INFORMATION TO USERS This reproduction was made from a copy of a m anuscript sent to us for publication and microfilming. While th e m ost advanced technology has been used to pho­ tograph and reproduce this m anuscript, the quality of the reproduction is heavily dependent upon the quality of the material submitted. Pages in any m anuscript may have indistinct print. In all cases the best available copy has been filmed. The following explanation of techniques is provided to help clarify notations which may appear on th is reproduction. 1. M anuscripts may not always be complete. When it Is not possible to obtain missing pages, a note appears to indicate this. 2. When copyrighted materials are removed from the m anuscript, a note ap­ pears to indicate this. 3. Oversize materials (maps, drawings, and charts) are photographed by sec­ tioning the original, beginning at the upper left hand com er and continu­ ing from left to right in equal sections with small overlaps. Each oversize page is also filmed as one exposure and Is available, for an additional charge, as a standard 35mm slide or in black and white paper format.* 4. Most photographs reproduce acceptably on positive microfilm or micro­ fiche b u t lack clarity on xerographic copies made from the microfilm. For an additional charge, all photographs are available in black a n d w hite standard 35mm slide format. • •For more information about black and white slides or enlarged paper reproductions, please contact the Dissertations Customer Services Department TUTA / T T Dissertation IVII Information Service University Microfilms International A Bell & Howell Information C om pany 3 00 N. Z eeb R oad. Ann Arbor, M ichigan 48106 8625028 H e a t l e y , R a n d a ll C h a r l e s GENETIC VARIATION OF C O R N U S FLORIDA IN A MICHIGAN PRO VEN ANCE TEST M ic h i g a n S ta le U n iv e r s ity University Microfilms international 300 N. Zeeb Road, Ann Arbor, Ml 48106 PH.D. 1986 PLEASE NOTE: In all cases this material has been filmed In the best possible way from the available copy. Problems encountered with this docum ent have been identified here with a check mark V . 1. Glossy photographs or p a g e s _____ 2. Colored illustrations, paper or print______ 3. Photographs with dark background_____ 4. Illustrations are poor copy_______ 5. Pages with black marks, not original copy______ 6. Print shows through as there Is text on both sides of p a g e _______ 7. Indistinct, broken or small print on several pages 8. Print exceeds margin requirem ents______ 9. Tightly bound copy with print lost in sp in e_______ yf 10. Computer printout pages with indistinct print______ 11. Page(s)____________lacking when material received, and not available from school or author. 12. Page(s)___________ seem to b e missing in numbering only as text follows. 13. Two pages num bered 14. Curling and wrinkled p a g es______ 15. Dissertation contains pages with print at a slant, filmed a s received_________ 16. Other_______________________________________________________________________ . Text follows. University Microfilms International GENETIC VARIATION OF CORNUS FLORIDA IN A MICHIGAN PROVENANCE TEST By Randall Charles Heatley A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Forestry ABSTRACT GENETIC VARIATION OF CORNUS FLORIDA IN A MICHIGAN PROVENANCE TEST By Randall Charles Heatley Genetic variation of fall color display, cold temperature hardiness and growth of Cornus florida seedlots from nineteen states, growing in Kalamazoo County, Michigan, were examined. hilltop, a hillside, a frost pocket, and in the shade of pine trees. colored, The trees were grown in four blocks; an open Fall color was rated for percent of leaves the color developed and the percent of leaves dropped. These ratings were used to calculate the number of days on which 70% of the leaves were bright red, but before 60% of the leaves abscised. Controlled cold temperature stress tests were used to determine the T50's of twigs from the provenance seedlots and from native trees. Hardiness of bracts and florets from native trees was determined by controlled cold temperature stress tests. Fall color display was found to vary by geographic origin of seedlots, rating. site in the provenance test and year of Seedlots from northern states had the longest red fall color display, while southern trees had very little if any. The open hilltop trees had the longest fall color displays. The frost pocket trees had shorter fall color displays because development. winter in j u r y delayed fall col o r Shaded trees were most delayed and also failed to develop uniform, bright red fall foliage color. The hardiest trees, northern states. Trees as measured by T50, from the were southernmost from states suffered frost injury in the autumn and were killed to the ground each winter. dogwood Outer bracts of flower buds from native were the least hardy flower bud structure. Flower buds dehardened in response to warm spring temperatures and were killed by subsequent cold weather. The tallest trees in the plantation were under the pines because trees in the other blocks suffered more winter injury. the b e s t Michigan, Ohio and Pennsylvania seedlots provided combination temperature hardiness. of fall color d i s p l a y and cold I dedicate this work to Suzanne and Scott Heatley for the love, patience and understanding they so unselfishly gave during the time I spent working to complete the requirements for my degree. ACKNOWLEDGMENTS I acknowledge, with gratitude, the thoughtful advice and helpful criticism provided by Dr. J. J. Kielbaso in guiding the research and completion of the dissertation. I am very appreciative of his many hours spent working with me as my major professor. I thank Dr. G. S. Howell for his help and advice, and for the use of his laboratory facilities for the cold temperature stress portion of the research. I am grateful to Dr. J. W. Wright for the help and guidance he gave while I was doing my research and writing the dissertation. I thank Dr. D. I. Dickmann for his helpful suggestions on preparation of the dissertaion. TABLE OF CONTENTS LIST OF TABLES...................................... LIST OF FIGURES INTRODUCTION . vii ...................................... LITERATURE REVIEWED ............................... ANNUAL ANNUAL ANNUAL ANNUAL GROWTH GROWTH GROWTH GROWTH Page v CYCLE - GENERAL DESCRIPTION . . CYCLE - PHYSIOLOGICAL . . . . CYCLE - GENETIC ................ CYCLE - ENVIRONMENTAL . . . . METHODS AND MATERIALS 1 3 3 9 17 22 ............................ 31 PLANTATION ESTABLISHMENT...................... FALL COLOR M E A S U R E M E N T ...................... COLD HARDINESS ASSESMENTS ................... 31 34 35 R E S U L T S ............................................ NURSERY DATA................................... FALL C O L O R ................................... FALL COLOR DISPLAY............................ GROWTH......................................... COLD TEMPERATURE HARDINESS ................... 41 41 43 60 63 67 DISCUSSION AND CONCLUSIONS......................... 77 APPENDIX............................................ 85 LITERATURE CITED ................................... 87 LIST OF TABLES Table Page 1. The ranges of seedlot germination percentages of Cornus florida seeds 28 months after being planted in autumn of 1972 in a nursery bed in East Lansing, Michigan...............................42 2. Main effects of geographic origin, site in plantation and year sampled on the mean number of days leaves were red on Cornus florida in 1979, 1980 and 1981 when grown In ^ p r o v e n a n c e test in Kalamazoo County, Michigan...........................62 3. Range of F values for 1979-1981 fall color ratings and height of Cornus florida growing in Kalamazoo County, Michigan..................................... 64 4. Comparison of mean annual growth and total height of Cornus florida growing on different sites in Kalamazoo County, Michigan....................... 66 5. Percent of replications of year old Cornus florida seedlings with topkill in a nursery at East Lansing, Michigan in 1974.................... 68 6. Temperatures (°C) on 4 test dates at which 50% of the twigs were killed (T50*s), for Cornus florida grown on an open hilltop in Kalamazoo County, Michigan..................................... 70 7. Temperatures (°C) on 4 test dates at which 50% of the twigs were killed (T50's), for Cornus florida grown in the shade in Kalamazoo County Michigan..............................................71 8. Temperatures (°C) at which 50% of the twigs were killed (T50's) by position in crown, for native Cornus florida growing in the shade in Kalamazoo County, Michigan...........................74 9. State of origin and length of fall color display of Cornus florida growing in a provenance test in Kalamazoo County, Michigan...........................85 v Table 10- Page Average number of days Cornua florida had red fall foliage color in 1979, 1980 and 1981 when grown on different sites in a provenance test in Kalamazoo County, Michigan..................................... 86 LIST OF FIGURES Figure 1. 2. Page Collection Sites for Cornus florida Seeds. . . . 32 Location of the Four Cornus florida Blocks at W. K. Kellogg Forest............................... 33 3. Percent color, leaf color and percent abscission of leaves of Michigan seedlots of Cornus florida growing on an open hilltop in a provenance test in Kalamazoo County, Michigan........................... 46 4. Percent color, leaf color and percent abscission of leaves of Michigan seedlots of Cornus florida growing in a frost pocket in a provenance test in Kalamazoo County, Michigan. . . . 47 5. Percent color, leaf color and percent abscission of leaves of Michigan seedlots of Cornus florida growing in the shade of pines in a provenance test in Kalamazoo County, Michigan........................ 48 6. Comparison of percent of leaves colored at weekly intervals on Michigan seedlots of Cornus florida growing on an open hilltop, in a frost pocket and in the shade of pines in a provenance test in Kalamazoo County, Michigan........................... 49 7. Comparison of leaf color at weekly intervals on Michigan seedlots of Cornus florida growing on an open hilltop, in a frost pocket and in the shade of pines in a provenance test in Kalamazoo County, Michigan.................. 50 8. Comparison of percent of leaves abscised at weekly intervals on Michigan seedlots of Cornus florida growing on an open hilltop, in a frost pocket and in the shade of pines in a provenance test in Kalamazoo County, Michigan........................... 51 9. Comparison of percent of leaves colored at weekly intervals on Michigan, Mississippi and Tennessee seedlots of Cornus florida growing on an open hilltop in a provenance test in Kalamazoo County, Michigan.............................................. 53 vii Figure Page 10. Comparison of leaf color at weekly intervals on Michigan, Mississippi and Tennessee seedlots of Cornus florida growing on an open hilltop in a provenance test in Kalamazoo County, Michigan. 54 11. Comparison of percent of leaves abscised at weekly intervals on Michigan, Mississippi and Tennessee seedlots of Cornus florida growing on an open hilltop in a provenance test in Kalamazoo County, Michigan.............................................. 55 12. Comparison of percent of leaves colored at weekly intervals on Michigan, Mississippi and Tennessee seedlots of Cornus florida growing in the shade of pines in a provenance test in Kalamazoo County, Michigan.............................................. 56 13. Comparison of leaf color at weekly intervals on Michigan, Mississippi and Tennessee seedlots of Cornus florida growing in the shade of pines in a provenance test in Kalamazoo County, Michigan. . 57 14. Comparison of percent of leaves abscised at weekly intervals on Michigan, Mississippi and Tennessee seedlots of Cornus florida growing in the shade of pines in a provenance test in Kalamazoo County, Michigan.............................................. 58 viii INTRODUCTION What ig the value of Cornus florida, dogwood, when grown in landscapes? "This growing in the Here are two assesments. tree is the best ornamental of all the flowering northern United States. It the natives has special interest every season of the year-in spring with flowers; in summer with good foliage not marred by insect or disease; in fall with brilliant red berries and vivid autumn color; in winter method of because of its picturesque horizontal branching" {Wyman 1965). "Ornamentally obliterated from the Midwest during winter 1976-77; flower buds and stems were killed; often the two were inner bracts were the only ones to develop, twisted and deformed; and these the southern-grown dogwood are too much a part of the northern landscape; plants need to be grown from northern seed sources; although winter 1977-78, was milder, injury appeared as bad; the pink flowering forms were also injured; after looking at dogwood falter the past two springs, we are opting for redbud" (Dirr et al 1978). These performance, genetic two descriptions of flowering dogwood, state variation ornamental traits the main reasons in the species. that give potential as a landscape plant. 1 for its investigating Wyman lists flowering and the dogwood many great However, Dirr et al (1978) 2 describes the major reason why the ornamental potential of the tree is not realized. Cold winter temperatures kill the flower buds and, if especially severe, can damage twigs. Are the southern dogwoods too much a part northern landscape as Dirr et al. (1978) suggest? of the How are southern flowering dogwoods different from northern dogwoods when grown in the north? Identification of some of the differences in hardiness and fall foliage color was why this work was undertaken. LITERATURE REVIEWED Deciduous trees growing in Michigan go through an annual growth cycle that can be described in four ways. general description provides an overview cycle. of the The growth The physiological description presents the results of research that identifies physiological processes taking place in the plant in each phase of the growth cycle. The third is genetic which describes how plants with different genotypes perforin when grown together. The environmental description summarizes how common environmental factors influence trees during the phases of the growth cycle. The cycle descriptions begin with flowering, since flowers are the major reason for using flowering dogwood in the landscape. ANNUAL GROWTH CYCLE - GENERAL DESCRIPTION Maps were presented by Reader (1975) summarizing flowering times of Cornus florida in the eastern United States for 1970, 1971, 1972 and 1973. early as mid-February in Florida, Flowering can be as or as late as May in Michigan. The description of the opening of dogwood flowers given by Sargent (1961) applies to trees from all parts of the range. bracts The flower stem begins to elongate and the four enclosing petal-like bracts the flower head open and expand. open, the 3 flowers are exposed. As the The 4 yellow, perfect flowers are in the center of what appears to be a white, pink, or red, 10 centimeter "flower" but is in reality an inflorescence. Growth V e g e t a t i v e buds b e g i n g r o w t h flowering occurs. in the sp r i n g when The tree enters a phase called the "grand phase of growth" during which it grows at the maximum rate for This the year. development can last rapid stem elongation and leaf from 4 to 12 weeks in fruit trees (Seeley 1978). Growth Cessation The grand phase of growth ends and terminal buds become quiescent then, according to Seeley (1978), winter scales form around the terminal buds in mid and late summer. Cessation of growth is considered to be the onset of dormancy. (1970). Dormancy is divided into two phases by Weiser Dormant quiescent stopped growing but will plants are those Dormant resting plants have stopped growing and entered a rest period. under conditions have start again when placed under conditions that induce growth. grow that that promote These will not growth until a requirement for exposure to cold temperatures has been met. Other phases which overlap the onset of dormancy are fall color development, cold temperatures• leaf abscission, and acclimation to 5 Fall Color Development Senescing dogwood leaves change from green to red. The development of reddish fall color in senescing Metasequoia l e aves was observed by Ida (1981). The first co l o r developed at the leaf tip in the area between the veins. leaf Color spread to the middle and base of the leaves until the entire leaf was colored. Leaf Abscission Leaf abscission ends the fall color display. Esau (1977), Kozlowski (1973) and Webster (1973) provide brief descriptions of leaf abscission. the leaf senesces. Leaf activity declines and Senescence triggers activity in cells of the abscission zone which is located at the base of the leaf petiole. This activity leads to a weakened between the leaf and twig. connection The connection is eventually broken in a layer of cells called the separation layer and the leaf falls. The wound on the twig, left by the fallen leaf, is sealed by the formation of a protective layer. The separation and protective layers are subdivisions of the abscission zone. Acclimation To Freezing Temperatures After growth has stopped, and while the leaves color and abscise, the tree acclimates to freezing temperatures. Weiser (1970) described the two stage acclimation process. The first stage occurs after growth has stopped, but then i 6 plateaus. The second stage is triggered by frost and leads to maximum cold hardiness. Acclimation may not always develop at the same rate in all plant parts. Azaleas tested by Alexander and Havis (1980) hardened first at the branch tips and last at the bases. Cain and Anderson (1976) found the bases of peach branches to be hardier than branch tips. differences between tree sectors. They also found Branches from the upper southwest side of the trees were hardier than those from the lower northeast side. studer et al (1978) found the roots of container grown ornamentals to be less hardy than stems. The roots of container grown juniper tested by Pellett and White (1969) did not harden below -12°C while tops survived temperatures as low as -39°C. When acclimation is complete, the tree has attained its maximum capability to survive low temperature stress during the winter. Winter Survival Winter survival of hardy plants was reviewed by Weiser (1970). Water freezes in hardy plants at temperatures between -2°C and -8°C in laboratory studies. Ice at these temperatures forms between the cells or in dead conductive cells such as vessels or tracheids. throughout the stem. Ice formation spreads As the temperature drops, the cells dehydrate as water moves to the ice in the intercellular spaces. Very hardy plants tolerate extreme levels of cell 7 dehydration and survive to very low temperatures. Plants or plant parts are killed when water freezes within living cells. The rate of temperature drop can influence the amount of injury. White and Weiser (1970) recorded temperature drops of 17°F in one minute in Thuja occidentalis foliage at sunset. Litzow and Pellett (1983) found rapid temperature changes in stem sections of white ash when shade was applied or removed on sunny, winter days. The potential injurious effects of such temperature fluctuations were summarized by Kramer and Kowzlowski (1979). Tolerance to low temperature by many different plants has been investigated by Pellett et a^l (1981) and Sakai (1982). Sakai (1982) tested various plant parts, such as leaves, twigs, xylem. plants. vegetative and flower buds, cortex, and Differences were found between, as well as within Pellett et al (1981) compared survival of stem sections of different plants. They presented their results in tables along with those of Sakai's earlier work, and hardiness levels reported in Hortus XIX and Rehder's Manual of Cultivated Trees and Shrubs. Plant survival temperatures reported by the separate sources were often the same or within a degree of each other. However, differences between the four sources were as much as 19°C for some species. Flowering dogwood was listed hardy to -29°C or -30°C. 8 Spring Loss Of Cold Hardiness The ability to withstand cold temperatures is gradually lost with the onset of warm spring weather. Warmer spring temperatures caused the loss of cold hardiness in apple trees tested by Howell and Weiser (1970). The trees were able to partially regain lost hardiness, but at a slower rate than it was lost. If a p l a n t dehardening is in rest, conditions it is less and h a r d i n e s s responsive is more to easily retained. Peach flower buds examined by Proebsting (1963) maintained a minimal Once rest was level of hardiness while satisfied by cold in rest. temperatures, warm temperatures induced the buds to deharden above the minimal hardiness level. Cornus stolonifera tested by Litzow and Pellett (1980) did not grow or lose more than a few degrees of cold hardiness until exposure satisfied the rest requirement. to cold temperatures Rest was satisfied as early as November 28 and plants would deharden in response to warm temperatures. longer than Plants exposed to cold temperature, for times the minimum needed to satisfy rest, lost hardiness more rapidly. Dehardening of twigs may not be completed before the spring growth flush begins. hardiness levels of George and Burke (1977) tested shagbark hickory by freezing sections from September 1974 to September 1975. stem The plants rapidly dehardened in April and reached m i n i m u m hardiness 9 levels in late May, at which time the 1974 wood was hardier than the actively growing 1975 wood. Hardiness started to increase for both ages of wood in June and by mid-July they were equally hardy. II. ANNUAL GROWTH CYCLE - PHYSIOLOGICAL Growth Salisbury and Ross (1978) suggested that auxin moving down the stems from buds promotes cell elongation, and that cambial activity begins when auxin and gibberellins are released by developing leaves. Apical meristem activity lenthens stems and branches by adding cells to the stem apices (Kramer and Kozlowski 1979). Cells produced by plant meristems differentiate and mature, forming the various tissues that make up the plant. Growth Cessation And Acclimation To Freezing Temperatures Growth cessation, fall leaf abscission, and acclimation overlap. coloration, leaf They are considered together. In response to environmental factors, woody plants stop growing in mid to late summer. Alvim et a^ (1978) found that the abscisic acid content of sap and apices of Salix viminalis was highest on July 9. willow growing Growth measurements for under natural daylength showed a steady decline from that time until they stopped growing completely in early September. 10 Several relations Minnesota researchers have in a c c l i m a t i n g s t o l o n i f e r a . studied The water water permeability of Cornus stolonifera cortex cells was found to increase during the early stages of acclimation to cold temperatures (Mckenzie et al, 1974). below freezing but noninjurious Exposing the plants to temperatures increased hardiness but did not increase the water permeability of the cortex cells. A growth chamber study of acclimation in was c o n d u c t e d by P a r s o n s conducted twice, (1978). from North Dakota was used. were development The e x p e r i m e n t was once from July 14 to September 22 and from September 10 to November 4. of plants stolonifera used, one In both experiments a clone In each experiment two series under of cold hardiness conditions that induce and the other under non- inductive conditions. Parsons' (1978) data show that acclimating plants, as compared to non-acclimating plants, had reduced stomatal resistance within 6 days. The stomates on the undersides of the leaves remained more open for 5 to 7 weeks and then became less open. When stomatal resistance increased, hardiness levels were -12°C to -14°C. A similar trend was observed for transpiration. In the fall series of the experiment, acclimating plants always had higher root resistance to water movement. Stem water content of acclimating plants continued to 11 decline and was 50 to 55% that of the nonacclimating plants by the end of the experiment. Bray and Parsons (1981) also studied acclimation in the North Dakota clone of stolonifera natural conditions in Minnesota. acclimating under The plants started to develop red stem color on September 12 and fall leaf color by October 3. The first frost for that year, 1977, occurred on October 10 and leaf abscission occurred on October 26. On September 13, hardiness levels increased slightly from -6°C to -8°C. October 19. Hardiness leveled out near -20°C until Stomatal resistance remained constant until a slight increase was recorded on October 13 when hardiness levels were -18°C. From then, until readings were stopped, stomatal resistance increased. Stem water contents were first sampled on August 29th. Stem water content was expressed as grams of water per gram of dry weight. On August 29 stem water content was 3.9 grams of water per gram of dry weight and declined to 2.7 grams on September 5. On September 13 the stem water content was 1.8 grams of water per gram of dry weight. Further declines were not as large and stem water content reached .73 grams of water per gram of dry weight on October 26. Roots as well as leaves may have a role in reducing a plant's water content during acclimation. McKenzie et al (1974) tested root water conductivity of the North Dakota clone of C. stolonifera at different hardiness levels. The 12 root systems were tested while in soil and water. They were also tested after being killed by boiling. acclimating plants were compared Root systems of with those of nonacclimating plants. Water flow through the roots of nonacclimating plants was .28 ml of water in 30 minutes as compared to .08 ml of water for acclimating plants. When the stems were hardy to -45°C the flow of water was .01 ml in 30 minutes. The mechanism of root resistance to water flow may change as acclimation progresses. The resistance of dead root systems was nearly the same for plants hardy to -12°C and for nonacclimated plants. The water flow through the root system from a plant hardy to -45°C was half that through a root system from a tender plant after both were killed by boiling. McKenzie et al (1974) note that the roots from the plants hardy to -45°C were darker and suggest this may be due to suberization. While acclimation is occurring, the leaves change color as they senesce. Leaf senescence is controlled by genes located in the nucleus (Thomas and Stoddard, 1980). researchers suggest a number of initiation of senescence. competition r e g u l ators? for light, factors that These influence The list of factors includes: space, and e n v i r o n m e n t a l nutrients factors. and growth They p r o p o s e possible mechanisms by which senescence is induced. As chloroplast components are degraded or used up, the balance 13 of proteins may be altered to allow expression of senescence genes. Declining photosynthetic rates due to loss of chloroplast enzymes may reduce stored energy availability leading to the onset of senescence. A third suggestion is a change in cell pH. Wareing and Phillips (1978) suggest that an imbalance in relative levels of initiation of senescence. growth hormones is involved in The hormonal imbalance may be due to environmental factors or to competition between plant organs. Thimann (1980) suggested that leaf senescence occurs when kinetin can no longer keep the stomates open. Stomate closure then leads to an increase in abscisic acid and leaf senescence. The fall color display ends as the senescent leaves abscise. A model describing regulation of leaf abscission has been developed by Osborne (1973). Her model assumes that environmental factors create an internal imbalance of hormones. occurs. As a result of the imbalance, leaf senescence Protein and nucleic acid synthesis is reduced, auxin levels are lowered and membrane permeability is lost. A senescence factor leaks from its membrane-bound storage compartment and causes enhanced biosynthesis of ethylene. Abscission zone cells respond to the ethylene by increased production of proteins and nucleic acids. Enzymes increase their activity and degrade cell walls in the separation 14 layer. Cells grow on the proximal side of the abscission zone, providing the force to cause cell wall breakage and leaf abscission. Winter Survival Once the leaves have acclimation is complete, colored and abscised, and the physiological processes needed for winter survival become most important. Plants die when water freezes inside living cells or when too much water is withdrawn from cells (physiological intercellular ice crystals (Weiser, However, diffuse to form 1970). the water must be given sufficient time to through crystals. drought) the Harrison me mbrane et al to (1978) the intercellular used nuclear ice magnetic resonance spectroscopy to compare the freezing patterns water in the stems of hardy and tender C^_ stolonifera. Most of the freezing -20°C. of occurs at temperatures warmer than This amounts to freezing of 92% of stem water in tender stems versus 54% for hardy stems. They demonstrated the importance of freezing rate above the temperature at which about 70% of the freezable water has been frozen. Slow freezing rates above this threshold enhanced survival by allowing migration of water from cells to extracellular ice. G i v e n e n o u g h e x p o s u r e to — 20°C the s t e m s could withstand -196°C when placed in liquid nitrogen. Supercooling is used by plants to avoid cold temperature injury. It allows water to remain unfrozen at 15 temperatures freezes at well below freezing. temperatures near Supercooled -40°C, the water homogeneous nucleation point, (George et al. 1974). Supercooling has been found by Burke and Stushnoff (1979) to occur in the flower buds of several species of prunus, in flower buds of florida by Sakai (1979), and both flower and vegetative buds of Pyrus by Rajashekar et al (1982). Ashworth (1982) and Ashworth and Rowse (1982) allowed twigs from several species of Prunus to take up dye. In species in which supercooling occurred in flower buds, no dye was seen in the flower buds. In species which did not supercool, dye moved into the flower buds. supercooled, In species that undifferentiated xylem prevented ice crystals from coming into contact with supercooled water. George and Burke (1977) demonstrated the occurrence of s u p e r c o o l i n g in x y l e m ray p a r e n c h y m a of C a rya S^ata. Ashworth et al (1982) and Burke and Stushnoff (1979) found supercooling to occur in the xylem parenchyma of Prunus, and Rajashekar (1982) found supercooling in xylem parenchyma of Pyrus species. George et al (1974) tested 49 woody species growing in Minnesota and found supercooling. Hong et al (1980) reported that the death of ray parenchyma in apple xylem coincided with the beginning of the low temperature exotherm, when supercooled water begins to freeze. Cellular dehydration is another strategy used by plants to survive cold temperatures. As water moves from cells to 16 intercellular ice crystals the cells become more dehydrated. Species such as C^_ stolonifera t olerate e x t r e m e cell dehydration and thus withstand extremely cold temperatures (Harrison et al 1978). Ashworth et all (1983) found this mechanism for winter survival in the bark of peach and apricot. Controlling the tissues in which ice forms enhances flower bud survival in some Rhododendron species. Graham and Mullin (1976) examined the hardiness of azalea flower buds. They found hardier buds were able to lose water more quickly than less hardy buds and suggested that perhaps the bud scales served as ice nucleation sites. In a series of papers by Ishikawa and Sakai (1981), Kaku et al (1980) and Kaku et al (1981) such a water migration was found but was dependent on bud morphology. Water migration to bud scales was found in species that had the scales interleaved with the florets. The growing conditions occurring during the previous summer can influence winter survival. Dennis and Howell (1974) found that defoliation reduced the hardiness of tart cherry bark and buds. The later in the season the defoliation occurred the less impact it had on hardiness development. 17 III. ANNUAL GROWTH CYCLE - GENETIC Plants of the same species, but with different genetic make up, have different responses when grown on a common site. Wright (1976) listed differences between northern and southern trees of the same species. However, not all observed differences may be due to the genetic diversity of the species and the same experiment in another location may give different results (Wright 1973). Flowering Variation may not be as great in flowering as in some other traits. Smithberg and Weiser (1968), comparing races of C^ stolonifera, red osier dogwood, of diverse origin but growing in Minnesota, found little difference in time of flowering. Trees of Populus deltoides, eastern cottonwood, growing in a Nebraska provenance test were observed by Ying and Bagley first, (1976). Trees from the north and west but the authors noted exceptions. flowered The range of flowering times was from April 7 to 21. Most clones flowered between April 11 and April 14. Wright et al (1976) found fruiting to be heaviest on western and southern European varieties (Pinus sylvestris). of Scotch pine Within a seedlot, flowering was heavier on the taller than the shorter trees. Fruiting on white pine (Pinus strobus) was heaviest on slow growing northern types. 18 Stem Elongation Differences were more apparent when the growth of trees from different origins was compared. Juglans nigra grown in Indiana and Illinois were measured regularly for 10 years (Bey 1979). At first, northern sources grew more rapidly than southern ones. The growth rate of southern sources increased gradually over time and eventually exceeded that of the northern trees. Pinus banksiana, jack pine, seedlots of southern origin were taller than those from farther north when all were grown in Michigan (Canavera and Wright, 1973). Scotch pine from central Europe was found by Wright et al (1966, 19 76) to grow faster than other European sources when grown in Michigan. Douglas-fir, white fir, climates (Abies grew susceptible (Pseudotsuga menziesii; Wright 1971a) and to concolor; Wright 1971b) f aster in M i c h i g a n but winter injury than from were slower milder als o mo re growing, but hardier trees from colder climates. Eastern cottonwood demonstrated increasing growth rate with more southerly origin (Ying and Bagley, 1976). The growth superiority of southern sources declined when they could not withstand cold temperatures and suffered winter injury. 19 Growth Cessation Early growth cessation may be a factor in the slow growth of trees of northern origin. Northern ecotypes of Picea abies stopped growing sooner than those of southern origins when all were grown in one location (Heide, 1974). The early cessation of growth of northern trees was observed by Pauley and Perry (1954) in a provenance test of three Populus species. clones ranged from Growth cessation of P^_ balsamifera June 20 to September 19. Growth cessation dates ranged from August 15 to October 18 for P. deltoides and from June 20 to October 28 for P^ trichocarpa. The northern, or cold climate races of red osier dogwood observed by Smithberg and Weiser (1968) stopped growing sooner than did those from more southern or warmer climates. Fall Color Development Townsend (1979) summarized observations made of red maple (Acer r u b r u m ) from different parts of the natural range. The northernmost sources had the most brilliant red fall color and were also most reliably red in different plantations from year to year. Autumnal f oliage coloration varied between years of observation. Autumnal foliage coloration was seen in northern sources of jack pine by Canavera and Wright (1973) and in Sc otch pine (Wr igh t et al 1966, 1976). In jack pine seedlings the color change was from green to purple, but in 20 later years it was from green to yellow. The color change in Scotch pine was from green to yellow. Leaf Abscission Smithberg and Weiser (1968) observed leaf abscission to be earlier on races of red osier dogwood from northern, or cold climates as compared to southern or mild climate races. Populus from northern sources were observed by Ying and Bagley (1976) to abscise leaves earlier than southern sources. Acclimation To Freezing Temperatures Northern races of red osier dogwood observed by Bray and Parsons (1981) and McKenzie et al (1974) had reduced stem water content sooner than did those of southern or mild climate origin. Bray and Parsons (1981) found that northern sources of C. stolonifera had increased stomatal resistance earlier than plants from southern or mild climates. A second year of observation did not reveal such distinct differences. Hardiness developed first in northern, or cold climate, sources of red osier dogwood (Smithberg and Weiser 1968) and white pine, Pinus strobus (Maronek and Flint 1974). Winter Survival Winter survival has been observed in a number of range tests. The species or genera in which northern sources have been observed to have better winter survival than southern sources include: black walnut (Bey 1979); Austrian pine, 21 Pinus nigra (Wheeler et al 1976); Scotch pine (Wright et al, 1966, 1976); 1969); ponderosa pine, Pinus ponderosa (Wright et al Douglas-fir (Wright et al 1971a); white fir (Wright et al, 1971b) and Populus (Ying and Bagley 1976) Observations of increased cold hardiness of northern species based on controlled freezing tests include; oak, Quercus rubra Flint 1974); (Flint 1972); white pine (Maronek and red osier dogwood (Smithberg and Weiser 1968); and L i q u i d a m b a r s t y r a c i f l u a McMillan, (Sweetgum; Williams and 1970). Supercooling of water in xylem parenchyma, s urv i v a l red strategy, is limited temperatures are above -40°C. to areas as a winter where winter Burke and Stushnoff (1969) found that Prunus species with natural ranges extending into regions with temperatures below -40°C did not use supercooling of water in xylem parenchyma as a survival strategy. Spring Loss Of Cold Hardiness Dehardening also varies with geographic origin. The pattern of dehardening in three origins of white pine was presented by Maronek and Flint (1974). deharden at about the same time. All trees began to Northern trees developed greater hardiness but lost it at a faster rate. By the end of March, northern and southern origins were hardy to about -30°C and afterwards the patterns of dehardening were the same. 22 Bud Break Differences observed in the pattern of bud break are variable. Perry and Wu (1960) investigated bud break in red maple from various origins grown in Florida plantations. With one month of chilling, Florida sources broke bud on February 20 while Ohio and New York sources did not break bud until April 22. Branches from races of stolonifera grown in Minnesota and observed by Smithberg and Weiser (1968) broke bud over a 17 day period when brought indoors and held at 50°F day and 35°F night temperatures. Day temperatures of 70°F promoted such rapid bud break that differences between races could not be detected. Y i n g and reported that leaf flushing dates of Bagley (1976) deltoides clones varied from 0 to 11 days with an average of 4 days. Eriksson et al (1978) made crosses between trees from France and Sweden. abies When tested for date of bud flush the progeny mean was close to the mean for French x French and Swedish x Swedish crosses. However, when individual progeny were examined, trees could be found that were closer to one or the other intracountry cross. IV. ANNUAL GROWTH CYCLE - ENVIRONMENTAL The progression of the growth cycle is regulated by environmental factors such as daylength and temperature. The w arm temperatures and lengthening days of spring and early summer promote plant growth. In the late summer and 23 autumn, progressively shortening days and co o l e r temperatures reduce and finally stop growth. Flowering Reader (1975) developed a model to predict flowering time in flowering dogwood. The data used were the flowering times of the species in Wauseon, Ohio from 1883 to 1912. The model predicted that flowering occurs when 379 heat units are accumulated with a base temperature of 2.2°C. Volunteers, recruited to make observations, were asked to record the flowering times of flowering dogwood at various lo c a t i o n s throughout the natural r ang e of the tree. Observed flowering times were close to the predicted times, except in the northern part of the range where the predicted dates were two weeks earlier than those that actually occurred. Growth Cessation Shortening of the photoperiod has been found to be the environmental factor that causes growth cessation in Salix viminalis (Alvim et al 1978), Picea. abies (Heide, Acer saccharum Perry, 1954). (Olmsted 1951), and Populus 1974), (Pauley and In each case, growth stopped when plants were exposed to artificially shortened days or shortening natural days. Artificially lengthened days caused continued growth. The photoperiodic response may be temperatures. modified by low Low temperature caused growth cessation in Norway spruce under continuous light according to Heide 24 (1974). Temperatures within the range of 12°C to 24°C had little modifying effect on growth responses to photoperiod. However, 4°C under continuous light did eventually cause growth cessation. Fall Color Development Shortening photoperiod was found by Olmsted (1951) to trigger loss of green color in sugar maple (Acer saccharum). Plants maintained maintained green under leaves artificially longer long daylengths than plants exposed to natural day length. Leaf Abscission Several environmental factors have been listed by different researchers as influencing leaf abscission. the factors Among are: light, including photoperiod; temperature; rainfall; wind; and nutrition. Alvim et al (1978) grew Salix viminalis under natural daylength and also provided lighting to extend the photoperiod. The long-day plants continued to grow until cold w e a t h e r c aus ed g r o w t h cessation. Matzke (1936) observed that street trees next to street lights retained leaves Olmsted longer on the side closest to the (1951) observed variety of conditions. sugar maple street seedlings light. under a In each case longer than normal photoperiods delayed the onset of senescence and abscission. Light exposure also determines photosynthate that may be available. the amount of Addicott and Lyon (1973) 25 suggest, that when carbohydrates are plentiful, cell walls are thicker and abscission is more difficult. when low light levels lead to insufficient carbohydrate levels for enzyme synthesis, abscission may be impeded. Such effects may be localized, as shaded leaves are abscised more readily than leaves exposed to full sun. Thomas and Stoddart (1980) suggest that as light penetrates a plant canopy, the leaves absorb some wavelengths changing the light intensity and quality, thus influencing the abscission of shaded leaves. Temperature may have a disruptive influence on abscission. Both Addicott and Lyon (1973) and Thomas and Stoddart (1980) suggest that low temperatures which disrupt the m etabolism of the leaf but do not destroy cellular integrity can induce leaf abscission. However, if cell structures are destroyed during freezing and thawing, a leaf may not be able to produce the energy or enzymes needed to complete abscission. Destruction of the abscission zone by freezing also has an adverse influence on leaf abscission. Addicott and Lyon (1973) described the role of moisture during normal autumnal moisture from leaf absci ssi on. senescent tissue causes The shrinkage loss of which, combined with adequate moisture in tissues on the twig side of the a b s c i s s i o n separation. zone, ca u s e t e n s i o n s w h i c h lead to Where moisture levels are low, separation of senescesed leaves may be delayed until adequate moisture is available. 26 Acclimation To Freezing Temperatures Photoperiod has been acclimation in plants. Long days delay acclimation in C. s t o l o n i f e r a as has b e e n Fuchigami et aJL (1971), found to have an influence on found by Che n and Li (1978), and van Huystee et al (1967). A delay in acclimation caused by long days has been seen in apple by Howell and Weiser (1970), maple and viburnum by Irving and Lanphear (1967), and Douglas-fir by Timmis and Worral (1975) and van den Driessche (1970). However, the long day delay of acclimation modified by low temperature. can be Irving and Lanphear (1967) were able to induce acclimation by a combination of long days and low temperature. Howell and Weiser (1970) trained apple trees to two leaders and exposed trees to either long or short days or a combination, in which one leader was given long days and the other short days. Trees totally under long days acclimated last and those totally under short days first. plants were intermediate. Therefore, The split exposing half the plant to long days influenced the other half exposed to short days. The authors cited above as having reported a long day influence have also investigated the effect of short days. In each case acclimation. short days have been found to induce 27 Combinations acclimation. short days, of e n v i r o n m e n t a l factors influence Chen and Li (1978) examined the effects of low temperature and water stress and found the effects of each to be additive. Fuchigami et al (1971) exposed red osier dogwood (C. stolonifera) to short days and low temperatures. Some plants were completely or partially defoliated by hand, others were allowed to retain all their leaves. partially defoliated plants, foliated branch was girdled. either the on p a r t i a l l y d e f o l i a t e d plants. with defoliated or Completely defoliated plants did not acclimate nor did the defoliated, branches On some girdled branches P lan ts w i t h leaves on girdled plants, leaves, and defoliated branches on ungirdled plants acclimated normally. Temperature effects have been investigated by the same workers listed as having examined the effects of light. In each case, low temperatures promoted, and high temperatures inhibited, acclimation. Howell and Weiser (1970) trained young apple trees to two leaders as in their previous experiment. Trees were either exposed totally to naturally occuring temperatures or exposed to natural temperatures but not below 4.5°C. tr ees had one le ade r exposed to naturally Some occuring temperatures and one leader exposed to natural temperatures but not below 4.5°C. photoperiods. All plants were exposed to natural Plants and branches exposed to natural day 28" length and temperatures acclimated to — 55°C. branches never exposed acclimated to -30°C. to temperatures Plants and below 4.5°C None of the plants receiving split treatments were intermediate. Light exposure has an effect on acclimation. et al (1971) stolonifera. covered Plants the leaves of either had all Fuchigami acclimating leaves covered, leaves covered, or leaves on half the plant covered. c. no Plants completely covered at the beginning of the experiment did not acclimate. At the end of the experiment, the hardiest tissue was on the branches on split treatment plants that had leaves covered at either week 1 or week 4. The second hardiest tissue was that from plants completely uncovered for the duration of the experiment. on the split treatment plants were The uncovered branches less hardy than the uncovered plants and much less hardy than the portion of the plant that had been covered. Howell and Shaulis (1980) tested the hardiness of grape canes grown under field conditions. exposed to the Canes which were well sky in the previous growing season were hardier than less exposed canes. Hardiness evaluations of Douglas-firs under various light intensities and photoperiods were made by van den Driessche (1970). Hardiness of plants given either long or short photoperiod increased with exposure to increasing light intensity. He a lso i n v e s t i g a t e d the e ffe ct of 29 moisture supply on acclimation of Douglas-fir seedlings (van den D r i e s s c h e 1969). Moisture level had no eff ect on acclimation except that plants required adequate moisture in order to acclimate properly. Water stress can increase hardiness levels. Chen and Li (1978) found C. stolonifera increased in hardiness when water-stressed. Winter Survival Light exposure during influences winter survival. the previous growing season Flore et aJL (1983) provided sour cherry and peach with artificial shading at different percentages of full sun. Hardiness of wood, flower buds, and vegetative buds was reduced by growing plants under 21% or 9% of full sun. Spring Loss Of Cold Hardiness Howell and Weiser (1970) examined dehardening in apple and found dehardening and rehardening to be closely related to air temperature. Dehardening of black walnut seedlings was studied by Murray and Byrnes (1975). were brought conditio ns, conditions. inside, than did dehardened seedlings Seedlings that fa s t e r u n d e r w a r m e r with cooler growing 30 Bud Break Longer exposure to cold temperature hastens bud break in Wu, sugar maple (Olmsted, 1960). Plants given 1951) and in red maple (Perry and insufficient exposure to cold temperatures while dormant did not break bud normally when exposed to warm temperatures. METHODS AND MATERIALS Plantation Establishment T he flowering dogwood t ree s in the study w e r e plantations at W. K. Kellogg Forest in Augusta, Ml. in They were established from seeds collected for J. J. Kielbaso and J. W. Wright by cooperators in other states. Figure 1 shows the location of the seed collection sites, the W. K. Kellogg Forest and East Lansing, Michigan. The uncleaned seeds were planted in nursery beds during the fall of 1972 and grown for two years at the Michigan State University Michigan. The Tree Research Center -at East seedlings were planted at W. Forest in May of 1975. Lansing, K. Kellogg The experiment consisted of ten replications divided among four blocks. Figure 2 shows the location of the~four blocks at W. K. Kellogg Forest. The rows in all blocks ran north and south. Bl o c k one, consisting of n ear ly established on an open, level site. 300, trees was Paraquat was used to kill weeds prior to planting, and then trees were planted by machine. Block two was established on a hill top at the south end and sloped down into an area of low elevation at the north end. The elevation changed abruptly in the middle of the block. in the planted. Paraquat was used in this block to kill weeds planting strips before the trees were machine Trees growing in the low part died back to the ground in most winters. Block three was located entirely in 31 32 •• Figure 1. Collection Sites for Cornus flo rid a Seeds. Collection sites axe denoted with •, W. K. Kellogg Forest with a □, and East Lansing, Michigan with a 33 W. K . KELLOGG AUGUSTA FOREST M ICHIGAN KALAMAZOO COUNTY ROSS TWP. — T I S . R S W ' 27 B ST O t J LEGEND com partm ent aUtLOINGS SCALEl — Figure 2. < l in e s ROAOS T R A IL S •**ia r « T Location of the Four Cornus flo r id a Blocks a t W. K. Kellogg Forest 34 a low area. No herbicide was used prior bo planting the trees with a machine. ground each winter. Most of these trees died back to the The few trees on relatively high spots in the block were the exception. Block four, c o n s i s t i n g of n e a r l y 300 trees, was established in the shade under a stand of Scotch pine. The pines were planted in 4-tree plots that had been reduced to three by removing the weakest tree. The dogwoods were planted by hand in the vacated spots. The pines had kept herbaceous vegetation to a minimum so that no herbicide was used prior to planting. Unfortunately, the pines began to die out as a result of spittlebug and Diplodia sp. and, by the end of established. 1981, considerable undergrowth had become Brush encroaching on the dogwoods was cut to reduce competition and to facilitate observations. Paths were mowed down the center of each 4-tree Scotch pine plot to allow easier access to the planting. It was expected that the frost pockets in blocks two and three would provide conditions that would test the cold hardiness of the trees. The lowest parts of the frost pockets were shaped like basins so cold air was trapped. Fall Color Measurement During the autumns of 1979-1981 the trees were rated for the amount of foliage showing fall color, the intensity of color displayed, and the number of leaves dropped. tree was rated 0 through 9 for each characteristic. Each The 35 ratings began in early September and continued at 6 to 8-day intervals until early November, when all but the trees from southernmost sources h ad dropped standardize tree appearance, their leaves. To the southeast side of each unshaded tree was observed in the morning. Cold Hardiness Assesroents Two sets of trees were selected for cold temperature hardiness determinations. The first set consisted of trees from high ground in the provenance test and contained trees from 15 of the 19 states represented in the planting. Texas, Louisiana and Mississippi trees died to the ground each winter, thus demonstrating their lack of adaptation to Michigan growing conditions. Minimum-maximum thermometers were placed in each block. Temperatures there were higher than those forest h e a d q u a r t e r s building. recorded at the S e p t e m b e r and O c t o b e r temperatures were 2.5°C higher than at the headquarters building but winter temperatures varied by only 1°C. The thermometers were read each time readings were made or samples collected. Samples were taken from the south side of selected trees growing in the large open block and from trees growing under the Scotch pine. All samples were taken at the same height and from the same tree sector. The second set of hardiness assesment trees was made up of ten native flowering dogwoods, all growing in the 36 understory nearby on the forest. Each tree crown was sampled on the north and south side, at high, m edium and low points. High, medium and low correspond to the upper, middle and lower third of the crown. Twigs from the central portion of these sectors were taken as long as available. Then twigs closest to the center were used. Some of the tree and sector combinations were not represented in the later sets of low temperature suitable twigs in th ose stress tests because all sec tor s had been previously collected. The terminal twig clusters were pruned and placed in styrofoam coolers (fall) or plastic bags (winter) then placed in the trunk of an automobile for transportation to the Michigan State University Horticulture Research Center. After the two hour trip the samples were held in a cooler at 1.7°C. When the native trees were sampled, the twigs were bearing varying numbers of flower buds. No attempt was made to collect twigs that had flower buds, but when they were available, the buds were frozen along with the twigs. The flower buds were selected for uniformity and apportioned equally among the treatments at each low temperature stress test. The stems were cut to a uniform length and then the samples were prepared as will be described later for twigs. The length of the stems left attached to the flower buds 37 varied between stress tests b ut was uniform for an individual test. The samples were prepared for freezing the same day they were collected. Terminals were taken from the cooler one seedlot and block at a time. The bottom 3 inches of the most recent year's growth was cut off for cold temperature stress testing. Twigs from the same seed source were taped together with masking tape and labeled with accession number and block identifier. Twigs from the ten native trees were labeled with tree number, side of tree, and tree sector from which they were collected. The bases of test twigs were placed in contact with damp gauze to eliminate supercooling (Mckenzie packets, and Weiser in aluminum foil and then returned to the walk-in cooler. The controls were cut, 1975), wrapped labeled then placed directly into a high humidity jar. Two packets per seed source were prepared for each test temperature. One packet contained twigs from the trees on the open hilltop, trees. the other packet held twigs from shaded Twigs from all the native trees, for each test temperature, were placed in one packet. Thermocouples were taped to three inch long segments that had been saved during sample preparation. stem The twigs were selected to provide a uniform size and did not represent any particular seed source. These twigs were then 38 wrapped for freezing as has been described for the other twigs. Two strips of masking tape were stretched lengthwise within, and fastened to the ends of, styrofoam boxes. packets were suspended from the strips of tape, The and two packets containing thermocouples were placed in each box. The covers were placed on the box and taped down. Once prepared, the boxes were held overnight at 1.7°C until the twigs were frozen the next day. A walk in freezer was used when stress temperatures were above -30°C. freezer was used. Below -30°C, All freezing setting the freezers to 1.7°C, freezer, and plugging the connected to a potentiometer. a Revco stress Ultralow tests chest began by placing the boxes in the thermocouples into wires The temperature was dropped 3°C per hour. Twig temperature readings were taken every 15 minutes. After exotherms occurred the freezer temperature was held at -10°C until twig temperatures were one or two degrees apart, then the temperature was again dropped 3 degrees C per hour. reached When both thermocoupled twigs in a box a predetermined test temperature, the box was removed and placed in the 1.7°C cooler to thaw gradually. Twigs were unwrapped and placed in the humidity chamber with control twigs on the next day. All test temperatures were separated by intervals of 3°C. The temperatures were selected so that all twigs would 39 be expected to survive the warmest temperature, would be killed by the coldest. but all The first test included the temperature of -3°c, but since exotherms did not occur until -4°C, -3°C was not used again. Three weeks after the controlled freezing stress, the twigs were microscope. sliced open and examined under a dissecting Browned twig tissues were rated injured, no attempt made to rate the severity. with At the same time, flower buds were dissected and rated for injury to bracts, florets or pedicel. The bud was rated as injured if any of the floral parts was browned. The temperature at which, twigs die is called the T50. theoretically, 50% of the These T 5 0 ’s were determined according to the method of Proebsting et al (1966). The Spearman-Karber formula described by Bittenbender and Howell (1974) was used determinations. all twigs, to bring greater precision to the T50 When the warmest test temperature killed s u r v i v a l of all t w i g s at the next h i g h e s t temperature was assumed in calculation of the T50's. all twigs survived the lowest temperature, When complete kill by the next coldest temperature, was assumed. Separation of the T50's was done by pairwise comparisons, using Chi square at p=.05 (Meddis, 1975). 40 Growth Measurements Tree height and terminal growth were measured in late autumn in 1979-1981. The tallest terminal was selected for terminal growth measurements. RESULTS Nuraery Data Nursery data were collected and provided by Kielbaso who made observations during 1973 and 1974. 28 months, percent. overall germination J. J. After ranged from one to five This resulted in a limited number of trees to be planted out. The ranges of percent seedlot germination and second year germination as a percentage of the first year are listed in Table 1. There were two additional Missouri trees the second year. The zero is due to rounding down to the nearest 10 percent. The numbers of trees from Illinois, Louisiana, Texas, and Maryland declined the second year, presumably from winter kill or other nursery loss, with little or no second year germination. In some replications, no seed germinated in either year, nor was germination evenly spread among replications. The poor germination may be due to the planting of uncleaned seed. The U. S. D. A. Agriculture Handbook 450, Seeds of Woody Plants of the United States does suggest prompt planting of cleaned or uncleaned seed in in the fall. 41 42 Table 1. The ranges of seedlot germination percentages of Cornu 3 florida seeds 28 months after being planted in autumn of 1972 in a nursery bed in East Lansing, Michigan.a State of Origin Northern and Northern Appalachian States MI OH PA VA WV Central and Coastal Plain States CT IL IN KY MD MO NJ Southern States AL GA LA MS TE TX Seedlot % Germination Second-year Germination as a % of the first 0 2 1 0 0 10 2 5 10 3 50 60 140 30 190 0 2 0 0 1 0 2 20 13 8 6 13 4 4 60 -10 20 50 *30 0 90 0 0 0 0 1 0 10 9 10 5 4 5 30 20 -10 60 10 -30 a ) Nursery data were collected and provided by J . J • Kielbaso. 43 Fall Color Leaves -that turned completely red followed a common pattern. Small, rust colored, interveinal specks formed near the leaf tips, and then spread to the margins of lower parts of the leaf. When leaf tips were solidly colored, lower leaf parts were speckled nearly to the mid-rib, but remained green near the veins. Leaves then became entirely red but with a distinctly green cast. A clear shade of red developed when the greenish cast disappeared. The shade of red that developed varied by tree and growing conditions but was seen only on leaves or leaf parts exposed to full sun. Leaves at the bases of the lowest branches, but exposed to sunlight, colored first. Color development spread toward the lowest branch tips and up to basal leaves of the next higher branches. color development The branches were in different stages of until the leaves on all branches had colored. Leaf abscission ended the fall color display. Leaf abscission moved up the trees following the same pattern as color development. between trees* However, leaf abscission timing varied The leaves of some trees abscised in 3 to 4 weeks while others went from 0 dropped to 90% in one week. The shade under the Scotch pine virtually eliminated bright red fall color development. late and were various mixtures Shaded trees colored of red and yellow. The understory dogwoods turned red only when some of the Scotch 44 pine died, creating openings. Trees that were winter killed to the ground had delayed fall color development. Shaded leaves turned yellow and then abscised in the second or third week of October, on leaves exposed to sun. dense, multi-stemmed foliage. the time of peak red color This event was most pronounced on trees, with considerable Most foliage on open trees was exposed to sun. Abscission of shaded leaves was rapid. that was shaded obvious at one disappeared by the next, effect was Yellowed foliage r a t i n g time, had completely a period of 6 to 8 days. so pronounced that color ratings The were often lowered. The effect of site on fall color is demontrated Figures 3 through 8. in These graphs show the mean ratings for percent of the leaves colored (% color), the leaf color developed and the percent of leaves abscised (% drop) for Michigan seedlots. The rat i n g s were made at w e e k l y intervals from September 6, 1980 to November 1, 1980. color was Leaf just starting to develop on some trees at the first rating, and nearly all trees had abscised their leaves by the last rating. The percent color and percent leaf drop ratings correspond to tens of percents. (90%), were (%dropped). ornamental considered The color were 7, Trees rated as 9 fully colored (% color) rat i n g s 8 and 9. bright red, dark red and maroon. or bare c o n s i d e r e d to be mo s t These were, respectively, 45 Figure 3 shows the mean ratings for the Michigan seedlots growing on the open hilltop. These trees colored quickly and color rating. At the fifth week the leaf color intensified to bright red. were showing about 35% at the first Leaf abscission began at week 4 but accelerated rapidly between week 6 and 8. This pattern was typical of trees from the northern group (as listed in Table 1) plus those from Connecticut and New Jersey. The mean ratings for the Michigan seedlots growing in the frost pocket are shown in Figure 4. In 1980 the trees in the frost pocket had delayed color development and in the color intensification to hilltop trees. red, as compared to the open Leaf abscission was not delayed in the frost pocket. The beginning of fall color development was even more delayed on the Michigan seedlots growing in the shaded block (Figure 5). The trees had not begun to color at the first rating but reached complete color only slightly later than trees in the other two blocks. The most pronounced effect of shade on fall color display was the elimination of red fall color. As can be seen from Figure 5, no red fall color (rating 7 or higher) developed on dogwood growing under the pines. The percent of leaves colored for the Michigan seedlots growing on the open hilltop, in the frost pocket and under the pines are shown in Figure 6. The trees in the frost 146 Percent color - Color - Percent leaf drop color Weekly ratings starting 9/6/80 Figure 3. Percent color, leaf color and percent abscission of leaves of Michigan seedlots of Cornus florida growing on an open hilltop in a provenance test in Kalamazoo County, Michigan. 1*7. drop c o lo r •* Percent color - Colof - Percent leaf c o lo r Mi. Weekly ra tin g s s ta r tin g 9 /6 /8 0 Figure U. Percent color, leaf color and percent abscission of leaves of Michigan seedlots of Cornus florida growing in a frost pocket in a provenance test in Kalamazoo County, Michigan. 1*0 leaf drop }••*.% color f^ ^ Percent color/Color/Percent -