterns while xylem produced nonequilibrium pat- terns. This indicated a greater interaction between water and cellular components in the bark than in the xylem. Freezing patterns of corresponding tissues in 'Redhaven' and 'Siberian C' were similar. Inner bark, xylem and vegetative bud injury were assessed for 7 parents and 5 progenies differing in hardiness. Mean progeny inner bark and xylem hardiness could be predicted from average parental performance. Relative progeny vegeta- tive bud hardiness did not correspond to that of bark or xylem, and it could not be predicted from parental perfor— mance. Injury, as measured within dates and tissues, was highly heritable, while environmental variation estimates were very low. Based on individual tree observations, corre- lations among injury ratings for different tissues, tempera— tures and dates were low, indicating poor repeatability of individual genotypes. Correlations based on family means were higher. The low individual tree correlations indicate that selection based on individual phenotypic performance in a single test might be ineffective. This thesis is dedicated to my parents for their years of devotion and encouragement ACKNOWLE DGMENTS I would like to express my sincere gratitude to Drs. G. S. Howell, F. G. Dennis, C. E. Cress and C. R. Olien for serving on my committee and for their guidance and many helpful suggestions during the course of this research. I would like to thank Dr. R. E. C. Layne of Agriculture Canada for the use of his plant materials and for his valuable review of this manuscript. Special thanks are extended to Dr. Robert L. Andersen for serving as my major professor and torfiri and his wife Judy for making me welcome in their home many times during my years at Michigan State University. *4 H . '4 TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . . . . . . . . . . INTRODUCTION . . . . . . . . . . . . . . . . . . . . . LITERATURE REVIEW . . . . . . . . . . . . . . . . . . Hardiness differences among tissues . Effects of cultural practices . Factors affecting recovery . . . Mechanisms of freezing injury . Histological . . . . . . . . Cytological effects . . . . Measurement of freezing processes Deep supercooling . . . . . . . . Genetics of cold hardiness . . . . SECTION I: COMPARISON OF RELATIVE FREEZING IN 'VEL- VET]. 'REDHAVEN' AND 'SIBERIAN C' PEACHES FOL- LOWING CONTROLLED FREEZER TESTS AT SELECTED DATES DURING TWO WINTERS . . . . . . . . . . . . INTRODUCTION 0 C O O O O O O O O 0 O O O O O O 0 MATERIALS AND METHODS . . . . . . . . . . . . . Plant material . . . . . . . . Sample preparation . . . . . . Freezing procedure . . . . . . Tissue browning ratings . . . Callusing and TTC tests . . . Moisture content of excised twig internodes Moisture content of bark and xylem . . . . . Effect of varying moisture content on tissue hardiness . . . . . . . . . . . . . . . . . O I O O 0 O O O O O O O O O O O O O O O 0 RESULTS . . . . . . . . . . . . . . . . . . . . Cultivar browning ratings compared at 13 dates . . . . . . . . . . . . . . Callusing and TTC tests . . . . . . . . Relationship between moisture content and injury . . . . . . . . . . . . . . . . . . iv Page vii ix 20 20 21 21 22 22 24 25 27 29 29 33 SECTION 2: SECTION 3: Determination of bark and xylem moisture Effect of temperature and adjustment on hardiness. . . . . . . . . . DISCUSSION . . . . . . . . . . Seasonal hardiness comparisons . Relationship between moisture content and inj ury O O O O O I O O O O 0 Differences in freezing characteristics of bark and xylem . . . . Phenological characteristics of hardiness in 'Siberian C' . . . . . . Summary . . . . . . . . . . COMPARATIVE FREEZING PATTERNS OF BARK AND XYLEM OF 'SIBERIAN C' AND 'REDHAVEN' PEACH WIGS O O O O O O O O O O O C 0 INTRODUCTION . . . . . . . . . . MATERIALS AND METHODS . . . . . Plant material . . . . Freezing patterns . . Contact resistance . . . Dye flow experiments . Moisture content . . . RESULTS AND DISCUSSION . . . . . HYBRIDS OF COMMERCIAL AND EXOTIC TYPES O O I O O O O O O O O O 0 INTRODUCTION- . O O O O O O O O 0 MATERIALS AND METHODS . . . . . Plant materials. . . . . . Sampling procedure . . . . Freezing of twigs . . . . Evaluation of injury . . . Crop load and ripening date Canker ratings . . . . . . . Bloom type . . . . . . . . . Data analysis . . . . . . . Heritability estimates . . V PEACH O O 0 O O O O O O O O 0 O O O O 0 O O I INHERITANCE OF WOOD HARDINESS AMONG GENO- Page 43 43 46 46 49 51 52 54 55 SS 56 56 56 58 59 59 60 69 £59 '71 771 772 773 7’3 774 7W5 765 763 77' RESULTS . . . . . . . . . . . . . . . . . . . . 78 Parent injury ratings cambium . . . . . . . . . . . . . . . . 78 xylem . . . . . . . . . . . . . . . . 78 vegetative buds . . . . . . . . . . 80 Mean family hardiness . . . . . . . . . . . 80 cambium . . . . . . . . . . . . . 1 . . 80 xylem . . . . . . . . . . . . . . . . . 80 vegetative buds . . . . . . . . . . . . 82 Crop load 1976 . . . . . . . . . . . . . . 82 Ripening date 1976 . . . . . . . . . . . . 83 Canker ratings . . . . . . . . . . . . . . 83 Bloom type . . . . . . . . . . . . . . . . 83 Population dispersion . . . . . . . . . . . 84 Genetic analysis of progenies . . . . . . . 84 DISCUSSION . . . . . . . . . . . . . . . . . . 9O SUMRY MD CONCLUSIONS 0 O O O O O O O O O O O O O 0 lo 3 APPENDICES . . . . . . . . . . . . . . . . . . . . . 107 BIBLIOGRAPHY o o o c o 0 O O O 0 0 0 9 ° ' ° ‘ ° ° ° 119 c T] l Table LIST OF TABLES SECTION 1 Collection dates, injury rating dates, number of temperatures used and temperature ranges used for comparing relative hardiness of 'Vel- Vet', 'Redhaven' and'Siberian C'peach twigs during 2 winters . . . . . . . . . . . . . . . Correlations between callus proliferation and inner bark ratings following freezing of twigs of 3 peach cultivars . ... . . . . . . . . . Color of 'Redhaven' and 'Siberian C' twig tissues treated with triphenyl tetrazolium chloride following controlled freezing on March 7, 1977 . . . . .,. . . . . . . . . . . Moisture content of peach twigs at each test date 0 ‘0 o o o o o o o o 01 o o o o o o o o 0 Correlations between internodal moisture con- tent and mean tissue injury ratings of all trees for each sampling date during 1975-76 . Moisture content of bark and xylem of 'Redha- ven' and 'Siberian C' at selected dates in 1976-77 . . . . . . . . . . . . . . . . . . . Effect of moisture and temperature pretreat- ments on moisture content and injury to 'Redhaven’ and 'Siberian C' inner bark and xylem following freezing to -26.1 C on November 5, 1976 . . . . . . . . . . . . . SECTION 2 Sample data showing calculations used to estimate the amount of liquid water at a given temperature . . . . . . . . . . . . . . . . Calculation of contact resistance of longitu- dinally split twig sections in 6 separate experiments . . . . . . . . . . . . . . . vii Page 23 36 39 41 42 42 44 58 60 Table Page SECTION 3 l. Injury ratings of parents' cambium, xylem and vegetative buds subjected to controlled freez- ing tests on 3 dates . . . . . . . . . . . . . . 79 2. Mean freezing injury of cambium, xylem and vegetative buds tested on 3 dates and crOp load, ripening date, and canker ratings for 5 progenies . . . . . . . . . . . . . . . . . 81 3. Inheritance of bloom type in parents and prog- enies. Showy bloom (Sh) is dominant to non- showy bloom (sh) . . . . . . . . . . . . . . . . 84 4. Progeny and parent broad-sense heritability estimates for cold injury assessed for individ- ual tissues, temperatures, and dates . . . . . . 85 5. Deviations of progeny mean injury ratings from midparent ratings calculated for each tissue, temperature and date . . . . . . . . . . . . . . 87 6. Correlation coefficients between individual tree ratings across all progenies for injury, crOp load, ripening date and canker ratings . . 88 7. Correlation coefficients among progeny means for injury, crop load, ripening date and canker ratings 0 O O O O O O I O C O O O O O O O O O O 89 APPENDIX A 1. Means and standard errors for cambium and xylem injury ratings to 'Velvet', 'Redhaven' and 'Siberian C' at each sampling date and test temperature . . . . . . . . . . . . . . . . 107 2. Means and standard errors for callus regrowth ratings for 'Velvet', 'Redhaven' and 'Siberian C' at each test temperature on 3 dates, 1975-77 . . . . . . . . . . . . . . . . . . . llO viii Figure LIST OF FIGURES Page Section 1 Mean inner bark and xylem injury ratings of 'Velvet', 'Redhaven' and 'Siberian C' averaged over test temperatures at.which at least 1 cultivar had a mean injury rating greater than 1.0 and less than 5.0 . . . . 31 Changes in temperature response patterns of inner bark and xylem of 3 peach cultivars from late fall to winter . . . . . . . 35 Callusing of 'Redhaven' (left in each set of 6) and 'Siberian C' (right) following exposure to stress temperatures ( C) of (upper left to lower right): ~17.8, -20.6, -23.3, -“8.9, and -32.8. Twigs collected on November 5, 1976 . 38 Section 2 Typical freezing patterns of bark and xylem from acclimated peach twigs. Patterns for sucrose and cellulose model systems (from Olien, 1977) are given for comparison . . . 64 Appendix B Progeny distributions according to cambium injury class. A to D are: January 16, 1976 (-32.2°C); November 18, 1976 (-26.l), (-31.7); February 7, 1977 (-33.8), respectively . . 112 Progeny distributions according to xylem injury class. A to C are: January 16, 1976 (~32.2°c); November 18, 1976 (-26.1); February 7, 1977 (~33.8), respectively . . 114 Figure 3. Progeny distributions according to vegetative bud injury class. A and B are: November 18, 1976 (-31.7OC) and February 7, 1977 (-33.8), respectively . 4. Progeny distributions according to over- all mean injury ratings of each seedling as assessed by averaging individual seedling performance across all tissues, temperatures and dates . . . . Page 116 O C 118 INTRODUCTION Freezing damage to peach flower buds and woody tissues is a major factor limiting peach production in most of North America. Injury to woody tissues is sometimes dramaticefor a single freeze may kill thousands of trees (Bradford and Cardinell, 1952). Such freezes may occur only a few times in a century but low temperature stress causes some injury nearly every winter. While such chronic damage may appear innocuous, it often causes increased sus- ceptibility to diseases and may reduce productivity. Freezing damage has been implicated in increasing tree sus- ceptibility to perennial canker (Weaver, 1963; Tekauz and Patrick, 1974; Layne et a1., 1976). In southern produc- tion areas freeze damage has been associated with peach tree short life (Daniell and Crosby, 1971; Nesmith and Dowler, 1976; Brittain and Miller, 1976). Loss of whole trees and major scaffold limbs can reduce orchard bearing surface to such a degree that, as an economic unit, the productive life of an orchard may be only 7 or 8 years (Brittain and Miller, 1976). Winter injury also causes occlusions in xylem vessels (Daniell and Crosby. 1968) and causes wood to become brittle, resulting in limb breakage under heavy crop loads (Campbell, 1948; Campbell and Hadle, 1960). Improved wood hardiness is essential for the success- ful long—term commercial production of peaches in Michigan and other regions having serious winter injury and should be of prime concern in breeding new cultivars. Genealogi- cal information derived from cultivar descriptions reveals that most commercial freestone peach cultivars have been derived from a limited number of progenitors (Hedrick, 1917; Savage and Prince, 1972). Hesse (1975) notes the severely restricted base used for cultivar improvement in the United States, and the tendency for breeding programs to become highly inbred. Recently, very cold hardy germplasm has been obtained from nothern China (Pieniazek, 1968). This material offers a new source of genes to peach breeders and is being used in breeding programs in Michigan, New York and other states, and in Ontario, Canada. Of this material, 'Siberian C' has been the most extensively studied (Chaplin and Schneider. 1974; Layne, 1974, 1976; Ormrod and Layne, 1974, 1977; Quamme et al., 1975; Layne et al., 1976; Layne et al., 1977). Under Canadian conditions, 'Siberian C' rootstock causes earlier than usual defoliation of scion cultivars, increases scion cold hardiness of phloem, cambium and xylem in fall and enhances midwinter flower bud hardiness (Layne,et al., 1977). The roots of ‘Siberian C' are also very cold hardy (Layne, 1974), and they increase scion survival through their own survival (Layne et al., 1976; Ormrod and 3 Layne, 1977). In Kentucky, hardiness as assessed by electrolytic conductance tests indicated 'Siberian C' transmitted less hardiness to 'Redhaven' scions than did 'Harrow Blood‘ but was generally better than 'Rutgers Red Leaf‘. 'Siberian C' is presently used as a rootstock but breeders are interested in utilizing it as a source of cold hardiness for both scion and rootstock breeding programs. However, its fruits are very small, soft, bitter, and white-fleshed and have no commercial value. Therefore, its valuable hardiness characteristics must be recombined with the high fruit quality of commercial cultivars via hybridization. Very little is known about the physiologi- cal or genetic features which cause 'Siberian C' to possess superior hardiness. My thesis research focused on 3 related questions: 1. When does 'Siberian 0' exhibit its superior cold hardiness? 2. What are some physiological causes of this superiority? 3. What breeding value does this germplasm have for creation of commercially adapted scion cultivars with improved cold resistance? LITERATURE REVIEW Siminovitch et a1. (1968) define winter hardiness as "the capacity of living cells to respond without in- jury to the sum total of freezing stresses experienced over the winter”. For perennial crops the definition must be expanded to include many winters. In its general defini- tion winter hardiness becomes an exceedingly complex trait. There is voluminous literature dealing with many facets of cold hardiness. Many good books and reviews cover various aspects of the subject (Chandler, 1954; Olien, 1967; Mazur, 1969; Weiser, 1970; Stushnoff, 1972; Levitt, 1972; Burke et al., 1976). Since plant breeders have finite resources in terms of money, land, time and plant material, efficient use of these resources is necessary for maximum breeding progress. The relative importance of plant characters contributing to cold hardiness needs to be known in order to develop suitable screening techniques and appropriate breeding approaches. Stushnoff (1972) has classified the seasonal hardiness cycle in terms of (a) time of tolerance develop- ment, (b) rate of development, (0) level of intensity developed, (d) retention of tolerance, (e) onset of loss of tolerance, (f) rate of loss, and (g) ability to regain tolerance. He points out that in some situations it may 4 5 be more useful to breed for a timing mechanism rather than for maximum level of expression. This may be true for peaches in northern regions where much injury often occurs in late fall (Bradford and Cardinell, 1926; Rollins et al., 1962; Andersen, 1974). Differences in cold injury among geographic clones of Cornus stolonifera during late fall are associated with differences in phenological responses (Smithberg and Weiser, 1968). Evidence for a two stage acclimation mechanism exists in apple (Howell and Weiser, 1970a). The first stage is induced by short photoperiod and the second by killing frost. Photoperiod has an effect on flower bud hardiness at warm temperatures, but temperature effects overcome photoperiod effects, especially at lower but above freezing temperatures. Therefore, peaches appear to follow the general acclimation pattern for woody plants (Ormrod and Layne, 1974). Ambient air temperature prior to stressing markedly affects injury. Ketchie and Beeman (1973) find high negative correlation between cold resistance of apple bark and air temperature during the preceding 7 days. Short term changes in cold resistance of apple bark during spring dehardening are related to air temperature of the preceding day (Howell and Weiser, 1970b). Peach and blue-r berry flower buds also deharden upon exposure to warm temperature (Proebsting, 1970; Bittenbender and Howell, 1975). The 'Latham' raspberry, a classic example of a 6 plant considered very hardy during prolonged cold, but loses its hardiness very quickly under fluctuating temperature conditions (Brierley and Landon, 1946, 1954). Blake (1935) describes differential varietal response to several environ- mental conditions. Some cultivars are resistant to con- sistently low winter temperatures while others are better at resisting variable temperatures. Prunus davidiana and 2. kansuensig are extremely hardy during midwinter but are subsequently injured because they begin growth very early in the spring. Hardiness differences among tissues Different tissues and regions of the tree do not acclimate at the same rate or attain the same maximum level of hardiness. During the summer all tissues are quite sensitive. Flower buds begin to develop hardiness early in the fall. Xylem also develops some degree of hardiness earlier than bark tissues but it acclimates more slowly than flower buds. By early winter the situation is reversed. Now, flower buds are most sensitive, xylem is intermediate, phloem tissues are slightly hardier while cortex and cambium are most resistant in most field condi— tions (Chandler, 1913; Bradford and Cardinell, 1926; Fogle and Overley, 1954; Edgerton, 1960). The early acquisition of flower bud hardiness might be explained by the fact that they undergo deep supercooling (Quamme et al., 1975). 7 During deep supercooling, isolated water may remain liquid at normally subfreezing temperatures down to the homogeneous nucleation point of pure water which is about -380 to -40°C (Fletcher, 1962). Xylem tissues of many species can also deep supercool (Quamme et al., 1972a, 1973; George et al., 1974). Vegetative buds are usually hardier than flower buds and are usually not killed except when wood and bark tissues are also severely injured (Dorsey and Strausbaugh, 1923). However, late fall freezes can-kill vegetative buds while not injuring flower buds (Overley and Overholser, 1936). Even the same tissue can exhibit very different levels of resistance in different parts of the tree. In fall, bark tissues are often injured in the trunk and major scaffolds, yet identical tissues in twigs may not be injured (Chandler, 1913; Bradford and Cardinell, 1926; Blake, 1938; Fogle and Overley, 1954; Edgerton, 1960). Trunks may also suffer severe sunscald or southwest injury while twigs are not damaged (Savage et al., 1976). Narrow crotch angles also increase injury (Blake, 1935; Edgerton, 1960). Apical sections are less hardy than basal sections of the same twig, and twigs from lower shaded portions of the tree are less hardy than those from upper well-exposed areas (Cain and Andersen, 1976). 8 Effects of cultural practices Cultural practices have a dramatic effect on hardi- ness of fruit trees. Practices such as late nitrogen application and clean cultivation, which stimulate late growth, increase the chance of winter injury in northern regions (Bradford and Cardinell, 1926; Edgerton, 1960; Rollins et al., 1962). However, late fall nitrogen appli- cations in southern regions may decrease spring cold injury by delaying bud development (Chandler, 1913; Savage et al., 1976). Fall pruning (Bradford and Cardinell, 1926; Potter, 1938; Edgerton, 1960; Nesmith and Dowler, 1976), heavy fruit load, and late fruit maturity may reduce hardiness (Chandler, 1913; Potter, 1938; Edgerton, 1960). Factors affecting recovery Environmental and physiological factors affect re- covery. A cool moist spring aids recovery, whereas hot dry weather creates additional stress which can kill the tree (Bradford and Cardinell, 1926; Potter, 1938). Nitro- gen and potassium fertilizer applications may aid recovery (Potter, 1938). Some varieties such as 'Redhaven' may recover better even though suffering more apparent injury than other cultivars (Fogle and Overley, 1954). Injured barley tissues can produce toxic substances which cause secondary degeneration of adjoining uninjured tissues (Olien and Marchetti, 1976). 9 Mechanisms of freeging injugy Histological effectg. Directly or indirectly, freezing injury involves water transitions and associations within the plant. The amount and distribution of water greatly affect freezing processes. Freezing may occur intracel- lularly or extracellularly. Intracellular freezing is a nonequilibrium process and is nearly always fatal (Mazur, 1969; Levitt, 1972). If the cell membrane is not suffi- ciently permeable, if supercooling diSplaces the system far from equilibrium, or if freezing rate is sufficiently rapid, sudden intracellular freezing may result (Mazur, 1969). Nonacclimated plants may completely lack tolerance to ice formation in their tissues, while acclimated plants can tolerate ice formation and increase in resistance to intracellular freezing (Single and Olien, 1967). The pro- toplasmic membrane provides a barrier against ice inocula- tion into the cell (Asahina, 1963). Plots of tritiated water flux from dead and living cortex cells of Cornus stolonifera verified that membranes limited water flux (McKenzie et al., 1974). During acclimation, protoplasmic permeability increases (Krasavtsev, 1967; McKenzie et al., 1974). Reduction or elimination of ice nucleating centers in the cell may also be involved in the acclimation pro- cess (Burke et al., 1976). Ice crystal growth can cause histological damage by physical disruption of tissues (Olien, 1968). Adhesion energies can cause ice to adhere 10 to regions of the protoplasmic membrane resulting in mechani- cal stress (Olien, 1974). Cytological effects. In acclimated tissues, when freezing occurs as an equilibrium process ice forms first in the extracellular spaces because of the protoplasmic mem- brane barrier (Asahina, 1967; Siminovitch, 1967; Levitt, 1972). Water is withdrawn from the cell to the growing ice nucleus as a result of an extracellular vapor pressure deficit (Levitt, 1972). Injury is associated with this de- hydration or desiccation but the exact reasons for injury are not known. Removal of water causes stiffening and coagulation of the protoplast (Siminovitch et al., 1968). Reduction of cell volume to a critical level and/or exceed- ing a maximum tolerable osmotic pressure may be important (Meryman, 1970; Williams and Williams, 1976). Cold resis- tant tissues often tolerate having a greater proportion of their water frozen than nonresistant tissues (Burke et al., 1974, 1975, 1976). Electrolytes may precipitate as solute concentration increases if their solubilities differ, and such differential precipitation may result in large pH changes (Mazur, 1969). Formation of intermolecular disul- fide bonds due to close association of protein sulfhydryl groups in dehydrated cells may cause protein precipitation (Levitt, 1962). Changes in starch, sugars, water soluble protein, RNA, lipids and other cellular components have been associated with development of freezing resistance (Siminovitch et al., 1967, 1968). Samygin (1963) found that ll hardened kale cells die without deformation of the proto- plasts, suggesting death results from dehydration rather than mechanical damage. Certainly no one mechanism can explain all types of freezing injury. Measurement of freezing processes. Several methods have been used to measure water distribution in plants. Calorimetric methods can be used to measure the amount of heat given off during freezing and thus the amount of ice forming (Krasavtsev, 1966; Olien, 1974; George and Burke, 1977a). Both exotherm analysis and differential thermal analysis also detect gross rises in temperature at points where relatively large quantities of water freeze (Quamme, 1972a, 1972b, 1975; George and Burke, 1977a). Pulsed nuclear magnetic resonance spectroscopy can be used to directly measure liquid water content as a function of temperature (Burke et al., 1974; George and Burke, 1977a). Electrophoretic methods have been used to measure the ex- tracellular water content (Olien, 1961). Using electrophoretic techniques, Olien (1961) found 3 basically different freezing patterns. Reversible equilibrium freezing develops in hardened tissues where no sudden readjustment of water content occurs. Equilibrium freezing is controlled by regulating temperature. When the system is displaced farther from equilibrium, qualitative changes occur which result in nonequilibrium freezing. Nonequilibrium freezing is associated with a sudden read» justment of liquid water between the protoplasts as freezing 12 progresses, and is not controllable by temperature regu- lation. A nonequilibrium pattern characterized by an abrupt rise in water content with freezing is typical of non- hardened plants. The rise in water content is due to rupture of the protoplasts (Olien, 1964). Water content and structural features affect the freezing pattern. Ice tends to form explosively in water not closely associated with living protoplasts. This often occurs in senile or dead cells such as xylem vessels and in boundaries between different tissue types. At high moisture levels nonequili- brium freezing may occur in all tissues (Olien, 1964). Different freezing processes can occur simultaneously in different regions of a single plant (Olien, 1967). The type of freezing process affects the size and dis- tribution of ice crystals. Fine ice structure is often found in undifferentiated tissues, while large ice masses often occur in highly differentiated tissues and at transitional zones. Injured areas often contain large perfect crystals while uninjured areas correSpond to regions containing im- perfect crystalline structures. There appear to be varietal differences in crystal size and dispersion (Olien, 1964, 1967, 1971; Olien et al., 1968). Cooling rate, minimum temperature reached, and nature and concentration of the solute determine the type of ice formation which develops (Luyet, 1970). Because small imperfect ice crystals have a high surface free energy they can recrystallize into larger, more perfectly structured crystals having a lower 13 surface free energy. Crystals do not have to be in direct contact with liquid water but will grow as long as water vapor is provided. Polysaccharide polymers extracted from cereals can affect crystal structure and distribution by competing with water for positions in the ice lattice. The specific molecular configuration of a polymer determines its effec- tiveness and is a varietal characteristic (Olien, 1965, 1967). Structural features of individual cell types and the degree of association between water and cellular components are important in determining the type of freezing pattern produced by a tissue. Water is dynamically oriented in polylayers around cellular components. The physical properties of the water vary with the distance from the interface. Fluidity and freezing point decrease as the interface is approached. During freezing, solutes, hy- drophylic substances and the intervening liquid water com- pete for positions in the ice lattice (Olien, 1973). In xylem vessels where cells are dead or senile most water is not closely associated with cellular components and non- equilibrium patterns are common (Olien, 1964; Dennis et al., 1972). In supercooled wheat stems structural features cause .freezing boundaries to be immobilized at the nodes (Single and Olien, 1967). Deep supercooling. Flower buds and xylem ray cells of some woody plants undergo deep supercooling. In such 14 cells the liquid water content is a weak function of temp- erature (Burke et al., 1974). The system is displaced far from equilibrium. Freezing occurs as an explosive non— quilibrium process and is probably intracellular (Burke et al., 1976). Azalea florets supercooled to as low as -43° during midwinter (Graham, 1971), while apple, pear, blueberry and grape xylem have exotherms between -30° and -48° (Quamme et al., 1972a, 1972b, 1973; Quamme, 1976; Pierquet et al., 1977). Twenty-five of 49 temperate woody species had exotherms between -4l° and -47° (George et al., 1974). Freezing in these tissues occurs near the homogeneous nucleation point of pure water which is around -38° to -400 (Fletcher, 1962; Rasmussen and MacKenzie, 1972). Deep super- cooling depends upon structural features of the tissues. Very hardy tissues capable of surviving freezing in liquid nitrogen do not exhibit deep supercooling (Burke et al., 1976)./ Plants without low temperature exotherms have dif- fuse porous or nonporous xylem, while plants that deep supercool tend to be ring porous.“ State, living or dead, has little effect on the low temperature exotherm, but structural integrity is important. Finely ground samples exhibit no exotherm, nor do freeze dried twigs unless they are rehydrated (Quamme et al., 1973). Thin ( 0.5 mm) xylem sections do not supercool indicating that morphologi- calgfeatures are important (Burke et al., 1976). In azalea, the floret deep supercools but adjacent stem and bud scales do not (Graham, 1971; George and Burke, 1977b). Deep 15 supercooled water may remain stable for long periods of time (George and Burke, 1977b). Therefore, some physical or thermodynamic barrier must separate this water from other cellular water, or else it would slowly migrate to exterior ice crystals due to vapor pressure differences. Moisture content has a dramatic effect on freezing patterns as well as on hardiness. In xylem, a decrease in tissue hydration allows supercooling to lower levels (George and Burke, 1977a). In barley crowns nonequilie brium freezing processes occur when water content is greater than 2.6 g of water per g dry weight (Olien, 1964). A close relationship exists between crown moisture and regrowth of frozen wheat and barley cultivars; small changes in crown moisture result in a large difference in plant survival. Different freezing processes are observed at low moisture levels (Metcalf et al., 1970). Lumis et a1. (1972) attribute differential survival of two azalea species to differences in moisture content. Splits in the vascular and cortex tissues caused by ice crystals are eliminated when stem moisture is reduced from 54 percent to 46 percent. However, in this case freezing patterns of the two species differ little despite differences in moisture content. Reducing moisture content of blueberry flower buds increases hardiness, while increasing moisture content reduces hardiness (Bittenbender and Howell, 1975). 16 Moisture content appears to be a varietal character- istic which is relatively stable throughout the winter. Chandler (1913) found no consistent differences in mois- ture content of 'Elberta' peach buds, bark and entire twigs, from November to May. Johnston (1923) found mois- ture content of 'Elberta', 'Late Crawford' and 'Greensboro' peach flower buds to be negatively correlated with hardiness. Moisture content was stable from September to mid-November, rose slightly and remained stable until mid-February, then it slowly began to rise as buds developed. Moisture content of 2- to 3-year-old apple twig tissues is not constant for individual complex tissues (bark, outer xylem, inner xylem, etc.) from December to January but the total twig moisture content is practically constant (Traub, 1927). Winter moisture content of the hardy 'Duchess' apple remains practically constant while that of the less hardy 'Jonathan' shows a tendency to fluctuate more (Hildreth, 1926). Wildung et a1. (1973) find apple root hardiness is negatively correlated with soil temperature and moisture. Stem moisture content of azaleas remains relatively constant during midwinter (Lumis et al., 1972). In contrast, moisture content of plum twigs fluc- tuates markedly during the winter and is positively corre- lated with air temperature at collection. Artificial de- hydration results in greater water loss from semihardy twigs than from hardy twigs (Strausbaugh, 1921). l7 Genetics of cold hardiness Most knowledge of the genetics of cold hardiness is empirical and has been.gained by observation of cultivars after test winters. Because of the dynamic nature of cold hardiness, genotype by environment interactions are impor- tant. As cited earlier, different genotypes may acclimate and deacclimate at different rates and some may perform well in some climates but not in others. To be useful, genetic potential for cold hardiness must be expressed phenotypi- cally in a wide variety of environmental conditions occurring over many seasons. Tissue by genotype interactions may also be important Some cultivars, e.g. 'Elberta', are wood hardy but lack bud hardiness, while others, e.g. PChinese Cling', arebud hardy but very tender in wood (Chandler, 1913). Both Blake (1938) and Layne et a1. (1977) find that hardiness of the lower trunk and bark are not always correlated with hardiness of peach flower buds. However, phloem, cambium and xylem injury are highly correlated with each other (Layne et al., 1977). Flower bud injury in interspecific plum hybrids is correlated with wood injury and twig dieback. No individuals have hardy flower buds and tender bark or wood (Dorsey and Bushnell, 1925). Inheritance of cold hardiness in wheat is quantitative in nature. Both high and low intensity freezing are controlled by partially dominant genes which are mostly additive in 18 their effect. Changes in relative ranking of some varieties under different freeze intensities indicate different genes may operate under different freezing conditions. Broad and narrow-sense heritabilities are estimated as 77.09 and 63.90 percent respectively. No maternal dif- ferences have been found (Gullord, 1974). Some progenies of interspecific plum hybrids show bi- modal groupings (Dorsey and Bushnell, 1925). When both parents are hardy nearly all the progeny are as hardy as the parents. When one parent is hardy a large proportion of the F1 offspring are hardy. Winter hardiness is dominant in crosses between cultivated and hardy wild strawberries (Hildreth and Powers, 1941). Lapins (1962) finds injury to apple seedlings is approximately normally distributed. One progeny gives some evidence of maternal influence or specific combining ability. Morrison et al. (1963) rate hardiness and vigor of over 130,000 apple seedlings from crosses involving 24 parents. Progenies were grown at 8 locations in Alberta, Saskatchewan and Manitoba. No pro- geny was outstanding at all locations. Some trees of all progenies had satisfactory ratings. They suggest that selection of parents should be based on average performance of their progenies. Mowry (1964) finds peach flower bud hardiness to be quantitatively inherited. Progenies generally have normal distributions but 'Redskin' progenies are skewed, indicating that 'Redskin' may have prepotency for l9 transmitting hardiness. Hardier parents transmit more hardi- ness than tender parents, and Mowry suggests that parents be chosen on the basis of their individual performance. Watkins and Spangelo (1970) have studied inheritance of 5 traits and 2 composite traits contributing to plant sur- vival in 2 apple dialleles. Additive genetic variance accounts for 59 percent to 100 percent of the total genetic variance. Fejer (1976), using other apple dialleles. finds Specific combining ability to be slightly higher than reported by Watkins and Spangelo, but additive variance is still high. Watkins and Spangelo (1970) conclude that breed- ing progress can be achieved by exploiting additive genetic variance, total genetic variance or by progeny selection. Potential parents can be screened phenotypically and so eliminate the need for genetic evaluation. SECTION I COMPARISON OF RELATIVE FREEZING INJURY IN 'VELVET', 'REDHAVEN' AND 'SIBERIAN C' PEACHES FOLLOWING CONTROLLED FREEZER TESTS AT SELECTED DATES DURING TWO WINTERS INTRODUCTION Phenological response of a plant in a specific environment is an important aspect of hardiness. As stated earlier, many North American peach breeders are interested in utilizing 'Siberian C' as a parent in root- stock and scion cultivar breeding programs. Much is known about its fall and midwinter influence on increas- ing wood and flower bud hardiness of budded scion cultivars. However, little is known concerning its behavior as a scion cultivar, especially during late winter and spring. Nur- serymen and growers have expressed concern about its spring hardiness status, because as a scion cultivar under northern conditions it blooms earlier than most commercial cultivars. This study was undertaken to determine the hardiness of 'Siberian 0' relative to a medium hardy and a tender commercial cultivar over the course of the dormant season, and also to determine the reasons for its superior hardiness. 20 21 MATERIALS AND METHODS Plant Material 'Velvet', 'Redhaven' and 'Siberian C' peaches were selected to represent tender, medium hardy and very hardy genotypes, respectively, with regard to field observations of tree hardiness. During the 1975-76 winter (year 1) the 3 cultivars were tested on 7 dates. During the 1976-77 winter (year 2) only 'Redhaven' and 'Siberian C' were tested because the 'Velvet' orchard had been removed. Six tests were conducted in year 2. All trees were in commercial orchards near Hartford, Michigan. 'Velvet' and 'Redhaven' were located on comparable sites within one-half mile of each other. The ’Siberian C' seed orchard was within a mile of the other orchards. Preceding year 1, 'Velvet' and 'Redhaven' bore a nearly full crop. 'Siber- ian C' bore a heavy crop and was not thinned, which re- duced the amount of vegetative growth. In year 2, 'Red- haven' bore a very light crop while 'Siberian 0' had no crop due to spring frosts. In year 2 'Siberian C’ grew more than 'Redhaven'. 22 Sample Preparation Current season twigs 30 to 70 cm long were removed from the upper exposed portion of randomly chosen trees (Cain and Andersen, 1976). Excised twigs were placed into plastic bags contained in a large plastic can. During midwinter twigs were packed with snow or ice to prevent excessive warming during transport to the laboratory. Twigs were usually frozen within 5 to 15 hours after ex- cision. Before freezing, 20-twig groups of each cultivar were placed in plastic bags with the basal ends left ex- posed. Four to 9 such samples were frozen at each date (Table 1). All samples were placed in a programmable walk-in freezer. Temperature change was controlled by either an evaporator pressure regulator (EPR) valve acti- vated by 2 time clocks or by a Partlow cam-controlled sole- noid valve. In year 1 only the Partlow system was used. During year 2 the EPR system was utilized, as it gave a smoother temperature decrease and reduced temperature fluctuations associated with cycling of the solenoid system. Fans inside the freezer prevented air stratifi- cation. FreezingProceggre The general freezing procedure was to lay the pre- pared samples on a wooden rack in the chamber. The exposed 23 Table 1. Collection dates, injury rating dates, number of temperatures used and temperature ranges used for comparing relative hardiness of 'Velvet‘,‘Redhaven‘ and ‘Siberian C' peach twigs during 2 winters. Date of Date of injury Number of Temperature collection evaluation temperatures range Maximum Minimum 1225:2é Nov 29 Dec 7 6 -23.3 -4o.6 Dec 12 Dec 17 7 ~23.3 -4l.l Jan 1 - 6 -2601 ‘LFOQ 6 Feb 7 " 6 -2601 '4006 Mar 8 mar 19 9 -1708 -u0.0 Apr 10 Apr 18 8 -3.9 ~23.3 1226—22 Oct 3 Oct 12 6 -6.7 -21.1 Nov 5 Nov 11 6 -l7.8 -32.8 Jan 6 Jan 16 8 Fieldy -26.1 -43.9 Mar '7 Mar 15. 8 Field ~23.3 -38.5 mar 26 - 9 Field -21.7 - 0.0 Apr 13 Apr 18 8 Field -3.9 -20.0 zExact date not known yField control 24 twig bases were covered with snow or crushed ice to induce ice nucleation and prevent excessive supercooling. All samples were allowed to equilibrate at -2.20C for 2 to 4 hours before the temperature was decreased at 1.70 to 2.80 per hour. Samples were held for at least 2 hours and occasionally up to 10 hours at each test temperature. Generally samples were removed at 2.80 intervals. Twig temperature was monitored with 24-gauge copper-constantan thermocouples inserted into the twigs and attached to a Honeywell Electronik Potentiometer. Chamber air tempera- ture was recorded on a Partlow recorder and checked against a low temperature alcohol thermometer. Generally there was little difference between air temperature and twig tempera- ture. Twenty-four to seventy-two hours were required to reach the minimum temperature. Frozen samples were held at 00 chamber until all samples appeared to be thawed. Thawed samples were left in warm humid conditions to oxidize for 2 to 10 days. Preliminary results indicated tissue browning was complete after 48 hours and no visible changes could be detected for several weeks if twigs re- mained under cool humid conditions to prevent desiccation and growth of microorganisms. Tissue browningratingg A thin cross section of each twig was rated separately for browning of cambium and xylem to ascertain the relative 25 level of injury in each cultivar. Cambium was rated on a 1 to 5 scale (Cain and Andersen, 1976). Xylem ratings used were: 1- no injury; 2- slight browning of the xylem near the pith; 3- no more than half the area brown; 4- completely brown except at the outer edge adjoining the cambium region; 5- completely brown and considered dead. Cultivars were com- pared by averaging injury across temperatures at which at least one cultivar had a rating greater than 1.0 or less than 5.0. Calluging and TTC Tests In addition to browning ratings, other viability tests were performed on selected dates to check the reliability of the browning ratings. After assigning browning ratings to twigs on sampling dates December 10, 1975, March 7, 1977 and April 13, 1977 the twigs were dipped in a 100 ppm beno- myl solution and the bases were placed into sand in a green- house with intermittent misting to prevent desiccation. When sufficient callus had formed, usually 23 to 35 days after initial harvest, each twig was rated for extent of callusing based on a 1 to 5 scale, 1 indicating prolific callus, 5 indicating death. Data were analyzed in a manner similar to browning data. A tetrazolium chloride (TTC) test was performed on 2.0 to 5.0 mm twig sections removed from basal portions of twigs used on March 7, 1977. The twig sections were placed 26 in test tubes and covered with approximately 2 ml of a 0.6% W/V 2,3,5-triphenyl-2H-tetrazolium chloride solution containing 0.05 M Na2P04-KH2P04 buffer (pH 7.4) and 0.05% Y/V ortho X-77 wetting agent. Tubes were placed in a bell jar and vacuum infiltrated at 24°C for 15 hours. Five to 6 twig sections were used per cultivar and temperature. Infiltrated twig sections were dissected and examined microscopically. Extent, intensity and location of red coloration were recorded, but the amount of TTC was not quantified spectrophotometrically. Moisture Content of Excised Twig Internodes At each test date moisture content of each cultivar was measured. The basal 3 to 4 internodes were discarded and the next 4 internodes were used as the sample because this same region was used for browning determinations. The nodal sections with vegetative and flower buds were discarded. Five twigs per replicate were used. Three to 9 replicates were used per date (Table 4). In year 1 the sections were placed in cork stoppered test tubes during weighing; during drying they were transferred to weighing tins. Because of excessive moisture changes during trans- fer and weighing, in year 2 twigs were placed into glass weighingjars whose covers were sealed with silicone grease. Twigs were sectioned at 4.40 to reduce moisture loss during sectioning. After recording fresh weight,samples were , 'EHIIILIIirIIIIInI 27 dried at 700 to constant weight. Moisture content was expressed as percent dry weight. Data were analyzed within dates via analysis of variance (AOV). In year 1 means were compared using Dun- can's Multiple Range or its approximation when there were uneven replicates (Steel and Torrie, 1960). Two treatment contrasts (year 2) were compared using the F values from the AOV. Browning injury was correlated with moisture content. Moigture Content of Bark and Xylem To determine if differences in tissue moisture content could account for differences in injury, moisture content of bark and xylem tissues of 'Redhaven' and 'Siberian C' was determined on October 3, 1976, January 6, 1977 and March 7, 1977. Basal internode sections were used as in whole twig moisture determinations. The bark and xylem were separated and placed into weighing jars. On October 3, 1976, 3 replicates per cultivar were used while on January 6, 1977 and March 7, 1977, 4 replicates were used. Effect of Varying Moisture Content on Tissue Harginegg On November 5, 1976 samples of 'Redhaven' and 'Siberian C' were placed under 2 moisture and 2 temperature regimes in a factorial design to determine the effect that these 28 environmental variables had on hardiness. For the low moisture treatment, twigs were kept sealed in dry plastic bags. High moisture twigs were allowed to stand in about 5 cm of water for 24 hours. High and low test temperatures were 27.300 and 1.50 respectively. Fifteen twigs of each cultivar per treatment combination were frozen to -26.l°. Inner bark and xylem were rated via browning. Another 15 twigs of each treatment combination were used for moisture determination. Four basal internodes from each of 5 twigs constituted a replicate, giving 3 replicates per cultivar per treatment combination. Differences among means were tested using Duncan's multiple range at p = 0.05. 29 RESULTS Cultivar browning ratings compared at 13 dates Inner bark and xylem ratings of 'Siberian C' were much lower than comparable ratings of either 'Redhaven' or 'Velvet' at all 13 test dates during both winters (Figure 1). Average injury to 'Velvet' was slightly higher than 'Redhaven' for both tissues at all dates during year 1, except for inner bark on December 10, 1975, when both means were 1.5 (Figure 1). Inner bark ratings for 'Velvet' were 0.0 to 0.3 units higher than for 'Redhaven' and 0.6 to 2.1 units more than 'Siberian C'. Xylem ratings for 'Velvet' ranged from 0.1 to 0.5 and 0.6 to 1.7 units above 'Redhaven' and 'Siberian C' respectively. 'Redhaven' inner bark was 0.3 to 1.8 units greater than 'Siberian C' while its xylem was 0.2 to 1.9 units higher. In year 1, for both inner bark and xylem, the greatest difference between 'Siberian C' and the other cultivars occurred during the fall, especially in November. Relative differences were much more variable in xylem than in bark tissues. In year 2, differences between 'Redhaven' and 'Siberian C' were more uniform for both tissues. Injury ratings to both tissues in year 1 were generally 30 Figure 1. Mean inner bark (A) and xylem (B) injury ratings of 'Velvet', 'Redhaven' and 'Siberian C' averaged over test temperatures where at least 1 cultivar had a mean injury rating greater than 1.0 and less than 5.0. 31 U : Volvo! 'Redhavon ' 1. 7‘37“ .. ..u Pl A . 6 9 "““‘ 2 7"."" r n c M . ""“‘ n 4 "“‘y“ I11 . u M . ‘3‘3‘ 6 J '7’" ' n “a u _ 5 9 "‘ n 4. 5".“ u ... 3 2 "““““’ o. 2. i"“““ m ... 7“ w 7“ r J - 33333....3... Mm .................. "‘ I I," - --Huuu u n ..J .. ...... I -.. M . c 0 ll 7‘ 7 M D. "“” b a a m C n I: uuuuu .N fl D O n n n C v‘ m 2 .0. M u "I‘ V n .33 a 3 o ... 33.33.33.333 in. .33....3 J _ w R s u ... 3 I. m ' 1°- . ll” I‘ll s C ..u. nnnnnnn r! ...-33.3.... no. . ............... 9 3"! 2 3 """ v Q ...............3............................ N. m. 3....................................... nu "“" 9 7 7"““" v o. ......3......3.33.33.33.33.33.......... m .... ........33.333.333....333.333.33.-.... 4 3 2 .l 5 4 3 2 1 >5?— xcam 3::— >.:.:. Eaton Apr 13 -13.!) Mcr26 -26.3 Mar 7 '27.5 Jan 6 -30.3 On 3 Nays _|2.2 -24.9 Apr IO -l7.7 Mar O -26.1 Data Temp (0c) Feb 7 -30.3 Jan IO -30.3 New 29 00: IO -33.l ~30] Nov 9 - 28.5 32 lower than in year 2 (Figure 1, Appendix A). In year 2, field injury occurred in both tissues prior to January 6, 1977. Overall, inner bark tissues sustained less damage than xylem tissues. Xylem of all cultivars was always killed at temperatures below -37.2°C (Appendix A-Table 1). When fully acclimated, inner bark often suffered only slight damage at -41.l° (Appendix A). During both winters, tissues of all genotypes did not deacclimate as rapidly as expected. On both April 10, 1976 and April 13, 1977, flower buds were in pink tip, and developing leaves were about 1.3 cm long. Bark and xylem were assumed to be fully active. However, the average level of injury to both tissues was low (Figures 1A and B). In fact, tissues of all genotypes suffered only minor to moderate damage at -23.30 (Appendix A-Table 1). Shoots of all cultivars had partially defoliated on November 9, 1975 with 'Siberian C' being somewhat more advanced. At this time, inner bark of 'Velvet' and 'Red- haven' were completely killed at -33.3OC (Figure 2), while 'Siberian C' already exhibited considerable resistance to temperatures as low as -37.2°. During midwinter, all cul- tivars exhibited greater resistance. 'Velvet' and 'Red- haven' suffered moderate injury, while 'Siberian C' was only slightly injured at -41.10, which was the lower limit of the freezing chamber. Although the magnitude of the differences between 'Velvet' and 'Redhaven' compared with 'Siberian C' decreased during midwinter, 'Siberian C' 33 always exhibited less injury. During fall, hardiness of the xylem was quite similar to that of the inner bark (Figures 2A and B). 'Velvet' and 'Redhaven' had considerable damage at -23.3OC and were killed by -28.90. 'Siberian C' was not injured at -23.3°, but was killed between -33.30 and -37.20 (Figure ZB). By midwinter, initial injury of all cultivars occurred at lower temperatures. Although a considerable difference was still apparent in temperature of initial injury, 'Siberian C' had a killing temperature only slightly lower than that of 'Velvet' or 'Redhaven'. In midwinter, bark differed considerably from xylem in response to temperature» While injury changed little with temperature in the inner bark, xylem still exhibited an abrupt increase in injury within a narrow temperature range (Figures 2A and B). Callusing and TTC tests Proliferation of callus about the cut base of the twig was used as another criterion of viability. Figures used to obtain the correlations were mean injury ratings at each test temperature for each date. Correlation coefficients between browning ratings and callus ratings for each culti- var and the overall correlation were significant at the 0.05 level (Table 2). The correlations indicate good agreement between the 2 methods of injury assessment. Callus tests provided additional proof that inner bark tissues were 34 Figure 2. Changes in temperature response patterns of inner bark and xylem of 3 peach cultivars from late fall to winter. Inner Batk hfiury Injury Xylenz 35 EVolvot 'Rodhavon I ‘Siborionc f Fall Nov 9.1975 w Wintot Doc 10,1975 5 4 3 EVOWot Iflodhovon ‘Siberian C fFoll Nov 9,1975 w Wintot Doc10.1975 -4OJ) Table 2. of 3 peach cultivars. 36 Correlations between callus proliferation and ° inner bark browning ratings following freezing of twigs Date Velvet Redhaven Siberian 0 Overall 1‘ Hz r n r n r n 12-10-75 .94** 7 .86* 7 .891” 7 .69** 21 ** 3-7-77 - - .98** 7 .84* 7 .93 in 4-13-77 - - .85* 6 .89H 7 .83** 13 Z *Significant at 5% **Signifcant at 1% n is the number of observations involved, r is the corre- lation coefficient. 37 alive at -4l.1oC. 'Redhaven' forms more callus than 'Siberian C' when both are uninjured (Figure 3), but pro- gressively less than 'Siberian C' as temperature decreases. At the lowest temperatures, 'Redhaven' produces no callus, while 'Siberian C' continues to produce some even at -410. Thus, inherent callusing capacity is important in comparing genotypes. A TTC reduction test performed on twigs from the March 7, 1977 sampling date provided another method of injury assessment (Table 3). The extent of TTC reduction (red coloration) agreed closely with browning ratings. Because TTC tests were not conducted until March 17, 1977, tissues had already undergone oxidative browning. The brown pigments reduced the ability to discern faint differ- ences in pink coloration, especially in more severely injured tissues. Lack of red color in the xylem at -31.7OC to -34.40 indicated that death occurred in this range. This is the same range in which xylem death occurs as assessed via browning (Figure 2B). Inner bark tissues showed moderate red color at -38.50, indicating that they were still alive but had been injured. These observations agree with browning and callusing results. Relationship between moisture content and injury Moisture content of basal internodes was measured at each test date. Relative cultivar order within each date 38 Figure 3. Callusing of 'Redhaven'(1eft in each set of 6) and 'Siberian 0' (right) collected on Nov. 5, 1877 following exposure to stress temperatures ( C) 8f (upper left t8 lower right): -lg.8°, -2o.6 , —23.3°, —26.1 , -28.9° and -32.8 . Table 3 . 39 Color of ‘Redhaven' and 'Siberian C' twig tissues treated with triphenyl tetrazolium chloride following controlled freezing on March 7, 1977. Tissue Temperature Cultivar Cortex Outer Cambium Xylem (0c) phloem region F ield . Redhaven Dark redz Dark red Dark red Light pink c ontrol Siberian 0 Dark red Dark red Dark red Light pink - 2 3 . 3 " Redhaven Dark red Dark red Dark red Light pink Siberian C Dark red Dark red Dark red Red -26.l Redhaven Dark red Pink, some Dark red Light pink browning some brown Siberian C Dark red Dark red Dark red Red -28.9 Redhaven Dark red Pink to Dark red Pink near brown cambium Siberian C Dark red Red ‘~ Dark red Light pink -3l . 7 Redhaven Red Brown Red to Brown light red Siberian C Red Red Dark red Brown, pink at cambium -34 . 4 ‘ Redhaven Dull Brown Light pink Brown brown to dark red Siberian C Brown to Pink to Dark red Brown light pink brown -36 . 7 Redhaven Brown Brown Dark red Brown near xylem only Siberian C Light red Brown Dark red Brown to brown - 38 . 5 Redhaven Brown Brown Red near Brown xylem only Siberian C Brown Brown Dark red Brown with red to areas light pink ZLighter red and more browning indicate greater injury . 40 is quite consistent, 'Velvet' having the highest moisture content, 'Redhaven' being intermediate and 'Siberian C' having the lowest moisture content (Table 4).) During year 1, moisture change occurred as twig sections were trans- ferred between stoppered test tubes and weighing tins. In year 2, glass weighing jars were used; as a result, between date differences were reduced. On certain dates, e.g. November 29, 1975, twigs were collected while wet from rain,snow or frost. Much of the external water was wiped off but external residues may have masked internal moisture content differences. However, precipitation and tempera- tures preceding and during collection did not explain all such differences. Within test dates, a strong relationship exists be- tween moisture content and cultivar injury ratings (Table 5). Each correlation coefficient is based on only 3 observa- tions, thus, only 1 degree of freedom. Correlations were not calculated for year 2 because only 2 cultivars were available. However, comparison of moisture content data in year 2 (Table 4) with injury ratings (Figure 1) shows the same consistent relationship; 'Redhaven' was less hardy than 'Siberian C' and generally had a higher tissue moisture content. Correlations between xylem injury and moisture content were usually slightly higher than those between inner bark injury and moisture (Table 5). 41 Table 4. iMoisture contentZ of peach twigs at each test date. Cultivar Date velvet Redhaven Siberian C x ny x n i n x, 1975—75 6 NOV 9 4 78.01a 4 74.84b 4 63.81c Nov 29 4 96.49a 4 93.27a 4 93.42a Dec 12 4 88.11a 5 86.30b 5 84.610 Jan 1 3 93.53a 5 91.23b 3 87.52c Feb 7 5 87.45a 5 82.48ab 5 79.72b Mar 8 w 5 91.68a 5 88.08b 5 85.54c 1976-77 Oct 3 10 80.65a lO 79.04b NOV 5 6 81.193 6 81.24a Jan 6 9 83.97a 9 80.55b Mar 7 9 85.53a 9 79.84b Mar 26 9 80.37a 9 80.02a Apr 13 9 95.24a 9 93.10a zPercent dry wt yNumber of replicates XWithin rows, means followed by the same letter are not significantly different at P 05 by Duncanfls multiple range test. ' wWithin rows, means fellowed by the same letter are not significantly different by F 05. 42 Table 5. Correlations between internodal moisture content and mean tissue injury ratings of all trees for each.sampling date during 1975-76. Date Cambium Xylem * NOV 9 .997 .990 Nov 29 .602 .685 Dec 12 .860 .904 Jan 10 .996 .989 Feb 7 .909 .969 Mar 8 .872 .918 *Significant at P 05 Table 6. iMoisture content2 of bark and xylem of 'Redhaven' and 'Siberian C' at selected dates in 1976-77. Tissue Date Cultivar Bark xylem Oct 3y Redhaven 124.70a 64.620 Siberian C 115.10b 67.98c Jan 6 Redhaven 122.23a 56.99d Siberian C 116.14b 60.65c Mar 7 Redhaven 131.77a 53.930 Siberian.C 114.00b 54.46c zPercent dry wt yWithin dates, means followed by the same letter are not significantly different at P 05 by Duncan's multiple range test. 43 Determination of bark and xylem moisture On 3 selected dates during year 2 , moisture content was determined separately for bark and xylem tissues (Table 6). Bark tissues of 'Redhaven' contained signifi-- cantly more water than those of 'Siberian C' at all 3 dates. 'Redhaven' xylem tissues had slightly, but not signifi- cantly, lower moisture content than 'Siberian C'. Moisture content of bark tissues was approximately twice that of the xylem. Effect of temperature and moisture adjustment on hardiness Both temperature and moisture pretreatment signifi- cantly affected hardiness (Table 7). Twigs took up more water at high temperature than at low temperature, and 'Redhaven' took up more than 'Siberian C' at both tempera- tures. Twigs stored at high temperature suffered more injury than those stored at low temperature. 'Siberian C' xylem suffered significantly less injury than fRedhaven', but moisture content did not affect xylem injury. Cambium showed significant temperature by cultivar, and moisture by cultivar interactions. High storage temperature increased injury more in"Siberian C' than in 'Redhaven', while soake ing increased injury more in 'Redhaven' than in 'Siberian C'. ’Siberian C' xylem pretreated at high temperature and moisture had less injury than 'Redhaven' pretreated at low 44 Table 7. Effect of moisture and temperature pretreatments on moisture content and injury to 'Redhaven' and 'Siberian C' inner bark and xylem following freezing to -26.l°C on November 5, 1976. Pre-treatment Injury rating Moisture content Cultivar Temp Moisture Inner bark Xylem (% d.w.) Redhaven 0.5 high 4.313z 4.5c 90.78c low 3.0d 4.3c 82.lla 27.6 high 5.0g 5.0d 98.99e low 3.6e 5.0d 81.81a Siberian C 0.5 high 1.9b 1.6a 89.04b low 1.3a 1.8a 81.78a 27.6 high 3.1d 2.2b 95.87d low 2.3c 1.9ab 81.21a Main effects Effect: Redhaven 4.0aY 4.7a 88.42a Siberian C 2.2b 1.9b 86.98b 0.5 2.6a 3.0a 85.93a 27.6 3.5b 3.5b 89.47b high 3.6a 3.3a 93.67a low 2.6b 3.3a 81.73b zWithin columns, means followed by the same letter are not significantly different at P 05 by Duncan‘s multiple range test. ° yWithin each column of each main effect, means followed by the same letter are not significantly different at F 05. 45 temperature and moisture. Similarly treated inner bark of 'Siberian C' had no more injury than 'Redhaven' bark held at low temperature and moisture, even though twig moisture content of 'Siberian C' was 13.8 percent higher. 46 DISCUSSION Seasonal hardiness compariggng At all test dates inner bark and xylem of 'Siberian C' suffered less injury than similar tissues of 'Redhaven’ or 'Velvet'. This supports conclusions by Ormrod and Layne (1974) and Layne (1974) that 'Siberian C' is relatively cold hardy compared to other rootstocks and scion cultivars grown under Canadian conditions. The average injury to 'Velvet' was slightly greater than for 'Redhaven'. This agrees with the general description by peach growers that 'Velvet' is a ”tender" tree while 'Redhaven' is "hardy", although not the hardiest of the commercial types. The relationship between temperature decline and injury changes as plants acclimate from late fall to mid- winter. Fall temperature response patterns of inner bark and xylem are quite similar (Figures 2A and B). They are characterized by a rapid rise in injury over a narrow temperature decline. In contrast, winter temperature re- sponse patterns of the inner bark and xylem are quite dif- ferent (Figures 2A and B). The major change in the xylem pattern is a shift in the overall pattern to a lower series of temperatures. Injury still increases rapidly over a 47 narrow temperature range. The inner bark also shows a downward shift in the temperature at which initial injury occurs but injury levels rise very slowly with declining temperatures. The characteristic changes in temperature response patterns may indicate that different stresses are develop- ing in the inner bark and xylem at different dates. The rather abrupt killing of the partially acclimated inner bark and xylem indicate direct injury by ice. In non- acclimated or partially acclimated wheat, freezing often occurs in a nonequilibrium pattern. In such cases proto- plasts are disrupted by crystallization, causing irreversi- ble damage (Olien. 1964).. Intracellular freezing may occur in such a situation and is nearly always fatal (Mazur, 1969). Xylem ray cells of many woody plants including black cherry (Prunus serotina Ehrh.) exhibit deep supercooling (Quamme, 1972a, 1972b. 1976; George et al., 1974; George and Burke, 1977a). Freezing occurs near the homogenous nu- cleation point of water, which is approximately -38 to -40°C (Fletcher, 1962; Rasmussen and MacKenzie, 1972). The abrupt rise of xylem injury over a narrow temperature decline and the fact that xylem of all cultivars never survived temperatures below -37.2° (Appendix A-Table 1) suggest that peach xylem undergoes deep supercooling. The slow increase in cambium injury with temperature decline suggests that the cambium is undergoing equilibrium freezing. During an equilibrium freeze ice formation occurs fi“. 48 extracellularly because the cell membrane provides an effective barrier against intracellular ice inoculation (Asahina, 1963). Water is withdrawn from the cell to extracellular ice nuclei near the cell wa11,due to an extracellular vapor pressure deficit. This results in protoplasmic dehydration (Asahina, 1963; Krasavtsev, 1966; Mazur, 1969; Levitt, 1972). Injury is associated with protoplasmic dehydration (Burke, 1974). Cold resistant tissues seem to tolerate having a greater proportion of their water frozen than non-resistant tissues (Burke et al., 1975). Inner bark injury during midwinter may be due to desiccation effects rather than direct effects of ice for- mation. 'Siberian C' may be able to tolerate removal of greater amounts of its cellular water than 'Velvet' or 'Redhaven'. or the temperature for removal of a critical amount of water may be lower for 'Siberian C' than for 'Velvet' or 'Redhaven'. Hardiness of inner bark was unexpectedly high. especially during year 1 (Figure 1A). Viability, however, was verified by callus regrowth (Table 2) and TTC tests (Table 3). Wood injury and even complete tree death often occur at warmer temperatures in the field (Chandler, 1913; Bradford and Cardinell, 1926; Fogle and Overley, 1954). However, under some conditions commercial culti- vars have survived field temperatures as low as -35.6°C (Campbell, 1948) and hardy Chinese germplasm has survived -38° (Pieniazek et al., 1968). However, temperature alone 49 gives no indication of the types of stresses developing in the plant (Olien, 1974a). Thus. absolute comparisons between injury levels to specific tissues produced in dif- ferent situations should be interpreted with caution. Still, relative ranking of the three cultivars according to freezer tests agree with field performance rankings. The greater injury in year 2 as compared with year 1 may have been a result of the field injury occurring in year 2. This injury may have made the tissues more susceptible to damage during subsequent freezer tests. Olien (1974b) has shown that cyclic freezing and thawing greatly increases injury from ice crystal growth, which raises the killing point of barley crowns. Relationship between moisture content and injury Although tissues of the hardy cultivar consistently contained less water, seasonal hardiness patterns within cultivars did not parallel moisture content (Table 4). Changes in moisture content did not follow any discernible pattern from late October to mid April. Moisture content of each cultivar remains nearly constant throughout the dormant season (Table 4). Chandler (1913) likewise found moisture content of 'Elberta' twigs to be fairly constant throughout the winter. His moisture estimates were similar but slightly higher than moisture estimates in Table 4. 50 However, he included buds, used a different cultivar and used a higher drying temperature. His bark moisture estimates are very similar to bark moisture estimates in Table 6. Stem moisture content of two azalea clones remained rela- tively constant during midwinter (Lumis et al., 1972). Winter moisture content of the hardy 'Duchess of Oldenburg' apple was constant but the less hardy 'Jonathan' showed much greater fluctuation (Hildreth. 1926). Maximum difference in water content between cultivars was 14 percent CVelvet' and 'Siberian C' on November 9, 1975) but averaged 6.8 percent. 'Redhaven' averaged 3 percent higher than 'Siberian C' and 3 percent less than 'Velvet'. Such differences may be small compared to the total mois- ture content but can be very important biologically. Gullord (1974),using 9 wheat cultivars with a maximum dif- ference in content of 3.6 percent, showed that moisture content explained 68.9 and 72.3 percent of the variation in freezing hardiness during high and low intensity freezes, respectively. Moisture content was a heritable trait. Metcalf et al. (1970) demonstrated that a small increase in the percent crown moisture could result in a large decrease in survival of wheat and barley. Differential stem injury of 2 azalea clones was correlated with water content (Lumis et al., 1972). 51 Differences in freezing characteristics of bark and xylem The lower moisture content of'Siberian C' bark (Table 6) may be an important factor contributing to its superior hardiness,since, as pointed out above, small differences in moisture among cultivars may affect injury. Xylem behavior is more difficult to explain. Even though water content of the xylem was slightly, but not significantly, lower in 'Redhaven' than in 'Siberian C' (Table 6), 'Siberian C' xylem had less injury than 'Red- haven'. Location of this water within the xylem is not known. Differences in amount of water in mature xylem vessels may not be critical. This water freezes a few degrees below 0°C (Kitaura, 1967), but this freezing appears to cause no damage. In acclimated peach xylem, injury occurs only at low temperatures, indicating that xylem ray cells are deep supercooling. Moisture content may have minimal influence on the deep supercooling process. Effectiveness of the protoplasmic membrane as a barrier against ice growth into the cytoplasm (Asahina, 1963) might be more important. or xylem ray cell volume may be important, since the temperature at which homogenous nucleation occurs is a function of droplet size (Fletcher, 1962). Concentration and type of solutes present may also affect nucleation temperature. 52 Artificially increasing twig moisture content sig- nificantly increased injury to inner bark but not to xylem. Both temperature and cultivar affected moisture uptake and injury. Freezing stress is affected by the rate of freezing, the amountof water involved and the resultant amount of free energy released (Olien, 1971,1973). High freezing rates and moisture contents increase the probability that nonequilibrium freezing will occur (Olien, 1974). Moisture distribution and redistribution during freezing were not studied in this experiment. Much of the absorbed water probably occurred in the large xylem vessels as bulk water, only weakly associated with cellular constituents. 'Siberian C' scions defoliate and acclimate earlier in the fall and transmit these characteristics to budded scions. It also promotes fall and midwinter flower bud hardiness of budded scion cultivars (Layne et al., 1977). In Ontario, field survival of scions budded to 'Siberian C' seedlings was superior to those budded onto more tender stocks (Layne et al., 1976; Ormrod and Layne, 1977). The present study confirms the superior wood hardiness of 'Siberian 0' compared to 'Velvet' and 'Redhaven' during fall and midwinter. ’Siberian C' and other peaches from northern China evolved in a climate with long, consistently cold winters. Zagaja (1974) expressed concern that the trees might be susceptible to fluctuating winter and spring temperatures. 53 Continued observation of this Chinese germplasm indicates it has a chilling requirement comparable to most commercial types (S.W. Zagaja, personal communication). Since 'Siberian C', under northern conditions, blooms earlier than most cultivars, growers and nurserymen have been concerned that this stock and scions budded to it may be susceptible to flower bud and wood injury in early spring. Therefore, it was not known if the superior fall and midwinter wood hardiness of 'Siberian C' would be exhibited during late winter and spring. In this study, 'Siberian C' continued to have less injury than 'Velvet' or 'Redhaven' up to pink tip (Figure 1). Both inner bark and xylem of 'Siberian C' had less injury than 'Velvet' or 'Redhaven' on April 10, 1976 and April 13, 1977. Callus regrowth verified browning results on April 13, 1977 (Appendix A-Table 2). Injury to inner bark and xylem was uneXpectedly low at both dates, for all cultivars were expected to have deacclimated to a greater extent. Similarly, Howell (1970) found that apple bark remained somewhat hardy even during bud and leaf eXpansion. Blake and Steelman (1945) reported that peach flower buds in medium pink stage survived ~12.8°C. In Missouri, Chand- ler (1913) found no wood injury to ‘Elberta' twigs frozen to -24.7° on April 5, 1913. The reasons for survival at such low temperatures in early spring are not known. 54 This study indicates that 'Siberian C' is hardier than commercial cultivars in the spring. Also, 'Siberian C‘ has not caused early blooming of scion cultivars in rootstock trials at South Haven, Michigan (unpublished data) or at Harrow, Canada (R.E.C. Layne, personal communi- cation); in fact, reports from Georgia and California in- dicate that 'Siberian C' delays bloom of budded scion cul- tivars (S. Dowd and W. Krause, personal communication). Thus, 'Siberian C' does not appear to be unduly susceptible to fluctuating late winter and spring temperatures, and in- deed may be more tolerant than other rootstocks and scion cultivars. Summary 'Siberian C' was hardier than 'Velvet' or ’Redhaven' at 13 dates during 2 winters. Within test dates, cultivar injury was positively correlated with moisture content. Moisture content of bark was twice that of the xylem: how- ever, xylem suffered more injury than bark. Differential response of fully acclimated bark and xylem to temperature decrease suggested that bark underwent equilibrium freezing while xylem deep supercooled Because of its distinctly different geographic origin and its consistently superior freezing resistance compared to commercial genotypes, 'Siberian 0’ represents a new germplasm source for cold hardiness breeding. SECTION 2 COMPARATIVE FREEZING PATTERNS OF BARK AND XYLEM OF 'SIBERIAN C' AND 'REDHAVEN' PEACH TWIGS INTRODUCTION Electrophoretic mobility is a reliable means of studying water redistribution and resultant stresses aris- ing in plant tissues as ice forms (Olien, 1961). It has been used to study freezing patterns in winter cereals and woody perennials (Olien, 1961; Single and Olien, 1967; Dennis et al., 1972; Lumis and Mecklenburg, 1974). It pro- vides a simple means of continuously monitoring changes in extracellular liquid water during freezing and thawing. Three basic patterns have been found, as described in the literature review. This study was undertaken to determine the types of freezing patterns occurring in acclimated twig tissues of 'Redhaven' and 'Siberian C' peaches; 55 56 MATERIALS AND METHODS Plant material The twigs of 'Redhaven' and 'Siberian C' used were subsamples of those used to study seasonal hardiness (Sec- tion 1). Twigs were sealed in plastic and stored at 4°C for several days until used. The basal region of the twig used for hardiness ratings in Section 1 was used in these tests. Intact twig sections, longitudinally split twig wedges, and excised bark and xylem were tested. Bark sec- tions were prepared by removing a 3 to 5 mm strip of bark, xylem sections by removing all bark tissues and excising a longitudinal wedge about 3 mm wide. All sections used were 40 to 50 mm long. Freezing patterns All sections were coated with silicone grease and wrapped with plastic film (Saran wrap) to prevent desicca- tion and to confine current flow to within the section. About 5 mm on each end was left exposed to provide a con- tact surface. The prepared section was placed on the freezing block (see Dennis et al., 1972, Figure l). A fine thermocouple was placed under the section and connected to a manual potentiometer to monitor temperature. An aluminum block cooled by an ethanol-ice bath and a thermoelectric 57 block were placed over the section. Temperature was con- trolled by regulating the ice bath temperature and by apply- ing current to the thermoelectric block. Electrical con- tacts with the twig section were made by coating the ex- posed section ends with a paste of finely divided moistened carbon. A flat platinum plate was inserted into the paste and connected to a 22.5 V dry cell. Cotton wicks inserted in water-filled test tubes and in contact with the paste prevented drying of the paste. An ammeter and voltage reg- ulator were included in the circuit. To prevent polariza- tion of cellular electrolytes, a low frequency (less than 5 cycles per second) direct current was used, as use of low frequency current gives the same conductivity data as dir- ect current (Olien and Chao, 1973). The test section was cooled to -2°C to -3°. Here temperature was stabilized and ice nucleation was induced by touching a cold probe to the tissue. Temperature change did not exceed 2.8°.per hour except during the initial exotherm after inoculation. Voltage (V) and amperage (C) were recorded at each temperature and used to calculate resistance (I) by the equation I = g. Resistance data were expressed relative to an arbitrarily selected prefreeze temperature. Observed values were corrected for the viscosity of bulk water at that temperature (NS). Viscosity was also expressed on a relative basis (NSr). The product of relative resistance and relative viscosity was used to estimate the relative extracellular liquid water content at a given temperature 58 (Mr). This was multiplied by the grams of water per gram of dry weight for the particular tissue in question to estimate the grams of liquid water per gram of dry weight in the tissue at a given temperature. Sample data are shown in Table l. Table 1. Sample data showing calculations used to estimate the amount of liquid water at a given temperature Temperature (° C) Parameter +1.7 -l.4 -l4.2 (a) Observed resistance 1.7600 1.2800 0.0975 (I) _ (b) Relative resistance 1.0000 0.7273 0.0554 (Ir) (c) Water viscosity 1.7000 1.8750 3.1380 (centipoise) (NS) ’ (d) Relative viscosity 1.0000 1.1029 1.8459 (NSr) (e) (b)x(d) Relative 1.00 0.80 0.10. extracellular water (Mr) 2 (f) (e)x g H20/ g dry wt 0.65 0.52 0.06 z . MOisture content = Contact resistance 0.6462 g HZO/g dry wt. Contact resistance was measured by preparing twigs as described above and measuring current flow (C) at a known voltage while varying the distance between contacts from 119 mm to 1.0 mm. The resistance (I) was then plotted as a function of distance and extrapolated to 0 distance to estimate contact resistance. The contact resistance was 59 expressed as a percent of the total predicted resistance at 40 mm because all sections tested were at least 40 mm long. Six separate tests were conducted. Dye flow experiments Amaranth, a negatively charged dye, was used to de- termine the path of current flow in the sections. The sections were prepared as for the freezing pattern deter- minations but a finely ground cellulose paste replaced the carbon paste and carbon electrodes replaced the platinum ones. A 1 percent dye solution was used to moisten the negative terminal. A 22.5 V battery connected with a var- iable resistance provided a stable current from 0 to 30 uA applied for up to 12 hours. Dye location and movementnwere determined by microsc0pic examination of the tissues. Moisture content Since preliminary experiments indicated that moisture content remained stable for several weeks when twigs were sealed in plastic bags and stored at low temperatures, moisture values determined for hardiness tests in Section 1 were used as estimates of moisture content in these exper- iments (see Table 6, Section 1). 60 RESULTS AND DISCUSSION Use of whole twigs resulted in unstable contacts. Contact resistance accounted for up to 60 percent of the total resistance. In longitudinally split twig sections contact resistance accounted for no more than 11 percent of the total resistance (Table 2) and had little effect on the freezing pattern. Therefore, data were not corrected for contact resistance. Twig wedges produced patterns similar to those of whole twigs. Contact resistance of 'Redhaven' and 'Siberian C' were similar and both cultivars produced identical freezing patterns. Table 2. Calculation of contact resistance of longitudin- ally split twig sections in 6 separate experiments Estimated Contact riggzigzce Slope R2 Resistance Resistance (at 40 mm) (% of total) 0.0824 0.0208 0.999 0.9144 9.0 0.0473 0.0096 0.998 0.4313 1.0 0.0455 0.0118 0.997 0.5175 8.8 0.0033 0.0159 0.996 0.6393 0.5 0.0241 0.0175 0.998 0.6759 - 0.0269 0.0172 0.997 0.7129 3.8 The current was proportional to the voltage applied within the range used in the tests. Therefore, the currents 61 used did not appear to cause cellular injury. Diffusion of indicator dye revealed that the main current path was through the outer xylem area near the cam- bium region. Very little movement occurred in the phloem and cortex. The current path in peach is much different than that in azalea stems where the cortex appears to be the main path (Dennis et al., 1972; Lumis and Mecklenburg, 1974). Rate of dye movement was proportional to current flow. When the cellulose paste was very wet, mass water flow and transpirational effects influenced dye movement. When the paste was quite dry, dye moved only 1 t0 2 mm into sections with no current flow. In bark sections dye moved throughout the entire phloem and cortex region. Freezing patterns of intact sections or twig wedges with bark attached were very similar to patterns produced when only xylem tissue was used. This evidence and that from the dye flow experiments indicate that twig wedges were essentially measuring freezing of xylem water. Water content can have a marked effect on the type of freezing pattern in a specific tissue (Olien, 1963, 1964). Some tissues of 'Siberian C' contained as much as 18 percent less water than similar tissues of 'Redhaven'; however, this difference did not result in qualitatively different freezing patterns. Freezing patterns of bark, xylem or twig wedges of 'Redhaven' were very similar to those of similar tissues of 'Siberian C.‘ Major differences in freezing processes were evident 62 in bark and xylem tissues (Figure 1). Freezing curves for sucrose and cellulose (from Olien, 1977) are given for com— parison. Data were corrected for mobility. Both bark and xylem were supercooled to -2.8°C before inoculation. Upon inoculation, temperature rose rapidly in both tissues as ice formed throughout the system releasing heat of fusion. During this initial freeze only about 14 percent of the bark tissue water froze, as compared with about 58 percent of the xylem water, before tissue temperature returned to that of the test environment (Figure 1). The slight rise in water content prior to freezing is probably due to water movement vfrom the carbon paste into the tissue. The initial moisture content of the bark was nearly twice that of the xylem (see Table 6, Section 1). However, bark water must be intimately associated with the living protoplast, while much of the xylem water is less closely associated with cellular components. Cellular substances can interact competitively with ice for water at interfaces. Cellular water orients itself around hydrophilic substances within the.cell. The degree of orientation is a function of the distance from the interface. When the system has a large interfacial area, as in the dispersed sucrose system, there is more opportunity for intimate association between water and other molecules. Where interfacial contact area is small, as in the cellulose system, there is less inter- action between water and other molecules. Bark water behaves much like the dispersed sucrose solution,whi1e xylem water 63 Figure 1. Typical freezing patterns of bark and xylem ‘ from acclimated peach twigs. Patterns for sucrose and cellulose model systems (from Olien, 1977) are given for comparison. 64 H20 Per G Dry Wt. G Liquid 4.0 #5 a.» ...O ..M .5 ..m ..m donate-9:20 A one ...—O ....M Lb >0:mw.02 Omm_oo>4_oz ....m 65 resembles that associated with the cellulose model. During equilibrium freezing of pure water associated in colloidal interfaces, net energy is dissipated by shifts in activation energies as ice crystals grow and the ice- liquid interface approaches the interface between liquid and the cellular components. Competitive structuring of water causes a lowering in the activation energy required for a water moelcule to escape from the ice lattice, re- sulting in a lowering of the activation limit for melting. This shifts the equilibrium towards melting and reduces the freezing point. Energy for adhesion arises primarily from the lowering of the activation limit for melting. The competitively structured interstitial water acts as an ad- hesive, binding hydrophilic substances to the ice. Such adhesions between ice and the protOplasmic membrane can cause distortion and disruption of the membrane, while ad- hesions between ice and soluble proteins or polysaccharides can protect vital regions by modifying structure and loca- tion of ice crystal growth (Olien, 1965). Structure formers and structure breakers can.affect transition patterns not only by increasing or decreasing ordering of water,but also by affecting the surface charac- teristics of the polymer or cellular component. They can also create complex interactions between themselves, ice, water, and cellular components. Addition of sucrose or other substances also produce colligative effects by reducing the density of water 66 molecules in the polylayer at the ice—liquid interface. The density of water molecules in the liquid phase is re- duced while the density in ice is not decreased. More water molecules are available to escape from the ice lattice than to enter it, resulting in net melting. When ice and water are separated by a vapor phase, the amount of water associated with each is determined by the vapor pressure, a density function. As temperature slowly decreases, the vapor pressure over ice decreases, producing a difference between the vapor pressure over water associated with plant components and that over ice. This difference is an index of the energy available for vapor desiccation. Water associated with the tissues moves to the ice lattice until the vapor pressures are again in equilibrium. As the water is removed from the cell to extracellular ice crystals, cells become dehydra- ted. Injury apparently results from removal of some crit- ical amount of water rather than by direct effects of ice, but the exact causes of injury are not clear. During nonequilibrium freezing,large amounts of available free energy allow explosive ice formation to occur in liquid water not closely associated with cellular components. This results in large perfect crystals which can physically disrupt tissues (Olien, 1964). In accli- mated peach xylem tissues this initial nonequilibrium freezing appears to cause little damage. Xylem injury as assessed by browning did not begin until -28.9°C to -31.7°. 67 and complete death did not occur until -37.2° (Figure 4, Section 1). The living xylem ray cells appear to undergo , deep supercooling similar to that in other woody plants (Quamme et al., 1973; Geroge et al., 1974; Burke et al., 1976). The water in the ray cells is probably isolated from.water in the xylem vessels by a freezing barrier which is related to the structural integrity of the tissues (Quamme et al., 1973; George and Burke, 1977a). The cum- ulative effect of repeated freezing and thawing might allow such stresses to become injurious at some point. Also, while initial freezing may not result in apparent damage, it could set up conditions in the tissue for increased damage at lower temperatures by limiting time for diffu- sion of water to sites where less injury would occur. The patterns of water redistribution during freezing indicate that cambium and xylem differ in their control of water. The resultant freezing stress is a function of freezing rate, amount of water involved and the free energy produced by freezing (Olien, 1971). Moisture content is a heritable trait in wheat (Gullord, 1974). Moisture con- tent of peach twigs is a relatively stable cultivar trait (Section 1). Differences in control of tissue water can affect hardiness by modifying stress energies. Kinetics inhibitors can modify freezing stress by controlling struc- ture and location of ice. The effectiveness of a specific polysaccharide inhibitor is related to its molecular struc- ture and is a cultivar characteristic (Olien, 1967). If 68 these and other component traits of overall plant hardi- ness are heritable, and genetic variation exists for them in peach germplasm, they should have simpler inheritance patterns than overall plant hardiness. Their identification would enable plant breeders to develop plants resistant to specific types of stresses, enhancing rate of breeding pro- gress. No differences were found in initial freezing patterns of similar tissues in 'Redhaven' and 'Siberian C.‘ Thereé fore, the superior freezing resistance of 'Siberian C' tis- sues is not explained by initial redistribution of water. The different freezing patterns of bark and xylem may con- tribute to observed hardiness differences between these tis- sues. Different genes may control resistance to these very different types of stresses in peach. Evidence for such separate genetic control exists in wheat (Gullord, 1974). Future research should explore other reasons for ob- served cultivar differences. 'Such differences might arise from differences in resisting or modifying adhesion or desiccation stress. Kinetics inhibitors might also modify structure and location of ice crystals. SECTION 3 INHERITANCE OF WOOD HARDINESS AMONG HYBRIDS OF COMMERCIAL AND EXOTIC PEACH GENOTYPES p hat» 6%... INTRODUCTION The need for improved wood hardiness in peaches is well documented (Chandler, 1913; Bradford and Cardinell, 1926; Campbell, 1948; Fogle and Overley, 1954). Although artificial freezing has long been used to assess hardiness (Chandler, 1913), most early cultivar improvement was the result of identifying hardy genotypes after test winters. A number of detailed studies of hardiness in other crops indicate that hardiness is a genetically complex trait (Dorsey and Bushnell, 1925; Watkins and Spangelo, 1970; Gullord, 1974; Fejer, 1976). Field observations and in- tuitive reasoning suggest that the same is true for peach. There are few detailed genetic analyses of quantitative traits in peach. French (1951) investigated several char- acters, and Hansche et a1. (1972) calculated heritability estimates for a number of traits for peaches used in the California Agriculture Experiment Station breeding program. Mowry (1964) concluded that inheritance of flower bud hard- iness in peach is quantitative in nature. Hardy parents tended to transmit a greater level of hardiness to their offspring than did tender parents. Some evidence of spe- cific combining ability was also indicated. Recently, very cold hardy peach germplasm has become available from 69 70 northern China (Pienazek et al., 1968; Zagaja, 1974). As described in the literature review, this germplasm has very poor quality fruits with no commercial value. Therefore, hybridization with genotypes possessing commer- cially desirable fruit is necessary to obtain recombinant genotypes with improved hardiness and commercially accept- able fruit. Because of the long breeding cycle involved, this process may require 20 or more years. The extent and rate of expected hardiness improvement must be balanced against the time and expense involved when determining merit and feasibility of such a project. $13-— This study was undertaken to examine inheritance of wood hardiness in crosses between commercial cultivars and recently introduced cold hardy peaches. Plant materials were provided by Dr. R.E.C. Layne of the Agriculture Can— ada Research Station, Harrow, Ontario, Canada. 71 MATERIALS AND METHODS Plant material Seven parents were used. Two, 'Harken' and 'Can— adian Harmony' are scion cultivars developed at the Harrow station, while 'Garnet Beauty' is an early ripening mutation of 'Redhaven' found in Ontario, Canada. All 3 are well adapted to northern growing conditions. 'Siberian C' and 'Harrow Blood' are rootstocks that are very hardy un- >4 der northern conditions (Layne, 1974, 1976; Layne et al., 1976, 1977). 'Siberian C' was selected at Harrow from seeds originating in China. 'Harrow Blood' is a chance seedling found near Harrow in 1938. 'Harrow 6116—256' and 'Harrow 6116-292' (henceforth designated 256 and 292 re— spectively) are cold hardy, yellow fleshed, freestone sib- lings selected from an open pollinated F2 progeny derived from the cross 'Valiant' x P. persica var Mandschurica. Mandschurica is a very hardy wild Chinese peach. Pollen for the initial cross was obtained from Poland by Harrow research personnel. The crosses were 'Siberian C' x 'Harrow Blood' (prog— eny 1), representing very hardy x hardy parents; 'Siberian C' x 'Harken' (progeny 2), representing very hardy X med- ium hardy parents; 'Garnet Beauty'x 'Harken' (progeny 3), 72 representing medium hardy x medium hardy parents; 'Canadian Harmony' x 'Harrow 6116-256' (progeny 4) and 'Harrow 6116- 292' x 'Harken' (progeny 5), both representing backcrosses to a medium hardy genotype from a medium hardy x very hardy cross. All progeny trees were planted in 1971. 'Siberian C' and 'Harrow Blood' were in a seed orchard while the commer- cial cultivars were in a test orchard. Tree age varied but all were 4 to 8 years old except 256 and 292 which were 2- year-old trees. All trees were grown under clean cultiva- tion with a winter cover crop of oats being sown in July. Samplingfprocedure Each progeny contained 50 seedlings, except;for proge- ny 5 with 29. Only healthy trees, free from major cankers, were selected initially. The same seedlings were sampled at each test date. For each sampling date and test temper- ature,each seedling was represented by a sample of 5 twigs, collected, when possible, from the upper exposed part of the tree (Cain and Andersen, 1975). Each sample was labeled, placed in a plastic bag and packed with snow in a large plas- tic can to prevent excessive warming during transport to East Lansing. Ten twigs were collected from each tree of the parent cultivars. On Jan. 16, 1976,5 trees of 'Harken', 'Siberian C' and 'Harrow Blood' and 4 trees of 'Canadian Harmony' and 'Garnet Beauty' were sampled. Parents 256 and 292 were not available on January 16, 1976. On November 18, 1976 and 73 February 7, 1977, 3 trees each of 'Harken', 292 and 256, 4 trees of 'Canadian Harmony' and 5 trees of 'Harrow Blood' and 'Siberian C' were sampled, the same trees being sampled on both dates. Freezing of twigs Twigs were kept in a cold room at 0°C while being separated into temperature replicates, then were placed into large plastic bags. Twig bases remained exposed and were covered with snow or crushed ice to induce ice inocu- lation and prevent excessive supercooling. Freezing was donezhia 2.4 x 3.0 m freezer. Temperature was controlled by a Partlow cam system on January 16, 1976, and by an evaporator pressure regulator valve on November 18, 1976 and February 7, 1977 (Section 1). All samples were allowed to equilibrate at -2.2°C for at least seven hours, then the temperature was lowered at l.7° to 2.8° per hour. Samples were held at test temperatures for at least 2 hours, then were transferred to 0°. Test temperatures were -32.2°on January 16, 1976; -26.1° and -3l.7° on November 18, 1976 and -33.8° on February 7, 1977. Twigs were placed under warm humid conditions for 36 to 48 hours to allow browning development before being evalu- ated. Evaluation of injury A thin cross section from the basal region of each of the S twigs per progeny tree, or 10 twigs per parent 74 tree, was examined under a binocular microscope. Cambium was rated on an arbitrary l to 5 scale (Cain and Andersen, 1976). Xylem was rated on an arbitrary 1 to 5 scale (Section 1). Vegetative buds were examined on November 18, 1976 at -3l.7° and on February 7, 1977 after exposure to -21.7° Observations (n1 5 buds per twig were taken on each seedling and 10 observations per twig were taken on each parent tree. A l to 4 scale was used as follows: 1- no injury; 2- less than one-half of the leaf primordia dead (brown); 3- more than one-half the leaf primordia killed but a few remaining green; 4- all leaf primordia brown (considered dead). Crop load and ripening date on August 6, 1976 the amount of fruit borne by each seedling was estimated on a l to 5 scale as follows: 1- no fruit; 2- a few scattered fruit; 3- a near commercial crop with no thinning needed; 4- needed a light thinning; 5- needed heavy thinning. None of the seedlings were thinned. The week of ripening was recorded relative to August 6, 1976 which was designated week 4 to avoid negative num- bers. Dates of fruit ripening before August 6, 1976 were estimated by the extent of deterioration of fruit which had fallen from the tree. The earliest ripening trees were estimated to be four weeks earlier than those ripe on August 6, 1976. Ripening dates after August 6, 1976 were 75 estimated by approximate maturity. The latest ripening trees were estimated to ripen at least four weeks after August 6, 1976. The effect of these variables on previous and subsequent wood hardiness was examined. Canker ratings Extent of canker was rated for each seedling on April 26, 1977. Two ratings, one from each side, were taken on each tree and averaged. All cankers were assumed to be symptomatic of infection by perennial canker fungus (Leucostoma spps.). A l to 9 rating scale developed by Dr. R.E.C. Layne of Agriculture Canada was used as follows: 1 - absence of visible cankers 2 - evidence of gum and/or a few small insignificant cankers 3 - evidence of distinct cankers, few in number and having little adverse effect on the tree 4 - distinct cankers with one capable of having a debilitating effect on the tree 5 - distinct cankers with more than one causing visible damage 6 - presence of cankers in the trunk and scaffold region 7 - large scaffold and/or trunk cankers likely to kill the tree 8 - large scaffold and/or trunk cankers killing many branches 9 - death of the tree from trunk and scaffold cankers. The data presented may not be truly representative of the populations, for trees were initially chosen because they had few serious cankers. 76 Bloom type On April 26, 1977 each seedling was classified as having shOwy or nonshowy bloom. Bloom type is a monogenic trait with nonshowy being recessive (Bailey and French, 1942). Where possible,this trait was used as one indication that seedlings were true hybrids. Data analysis Data on injury, crop load, ripening date and cankering were analyzed via analysis of variance (AOV) and means were separated using Duncan's Multiple Range test or its uneven replicate approximation (Steel and Torrie, 1960). Population distributions of each progeny for each trait measured are presented in Appendix B. As a population approaches the scale limits,it becomes distinctly skewed and its variance may be reduced. Nonnormality and heter- ogeneous variance can affect the mean separation tests. Test temperatures were chosen to induce a 3.0 levelof in- jury which allows a normal dispersion of ratings about the mean. Gullord et a1. (1975), rating freezing injury of wheat crowns, found that when the populations were kept well within scale limits all varieties had a normal distri- bution and homogeneous variance; however, populations be- came skewed when the population mean approached either scale limit. 77 Heritability Estimates Broad-sense heritability estimates of injury ratings were calculated for progenies and parents. Progeny esti— mates were calculated using the equation: h1238=(6\2+£ 2-Qz)/(§ 2+3 2) we BP wp where hgs is the broad-sense (BS) heritability estimate; 2 is the between progeny variance; 3 2 is within BP A WP progeny variance; and o 2 is within parental clone vari- WC ance used as an estimation of environmental variation. Broad-sense heritability estimates obtained from par- ent data were calculated as follows: 2—3 22.2 BC WC BC where OBCZ is between clonal parent variation and g 2 is WC' within clonal parent variation. 78 RESULTS Parent injury ratings Significant differences among parent cultivars for injury of the cambium region, xylem and vegetative buds were found at every test date (Table l). Cambium. 'Siberian C' consistently exhibited the least injury (Table l). 'Harrow Blood' sustained signif- icantly more injury than 'Siberian C' except on January '16, 1976. The backcross selections 256 and 292 were com- parable to 'Harrow Blood.‘ Significant differences existed among the 3 commercial cultivars, all of which were injured more than the rootstocks or backcross parents, except on February 7, 1977 when injury to 'Harken' was not signifi- cantly different from 256 and 292. 'Harken' and 'Canadian Harmony' were never significantly different from each other, and 'Garnet Beauty' always exhibited the most injury. xylem, At all dates 'Siberian C' suffered less in- jury than the commercial and backcross cultivars (Table l). 'Harrow Blood' was similar to 'Siberian C' on January 16, 1976 and February 7, 1977 but had more injury at both temp- eratures on November 18, 1976 when it suffered as much in— jury as the commercial cultivars. The two backcrosses 79 Table 1. Injury ratings of parents' cambium, xylem and vegetative buds subjected to controlled freezer tests on 3 dates Parents Tissue Siberian Harrow Canadian Garnet C Blood Harken Harmony Beauty 256 292 January 16, 1976 (-32.2°) Cambium 1.1az 1.2a 1.8bc 1.7b 2.1c - - Xylem 1.1a 1.5a 2.1b 2.9c 2.8c - - November 18, 1976 (-26.l°) Cambium 1.4a 1.8b 2.4c 2.4c 3.0d 1.5ab 1.8b Xylem 1.1a 3.1de 3.1de 3.4e 2.9d 1.4b 2.2c (-3l.7°) Cambium 2.3a 3.4c 3.3c 3.5c 4.0d 2.5ab 2.7b Xylem 3.6a 5.0b 5.0b 5.0b 5.0b 4.8b 5.0b Vegeta- 2.2b 1.7a 3.5c 3.6cd 3.9d 3.8cd 2.2b tive buds February 7, 1977 (~33.8°) Cambium 1.7a 2.2b 2.8cd 3.1d 3.8e 2.4bc 2.3bc Xylem 3.2a 3.3a 4.8b 5.0b 5.0b 4.7b 4.4b Vegeta- 2.5b 2.0a 3.4c 3.6c 3.7c 3.7c 2.8b tive buds zWithin rows, means followed by the same letter are not significantly different at P.05 by Duncan's multiple range test. 80 were similar to each other and intermediate between the hardy rootstocks and the commercial cultivars at -26° C on November 18, 1976. On November 18, 1976,at -31.7° and February 7, l977,a11 cultivars except 'Siberian C' were nearly or completely killed,thus no differences could be detected among them. At -26.l° on November 18, 1976, 'Harmony' had more injury than 'Garnet Beauty' while 'Har- ken' was intermediate. Vegetative buds. In contrast to woody tissues, vegetative buds of 'Siberian C' showed significantly more injury than those of 'Harrow Blood'. 'Siberian C' and 292 were similar while 256 suffered as much injury as the com- mercial cultivars. At -3l.7° on November 18, 1976, 'Har- ken' had less injury than 'Garnet Beauty'. This was the only significant difference between the commercial types. Mean family hardiness Cambium. In general, progeny l was the most hardy in all tests, while progenies 3 and 4 were least hardy (Table 2). Progenies 2 and 5 were intermediate and dif- fered significantly in only one case. Xylem. Hardiness of the xylem generally paralleled that of the bark tissues, but more injury occurred at the same temperature (Table 2). Progeny l was hardier than all others except on January 16, 1976, while progenies 3 and 4 were least hardy except on February 7, 1977 when 81 Table 2. Mean freezing injury of cambium, xylem and vege- tative buds tested on 3 dates,and crop load, ripening date and canker ratings for 5 progenies. Progeny Tissue l 2 3 4 5 Siberian C Siberian C Garnet Canadian 292 X X Beauty Harmony X Harrow Harken X X Harken Blood Harken 256 January 16, 1976 (-32.2°) Cambium 1.1az 1.3b 1.5c 1.5c 1.2ab (.ll) (.34) (.33) (.32) (.19) Xylem 1.2a l.5b 1.9c 2.3d 1.4ab (.23) (.41) (.46) (.52) (.38) November 18, 1976 (-26.l°) Cambium 1.5a 2.0c 2.1a 2.2a 1.9b k (.27) (.26) (.24) (.27) (.42) Xylem 1.7a 1.9b 2.6d 2.9e 2.3c (.34) (.45) (.41) (.46) (.69) (-3l.7 ) Cambium 2.8a 3.0b 3.0b 3.0b 3.0b (.27) (.21) (.25) (.20) (.19) Vegetative 3.0a 3.1ab 3.5d 3.3cd 3.2bc buds (.40) (.34) (.43) (.37) (.47) February 7, 1977 (-33.8°) Cambium 1.5a 1.9b 2.3d 2.1c 1.9b (.26) (.42) (.37) (.43) (.30) Xylem 3.2a 3.5b 4.4d 4.2c 4.6d (.49) (.58) (.48) (.52) (.41) Vegetative 2.5a 2.4a 3.6b 3.4b 3.4b buds (.47) (.38) (.40) (.58) (.42) 1976 crop loady 3.6bc 3.8c 2.7a 3.3b 3.4bc (1.05) (1.19) (1.21) (1.08) (1.08) 1976 ripe dateX 8.0c 6.3c 3.5a 6.4b 6.0b (0.00) (2.05) (2.59) (1.93) (1.83) Canker ratingW 6.3c 5.9c 7.0d 5.3b 4.5a (0.96) (1.88) (0.89) (1.78) (1.98) 82 2Within rows, means followed by the same letter are not significantly different at P.05 by Duncan's multiple range test. Figures in parentheses are standard deviations used as a measure of population dispersion. yCrop load, amount of fruit borne by each seedling, l = no crop to 5 = heavy crop. XRipe date, estimated weeks to fruit maturity after July 6. wCanker rating, 1 = no cankers to 9 = death of tree from cankers. progency 3 did not differ from progeny 5. Progeny 2 was intermediate between progeny 1 progenies 4 and 5 except on January 16, 1976 when progeny 2 did not differ significantly from progeny 5. Vegetative buds. Progenies l and 2 were similar and generally hardier than the other progenies. No differences were detected between progenies 3, 4 and 5 on February 7, 1977. On November 18, 1976, progeny 3 was slightly but not. significantly worse than progeny 4, while progeny 5 was in- termediate, not differing from either progeny l or 4. Crop load 1976. Progeny 3 had significantly fewer fruit than other progenies in 1976 (Table 2) and was the least hardy. Progenies l, 4 and 5 bore similar amounts, while yield of progeny 2 was significantly greater than that of progeny l, but not significantly different from that of 4 and 5. Very low negative correlations existed between crop load and hardiness (Tables 6 and 7). 83 Ripening date 1976. Large differences in ripening date were apparent among the progenies (Table 2). Fruits of progeny 1were estimated to be 435 weeks later than those of progeny 3, and much later than those of most commercial varieties. Progenies 2, 4 and 5 included a majority of midseason seedlings with similar ripening dates. Most seedlings in progeny 3 were early to midseason. Very low negative correlations existed between ripening date and hardiness (Tables 6 and 7). Canker ratings. Progeny 5 had fewest cankers while progeny 3 had the most (Table 2). Progenies l, 2 and 4 were intermediate, 4 having significantly less cankers than the other two. The 7.0 rating of progeny 3 indicates very serious cankering. No correlation existed between canker ratings and hardiness Bloom type. All seedling trees were evaluated for bloom type. 'Harken', 'Canadian Harmony', {Garnet Beauty' and progeny 3 seedlings all had nonshowy flowers, while 'Siberian C', 'Harrow Blood': 256, 292 and progeny 1 seed- lings had showy blooms (Table 3). Progenies with showy x nonshowy parentage (progenies 2, 4 and 5) all segregated in 1:1 ratios, indicating that 'Siberian C': 256 and 292 have the genotype Sh/sh. 'Harrow Blood' is Sh/Sh and the commercial cultivars are all homozygous recessive. 84 Table 3. Inheritance of bloom type in parents and progen- ies. Showy bloom (Sh) is dominant to nonshowy bloom (sh) Parent Genotype Phenotype Harken sh/sh nonshowy Canadian Harmony sh/sh nonshowy Garney Beauty sh/sh nonshowy Siberian C Sh/sh showy Harrow Blood Sh/Sh showy 256 Sh/sh showy 292 Sh/sh showy Proqeny expected observed Siberian C X All showy All showy Harrow Blood Siberian C X 1:1 27 showy : 23 nonshowy Harken Garnet Beauty X All nonshowy All showy Harken Canadian Harmony 1:1 24 showy : 26 nonshowy X 256 292 X Harken 1:1 17 showy : 12 nonshowy Population dispersion. Standard deviations (SD) of each trait for each progeny indicated that no progeny had a consistently large or small variance (Table 2). Prog- enies whose mean approached the scale limits (progeny 1, January 16, 1976 cambium) tended to have lower variances. Population distributions are illustrated graphically in Appendix B. Genetic analysis ofeprogenies. Separate parent and progeny broad—sense heritability estimates were calculated for each tissue at each test date (Table 4). All estimates except for the cambium on November 18, 1976, -37.7°C were in the 0.65 to 0.96 range. Estimates based on parent data 85 Table 4. Progeny and parent broad—sense heritability estimates for cold injury assessed for individ— ual tissues, temperatures, and dates Date and tissue ProgenY Parent 22 o 0' test temperature ( C) hBS hBS we Jan 16 cambiumY -32.2 .65 .77 .04 Nov 18 cambium -26.1 .69 .82 .05 Nov 18 cambium -3l.7 .05 .84 .06 Feb 7 cambium -33.8 .73 .86 .06 Jan 16 xylem —32.2 .65 .80 .12 Nov 18 xylem -26.1 .92 .96 .04 Nov 18 xylem -3l.7 — .86 .05 Feb 7 xylem -33.8 .83 .84 .10 Nov 18 vegetative buds -31.7 .74 .94 .05 Feb 7 vegetative buds -33.8 .85 .84 .08 Jan 16 average injury —32.2 .69 .84 .05 Nov 18 average injury -26.1 .90 .95 .03 Nov 18 average injury -3l.7 - .94 .02 Feb 7 average injury -23.8 .86 .91 .05 Nov 18 total injury -3l.7 - .94 .02 Feb 7 total injury -23.8 .94 .96 .02 Overall cambium .82 .96 .01 Overall xylem .93 .99 .02 All tissues .94 1.0 .01 202 is within clonal parent variation taken as an WC estimate of environmental variation. yAverage injury is the average of the cambium and xylem; total injury is the mean of cambium, xylem and veg- etative buds; overall cambium and xylem is the mean rating across all dates and temperatures; all tissues is the mean rating across all tissues, temperatures and dates. 86 tended to be slightly higher in most cases. The extremely low heritability estimate for cambium on November 18, 1976, -31.7° is due to small differences between progenies (Table 2 and Appendix B). Environmental variation, as estimated from within parent variance, was consistently very small. Deviations of progeny means from their midparent means (Table 5) are generally small negative values, indicating progenies had slightly less injury than would be expected from parent performance. Values for progeny 4 were more positive than other progenies indicating that at several dates it was injured more than expected based on parent ratings. Comparing progeny means in Table 2 with parent means in Table 1 shows a number of cases where the progeny mean is lower than the least injured parent. Vegetative bud injury at -3l.7° on November 18, l976,is the only case in which the progeny had more injury than the most severely injured parent. Correlations between injury ratings for separate tissues and samples were based on individual tree ratings of all progenies (Table 6). Nearly all correlation coeffi- cients were significantly different from zero, but they tended to be very low. On November 16, 1976 a correlation of only 0.41 was obtained when cambium ratings at -26° and -31.7°C were compared. If progeny means were used (Table 7), values of correlation coefficients were increased, but Significance levels decreased because they were based on 5 observations rather than 229 (Table 6). 87 Table 5. Deviations of progeny mean injury ratings from midparent ratings calculated for each tissue, temperature and date . . Siberian C Garnet Canadian . Siberian C X Beauty Harmony 292 Tissue X Harrow X X X Harken Blood Harken 256 Harken JanuaryelG, 1976 (-32.2) cambium -o.1z -o.1 -o.5 - - xylem -0.1 -0.1 -0.5 - - November 18, 1976 (-26.1) cambium +0.1 -0.1 -0.6 +0.1 -0.1 xylem -0.2 -0.4 -0.4 +0.1 +0.1 (-3l.7) cambium +0.2 0.0 -0.6 -0.1 +0.1 vegetative +0.3 +1.0 -0.2 +0.4 -0.4 buds February 7, 1977 (-33.8) cambium -0.3 -0.5 -l.0 -0.6 -0.7 xylem -0.5 0.0 -0.5 -0.5 -0.2 vegetative -0.6 +0.3 0.0 +0.2 -0.1 buds z o I I u a I POSitive values indicate progenies sustained more injury than the parents, negative values indicate the opposite. 88 .Hoo. u m can as. u m pm DCCOHMHGmHm out _Hm. >Hm>fluommmwu M pom _ma._ M mcoflumaouuoo Hag» .COADMHCHHOU mom mcoaum>ummno mmmm mCHumu so. No.| we. no. Ho. No.| mo.| mo. mo. uoxcmo mutt wQHH om.| vm.| mv.l wN.I wH.| mm.| em.| mH.| NN.I wsma pmoH mono mm.| om.| NN.| oa.l vo. mm.| mH.| mH.| mH.I mmma moon ow. mm. we. ea. mm. mm. mm. em. w>fiumpmmm> ow. mm. mm. mm. mm. mm. Hm. Emaxx mm. mm. ov. ow. mm. mm. Esanemm 1a.mm-c a bma mpsb mo. om. mm. mm. mm. o>abmbmao> om. av. ca. so. EdHQEmo A>.HMIV ma >oz em. om. mm. Ewawx mm. vm. Esflbfimo AH.mmuv ma >oz waw. Emawx Am.mmuv ma ems mpsn mosh m>apppwm®> Ewawx fisflnfimo m>Humummw> ESfiQEmo Ema>x Esflbfimo Emamx ESHQEMO Am.mmuv as.amnv Aa.mm|v Am.mmnv puma .5 9mm mhma .mH >oz osma .ma >oz whoa .ma emu Aoov musumummfimu pom mpmo mmcflumu mesmo was mutt mcfigwmfln .pmoH mowo .musflafl MOM moflemmoum Ham mmouom museums coup Hmspfl>atcfl gmosuwa nmucwwowwwooo coflumfionuoo .m wants —'— | '1 .~.c ,_ 89 .~o>ou wo:uu«uu:mua unsoutzq mononucuuem :« nwsna>> .:oduu~ouuoo Hum accuuu>uouno mu .mm.. .on.. .v~.. .am.. .as.. .ma.. .ma.. .na.. .e.a. .na.. ..h.. oeauuu mm.» ev.s e~.: .n.- n~ «a.. «a.. m~.u o.= me. an. unseen .-.. ..~.. .o~.. .no.. .ma.. .m~.. .a~.. .sa.. .5... .m~.. «are be. no.1 as.- an.u «a.. .s.; ~o.n as.» n..: ob.u uaau mama .o~.. .«~.. .-.. Abo.. .om.. xe~.. ..v.. .mn.. .en.. enoa ~a.- so.- as.- no.1 ev.- as.: m..: am.- am.: mono shad .«e.. .m~.. ..c.. .«n.. ..o.. .an.. Aa~.. .en.. near .a. .n. as. an. as. am. so. an. u>auaboaa> .w~.. .me.. .a~.. .o~.. .m~.. .n... .nn.. as. be. on. as. mm. s.. «m. soaax .ae.. .mc.. .so.. .ne.. .m~.. .no.. we. nu. me. am. an. «a. asummfiw a.nn-. A not .ua.. ..o.. .aa. .m~.. .mc.. me=n on. em. «a. as. «a. o>abuuemo> .v_.. .aoo.. .o~.. .ao.. an. em. on. va. Esméeco .s.~mu. ad >oz .o~.. ..o.. .ma. as. an. we. sous: .me.. .mo.. as. no. Esuban .~.u~-. ma >oz .ac.. smm. sawmm ~n.~n:. a. bun o>«“W”Mmo> soda: E:«bfiuo,0>«wmuuau> saunsuo and»: E:«£Eu0,30~>x asunfiau cane coca .o.nau. .n.~n-. .n.w~|. .~.~n-. oaqu mono pang .s new mama .aa >oz mama ..a >oz beau .oa are read was" . .ua. ousunhoniou v:¢ can: nucwaau nox:nu 6:: case and» .eoo~ noun .>H:H:u new names anemone m=o2unan:o«0auuaoo noduuaouuoo .n wanna 90 DISCUSSION The relative hardiness of 'Velvet', 'Redhaven' and ‘Siberian C'r (Section 1) and parents used in this study, as assessed by controlled freezer tests, matched their field hardiness rankings. This indicates that controlled freezing is a reliable method of assessing relative hardi- ness of different genotypes. For all tissues at all dates tested, 'Siberian C' was significantly hardier than any of the commercial cultivars used. 'Siberian C' generally suffered less injury to cam- bium and xylem than did 'Harrow Blood': which is in agree- ment with Layne's (1974) and Ormrod and Layne's (1974) ob— servations on wood hardiness of these rootstocks. These results agree with the effects these rootstocks have in in- creasing flower bud, phloem, cambium and xylem hardiness of scion cultivars budded onto them, as measured by browning tests (Layne,et al., 1977). However,they disagree with find- ings of Chaplin and Schneider (1974) who, using electroly- tic conductance tests on 1-year-old trees found that 'Har- row Blood' promoted greater wood hardiness of budded scion cultivars than did 'Siberian C.‘ The greater hardiness of 'Harrow Blood' vegetative buds compared to 'Siberian C' may indicate a different 91 physiological stress is occurring in this organ in compar- ison with stresses occurring in cambial and xylem tissues. Data presented in Section 2 showed that stresses arising in the xylem are different from those in bark tissues. Layne et a1. (1977) found that injury ratings of phloem, cambium and xylem were closely correlated with each other in the fall, but were not correlated with flower bud injury on the same shoots. Blake (1938) also noted that hardiness of the lower trunk and bark tissues was not always correlated with fruit bud hardiness. Flower buds undergo different freezing processes than adjacent stem tissues (Graham, 1971; Quamme et al., 1975). The inconsistency in relative hardiness of various tissues in different genotypes may indicate that separate genes control reaction to stresses occurring in different tissues. Gullord (1975) has shown that different groups of genes may be responsible for reaction to differ- ent types of stresses in wheat. The hypothesis that separate groups of genes are con- trolling stress reactions in different tissues is not sup- ported by progeny data. If 'Harrow Blood' has more favor- able alleles for vegetative bud hardiness than 'Siberian C' it should contribute more favorable alleles to its progeny than ‘Siberian C.‘ ‘Harken', having more injury than either 'Siberian C' or 'Harrow Blood', should have fewer favorable alleles to contribute. Therefore, the 'Siberian C' x ‘Harrow Blood' progeny should have more favorable alleles than the 'Siberian C' x ‘Harken' progeny and thus 92 should have a lower injury rating, however, both have sim- ilar injury. A similar inconsistency occurs with the back- cross parents. Specific combining ability could account for such discrepancies. All interactions between tissues, dates and genotypes must be interpreted cautiously, since even within one date and tissue, data obtained at different temp- eratures yielded different results. Different stresses de- veloping within tissues at different temperatures may ex- plain such differences on a physiological basis, but dif- ferences could also be due to poor repeatability of the rating system. Repeatability of the rating system, using different plant material, was previously found to be about 0.8 (Cain and Andersen, 1976). Based upon parent performance, vegetative buds of progeny 4 were expected to have more injury than those of progeny 5; however, ratings were nearly identical. Again vegetative bud response differed from that of cam- bium and xylem. This again suggests that genetic control of hardiness in vegetative buds differs from that in cam- bium and xylem. Except for vegetative buds, the 'Siberian C' x 'Har- ken' (very hardy x medium hardy) cross was intermediate be— tween 'Garnet Beauty' x ‘Harken' (medium hardy x medium hardy) and 'Siberian C' x 'Harrow Blood' (very hardy x hardy). The standard deviations (Table 2) of each progeny are reasonably similar except when scale limits are approached. Cambium ratings show somewhat less dispersion 93 than those of xylem or vegetative buds. Distributions ap- proximated the normal distribution except where scale limits are approached (Appendix B), which is in agreement with the data of Gullord et a1. (1975). Parent and progeny mean per— formance and generally normal distributions of progenies indicate that inheritance of resistance to freezing stress is quantitative. This is to be expected in a physiologically complex trait and agrees with other reports (Watkins and Spangelo, 1970; Federer, 1976; Dorsey and Bushnell, 1925). Deviations of progeny means from midparent values were usually small negative values (Table 5). This may in- dicate presence of some dominant genes for hardiness, in agreement with the data of Hildreth and Powers (1941), who found winter hardiness in strawberries to be dominant. Gul— lord (1974) found low moisture content and high and low in— tensity freezing hardiness in wheat to be controlled by partially dominant genes mostly additive in their effect. However, hardiness appears to be recessive in crosses be- tween other hardy Chinese peaches and commercial types, with most seedlings tending to be closer to the commercial parent in hardiness (Zagaja, personal communication). In a number of cases, progeny means are lower than the least injured parent (Tables 1 and 2); this may indicate presence of overdominance or transgressive segregation. However, com— parisons between parent and progeny data should be inter- preted very cautiously because differences may be due to physiological differences resulting from divergent cultural 94 practices and tree age. These preliminary findings should be investigated further when more suitable populations become available for genetic analysis. Cultural practices are known to affect subsequent winter hardiness (Hildreth, 1926; Cooper, 1953; Edgerton, 1960; Chandler, 1913). Significant differences exist among progenies regarding crop load and ripening date (Table 2.) These differences were not highly correlated with differ- ences in injury on either an individual tree basis (Table 6) or a progeny basis (Table 7). Since none of the fruits were removed, excessive crop set on some seedlings was expected to reduce hardiness. It was suspected that late maturity might also reduce hardiness. The small neg- ative correlations for both crop size and ripening date vs. injury ratings may be explained by the fact that both 'Si- berian C' and 'Harrow Blood‘ ripen later and their flower buds are hardier than the commercial parents, and these traits are transmitted to their offspring. Hansche (1972) estimated narrow—sense heritability for ripe date and crop to be 0.84 and 0.08 respectively. Because no estimate of environmental variation was available in my study, no her- itability estimates were calculated for these traits. How— ever, large differences were noted among progenies for both traits (Table 2). Winter injury has been implicated in increasing sus- ceptibility of peaches to perennial canker (Weaver, 1963). Trees with more injury were therefore expected to have more 95 serious cankering. Tables 6 and 7 show no correlation be- tween canker ratings and injury ratings either on a prog- eny or single tree basis. Backcross progenies having commercial and exotic parentage have fewer cankers than progenies having entirely commercial or exotic parents (Table 2). Several explanations are possible. The Meeer churica ancestor of progenies 4 and 5 may contribute some resistance apart from improved hardiness. 'Siberian C' and 'Harrow Blodd' have narrow crotch angles of scaffold limbs, leading to more bark inclusions in these crotches which provide sites for canker infection. Narrow crotches are also more susceptible to winter injury than are wide crotches (Weaver, 1968; Blake, 1935). Hybrids between com- mercial and exotic types may have fewer cankers because recombination has incorporated improved hardiness of the exotics with wider crotch angles of the commercial types. Weaver (1968) found large differences in crotch angle be- tween cultivars. Crop load, ripening date and canker ratings were used in a multiple regression to see if they affected pre- vious or subsequent tissue injury ratings. All R2 values were 0.15 or less, and therefore accounted for virtually no variation among injury ratings. Broad-sense heritability estimates the relative im- portance of genetic and environmental influence on pheno- typic expression of a trait. Narrow-sense heritability is the ratio of the additive portion of the genetic variance 96 to the total variance; therefore, broad-sense heritability is always equal to or greater than narrow-sense heritabil— ity. A trait's heritability may be increased by either in- creasing genetic variation or by reducing non-genetic var- iation. Non-genetic variation can be decreased by increasing accuracy and precision of measurements through increased replication or improved sampling procedures and experimental technique. Cain and Andersen (1976) suggest sampling procedures for minimizing non—genetic wood hardi— ness variation in peach. Hansche et a1. (1972), estimating narrow-sense heritabilities for several traits in peach, found ripe date (.84) was very high, bloom date (.39), amount of bloom (.38) and fruit cheek (.26) were high, fruit firmness (.13) and acidity (.19) were moderate, and percent soluble solids (.01) was very low. In this study, broad-sense heritability estimates (Table 2) indicate that within dates and tissues, injury (browning) is a highly heritable trait. Heritability es— timates of composite traits were also high. The environ- mental variation was consistently very low and may have been underestimated,since seedlings were subjected to different cultural practices and growth response may have varied. Environmental variation was much lower than that found in previous experiments using different material (Cain and Andersen, 1976). The amount of additive genetic variance is not known. Genetic variance for winter survi- val of apples was found to be mainly additive (Watkins and 97 Spangelo, 1970), but significant specific combining abil- ity effects were also found in this species (Fejer, 1976). In wheatrgenetic variation for high and low intensity freezing was found to be largely additive in nature (Gul- lord, 1974). The progenies had full and half-sib rela- tionships and narrow-sense heritabilities were calculated. However, because estimates were based on only a single de- gree of freedom, they were not accurate enough to be use- ful and are not presented. Phenotypic correlations of injury to individual tis- sues between each date and temperature were used to esti- mate repeatability of individual trees (Table 6) or prog— enies (Table 7). Heritability is a measure of the repeat— ability of the genotypic expression. The correlations pre- sented, while they are not heritability peg ee, contain both the genetic and environmental variation and estimate phenotypic repeatability over time, temperature or tissue. Correlation coefficients were higher when based on progeny performance. This was expected, as progeny means were based on many more observations than individual tree means. Correlations within similar tissues were generally higher than correlations between different tissues. The low cor- relation coefficients obtained for individual trees indi- cate repeatability of individual seedlings is quite low when measured across tests. Watkins and Spangelo (1970) obtained correlation coefficients ranging from 0.84 to 0.50 for several measures of tree survival in two apple 98 dialleles. Fejer (1976), using other apple dialleles, also found low correlations between electrical impedance values and other measures of plant injury. Correlation coefficients for impedance values measured at three dif- ferent dates were also low. French (1951) has found low year to year correlations in other quantitative traits in peaches. Considering the limited environmental variation for an individual tissue sample, the lack of correlation may indicate presence of large genotype by environment inter- actions. These interactions may reflect the inability of the rating system to discriminate between genotypes in which injury actually differs or they may indicate true differential response of individual tissues to stresses produced at different temperatures and dates. In these experiments,separation of interactions due to experimental technique from those due to physiological and genetic causes was not possible. Large numbers of values near the scale limits during some tests, and the narrow range of some progenies undoubtedly also contributed to the low correla- tions. While important cultivar differences for tree survi- val occur among commercial peaches (Chandler, 1913; Fogle and Overley, 1954; Campbell and Hadle, 1960), wild peaches from northern China are far hardier then commercial types. This germplasm represents a source of new alleles and/or increased gene frequencies for wood hardiness improvement. 99 Since these wild peaches have small, low quality fruit, their favorable hardiness genes need to be recombined with the excellent fruit quality characters of commercial cul— tivars. The question, then, is: what is the most effici- ent breeding method for recombining these two groups of characters into new superior cultivars? After the initial crosses are made to produce an F1 progeny, several alter— native methods are available. Logical alternatives would be backcrossing the hy- brids to commercial cultivars or intermating of the hybrids using some form of recurrent selection. Allard (1960) states that in backcrossing the character being transferred should be highly heritable since the certainty with which the character can be identified in segregating populations determines the speed and efficiency with which it can be transferred. A useful intensity of the charaCter must be maintained through several generations of backcrosses. The high broad—sense heritabilities found in this study suggest that a high level of cold resistance could be maintained in a backcrossing program. Yet, conflicting low correlations indicate that many misidentifications might be made. The mean performance of the two backcross populations (Table 2) indicates that the overall perfor— mance of progeny 5 was superior to the commercial cross (progeny 3) while that of the other backcross progeny (progeny 4) was not. Parents 256 and 292 were selected primarily for a high level of cold hardiness combined with 100 commercial characteristics, leading to a rapid approach toward restored commercial fruit type. If selection were to be based primarily on hardiness, more improvement might be expected in cold resistance while slower return to com— mercial fruit quality might be expected. Linkage groups are a major factor in determining how readily cold hardiness and high fruit quality characters can be recombined. Hybrid progeny with the greatest cold resistance often have the poorest fruit quality, and yiee yegee (S. Zagaja and R.E.C. Layne, personal communication). Since maximum chance for recombination between chromosomes from opposite parents occurs at Fl synapsis, growing large F2 populations before beginning a backcross program would aid in producing desirable recombinant types. The alternative to the backcrossing procedure is some form of recurrent selection. Here the Fl constitutes a base source population from which selections can be inter- mated to produce a second cycle. Unless very large popu— lations were grown, this method would probably result in slower approach to the reconstituted commercial type, but it would provide a greater chance for recombination between the genomes. Andersen (1970) has discussed the relative merits of several recurrent selection schemes as they relate to clonally propagated crops. Recurrent mass selection based on clonal family performance (RMSCF) was inferior to simple recurrent mass selection (RMS) except when heritability was 101 very low, and even then gains resulting from RMSCF could be nullified if clonal propagation lengthened the breeding cycle substantially. Another disadvantage of RMSCF is that fewer genotypes can be grown on the same unit of land. The high heritabilities for tissue injury found here indicate that RMS would be superior to RMSCF. Even though correla- tion coefficients of injury ratings between tissues, temper- atures or dates were quite low, the mean performance of an individual seedling tested over an entire winter should increase the accuracy of the estimate of net genetic worth without lengthening the breeding cycle. The low environ— mental variation (Table 3) indicates that, if plots are reasonably uniform, replicating over locations (within a limited climatic region) would not significantly improve accuracy of estimates of genotype performance. This is also supported by earlier work (Cain and Andersen, 1976) which showed that between—tree variation was only a minor contributor to overall environmental variation. Other alternatives to RMS are recurrent full-sib family selection (RFSFS), recurrent selection among half- sib families (RSHSF) and selfed progeny family selection (SPS). RFSFS is superior to RMS only when heritability is very low, while RSHSF has utility only where large full—sib families are difficult to obtain, which is not true in this case. SP8 is more competitive with RMS at higher levels of dominance and at higher gene frequencies. The small number of families involved in this study 102 and the conflicts between the heritability estimates and the correlation data make it impossible to state with reasonable assurance which breeding approach would produce most rapid improvements in peach tree cold resistance. More suitably designed populations are needed to ob— tain more accurate estimates of the heritability and mag- nitude of genotype by environment interactions. Also, more physiological research is needed to identify components of the complex trait of cold hardiness. Such components should have simpler modes of inheritance and be easier to identify and manipulate genetically. 103 SUMMARY AND CONCLUSIONS Freezing tests were conducted on 'Velvet', 'Red- haven' and 'Siberian C' peach cultivars on 13 dates during the winters of 1975-76 and 1976-77. 'Siberian C' suffered least injury at all test dates, 'Velvet' suffered most and 'Redhaven' was intermediate. This agrees with field as- sessment of the relative hardiness of the cultivars. Xylem tissues were injured at higher temperatures and over a narrower range than the inner bark. During acclim— ation, the major change in xylem temperature response pat- tern was a downward shift in temperatures necessary to cause injury. In fall, inner bark injury increased rapid— ly over a narrow temperature range in a manner similar to the xylem. In midwinter, injury increased slowly as temp- erature declined. This shift may reflect a change in the type of freezing stress occurring at different seasons. Xylem appeared to deep supercool. Within test dates, twig injury was correlated with overall twig moisture content. 'Velvet' generally had the highest moisture content, 'Redhaven' was intermediate and 'Siberian C' had the lowest moisture content. Injury and moisture content were not correlated over dates because moisture content remained quite stable throughout the 104 winter while injury changed dramatically. Increasing twig moisture content and preconditioning twigs at warmer temp- eratures increased subsequent freeze injury to bark. In— creased moisture content did not affect xylem injury but high temperature increased xylem injury in both cultivars. Increasing the moisture content increased inner bark in- jury more in 'Redhaven' than in 'Siberian C', while high temperature pretreatment increased injury more in 'Siber- ian C' than in 'Redhaven'- The superior hardiness of 'Siberian C' was not explained by moisture content alone. Moisture contentcxfbark tissues was nearly twice that of the xylem. Equilibrium freezing occurred in bark tissues, nonequilibrium freezing in xylem tissues, indicat- ing that water in bark tissues was more closely associated with cellular components than was that in the xylem. Freezing patterns of corresponding tissues in 'Redhaven' and 'Siberian C' were similar. Thus, cultivar hardiness differences were not explained by differences in water re— distribution during initial freezing. Injury to inner bark, xylem and vegetative buds was determined for parents and progenies of crosses among medium hardy commercial cultivars, very hardy rootstock cultivars, and two backcross progenies. As a parent, 'Siberian C' inner bark and xylem suffered significantly less injury than the same tissues in all commercial cul— tivars tested, and generally had less injury than 'Harrow Blood'- Vegetative buds of ‘Harrow Blood': however, 105 suffered less injury than those of 'Siberian C'. Vegeta- tive bud performance in the progenies also did not corres- pond with that of inner bark and xylem, nor did relative ranking correspond with predicted ranking based on parent performance. Relative ranking of progeny means as to inner bark and xylem injury could be predicted based on average performance of the parents, with hardier parents producing hardier progenies. All populations had similar distribu- tions except where scale limits were approached. Progenies differed significantly in canker rating, ripening date and crop load, but these factors had no sig- nificant effect on injury. Backcross progenies had lower canker ratings than progenies of very hardy cultivars. Broad-sense heritability estimates indicated that within dates and tissues, freeze injury is a highly herit- able trait. Environmental variation estimates were very 1ow. Correlation coefficients among tissues, temperatures and dates based on individual tree observations indicated repeatability of individual genotypes was very low. Cor— relation coefficients based on progeny family means were higher. Relative ranking of parents was quite consistent over time for a given tissue. This suggested limited replication of individual seedlings may have been a prob- lem. The low correlations indicate many misidentifications 0f individual genotypes would be made if selection were based on individual phenotypic performance. This is in cOnflict with the high heritability estimates. The true 106 heritability of the trait would markedly affect the breed- ing procedure to be used. With high heritability, most rapid progress would be made by backcrossing or recurrent mass selection based on individual performance. Low her- itability would dictate that some form of family selection should be employed. APPENDICES APPENDIX A Means and standard errors for cambium and xylem injury ratings to 'Velvet', 'Redhaven' and 'Siberian C' at each sampling date and test temperature, 1975—77. Table 1. Siberian C cambium xylem Redhaven cambium xylem velvet cambium xylem Temp Date (00) SE 2 SE 1': SE 2 SE 0510 0110 0.3.70 12”...) 0577 0111 0582 1123 7.700 1000 0900 3.455 1700 1100 0600 2355 5000 1000 3000 Q ‘ Q . 3555 1.400 1100 H200 2u55 3932 3837 2233 .... 9 V O N ...... ...... ...... ooooooo ....... uuuuuuu nnnnnnn 0000000 ....... ooooooo 0000000 oooooo 000000 ...... ...... 999999 oooooo ....... 0000000 0000000 ....... ....... ooooooo ZField control 107 108 Table 1 (cont'd.) -37.2 2.7 .13 2.0 .00 2.6 .11 5.0 .00 2.0 .05 5.0 .00 440.0 3.1 .07 5.0 .00 2.9 .10 5.0 .00 2.1 .19 5.0 .00 55007583 00011112 ........ llOluuroQ/ ........ 11111111. 2u014196 1.” 111111 ....... 2&238709 ........ 11111121 00000199 00002211 0 0 O O 8 6 9 2 11111112 02331355 11111111 ........ 2O7232IIHIIH ...... O . 12122222 05007946 00001111 0 l 0 O 6 2 O 2 ll 1 l3 1 15 l 10 l 16 l 11 2 l3 3 18 3 12 21 19 21 09 OO OOOOOO 3 0 8 u 8 o ...... 13 l 23 2 2O 2 l9 3 07 )4 OO 5 3 7 8 9 9 O 21 l 31 1 OO 2 OO 3 OO )4 OO 5 4 7 O O O O 28 2 25 2 OO 5 00 5 OO 5 OO 5 l 8 O O O 0 007903 00001.]. ...... 001807 ...... 11113“. 00010.“. 000111 0 O 0 ll 3 0 111123 ““7300 111110 ...... 9nu790 ...... l3uuu5 017019 110110 . O O O O 0 82921.42 007lhfi000 00011000 ........ 00165000 ........ 11113555 891371115 001110111. ........ 22““1460 ........ 11112224 00300000 01120000 ........ 03860000 ........ 11135555 230581.450 11110110 l22333y45 m197u209 o o b o 0 8681.470?) 5.223334” F. _ _ _ _ __ 6 Jan 05880000 10011000 ........ 11287000 ........ llllu555 88““6188 llllllll ........ 188O7IIHO3 oooooooo 21122233 7.5700000 01102000 I O. ' O .I 0 16208000 oooooooo 11354555 5213771.!— 111111111 ........ 22038OIIH6 07807000 00011200 ........ 01278800 ........ 111121455 1577Ohahwro 21112111 ........ 06502802 ........ 21122233 34540000 11210000 ........ 45300000 ........ 119745555 01963820 21111010 Apr 13 Field -6.7 -9.4 -12.2 —15.0 -17.8 —20.0 109 Table 1 (cont'd.) WWWNNNNN I I I I I I I I 03-1:meth H O\ WNNNNHI—‘I—I 00.. o o u I EHKONNCDJEJ: o o n)n)FJFJF‘F‘F‘H I I I I O I I O u)m.:_: v I u x I“ I x \ . u ....\~\.::~.-“\=-== ~AD \ \ //l . x [I ».:_:_ tam o>..m.omo> x >coE.n: :w.umcmo cox.m: x auaaom «0:.aw uoo.m so..a: x o:~..~a.m :ox.m: x cm cca:09.w < V m N . ..l A... \ I“ II x \ \\\\\\. :ox.u: u x u NQNIwrwwI OWN-O—POI ON 0? 00 SIT . ‘ .... . 1 . .....n.....:...... Figure 4. Progeny distributions according to overall mean injury ' ratings of each seedling as assessed by averaging individual seedling performance across all tissues, temperatures and dates. Siberianc X Hatkon 118 50 .0 c« n' 38 Or. bf I'a SH V. n V. O .. m u a M 8 2 O n 5 9 n n. e n .4 2 o ax — g k In txk cl \ e r d 6 6x t \ a 1- 1 a n 8 4| 1 va‘ on M H n 6 6 H \“ I a H \\\ \ u G C H X‘ vA\ vIA ’ \ IUI. lull n:- \\v ” QQQ \\x \"A ‘\ C \ \ \ \ \ .\ \\ \ \\ $.33. ,' In I -’ ‘ .OOx ’ DOC. ’LT‘oo-oooouao I. I'-II'..‘ .lx ‘ I I Illl If, .00 Ilnm IIIIIIIIIIIIIIIIIIII II / ''''' 000.90. x 5...”..- .. . Jeri!!! I llllllllllllllllll I It...... ’fcooooooooooooooooooooow!roohho \vA IIIII “I .3 ’ o 000.0000 ...-CI... ’ \ on oauununnnnnuooo o III-IIIIIIIIIIII-IIlI-I-‘x VA \h’v’ .....0..... ‘ o.coo.oooooooooooooooooooooox OOOOO / 4O 30 O O 2 1 3.2 2.8 2.4 2.0 1.6 lniury BIBLIOGRAPHY BIBLIOGRAPHY Allard, R.W. 1960. Principles of plant breeding. Wiley & Sons Inc., New York. Andersen, R.L. 1970. Some quantitative genetic consider- ations pertinent to breeding asexually propagated crops. Ph.D. thesis, University of Minnesota. . 1974. Peach hardiness and varieties. Proc. 33rd Ann. Conv. Nat'l. Peach Council, St. Louis, Mo. Feb 10-13. pp. 43-49. Asahina, E. 1967. A principle of the frost-resistance mechanism in plant and animal cells. In Prosser, C.L. The cell and environmental temperature (English trans.) Pergamon Press, New York. pp. 15-23. Bailey, J.S. and A.P. French. 1942. The inheritance of blossom type and blossom size in peach. Proc. Amer. Soc. Hort. Sci. 40: 248-250. Bittenbender, H.C. and G.S. Howell. 1975. Interactions of temperature and moisture content on spring de-accli— mation of flower buds of highbush blueberry. Can. J. Plant Sci. 55: 447-452. Blake, M.A. 1935. Types of varietal hardiness in the peach. Proc. Amer. Soc. Hort. Sci. 33: 240—244. . 1938. Hardy rootstocks for the peach should extend well above the surface of the soil. Proc. Amer. Soc. Hort. Sci. 36: 138—140. , and C.H. Steelman. 1945. Preliminary invest- igations of the cold resistance of peach fruit buds at the pink bud stages of development. Proc. Amer. Soc. Hort. Sci. 44: 37-41. Bradford, F.C. and H.A. Cardinell. 1926. Eighty winters in Michigan orchards. Mich. Agr. Exp. Sta. Bul. 149: 1-103. Brierley, W.G. and R.H. Landon. 1946. Some relationships between rest period, rate of hardening, loss of cold resistance and winter injury in the Latham raspberry. Proc. Amer. Soc. Hort. Sci. 47: 224-234. 119 120 and . 1954. Effects of dehardening and rehardening treatments upon cold resistance and injury of Latham raspberry canes. Proc. Amer. Soc. Hort. Sci. 63: 173-178. Brittain, J.A. and R.W. Miller. 1976. Managing peach tree short life in South Carolina. Clemson Univ. cir. no. 568. Buchanan, D.W., R.H. Biggs and J.F. Bartholic. 1976. Cold acclimation of Florida peach and.nectarine cultivars. HortScience 11: 398-400. , , and . 1974. Cold hard- iness of peach and nectarine trees growing at 29-30° N latitude. J. Amer. Soc. Hort. Sci. 99: 256—259. Burke, M.J., R.G. Bryant and C.J. Weiser. 1974. Nuclear magnetic resonance of water in cold acclimating ed sier dogwood stem. Plant Physiol. 54: 392-398. , M.F. George and R.G. Bryant. 1975. Water in plant tissues and frost hardiness. In R.B. Duckworth (ed.) Water relations in foods. Academic Press, New York. pp. 111-135. , L.V. Gusta, H.A. Quamme, C.J. Weiser and P.H. Li. 1976. Freezing and injury in plants. Ann. Rev. Plant Physiol. 27: 507—528. Cain, D.W. and R.L. Andersen. 1976. Sampling procedures for minimizing non—genetic wood hardiness variation in peach. J. Amer. Soc. Hort. Sci. 101: 668-671. Campbell, R.W. 1948. More than thirty peach varieties survived minus thirty-two degrees fahrenheit. Proc. Amer. Soc. Hort. Sci. 52: 117-120. , and F.B. Hadle. 1960. Winter injury to peaches and grapes. Proc. Amer. Soc. Hort. Sci. 76: 332— 337. Chandler, W.H. 1913. Killing of plant tissues by low temperature. Univ. of Mo. Res. Bul. 8: 141—309. . 1954. Cold resistance in horticultural plants: A review. Proc. Amer. Soc. Hort. Sci. 50: 552-572. Chaplin, C.E. 1948. Some artificial freezing tests of fruit buds. Proc. Amer. Soc. Hort. Sci. 52: 121- 129. , and G.W. Schneider. 1974. Peach rootstock/ scion hardiness effect. J. Amer. Soc. Hort. Sci. 99: 231—234. 121 Daniell, J.W. and F.L. Crosby. 1968. Occlusion of xylem elements in peach trees resulting from cold injury. Proc. Amer. Soc. Hort. Sci. 90: 128 - 134. , and . 1971. The relation of physi— ological stage, preconditioning and rate of fall of temperature to cold injury and decline of peach trees. J. Amer. Soc. Hort. Sci. 96: 50-53. Dennis, F.G. Jr., G.P. Lumis and C.R. Olien. 1972. Com- parative freezing patterns in stems of cherry and azalea. Plant Physiol. 50: 527—530. Dorsey, M.J. and P.D. Strausbaugh. 1923. Plum investi- gations. I. winter injury to plum during dormancy. Bot. Gaz. 76: 113-143. , and J. Bushnell. 1925. Plum investigations. II. The inheritance of hardiness. Minn. Agr. Expt. Sta. Tech. Bul. 32. Edgerton, L.J. and M.B. Hoffman. 1952. The effect of thinning peaches with bloom and postbloom sprays on the cold hardiness of the fruit buds. Proc. Amer. Soc. Hort. Sci. 60: 155-159. . 1960. Studies on cold hardiness of peach trees. Cornell Agr. Expt. Sta. Bul. 958. Fejer, 8.0. 1976. Combining ability and correlations of winter survival, electrical impedance and morphol- ogy in juvenile apple trees. Can. J. Plant Sci. 56: 303-309. Fletcher, N.H. 1962. The physics of rain clouds. Cam- bridge Univ. Press, London. Fogle, H.W. and F.L. Overley. 1954. Winter hardiness of stone fruit varieties in irrigated central Washing- ton. Wash. Agr. Expt. Sta. Bul. 553. George, M.F., M.J. Burke, H.M. Pellett, and A.G. Johnson. 1 74. Low temperature exotherms and woody plant dis— tribution. HortScience 9: 519—522. , , and C.J. Weiser. 1974. Super- cooling in overwintering azalea flower buds. Plant Physiol. 54: 29—35. , and . 1977a. Cold hardiness and deep supercooling in xylem of shagbark hickory. Plant Physiol. 59: 319—325. 122 , and . 1977b. Supercooling in over- wintering azalea flower buds. Plant Physiol. 59: 326—328. Graham P. 1971. Cold injury and its determination in selected Rhododendron species. M.S. thesis, Univer- sity of Minnesota. Gullord M. 1974. Genetics of freezing hardiness in winter wheat (Triticum aestivum L.). Ph.D. thesis, Michigan State University. , C.R. Olien and E.H. Everson. 1975. Evaluation of freezing hardiness in winter wheat. Crop Sci. 15: 153—157. Hansche, P.E., C.O. Hesse and V. Beres. 1972. Estimates of genetic and environmental effects on several traits in peach. J. Amer. Soc. Hort. Sci. 97: 76— 79. Hedrick, U.P. 1917- The peaches of New York. J.B. Lyon 00., Albany, New York. Hesse, C.O. 1975. Peaches. In Janick, J. and J.N. Moore (eds.) Advances in fruit breeding. Purdue Univ. Press, Lafayette, Indiana. pp. 285-335. Hildreth, A.C. 1926. Determination of hardiness in apple varieties and the relation of some factors to cold resistance. Minn. Agr. Expt. Sta. Tech. Bul. 42. Hildreth, A.C. and L. Powers. 1941. The rocky mountains strawberry as a source of hardiness. Proc. Amer. Soc. Hort. Sci. 38: 410—412. Howell, G.S. and C.J. Weiser. 1970a. The environmental control of cold acclimation in apple. Plant Phys- iol. 45: 390-394. and . 1970b. Fluctuations in the cold resistance of apple twigs during spring de- hardening. Proc. Amer. Soc. Hort. Sci. 95: 190- 192. Johnston, E.S. 1923. Moisture relations of peach buds during winter and spring. Univ. Maryland Agr. Expt. Sta. Bul. 255. Ketchie, D.O. and C.H. Beeman. 1973. Cold acclimation in 'Red Delicious' apple trees under natural conditions during four winters. J. Amer. Soc. Hort. Sci. 98: 257-261. 123 Kitaura, K. 1967. Supercooling and ice formation in mulburry trees. In Asahina, E. (ed.) Cellular in- jury and resistance in freezing organism. Bunyeido Printing Co., Sapporo. pp. 143-156. Krasavtsev, O.A. 1967. Frost hardening of woody plants at temperatures below zero. In Asahina, E. (ed.) Cellular injury and resistance in freezing organisms. Bunyeido Printing Co., Sapporo. pp. 131-141. Lapins, K. 1962. Artificial freezing as a routine test of cold hardiness of young apple seedlings. Proc. Amer. Soc. Hort. Sci. 81: 26-34. Layne, R.E.C. 1974. Breeding peach rootstocks for Can— ada and the northern United States. HortScience 9: 10-12. . 1976. Influence of peach seedling rootstocks on perennial canker of peach. HortScience 11: 509-511. , G.M. Weaver, H.O. Jackson and F.D. Stroud. 1976. Influence of peach seedling rootstocks on growth, yield and survival of peach scion cultivars. J.Amer.Soc. Hort. Sci. 101: 568—572. , H.O. Jackson and F.O. Stroud. 1977. Influ- ence of peach seedling rootstocks on defoliation and cold hardiness of peach cultivars. J. Amer. Soc. Hort. Sci. 102: 89—92. Levitt J. 1962. A sulfhydryl—disulfide hypothesis of frost injury and resistance in plants. J. Theor. Biol. 3: 355—391. . 1972. Responses of plants to environmental stresses. Academic Press, New York. Lumis, G.P., R.A. Mecklenburg and K.C. Sink. 1972. Fac- tors influencing winter hardiness of flower buds and stems of evergreen azaleas. J. Amer. Soc. Hort. Sci. 97: 124—127. , and . 1974. Freezing patterns in twigs of evergreen azalea. J. Amer. Soc. Hort. Sci. 99: 564-567. Luyet, B.J. 1970. Physical changes occurring in frozen solutions during rewarming and melting. In Walsten— holme, G.E.W. and M. O'Connor (eds.) The frozen ‘ cell. J. & A. Churchill Publications, London. pp. 27-50. 124 Marshall, H.G. 1976. Genetic changes in oat bulk popula- tions under winter survival stress. Crop Sci. 16: 9-15. Mazur P. 1966. Physical and chemical basis of injury in single-celled micro-organisms subjected to freezing and thawing. In Meryman H.T. (ed.) Cryobiology. Academic Press, New York. pp. 214-315. . 1969. Freezing injury in plants. Ann. Rev. Plant Physiol. 20: 419-448. Mckenzie, J.S., C.J. Weiser, B.J. Stadelmann and M.J. Burke. 1974. Water permeability and cold hardiness of cortex cells in Cornus stolonifera Michx. - A preliminary report. Plant Physiol. 54: 173-176. Meryman, H.T. 1970. The exceeding of a minimum toler- able cell volume in hypertonic suspension as a cause of freezing injury. In Wolstenholme, G.E.W. and M. O‘Connor (eds.) J. & A. Churchill Publishers, London. Metcalf, E.L., C.E. Cress, C.R. Olien and E.H.Everson. 1970. Relationship between crown moisture content and killing temperature for three wheat and three barley cultivars. Crop Sci. 10: 362-365. Morrison, J.W., C.R. Ure, R.H. Anderson, R.B. Harris and P.D. Hargrave. 1963. Hardiness and vigor of apple seedlings grown on the Canadian prairies. Proc. Amer. Soc. Hort. Sci. 83: 113-119. Mowry, J.B. 1964. Inheritance of cold hardiness of dor- mant peach flower buds. Proc. Amer. Soc. Hort. Sci. 85: 128-133. Nesmith, W.C. and W.M. Dowler. 1976. Cultural practices affect cold hardiness and peach tree short life. J. Amer. Soc. Hort. Sci. 101: 116-119. Olien, C.R. 1961. A method of studying freezing stresses occurring in plant tissues during freezing.Crop Sci. 1: 26-28. . 1964. Freezing processes in the crown of 'Hudson' barley, Hordeum vulgare (L., emend. Lam.) Hudson. Crop Sci. 4: 91-95. . 1965. Interference of cereal polymers and related compounds with freezing. Cryobiology 2: 47-54. 125 . 1967a. Preliminary classification of poly- saccharide freezing inhibitors. Crop Sci. 7: 156-157. . 1967b. Freezing stresses and survival. Ann. Rev. Plant Physiol. 18: 387-408. , B.L. Marchetti and E.V. Chomyn. 1968. Ice structure in hardened winter barley. Quart. Bul. Mich. Agr. Expt. Sta. 50: 440-448. . 1970. Freezing of plant tissues and the pro- tective effect of naturally occurring substances, Symposium, Michigan State Univ. March 23-25, 1970. . 1971. A comparison of desiccation and freez- ing as stress vectors. Cryobiology 8: 244-248. . 1973. Thermodynamic components of freezing stress. J. Theor. Biol. 39: 201-210. , and Su-En Chao. 1973. Liquid water content of cell walls in frozen tissues evaluated by elec— trophoresis of indicators. Crop Sci. 13: 674-676. . 1974a. Energies of freezing and frost des- iccation. Plant Physiol. 53: 764-767. . 1974b. Winter hardiness in barley. Mich. State Univ. Res. Rep't. 247. , and B.L. Marchetti. 1976. Recovery of hard- ened barley from winter injuries. Crop Sci. 16: 201-204. . 1977. Barley: patterns of response to freezing stress. U.S.D.A. Tech. Bul. 1558. Ormrod, D.P. and R.E.C. Layne. 1974. Temperature and photoperiod effects on cold hardiness of peach scion-rootstock combinations. HortScience. 9: 451- 453. Ormrod, D.P. and R.E.C. Layne. 1977. Scion and rootstock influence on winter survival of peach trees. Fruit Varieties J. 31: 30-33. Overley, F.L. and E.L. Overholser. 1936. Low temperature injury of fruit trees in central Washington during 1935-36. Proc. Wash. State Hort. Assn. 32: 147-151. Pienazek, S.A., S.W. Zagaja and A. Wojniakiewicz. 1968. Winter—hardy peaches (P. vulgaris L.) from China. Bul. Acad. Pol. Sci. 16: 715—717. 126 Pierquet, P., C. Stushnoff and M.J. Burke. 1977. Low temperature exotherms in stem and bud tissues of Vitis riparia Michx. J. Amer. Soc. Hort. Sci. 102: 54-55. Potter, G.F. 1938. Low temperature effects on woody plants. Proc. Amer. Soc. Hort. Sci. 36: 185-195. Proebsting, E.L. Jr. 1970. Relation of fall and winter temperatures to flower bud behavior and wood hardi— ness of deciduous fruit trees. HortScience 5: 22— 24. Quamme, H., C. Stushnoff and C.J. Weiser. 1972a. The re- lationship of exotherms to cold injury in apple stem tissues. J. Amer. Soc. Hort. Sci. 97: 608-613. , , and . 1972b. Winter hardiness of several blueberry species and cultivars in Minnesota. HortScience 7: 500-502. , C.J. Weiser and C. Stushnoff. 1973. The mechanism of freezing injury in xylem of winter apple twigs. Plant Physiol. 51: 273—277. , R.E.C. Layne, H.O. Jackson and G.A. Spearman. 1975. An improved exotherm method for measuring cold hardiness of peach flower buds. HortScience 10: 521—523. 1976. Relationship of the low temperature exotherm to apple and pear production in North Amer- ica. Can. J. Plant Sci. 56: 493-500. Rasmussen, D.H. and A.P. Mackenzie. 1972. Effect of solute on ice-solution interfacial free energy; cal- culation from measured homogeneous nucleation temper- atures. In Jellinek H.H.G. (ed.) Water structure at the water-polymer interface,P1enum Press, New York. Rollins, H.A. Jr., F.S. Howlett and F.H. Emmert. 1962. Factors affecting apple hardiness and methods of measuring resistance of tissue to low temperature injury. Ohio Agr. Expt. Sta. Res. Bul. 901. Samygin, G.A. 1967. Protoplasmic dehydration as one of the causes of cell death from the formation of extra- cellular ice. In Prosser, C.L. (ed.) The cell and environmental temperature (English trans.) Pergamon Press, New York. pp, 44—52, Savage, E.F. and V.E. Prince. 1972. Performance of peach cultivars in Georgia. Univ. of Georgia Res. Bul. 114. 127 , R.B. Jensen and R.A. Hayden. 1976. Peach tree micro-climate and methods of modification. Univ. of Georgia Agr. Expt. Sta. Res. Bul. 192. Siminovitch, D., B. Rheaume and R. Sachar. 1967. Seasonal increase in protoplasm and metabolic capacity in tree cells during adaptation to freezing. In Prosser, C.L. (ed.) Molecular Mechanisms of Temperature Adap- tion Amer. Assoc. Adv. Sci. Pub. 84. , K. Pomeroy and M. Lepage. 1968. Phospholipid, protein and nucleic acid increases in protoplasm and membrane structures associated with development of extreme freezing resistance in black locust tree cells. Cryobiology 5: 202-225. Single, W.V. and C.R. Olien. 1967. Freezing processes in wheat stems. Aust. J. Biol. Sci. 20: 1025-1028. Smithberg, M.H. and C.J. Weiser. 1968. Patterns of vari— ation among climatic races of red—osier dogwood. Ecology 49: 495—505. Steel, R.G.D. and J.H. Torrie. 1960. Principles and pro- cedures of statistics. McGraw—Hill Book Co. Inc., New York. Strausbaugh, P.D. 1921. Dormancy and hardiness in the plum. Bot. Gaz. 71: 337-357. Stushnoff, C. 1972. Breeding and selection methods for cold hardiness in deciduous fruit crops. HortScience 7: 10-13. Tekauz, A. and Z.A. Patrick. 1974. The role of twig in— fections on the incidence of perennial canker in peach. Phytopathology 64: 683-688. Traub, H.P. 1927. Regional and seasonal distribution of moisture, carbohydrates, nitrogen and ash in 2-3 year portions of apple twigs. Univ. of Minn. Agr. Expt. Sta. Tech. Bul. 53. Watkins, R. and L.S.P. Spangelo. 1970. Components of genetic variance for plant survival and vigor of apple trees. Theor. and App. Gen. 40: 195-203. Weaver, G.M. 1963. Influence of rootstock susceptibility of peach to peach canker. Fruit Var. and Hort. Digest 17: 43—44. . 1968. Crotch angle development in peach trees as influenced by scion and rootstock cultivars. Can. J. Plant Sci. 48: 419—421. 128 Weiser, C.J. 1970. Cold resistance and injury in woody plants. Science 169: 1269—1278. Wildung, D.K., C.J. Weiser and H.M. Pellett. 1973. Temper- ature and moisture effects on hardening of apple roots. HortScience 8: 53-55. Williams, J.M. and R.J. Williams. 1976. Osmotic factors of dehardening in Cornus florida L. Plant Physiol. 58: 243-247. Zagaja, S.W. 1974. Breeding cold hardy fruit trees. Proc. Int'l. Hort. Cong. 19: 9-15. \_/' "I11'?11111111111111'55