LIBRARY Michigan State University This is to certify that the thesis entitled THE EFFECTS OF GENOTYPE AND VARIOUS CULTURAL FACTORS ON VEGETATIVE PROPAGATION OF BLACK LOCUST (Robinia pseudoacacia L.) presentecTEy Roy Allen Prentice has been accepted towards fulfillment of the requirements for M .S . degree in Forestry \lw ism / Major professor Date Nw, (311% 97 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES .—:—. RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. THE EFFECTS OF CENOTYPE AND VARIOUS CULTURAL FACTORS ON VEGETATIVE PROPAGATION OF BLACK LOCUST (Robinia pseudoacacia L.) By Roy Allen Prentice A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Forestry 1987 ABSTRACT THE EFFECTS OF GENOTYPE AND VARIOUS CULTURAL FACTORS ON VEGETATIVE PROPAGATION OF BLACK LOCUST (Robinia pseudoacacia L.) By Roy Allen Prentice Black locust (Robinia pseudoacacia L.) has many uses, including high quality dense lumber, honeybee forage, decay resistant fence posts, and animal feed. Since considerable genetic gains can be made by selecting trees with desirable characteristics, the need for effective vegetative propagation methods is acute. Several prOpagation alternatives were evaluated including the rooting of forced root suckers, rooting of root and stem cuttings, and grafting. Of particular interest was the effect of clone on propagation success. Also studied was the effect of benzylaminopurine on sprout production. Significant differences due to clone were found for sprout and root production characteristics in almost all studies. When root material from tall and short parents were compared, only roots per rooted segment and roots per sprout varied significantly. Hardwood stem cuttings and grafting techniques were effective for the propagation of mature plants. However, both techniques produced growth with lateral habit. ACKNOWLEDGMENTS I would like to express my sincere appreciation to Dr. James Hanover, committee chairman, for his guidance and support during the course of my research. I would also like to thank the other members of my committee, Dr. Donald Dickmann and Dr. Frank Dennis, for their assistance during the course of my program. A special note of thanks to my coworkers, Paul Bloese, Kevin Kell, and Ray Miller, whose assistance on this project was greatly appreciated. My appreciation also extends to many of my fellow students who helped me in innumerable ways. Thanks are also due to John Vigneron of Kellogg Forest, Michigan State University Department of Forestry, Fred Richey of the Michigan State University Department of Horticulture and John Royer of the Michigan Department of Natural Resources. Finally, I would like to give a big hug of thanks to Cindy and Minnette. Without their encouragement and unending support this project would not have been possible. iii TABLE OF CONTENTS LIST OF TABLES O O O O O O O O O O O 0 INTRODUCTION . . . . . . . . . . . . Chapter I. VARIABILITY IN ROOT AND SPROUT PRODUCTION OF TWENTY-ONE BLACK LOCUST CLONES Introduction . . . . . . . . . Materials and Methods . . . . . Results and Discussion . . . . Conclusions . . . . . . . . . . II. EFFECTS OF TREATMENT OF BLACK LOCUST ROOT SECTIONS WITH BENZYLAMINOPURINE Introduction . . . . . . . . . Materials and Methods . . . . . Results and Discussion . . . . Conclusions . . . . . . . . . . III. PROPAGATION OF MATURE BLACK LOCUST TREES FROM ROOT AND DORMANT STEM CUTTINGS Rooting of hardwood cuttings . Materials and Methods . . . . Results and Discussion . . . Conclusions . . . . . . . . . Outdoor root propagation studies Materials and Methods . . . . Results and Discussion . . . Conclusions . . . . . . . . . Crafting . . . . . . . . . . . Materials and Methods . . . . Results and Discussion . . Conclusions . . . . . . . . . IV. RECOMMENDATIONS . . . . . . . . . LIST OF REFERENCES . . . . . . . . . . . iv Page 10 17 31 34 34 36 38 41 42 43 46 49 51 53 55 57 60 60 61 63 68 72 LIST OF TABLES CHAPTER I Page Table 1. Characteristics of selected black locust clones . . . . . . . . . . . . . . . . . . . 12 2. Analysis of variance tables for root characteristics of 21 black locust Clones O O O O O O O O O O O O O O I O O O O 19 3. Root characteristics among root segments of 21 black locust clones . . . . . . . . . 20 4. Comparisons of root characteristics among clones from tall and short families . . . . . 22 5. Analysis of variance tables for three different dates of collection among 21 black locust clones . . . . . . . . . . . . . 24 6. Mean number of sprouts produced per root segment on three separate collection dates . . . . . . . . . . . . . . 25 7. Comparisons of sprout production among clones from tall and short families . . . . . 26 8. Sprout distribution on roots of 21 black locust clones . . . . . . . . . . . . 28 9. F-tests of percent rooting and mean number of roots per sprout produced by 21 black locust clones . . . . . . . . . . . 29 10. Mean percent rooting and mean number of roots on sprouts removed from 21 black locust clones . . . . . . . . . . . . . . . 30 ll. Comparisons of rooting of sprouts among clones from tall and short families . . . . 32 CHAPTER II Page Table 1. F-test of the effect of benzylaminopurine on the production of sprouts from black locust root segments . . . . . . . . . . . . 39 2. Sprouts occurring in the proximal and dis- tal halves of black locust roots treated with four levels of benzylaminOpurine . . . 39 3. Length of sprouts produced by root segments treated with four levels of BAP . . . . . . 40 CHAPTER III Table 1. Effects of clone, hormone treatment, and time of collection on rooting percent of hardwood stem cuttings of black locust . . . 50 2. Form and vigor of black locust produced by hardwood cuttings versus similar data for seedlings O I O O O O O O O O O O O O O O O 52 3. Analysis of variance tables for the effects of clone, root age, and root orientation on the rooting of black locust . . . . . . 58 4. Effects of age, orientation, and clone on rooting of root sections placed on outdoor nursery beds . . . . . . . . . . . . . . . . 59 5. Effects of scion clone on scion growth and morphology . . . . . . . . . . . . . . . 65 6. Effect of grafting technique on growth and branch angle of the scion . . . . . . . . . 66 INTRODUCTION In his book Michigan Trees, Otis (1931) lists many of the attributes and liabilities of black locust (Robinia pseudoacacia L.). He describes locust as being a moderately tall tree, reaching a height of thirty to sixty feet. Flowers appear after the leaves in May or June and are showy, abundant, and very fragrant. The wood is heavy, strong, close grained and very durable in contact with the soil. Locust will grow in almost any kind of soil, although it prefers the moist but well drained soils found in bottom- lands and along streams. Black locust exhibits very rapid juvenile growth, but is relatively short lived. Fowells (1965) recommends that all commercial cutting of black locust be completed before the tree is 40 years old. It is subject to attacks of wood boringinsects which tunnel in the sapwood and may weaken or kill the infested tree. The description of the species by Otis highlights many of black locust's positive aspects, such as its abundant flower production, the durability of its wood, its ability to grow on poor sites, and its rapid juvenile growth. Not mentioned is black locust's ability to fix atmospheric nitrogen, probably a primary reason for its adaptability to poor sites. Otis also points out one of black locust's present day liabilities, susceptibility to wood boring insect attack. 1 2 The locust borer (Megacyllene robiniae Forester) is a major obstacle to the full utilization of black locust. At present, there are no good cultural practices which control this insect. However, there is evidence that some degree of resistance to the locust borer can be attained through genetic selection. Variability associated with the family of origin of the attacked tree was observed in the incidence of injury caused by a twig boring insect (Mebrahtu and Hanover 1988). Mebrahtu and Hanover (1988) were not able to assess differential susceptibility to locust borer attack, as the plantations which were studied were only three years old. Locust borer does not ordinarily attack black locust until they are over four years old (Johnson and Lyon 1976). Future genetic evaluations of the species may identify families or individuals which exhibit resistance to locust borer attack. Also, variability among genotypes for other attributes of black locust may be detected. Traits which have already been identified as varying significantly among families include thorn length, form, and growth rate (Mebrahtu and Hanover 1988). To incorporate families with valuable traits into intensive cultivation systems, effective mass prOpagation techniques must be developed. Seed propagation of black locust is quite easy and has much promise as an economical pr0pagation method. However, the transmission of desirable traits from generation to generation may be low when sexual prOpagation methods are used. To ensure the maximum transmission of desirable genes, asexual prOpagation procedures must be used. One of 3 the goals of the research presented in this thesis was to evaluate various vegetative propagation methods for their potential use in black locust culture. Due to the attributes mentioned previously by Otis (1931), and additional qualities deemed to be of importance more recently, black locust is the subject of a great deal of international interest and research effort. Organizations studying black locust include the Michigan State University Department of Forestry (Hanover 1985, 1986), the University of Georgia Department of Forestry (Bongarten 1986; Kennedy 1983), the U.S. Department of Energy (Hanover 1985, 1986), and the national forestry agencies of Hungary and South Korea (Keresztesi 1980; Kim 1975). Interest in black locust stems from its value as a candidate for short rotation fiber production (Miller et al. 1987), its use as a nectar source for honeybees (Keresztesi 1977), the decay resistance of its wood (U.S.D.A. 1969), and its potential as a feed-stock for chemical production. Recently, locust foliage has been considered for use as an animal feed (Cheeke et al. 1983; Cheeke et al. 1985). Since develOping one variety which combines traits suitable for all of these widely varying uses will probably. be difficult or impossible, it may be necessary to develop separate varieties for each intended use. At present, it appears that the most practical way to develOp locust varieties is through vegetative prOpagation procedures, since propagation techniques which rely on controlled pollinations are very difficult (Bongarten 1986; Kim 1975), 4 and open pollinated seedlings will probably have low heritabilities for desirable combinations of traits. The development of varieties for specific uses is already practiced in Hungary and elsewhere (Keresztesi 1980; Steiner 1966). To obtain reasonable quantities of desired varieties, either older prOpagation methods must be evaluated, or new methods must be developed. The ideal propagation procedure should provide good quantities of propagules with relative ease. Also, it may be necessary to use different techniques for different types of plant material. PrOpagation procedures which work for juvenile plant material are often unsuccessful for older, more mature plant material. Since it may be advantageous to clone older trees with desirable phenotypes, techniques should be deve10ped which allow the prOpagation of trees of any age. There has been previous work on black locust vegetative propagation. Reresztesi (1983), and Swingle (1937), were able to successfully prOpagate black locust via root segments, while Stoutemyer et al. (1940) had success using dormant stem segments. Davis and Keathley (1988) reported success with in 11552 culture systems. However, much of the previous research fails to answer important questions. While Stoutemyer et al. (1940) had good success working With dormant stem cuttings, they did not describe the subsequent growth of the rooted cuttings. Problems with topophytic growth habit (the tendency of a plant part to grow according to its age or the position which it occupied 5 when part of a tree) of rooted stem cuttings, especially from mature material, has been mentioned by many authors (Wright 1976; Schaffalitzky De Muckadell 1959). Also, some authors have observed rooting variability due to clone (Donnelly 1971; Gabriel et a1. 1961), an aspect of hardwood cutting propagation not considered at all by Stoutemyer et al. (1940). Variable rooting success due to clone has been reported for in gitgg culture systems (Davis and Keathley 1988). Clonal response to chemical treatment was great with some clones easily subcultured while others could not be subcultured at all. Once in culture, the different clones produced differing numbers of shoots. Prepagation using root sections is reliable and fairly easy. Although this method has been used for a very long time, it has several disadvantages when compared with systems based on stem cuttings. Root prOpagation systems require relatively large amounts of plant tissue to produce usable plants. Also, roots can be difficult to obtain in quantity, and are quite easily damaged during collection. Collection of roots is best done when the plants are dormant (Hudson 1955; Eliasson 1971b; Schier 1972a; Sterrett et a1. 1968). However, this necessitates working during times of the year when poor weather interferes with root collection. These disadvantages make propagation by roots somewhat undesirable for commercial-scale production systems. Nevertheless, because of its long history and relative ease, 6 black locust propagation using roots must be the standard by which other propagation methods are judged. The research described in this thesis will deal with areas neglected by previous workers, especially the role of genotype in the rooting and sprouting of black locust. Additionally, techniques designed to increase production of plant material, and investigations into other aspects of locust propagation will be examined. Finally, practices will be recommended which will enable the black locust grower to produce vegetatively prOpagated material from a variety of plant parts. The following objectives outline the approach to the problem of vegetative prOpagation of black locust material taken in this research. Objectives of this thesis include: a) Examine the effect of genotype on shoot and root production of black locust root segments from twenty-one clones (Chapter one). b) Examine the effect of genotype on rooting success of shoots excised from root segments of twenty-one black locust clones (Chapter one). c) Examine the effect of treatment with the cytokinin benzylaminopurine on shoot production of black locust root segments (Chapter two). 7 d) Examine the potential for the use of roots, dormant stem cuttings, and grafting for the prOpagation of mature black locust trees. (Chapter three). CHAPTER 1 Variability in root and sprout production of twenty-one black locust clones INTRODUCTION Research on propagation of black locust using root cuttings has been previously conducted (Keresztesi 1983; Swingle 1937). Additional work, mostly pertaining to the effect of temperature and season of collection on shoot production from root sections, was conducted by Sterrett et al. (1968). None of the authors investigated the effect of genotype on sucker production, or attempted to excise and root shoots forced from root segments. Although research on shoot production from black locust roots is limited, there have been reports on shoot production in other species, including sweetgum (Brown and Kormanik 1967; Kormanik and Brown 1967), raspberry (Hudson 1955; Marston and Village 1972), apple (Robinson and Schwabe 1977), and Populus spp. (Brown 1935; Eliasson 1971a, 1971b; Farmer 1963; Maini and Horton 1966; Schier 1972a, 1972b, 1974, 1976, 1978a, 1978b, 1981; Tew 1970; Zuffa 1971). Only one of the researchers (Schier 1974, 1981), considered the 9 effect of genotype on sprout production and on the subsequent rooting of forced sprouts. Working with Populus tremuloides Michx., Schier (1981) found that there were significant differences in many sprouting and rooting traits due to clone. Specifically, Schier (1981) observed that the number of sprouts per cutting, the number of roots per sprout, and the mean length of the roots on the sprouts all varied depending on the clone tested. There were also general seasonal trends for all clones with root segments collected in the summer producing roughly half the number of sprouts as did root segments collected from dormant plants. However, sprout production response due to the time of collection was greatly affected by the clone observed. Studies will be described in this chapter which will investigate sprouting and rooting in black locust, using the previous work with Populus tremuloides performed by Schier (1981) as a model. Another aspect of sprout production research is the location on severed root segments where sprouting occurs. For most species, sprout emergence is not uniform along the entire length of the root segment, but occurs predominantly near the proximal end of the segment (Eliasson 1971b; Peterson 1975; Schier 1978a, 1981). This phenomenon is thought to be caused by the differential accumulation of indole acetic acid (IAA) in root segments. Sprout production is inhibited and root formation promoted by the concentration of IAA near the distal end of root segments. Bennett and Torrey (1965) concur with this explanation of 10 polar sprouting patterns and cite examples of this effect in various species, including Taraxacum, Cichorium, Chamaenerion, and Cochlearia. Among the experiments which they describe, Bonnett and Torrey (1965) note that exogenous application of IAA to the proximal end of root cuttings resulted in inhibition of bud formation, promotion of root formation, and suppression of the polar distribution of organ formation. In this chapter, observations on the distribution of sprouts on root segments of black locust will be described. Of special interest will be variability in sprout distribution along root segments due to clone. MATERIALS and METHODS Collection of Roots Root segments were collected from 21 different black locust trees. The trees were located in a young (three years from seed) progeny test composed of over 400 half-sib. families. To assess the effect of family vigor on sprouting and rooting characteristics, individuals used in this study were either from vigorous or non-vigorous families (with height as the indicator of vigor). One individual was selected from each of 10 separate families which ranked in the top 252 and one individual was selected from each of 11 separate families which ranked in the bottom 25% of the plantation on the basis of mean family height growth. Because of the difficulty of procuring roots of sufficient length and diameter from small trees, it was necessary to 11 collect roots from some relatively tall trees from short families. The greater height of these tall individuals relative to their families could be due to genotype or a site effect. Because it is very difficult to determine the relative effects of site and genotype on the growth characteristics of an individual tree, tall trees from short families were excluded from all high versus low vigor family comparisons. Table 1.1 displays some characteristics of the trees "+" indicates included in this study. A family ranking of that the family of the collected tree ranked in the top 252 of the plantation on the basis of height. A family ranking of "-" indicates that the family of the collected tree ranked in the bottom 25% of the plantation on the basis of height. In general, the root collection procedures which were used were similar to those described for aspen (Benson and Schwalbach 1970; Farmer 1963; Schier 1978b). Roots were excavated by hand during the first week of December 1986. Any roots which were damaged in the lifting process were discarded. Sufficient quantities of roots were collected so that a test could be constructed which contained four blocks, each with four cuttings per clone. Proximal and distal orientation of the cuttings in relation to the parent tree was noted to assess any sprout production gradients along the roots. The collected roots were placed in labeled plastic bags and were then transferred from the field to a cooler (4° C.). 12 Family mean Height of Family Accession height (cm) individual (cm) ranking 454* 155 195 + 455* 91 120 - 456* 146 228 + 457 142 198 + 458 108 198 - 459 149 162 + 460* 92 96 - 461 155 180 + 462 104 187 - 463 113 194 - 464 108 191 - 465* 110 155 - 466* 147 250 + 467* 154 200 + 468* 113 168 - 469 113 196 - 470* 165 252 + 471* 182 207 + 472* 100 160 - 473 168 171 + 474* 98 140 - * Indicates clones included in clones from tall families versus clones from short families comparisons. 13 Processing the Roots Procedures used to section and process the roots were as described for aspen by Schier (1978b). First, the roots were rinsed with water to remove excess soil The roots were then cut into 15 cm lengths. The orientation of the root cuttings with respect to the parent tree was maintained by making a small diagonal cut in the distal end of each cutting. The cuttings were soaked in a benomyl solution (1 g of active ingredient per 1 of water) for 30 minutes to reduce the incidence of fungal contamination. After the fungicide treatment, the ends of the cuttings were dried, dipped in melted grafting wax, and then dusted with talc to reduce the stickiness of the wax. The processed cuttings were placed in labeled plastic bags and stored at 4° C. Prepagation Bench Construction The prOpagation facilities used in this research were of two types. The first type consisted of benches with bottom heat but no mist capability. The second type consisted of benches with both bottom heat and misting capability. Both bench systems were constructed in a controlled environment greenhouse by placing wooden pallets on upright cement blocks. This created a bench surface which was approximately 60 cm above the ground. Each bench had a working area of about 3.85 m2 (1.4 by 2.75 m). Prior to initiating the study, all surfaces were washed with a 201 bleach solution (12 sodium hypochlorite) to reduce the risk of disease. 14 The sides and floor under the benches were insulated with one inch styrofoam board. Heat for the root zone was provided by two, 200 watt light bulbs attached to the underside of each bench. A small fan was placed under each bench to ensure adequate air circulation and uniform heat distribution in the root zone. Temperatures in the root zone were maintained between 18° and 21° C. Container Preparation Containers were constructed of polyethylene coated paper. Each container measured 15.25 cm long, 7.6 cm wide, and 7.6 cm deep. The size was selected so that each container was large enough to accommodate one experimental plot consisting of 4, 15 cm long root segments. To provide support, the containers were placed in plastic "milk case" type frames. Each frame measured 15.25 by 15.25 by 7.6 cm deep and was, therefore, able to hold two containers side by side. After the containers had been placed in the frames, they were filled with moistened number two grade (coarse) vermiculite. Vermiculite was chosen as it had been shown to be a good sprouting medium for aspen roots (Schier 1978b)- Establishment of the Cutting£_ On 6 January 1987, the root segments were placed in a randomized complete block design with four blocks and four root cuttings per block. The treatments consisted of the 21 separate clones. The root cuttings were blocked on the basis of diameter with the largest roots in block one. The 15 cuttings were placed horizontally on the surface of the vermiculite and gently pressed in. They were then covered with approximately 1 cm of moistened vermiculite. The roots were watered and treated with fungicides frequently to prevent desiccation and disease problems. Excising the Shoot Sprouts Root segments were lifted and shoots collected on three separate occasions (20 February, 23 March, and 26 May 1987). After being lifted from the vermiculite, the root cuttings were placed in a pan containing a benomyl (1 g active ingredient per 1 of water) solution to prevent desiccation and to control any fungi present. The collection of shoots was similar to the method described by Schier (1978b) and Zuffa (1971). For the purpose of data collection, each root section cutting was conceptually divided into quarters, the first quarter at the proximal end, the fourth quarter at the distal end. The number of shoots greater than 5 cm in length in each quarter was recorded. Also, the length of the longest shoot, the presence of any roots (on the shoots or the parent root), and the diameter of the root segments were recorded. Any shoot exceeding 5 cm in length was severed at its point of attachment with the parent root cutting, and placed in water. The segments from that plot were then returned to their containers on the prOpagation bench. 16 After all shoot cuttings for a particular plot had been severed from the parent roots, the basal 2 to 3 cm of each shoot was treated for 5 seconds with Indolebutyric acid (IBA) at 4000 ppm. This concentration of IBA was recommended by several authors (Bonaminio 1983; Hartmann and Kester 1968; Schier 1978b). The IBA solutions were formulated by mixing pure IBA crystals in a 50/50 solution of isopropyl alcohol and water (v/v). All sprouts in this study were treated with IBA. After treatment with IBA, the sprouts were placed in flats on a bench under intermittent mist. The rooting medium consisted of one part coarse vermiculite and one part perlite (v/v) as this mixture has been used with good success in the rooting of aspen sprouts (Farmer 1963; Schier 1978b). The sprouts were placed in the rooting medium in holes dibbled to a depth of approximately 3 cm with about 2.5 cm between sprouts. Overhead mist was Operated for 30 seconds every 5 minutes during the day and 30 seconds every 10 minutes during the night as recommended by Schier (1978b). After rooting had occurred (approximately 7 weeks after establishment) the sprouts were lifted and evaluated for root production. Data collected included the height of each sprout, the number of sprouts which rooted, the total number of roots produced per sprout, and the length of the longest root. All analyses were performed using the Number Cruncher Statistical System (NCSS) version 5.0 (develOped by Jerry Hintze) on plot averages of the character analyzed. Tests 17 of differences of means were performed using the Newman- Keul's range test (52 level unless otherwise noted). Data sets which exhibited a Poisson distribution (small value count data and percentages) were square root transformed prior to analysis. RESULTS and DISCUSSION Characteristics of Root Segments The primary purpose for collecting data on the number of roots and rooting percentage of the parent root segments was to assess the feasibility of producing plants with roots and shoots still connected to the parent root section. Tables 1.2 and 1.3 show the data collected from black locust root segments. There were highly significant differences (12 level) among the clones for percentage rooting and number of roots produced. The percentage of parent root sections which develOped roots (Table 1.3) varied from 19.22 for clone 458 (only 3 of 16 root segments deve10ped roots), to 1002 for clone 472 (all 16 root segments developed roots). Overall, the rooting percentage was 59.72, a higher rate than has been reported in previous outdoor prOpagation studies (262 observed in the study described in Chapter three of this thesis, and 24.62 reported by Swingle (1937)). The diameter of the root section had a fairly small effect on rooting percentage. Not only was there no significant block effect with diameter the primary blocking factor, but there were no significant effects detected when 18 an analysis of covariance was performed with root diameter as the covariant. This result contrasts with that of Swingle (1937) who found that root diameter had a large effect on rooting percent. However, Swingle (1937) worked with a very wide range of root diameters (0.15 to 2.54 cm), while this investigation used a more limited range of diameters (1.0 to 2.0 cm). Also, Swingle (1937) conducted his research in outdoor nursery beds. In the somewhat stressful outdoor environment, smaller, more marginal root segments may not have survived as well as they could have in the more protected environment of a greenhouse. In addition to rooting percentage, there were also very significant differences (12 level) among the clones for the mean number of roots produced per root segment (Table 1.2). Table 1.3 shows that the number of roots per root segment ranged from 9.1 (clone 461) to 0.6 (clone 458) with the overall mean of 3.9 (Table 1.3). Again, diameter was not a significant factor. Analysis of mean root number was performed on all root sections including those that did and did not produce roots. Root segments from clones which had low rooting percentages could appear to produce few roots when root producing root segments were averaged with a relatively large number of non-root producing root segments. To separate the characters of rooting percentage and number of roots produced per root segment, an additional analysis was performed considering only those root sections which 19 Table 1.2 Analysis of variance tables for root characteristics of 21 black locust clones. Y=Percent of root segments producing roots. Source d.f. M.S. F-ratio Significance Block 3 138.0 .21 ns Clone 20 2244.1 3.44 ** Error 60 651.4 Total 83 Y=Mean number of roots per root segment. Source d.f. M.S. F-ratio Significance Block 3 1655.5 3.07 * Clone 20 2580.9 4.78 ** Error 60 539.9 total 83 Y=Mean number of roots produced by segments producing roots. Source d.f. M.S. F-ratio Significance Block 3 3996.8 2.73 ns Clone 20 4189.2 2.86 ** Error 60 1465.6 total 83 * significantly different at the 52 level. ** significantly different at the 12 level. us not significantly different at the 52 level. 20 Table 1.3 Root characteristics among root segments of 21 black locust clones. Mean number of Percent Mean number of roots per Clone rooted roots per segment rooted segment 454 68.8 abc 1/ 6.5 abcde 9.4 ab 455 90.9 bc 8.3 de 9.1 ab 456 50.0 abc 4.7 abcde 9.4 ab 457 66.2 abc 3.0 abcd 4.5 a 458 19.2 a .6 a 3.1 a 459 62.2 abc 3.3 abcd 5.3 ab 460 70.6 abc 3.5 abcd 5.0 a 461 93.0 bc 9.1 e 9.8 ab 462 31.5 ab 1.5 ab 4.8 a 463 43.8 abc 1.0 a 2.3 a 464 68.5 abc 2.5 abcd 3.6 a 465 43.8 abc 2.0 abc 4.6 a 466 74.8 abc 4.0 abcde 5.3 ab 467 56.2 abc 3.4 abcd 6 0 ab 468 31.8 ab 1.9 abc 6.0 ab 469 87.0 bc 5.0 abcde 5.7 ab 470 62.2 abc 7.1 bcde 11.4 b 471 37.8 abc 3.3 abcd 8.7 ab 472 100.0 c 7.8 cde 7.8 ab 473 31.5 ab 2.1 abc 6.7 ab 474 31.5 ab .8 a 2.5 a Mean 58.0 3 9 6.2 1/ Means followed by the same letter are not significantly different at the 52 level. 21 produced roots. The results of this analysis are shown in the "Mean number of roots per rooted segment" column of Table 1.3. Since all four root segments of a clonal plot failed to produce roots in some cases, consideration of only those segments which did produce roots sometimes resulted in missing plots. Missing plots hindered the detection of differences among plot means when actual significant differences might indeed exist. Even with the confounding effects of missing plots, the data in Table 1.3 indicate that clones with higher rooting percentages also tended to produce greater numbers of roots, even when "Number of roots" data were adjusted for percent rooting. There were no significant differences among the two locust vigor classes with respect to either rooting percent or number of roots produced per root segment (Table 1.4). The only significant difference among the clones from tall and short families was the number of roots on root sections which produced roots. For this character, the clones from tall families were significantly (52 level) better than the clones from short families. The roots from short families had a slightly higher rooting percentage, but produced slightly fewer roots per root segment than roots from tall families. Sprout Production Sprout production differences were highly significant (12 level) among the clones on each collection date and for the Table 1.4 Comparisons of root characteristics among clones from tall and short families. Character Mean response of Mean response of measured tall clones short clones Significance Percent rooted 58.3 61.4 ns Mean roots per segment 4.8 4.0 ns Mean roots per rooted segment 8.4 5.8 * * significantly different at the 52 level. ns not significantly different at the 52 level. combination of all dates (Table 1.5). Since roots of trees from shorter individuals tended to be smaller in diameter, diameter was used as a covariate in the analysis. Use of diameter as a covariate was significant (52 level) for the number of sprouts present at the time of the first collection (20 February 1987) and for the total sprouts present for all collections. Based on this result, means for these two data sets were adjusted by the covariate diameter prior to performing the difference among means tests shown in Table 1.6 and the contrast between tall and short families shown in Table 1.7. In all cases, little sprouting occurred after the second sprout collection (Table 1.6). An interaction between clone and collection date occurred, with some clones producing the majority of their sprouts prior to the first 23 collection and others producing a substantial number of sprouts between the first and second collections. Because of this interaction, the total number of sprouts from all three collections is of most use in assessing sprout production differences attributable to clone. The "Total sprouts" column of Table 1.6 shows that mean sprout production per root segment varied from 7.1 (clone 463) to 2.2 (clone 462) with many significant differences among the clones within this range. Comparison of clones from tall families with clones from short families (Table 1.7) indicated no significant differences in sprout production between the two classes on any collection date or for the combination of all dates. Patterns of Sprout Production Sprout production patterns have been described for many species (Eliasson 1971b; Peterson 1975; Schier 1978a, 1981). Generally, few sprouts emerged near the distal end of the root segment, with increasing numbers of sprouts present towards the proximal end of the root segment. To test whether roots of black locust act in a manner similar to that described for other plants, numbers of sprouts were tallied along the length of the roots used in this study. Data on sprout number were collected from each quarter of each root segment. For the purpose of the analysis, the number of sprouts in the two most proximal quarters were Combined and the number of sprouts in the two most distal quarters were combined to produce sprout numbers present on 24 Table 1.5 Analysis of variance tables for three different dates of collection of root sprouts among 21 black locust clones. Collection 1: Source d.f. M.S. F-ratio Significance Diam. 1 365.1 4.33 * Block 3 53.4 .63 ns Clone 20 464.2 5.50 ** Error 59 84.3 Total 83 Collection 2: Source d.f. M.S. F-ratio Significance Diam. 1 27.7 .75 ns Block 3 31.4 .85 ns Clone 20 116.8 3.16 ** Error 59 37.0 Total 83 Collection 3: Source d.f. M.S. F-ratio Significance Diam. 1 2.34 .28 ns Block 3 .40 .05 ns Clone 20 43.66 5.18 ** Error 59 8.43 Total 83 All collections combined: Source d.f. M.S. F-ratio Significance Diam. 1 623.5 6.83 * Block 3 53.8 .59 ns Clone 20 505.4 5.53 ** Error 59 91.3 Total 83 * significantly different at the 52 level. ** significantly different at the 12 level. ns not significantly different at the 52 level. 25 Table 1.6 Mean number of sprouts produced per root segment on three separate collection dates. Number of Number of Number of Total Clone sprouts 2/20 sprouts 3/23 sprouts 5/26 sprouts 2'34 3.4 bcde 1/2/ 2.3 bc .8 b 5.9 cdef 455 3.2 bcde .8 ab .2 ab 4.2 bcdef 456 3.2 bcde 1.8 abc 0 a 5.0 bcdef 457 2.7 bcde 2.4 c .2 ab 5.3 ef 458 2.2 abc 1.6 abc .2 ab 4.0 abc 459 1.4 abc 1.2 abc .3 sh 2.9 abcd 460 3.5 de 1.3 abc .1 ab 4.9 def 461 3.6 cde 1.6 abc 0 a 5.2 bcdef 462 .4 a 1.4 abc .4 ab 2.2 a 463 3.3 bcde 2.4 be 1.4 c 7.1 f 464 2.2 abc 1.9 abc .4 ab 4.5 abcde 465 3.2 bcde .8 ab .4 ab 4.4 bcdef 466 4.8 e .5 a 0 a 5.3 bcdef 467 2.5 abcd 2.1 abc .4 ab 5.0 bcdef 468 1.2 ab 2.6 c .4 ab 4.2 bcde 469 1.8 abc 1.5 abc 0 a 3.3 abcd 470 2.6 bcde 1.1 abc 0 a 3.7 abcde 471 3.8 bcde 1.4 abc 0 a 5.2 bcdef 472 3.2 bcde 1.9 abc .7 ab 5.8 cdef 473 2.6 abcd 1.6 abc 0 a 4.2 abcde 474 .8 ab 1.8 abc _._2__ab 2.8 ab Mean 2.5 1.7 .3 4.5 1/ Numbers indicate observed mean number of sprouts per root segment. Tests of differences among means for number of sprouts on 20 February 1987, and total sprouts were performed on means adjusted by the covariate diameter. 2/ Means followed by the same letter are not significantly different at the 52 level. 26 Table 1.7 Comparisons of sprout production among clones from tall and short families. Character Mean response of Mean response of measured tall clones short clones Significance No. sprouts 2/20/87 3.3 2.5 ns (3.1)” (2.8) No. sprouts 3/23/87 1.5 1.5 us No. sprouts 5/26/87 .2 .3 ns Total no. sprouts 5.0 4.3 ns (4.8) (4 7) 1/ Figures in parentheses indicate mean number of sprouts adjusted with the covariate root diameter. Adjusted means were used to compute contrast F-tests when previous F-tests had shown the use of diameter as a covariate to be significant. ns not significantly different at the 52 level. 27 each root half. Since the patterns of sprout appearance were very similar for all collection dates, only data for the total number of sprouts for all collections are presented in Table 1.8. Summary of the data in Table 1.8 was done after an F-test indicated that differences in sprout production between the two root halves was highly significant (12 level). The data indicate that sprout distribution patterns in locust roots are similar to those in other plant species. Rooting of Sprouts Since sprout traits were similar among the three dates of collection (20 February, 23 March, and 26 May 1987), Tables 1.9, 1.10, and 1.11 were develOped based on the combined means of all collections. Table 1.9 gives the F- tests for rooting percent and number of roots per,sprout; there are very significant differences (12 level) among the clones for the measured traits. In the figures presented for "Mean number of roots per sprout" (Table 1.10), only Sprouts which develOped roots were included in the analysis. This procedure was followed to avoid biasing the results in favor of sprouts from clones with high rooting percentages.' Rooting varied from a low of 352 (clone 473) to a high of 962 (clone 461). All of the significant clonal differences in rooting were due to the relatively poor performance of clones 473 and 474. Differences among the clones in numbers of roots produced were more pronounced than were differences in 28 Table 1.8 Sprout distribution on roots of 21 black locust clones. Clone Proximal half Distal half 454 5.5 (93) 1/ .4 (7) 455 4.0 (95) .2 (5) 456 4.8 (96) .2 (4) 457 4.6 (87) .7 (13) 458 3.7 (92) .3 (8) 459 2.8 (97) .1 (3) 460 4.7 (96) .2 (4) 461 4.9 (94) .3 (6) 462 2.2 (100) 0 (0) 463 5.7 (80) 1.4 (20) 464 4.3 (96) .2 (4) 465 4.0 (91) .4 (9) 466 4.0 (76) 1.3 (24) 467 4.6 (92) .4 (8) 468 3.4 (81) .8 (19) 469 3.1 (94) .2 (6) 470 3.5 (95) .2 (5) 471 4.7 (90) .5 (10) 472 5.6 (97) .2 (3) 473 3.8 (90) .4 (10) 474 2.6 (93) .2 (7) Mean 4.1 (91) .4 (9) 1/ Numbers in parentheses indicate the percentage of sprouts occurring in each half of the root. 29 Table 1.9 F-tests of percent rooting and mean number of roots per sprout produced by 21 black locust clones. YBPercent of sprouts which develOped roots. Source d.f. M.S. F ratio Significance Block 3 .506 .87 ns Clone 20 3.800 6.53 ** Error 60 .582 Total 83 Y=Mean number of roots per sprout which develOped roots. Source d.f. M.S. F ratio Significance Block 3 350 1.52 ns Clone 20 1603 6.97 ** Error 60 230 Total 83 ** significantly different at the 12 level. as not significantly different at the 52 level. 30 Table 1.10 Mean percent rooting and mean number of roots on sprouts removed from 21 black locust clones. Mean Mean percent number Clone rooted of roots 454 92 b 1/ 8.6 ed 455 90 b 7.0 abcd 456 93 b 7.2 abcd 457 84 b 5.8 abc 458 78 b 5.4 abc 459 91 b 6.3 abcd 460 94 b 5.7 abc 461 96 b 7.9 bed 462 70 b 6.0 abcd 463 92 b 6.8 abcd 464 91 b 5.7 abc 465 90 b 5.1 abc 466 91 b 10.0 de 467 82 b 8.9 ed 468 84 b 6.6 abcd 469 82 b 4.0 a 470 72 b 11.7 e 471 76 b 6.4 abcd 472 82 b 5.4 abc 473 35 a 3.4 a 474 50 a 4.0 ab Mean 82 6.6 1/ Within columns, means followed by the same letter are not significantly different at the 52 level. 31 rooting percentages. This broader range of diversity is probably due either to differing response to exogenous IBA application when the sprouts were first established, or to differing endogenous levels of IAA in sprouts severed from the parent root. Exactly why the differences in root numbers occurred is difficult to determine from this study, but genotype obviously affects response. Table 1.11 presents the comparison between clones from tall families and clones from short families. Percent rooting did not differ significantly among the two groups. However, the number of roots produced by the two classes did differ significantly (12 level). Why the difference in the number of roots produced was so great is difficult to determine. It could be due to varying levels of endogenous hormones or could be a secondary result of some other factor such as the initial length of the sprout. Smaller sprouts would have reduced carbohydrate reserves and would be less able to produce relatively large numbers of roots. However, analysis of the two vigor groups for differences in initial sprout length indicated no significant differences (mean sprout length of clones from tall families was 11.4 cm, while the mean sprout length of clones from short families was 10.6 cm). CONCLUSIONS The sprouting and rooting procedures described in this chapter can be assessed for their potential for mass propagation systems. Using the methods outlined here, 15 cm Table 1.11 Comparisons of rooting of sprouts among clones from tall and short families. Character Mean response of Mean response of measured tall families short families Significance Percent rooted 84.7 81.8 ns Number of roots 8.8 5.6 ** ** significantly different at the 12 level. us not significantly different at the 52 level. long root cuttings produced an average of 4.5 sprouts per cutting of which an average of 822 developed roots. Therefore, it is possible to produce an average of 3.7 rooted cuttings per 15 cm length of root. Possibly, higher numbers of rooted cuttings per length of root could be produced if sprouts smaller than 5 cm are included in propagation efforts. However, work with very small sprouts would probably require a more sephisticated rooting environment than could be provided by the facilities used in this study. In addition to rooting excised shoots, roots and shoots can also be induced on field collected root segments. Although this approach yields only one plant per root segment, it has the advantage of being fairly simple, predictable, and less labor intensive than sprout rooting efforts. Almost 602 of the parent root segments used in 33 this study developed roots. With further tests of cultural techniques, it might be possible to achieve a similar degree of success using shorter and/or thinner root pieces than were used in this study, thereby making the prOpagation of one plant from each root segment an attractive alternative to rooting excised sprouts. CHAPTER 2 Effects of treatment of black locust root sections with benzylaminOpurine INTRODUCTION If black locust plants are to be prOpagated by forcing sprouts from roots, sprouts produced from a given length of root must be maximized. It is evident from the work of Bonnett and Torrey (1965), Peterson (1975), and Schier (1981) that some portions of root segments do not produce sprouts. Sprouting is usually inhibited by the relatively high concentrations of indole acetic acid (IAA) near the distal end of root cuttings. However, some researchers have found that many of the inhibitory effects of IAA can be counteracted by cytokinin application. Schier (1981), working with Populus tremuloides, tested the effect of the cytokinin benzylaminopurine (BAP) on sprout production of 15 cm root segments. He found that sprouting was promoted by soaking root segments for 24 hours in varying concentrations of BAP. The concentrations which Schier (1981) tested were 0 to 20 mg per 1 of water. Sprout production was maximized at 10 mg of BAP per 1 of water (18 34 35 sprouts per root segment at 10 mg versus 11.5 sprouts per segment for the control). Other researchers have also studied sprout production with cytokinins, mostly in herbaceous species. Bonnett and Torrey (1965) treated roots of Convolvulus arvensis in vitro with IAA and two different cytokinins (BAP and phenylaminopurine (PAP)). The cytokinin treatments inhibited root formation without changing the number of buds formed or the distribution of the buds along the root segments. Though this result appears to contradict the findings of Schier (1981), there were many differences in the conditions under which the two authors conducted their experiments. Bonnett and Torrey (1965) were culturing very short root segments (1.5 mm) of an herbaceous plant for a long period of time (6 weeks). Schier (1981) soaked fairly long root sections (150 mm) of a woody species for a short period of time (24 hours). Given the vastly different exposure times and different types of material used, it is not surprising that the two authors obtained different results. If the use of BAP does improve sprout production, its mode of action will most probably be to counteract the inhibitory effects of endogenous IAA. Any anti-IAA sprout promoting effect of BAP should result in changes in the total number of sprouts produced on a given length of root and could alter the IAA mediated pattern of sprout distribution on the root. These two characteristics, sprout number and sprout distribution pattern, were the primary 36 criteria used to evaluate the effectiveness of BAP in the present study. Additionally, the rooting ability of sprouts produced in the BAP treatments was used to evaluate whether changes in sprout vigor occurred in response to BAP treatment . MATERIALS and METHODS Most of the previous work on the effects of cytokinin application on sprout production of root segments had been performed in 11553. Schier (1981) is one of the few authors who reported work on in 2112 culture systems. Therefore, the experimental protocol described by Schier (1981) for aspen was chosen as the basis for this investigation because his material (relatively large root segments) was similar to the material studied here. Four concentrations of BAP were included in this test: 0, 5, 10, and 20 mg of BAP per 1 of water. Root cuttings were soaked for 24 hours in the BAP solutions and then were placed on sprouting benches and cultured in the manner described in Chapter one. Each BAP treatment was applied to three root segments in each of nine blocks. Since the quantity of root material from individual clones was limited, five blocks were made up from a bulk lot of roots from many clones. For four clones (454, 456, 458, and 464), there were enough root segments to allow an entire block to be made up of an individual clone. After 45 days in the vermiculite sprouting medium, the root sections were lifted and sprouts greater than 5 cm in 37 length were counted. After this initial sprout harvest, the parent root sections were returned to the prepagation bench to allow for further sprout production. Sprouts were harvested on two occasions, as previous studies (Chapter one) indicated that a third harvest yielded few additional sprouts. The study was established 29 May 1987, 184 days after the root segments were collected. Using root segments stored at 2° C. for an extended period could confound the results of this study. However, Schier and Campbell (1978) working with Populus tremuloides observed no significant loss in sprout production even when dormant roots were stored for as long as 175 days. As a measure of root viability, sprout production by the roots in this study was compared with sprout production from the roots established shortly after root collection (Chapter one). All analyses were performed using the Number Cruncher Statistical System (NCSS) version 5.0 (developed by Jerry Hintze) on plot averages of the character analyzed. Tests of differences of means were performed using the Newman- Keul's range test (52 level unless otherwise noted). Data sets which exhibited a Poisson distribution (small value count data and percentages) were square root transformed prior to analysis. 38 RESULTS and DISCUSSION Sprout Production It was expected that the most important result of BAP treatment of root segments would be increased sprout production. Also of interest was whether the rooting ability of sprouts forced from BAP-treated root segments differed from sprouts forced from non-treated root segments. Table 2.1 shows the F-test for the treatments examined in this study. The significant F-test (12 level) for block (Table 2.1) indicates that there were significant differences in sprouts produced among the blocks. Almost all of the significant variability due to block originated from the four blocks composed of known clones indicating that different clones performed differently. This point was amply demonstrated in Chapter one of this thesis. The non-significant result of the test of the treatment indicates that BAP did not affect shoot production. The mean number of sprouts produced for the various BAP treatments are shown in the "Total sprouts" column of Table 2.2. Patterns of Sprout Distribution BAP treatment had no significant effect on the distribution of shoots on the root segments (Table 2.2) as there was no significant interaction between treatment with BAP and the zone where sprouting occurred. Table 2.1 F-test of the effect of benzylaminOpurine on the production of sprouts from black locust root segments. Source d.f. MS F-ratio Significance Block 8 4.93 5.95 ** Treatment 3 .42 .50 ns Error 24 .83 Total 35 ** significantly different at the 12 level. us not significantly different at the 52 level. Rooting of Sprouts BAP treatment did not significantly affect rooting percentage (overall mean of 892), number of roots produced per cutting, or the length of the longest root produced on the cuttings. Sprouts from roots soaked for 24 hours in BAP at 10 mg per 1 of water were significantly shorter than Table 2.2 Sprouts occurring in the proximal and distal halves of black locust roots treated with four levels of benzylaminOpurine. BAP 1/ Proximal half np Distal half Total sprouts 0 1.96 (86) 2’ .33 (14) 2.29 5 2.11 (87) .22 (13) 2.33 10 1.44 (74) .52 (26) 1.96 20 2.15 (89) .26 (11) 2.41 Mean 1.924(85) .33 (15) 2.25 1/ mg of BAP per 1 of water. 2/ Numbers in parentheses are the percentage of sprouts occurring in each half of the root. Table 2.3 Length of sprouts produced by root segments treated with four levels of BAP. Sprout BAP 1/ length (mm) o 133 a 2/ 5 94 ab 10 82 b 20 99 ab 1/ mg of BAP per 1 of water. 2/ Numbers followed by the same letter are not significantly different at the 52 level. sprouts in the other three treatments (Table 2.3). Though sprout length for sprouts produced by concentrations of BAP of 5 and 20 mg per 1 of water are not significantly different than the control treatment, sprout lengths for these two BAP concentrations also seem to be depressed. Other Considerations One of the primary concerns when this study was designed was the effect of long term storage on sprout production by black locust roots. Although the roots did not visually appear to be affected by storage, the number of sprouts per root segment were reduced after long term storage. Sprout production from fresh roots averaged 4.5 Sprouts per root (Chapter one). After 184 days in storage, 2.25 sprouts were produced per root. Storage of black locust root segments for 184 days appears to be too long if the roots are to retain maximum sprouting capacity. 41 CONCLUSIONS Although at least one author (Schier 1981) has had success in promoting suckering from root segments by using exogenous hormone applications, this response was not observed in this study. None of the concentrations of BAP tested showed any effect on the characters measured with the exception of sprout length (Table 2.3). The roots used in this chapter behaved in all ways similar to untreated roots described in Chapter one. Long term storage of the root segments (184 days) decreased sprout production by the root segments. Although absolute numbers of sprouts were reduced by storage of root segments, differences in numbers of sprouts produced were significant (12 level) among the blocks composed of identified clones. CHAPTER 3 PrOpagation of mature black locust trees from root and dormant stem cuttings INTRODUCTION The problems associated with vegetatively prOpagating mature plants have received a great deal of attention in the past. Many changes are likely to occur in plants as they mature including the onset of flowering, less vigorous growth habit, reduced thorniness, and a decline in the ability to root adventitiously (Clark 1981; Davies 1983; Schaffalitzky De Muckadell 1954). Also, material taken from mature portions of plants may express weak apical dominance (Schaffalitzky De Muckadell 1954) producing plants that are often more prostrate than plants propagated from juvenile material. Black locust has many of the same traits noted above. Mature portions of black locust trees tend to be less thorny, produce most or all of the flowers (Schaffalitzky De Muckadell 1959), have slower growth (Otis 1931), and are more difficult to prOpagate (Swingle 1937). When Swingle (1937) conducted experiments on the establishment of locust by using root sections placed in a nursery bed, he found 42 43 that 52 of the sections from 60-year-old trees produced usable plants versus 252 of the sections from 5-year-old trees. Even though vegetative propagation of mature plants is often difficult, there are many potential advantages to prOpagating older trees. Beneficial traits might not become obvious until a tree has reached maturity, too late for selection and prOpagation of juvenile material. Such characters include reduced thorniness and increased flowering. There are several potential techniques for propagating mature plants with a reasonable degree of success. The discussion in this chapter will be concerned primarily with three of these methods; the direct rooting of hardwood cuttings, the outdoor propagation of root segments, and grafting stem material. ROOTING OF HARDWOOD CUTTINGS The method of vegetative prOpagation which is potentially the easiest is the rooting of hardwood stem cuttings. PrOpagation systems which use hardwood stem Cuttings have several advantages over systems based on other materials. Stem cuttings can be readily collected in quantity with almost all of the aerial portion of the plant serving as a source of prOpagation material. Even when fairly large amounts of material are removed during the collection process, damage to the tree is usually quite minor. There is a relatively long period of time when stem 44 material is at the prOper physiological state for collection (late fall to early spring). The requirements for collection, storage, and handling of dormant stem material are much less demanding than for other types of plant material. Finally, the actual propagation facilities used for rooting hardwood cuttings are often much simpler and less prone to failure than facilities required for other types of material. Because of their many advantages, propagation systems based on hardwood cuttings are probably among the oldest and most well researched. Stoutemyer et al. (1940) studied the effects of various chemicals on the rooting of hardwood cuttings of black locust. When cuttings were treated with either indoleacetic acid (IAA), naphthylacetic acid (NAA), or indolebutyric acid (IBA), rooting percentages varied from 502 to 902. When untreated cuttings were tested, rooting percentages were never greater than 12. Stoutemyer et al. (1940) state (unsupported by data) that hardwood cuttings collected from any portion of the plant and gathered at any time during the dormant season rooted equally well in the greenhouse. In contrast to the successes of Stoutemyer et al. (1940), Lee (1972) was able to root only 132 of the black locust stem cuttings with which he worked. However, it is not clear whether Lee used any root promoting chemicals in his studies. Without root promoting hormones, low rooting percentages are probable. One aspect of Lee's work is the inclusion of material from four different trees. Although 45 clone did not significantly affect rooting percent, it did affect the subsequent growth of cuttings which developed roots. Little evidence exists as to the need for cold temperature conditioning required for rooting of black locust stem cuttings. Of primary interest are the conditions required following fall bud formation to ensure prOper flushing and adventitious rooting of collected stem segments. Low temperature conditioning requirements have been described previously for other species (Kramer and Kozlowski 1979) but not for black locust. Since black locust has an indeterminate growth habit (Kramer and Kozlowski 1979), it may require little or no chilling to enable it to root adventitiously. Studies are described in this chapter which address the relationship between date of collection and rooting success in dormant stem cuttings. Another important aspect of the prOpagation of mature plant material concerns the form of the plants produced. Neither Stoutemyer et al. (1940) nor Lee (1972) discussed the form of black locust plants which resulted from their propagation efforts. Problems with t0p0phytic growth habit have been reported for other species propagated from stem material (Schaffalitzky De Muckadell, 1954). Measurements were made on black locust material propagated from stem cuttings in the studies described in this chapter. These measurements will be used to assess the physiological maturity of propagated material and to demonstrate possible problems related to mature growth habit. 46 MATERIALS and METHODS Evaluating hardwood stem cuttings for the purpose of propagating black locust focused on the effects of time-of- year of collection, clone, and IBA treatment on rooting success. This study involved the collection of hardwood stem cuttings every two months from November 1986, until just after bud break in, May 1987. Collection of Cuttingg On 4 November 1986, 19 January 1987, 9 March 1987, and 1 May 1987, black locust stems were collected from four clones. Two clones were located in the Dansville State Game Area (Dansville 1 and 2), Ingham County, Michigan (owned and managed by the Michigan Department of Natural Resources), a third was in the Sandhill Research Area, Ingham County, Michigan (owned and managed by Michigan State University), and a fourth was in the Jolly Road field research area of the Michigan State University Department of Horticulture, Ingham County, Michigan. On all dates of collection, stem material was collected from one tree per clone. Material was collected from the upper portions of trees which had at least 502 of the crown totally Open. All clones from which stem material was collected appeared healthy and vigorous. At each time of collection, stem material was removed from the trees in approximately half meter lengths and 'placed in labeled plastic bags. As soon as possible, the 47 stem sections were transferred to a refrigerated storage area (2° C). Preparation and Treatment of the Cuttingg Within a few days of collection, the stems were sectioned with hand pruners into 15 cm lengths and separated into three blocks on the basis of diameter. The thickest material was assigned to block one and the thinnest material was assigned to block three. After being sorted, the basal end of each cutting was wounded by making 2, 2 cm long cuts on either side with a sharp knife. All cuttings were treated by either dipping the basal 2 cm of the cuttings in a 50/50 solution (v/v) of isoprOpyl alcohol and water without hormone, or by dipping the basal 2 cm of the cuttings in a 50/50 solution (v/v) of isoprOpyl alcohol and water with pure IBA crystals added to make a 4000 ppm solution. A liquid dip of IBA was used because liquid hormone dips have been reported to be much more effective than powders for promoting root initiation on hardwood cuttings (Bonaminio 1983; Stoutemyer et al. 1940). After the IBA treatment the cuttings were transferred to a rooting bench with intermittent mist and bottom heat capabilities in a warm greenhouse (the design and construction of this bench is described in Chapter one of this thesis). On the rooting bench, the cuttings were placed in containers constructed of plastic coated paper. iEach container measured 5 by 5 by 15 cm. Containers used in 48 the study were filled with a medium consisting of peat and perlite mixed in equal parts. Because the containers used in this series of trials were not sturdy, they were supported in a hard plastic "milk case" type frame. Each frame measured 30 by 30 by 15 cm, and held 36 containers in a 6 by 6 arrangement. One stem cutting was placed in each container. Each treatment plot (clone by hormone) consisted of 6 containers, and therefore included 6 hardwood cuttings. As the hardwood cuttings flushed quickly in the warmth of the greenhouse, intermittent mist was used to prevent desiccation prior to root formation. Mist was operated for 5 seconds every 10 minutes during daylight hours. During the rooting period, the cuttings were drenched with fungicides frequently to inhibit attack by fungi. After 8 weeks on the mist bench, the cuttings were lifted and evaluated for root production. The number of roots and the length of the longest root present on each cutting was measured. The cuttings which produced roots were evaluated for various traits indicative of maturity, including stem form and thorn length. All analyses were performed using the Number Cruncher Statistical System (NCSS) version 5.0 (develOped by Jerry Hintze) on plot averages of the character analyzed. Tests of differences of means were performed using the Newman- Keul's range test (52 level unless otherwise noted). Data sets which exhibited a Poisson distribution (small value 49 count data and percentages) were square root transformed prior to analysis . RESULTS and DISCUSSION There were definite differences in rooting percent among the dates of collection (Table 3.1). Considering only the results of the 4000 ppm IBA treatment, mean rooting percent was highest at the January collection (682), cuttings collected either before or after January rooting less well. This trend also held for the rooting percent of cuttings not treated with IBA. January collected cuttings for two of the clones (Dansville l and 2) rooted nearly as well without as with IBA. Stem material collected in May, just after the spring bud flush, showed poor rooting for all clones. When comparisons were made among IBA treated cuttings from clones collected at what appears to be the prOper time Of year (January), there were significant difference in rooting percent (1002 for Dansville 2, 942 for Dansville 1, 562 for Horticulture, and 232 for Sandhill). Higher rates of IBA might have improved rooting of poor rooting clones. Higher hormone application rates or longer exposure times could supplement naturally low levels of IAA existing in these clones. However, if higher IBA rates did substantially improve the rooting percent of the Horticulture and Sandhill clones, these higher rates might be detrimental to the rooting of naturally good rooting 50 Table 3.1 Effects of clone, hormone treatment, and time of collection on rooting percent of hardwood stem cuttings of black locust. HARVEST DATE Clone IBA (ppm) 4 Nov 86 19 Jan 87 9 Mar 87 1 May 87 ------------- Percent rooted----------- Dans. 1 o 17 1/ 83 6 11 4000 78 a 2/ 94 a 67 a 17 a Dans. 2 0 6 88 6 0 4000 78 a 100 a 33 b 0 a Hort. 0 0 0 0 0 4000 6 b 56 b 0 c 22 a Sandhill 0 0 0 0 0 4000 22 b 22 c 0 c 6 a Mean 0 6 43 3 3 4000 46 68 25 11 1/ Values equal mean percent rooting of 18 hardwood stem cuttings. 2/ Values within each collection date for percent rooting at the 4000 ppm level followed by the same letter are not significantly different at the 52 level. 51 clones. An "Optimum" hormone concentration or exposure time may exist for each clone. On 18 September 1987, data were collected to assess the degree of maturity of plants prOpagated from the hardwood cutting trials (Table 3.2). All of the clones expressed mature traits, having shorter thorns, and more obtuse branch angles than juvenile seedlings. Another indication of plant maturity is flower production. The cuttings collected in January were scored on 26 February 1987 for the appearance of flowers on the shoots flushed in the greenhouse. Material collected from all of the clones produced flowers with percent of cuttings flowering ranging from 612 (Horticulture clone) to 972 (Dansville 2 clone). The flower production mean for all clones was 812. CONCLUSIONS Mature black locust plants can be propagated from dormant stem cuttings. Using the methods described in this chapter, rooting was obtained with mature plant tissue from four different clones. The rooting percentage of stem cuttings from clones tested in this study varied with time ' of collection with material collected in January rooting best. Stem material collected on other dates (especially prior to January) might perform as well as material collected in January if handling and/or prepagation 'conditions‘were altered. Treatment which exposed the cuttings to low temperatures (2° to 4° C) for one to two 52 Table 3.2 Form and vigor of black locust produced by hardwood cuttings versus similar data for seedlings. Mean Mean Mean length of thorn branch Clone growth (cm) 1/ length (mm) angle 2/ Dansville 1 44 0.30 81° Dansville 2 53 0.25 63° Horticulture 47 0.0 65° Sandhill 35 0.0 55° Seedling 3/ 93 8.2 10° 1/ Values shown are the mean total growth produced by all branches (18 September 1987). 2/ Values shown are the mean angle between the terminal branch the vertical. 3/ Values shown indicate the mean of measurements made on the top two branches of ten randomly selected three-year- old seedlings. All seedlings were tap pruned 8 July 1987. Values represent regrowth following pruning. 53 months could satisfy the cold conditioning requirements which black locust appears to have. Rooting percentages also varied over the different IBA levels. All clones tested rooted better following treatment with 4000 ppm IBA. Rooting percentages of stem cuttings collected in January and treated with 4000 ppm IBA ranged from 1002 (Dansville 2) to 232 (Sandhill). Although variation in rooting percentage due to clone was very significant (12 level) for all collection dates except May, it is possible that higher concentrations of IBA might improve rooting rates of poor rooting clones. Collection of hardwood cuttings from mature portions of older black locust trees resulted in prOpagated plants which exhibited many mature traits. These traits included small thorns, wide branch angles, and flower production. OUTDOOR ROOT PROPAGATION STUDIES The outdoor nursery bed culture of black locust is the current standard practice of most locust mass propagation systems (Keresztesi 1983). Since black locust sprouts from the roots quite readily, the propagation of locust using root sections is relatively easy. The primary limitation of root based systems is that they require fairly large amounts of root material for the production of each plant. Nevertheless, field propagation of black locust with root cuttings has been well researched and should be considered 'as one of the systems with the most potential for large scale prOpagation efforts. 54 Since black locust sprouts freely from its roots, there have been few treatments investigated for its field culture. The most universal recommendations have concerned the age and orientation of the root segments when they are placed in the ground. Swingle (1937) investigated the effects of root age, root diameter, and orientation on the rooting of black locust root cuttings. He found a considerable difference in the production of usable plants among roots collected near trees of various ages (52 prOpagation success from 60-year- old trees, versus 252 prOpagation success from 5-year-old trees). Many authors recommend that black locust root cuttings be placed vertically in prOpagation beds rather than horizontally (Keresztesi 1983; Pike 1972; Swingle 1937). Swingle (1937) stated that the benefits of vertical root placement were more due to the form of the resultant plant than to an actual increase in percentage of cuttings rooted. This observation by Swingle (1937) fits well with what has been demonstrated in this paper concerning the polar Sprouting habit of black locust root segments. The distal end of the cutting, where the bulk of the roots are produced, is placed "down," while the proximal end of the root cutting, where the bulk of the sprouts are produced, is placed "up". Because of the polar patterns of sprout and root production, vertical placement of the root cutting should result in a plant with the root and shoot system "in line" with the original placement of the parent root cutting. 55 MATERIALS and METHODS Roots for outdoor prOpagation trials were collected on two separate occasions, 15 April 1986, and 7 August 1986. On each date, root cuttings were collected from three different clones. One clone was located in the Dansville State Game Area, Ingham County, Michigan (owned and Operated by the Michigan Department of Natural Resources), a second clone was located in the Michigan Stated University Kellogg Forest, Kalamazoo County, Michigan, and a third was located in the Michigan State University Sandhill Research Area, Ingham County, Michigan. Collected roots were of two different age classes; old, mature plants, and young, juvenile plants growing on the same root system. Roots collected near stems Of trees which were less than 3-years- Old were considered young and roots collected near stems of trees greater than 20-years-Old were considered Old. After the roots were collected they were placed in labeled plastic bags and transferred to a cold storage facility (2° C). Soon after collection, the roots were sectioned into 10 cm lengths. The sectioned roots were placed in blocks based on their diameter, the thickest roots in block one and the thinnest roots in block four. Since the roots collected from young trees were relatively thin when compared to roots from Old trees, diameters of the roots were measured and evaluated for use as a covariate in all analyses. 56 Since many authors (Keresztesi 1983; Pike 1972; Swingle 1937) recommend that root cuttings be planted in a vertical rather than a horizontal orientation, root cutting orientation was a treatment evaluated in this trial. For the vertical placement of the roots, the distal end of the root was pointed down. Orientation of the root cuttings was maintained by making a small, diagonal cut in the distal end- of each root when the roots were collected. After the root segments had been assigned to blocks, they were placed in an outdoor nursery bed at the Michigan State University Tree Research Center (operated by the MSU Forestry Department). The same experimental design was used for studies established with roots from both collection dates. The treatments tested included the clone of origin (three levels), the age of the material (two levels), and the orientation of the cuttings (two levels). Treatments were arranged in a randomized complete block design with 4 blocks, 10 root cuttings per plot (clone by age by orientation). When the studies were established, horizontal roots were placed on the surface of the soil and pressed in to a depth Of approximately 2 cm. Vertically placed roots were pressed into the soil so that the proximal end was approximately 2 cm below the soil surface. Throughout the summer of 1986, the root cuttings were watered whenever necessary to maintain adequate soil moisture levels. At the end of the 1986 growing season, all roots were lifted and evaluated for various traits. Data collected 57 included the number of roots and shoots produced, the length of the longest root and shoot, and the diameter of each root cutting. All analyses were performed using the Number Cruncher Statistical System (NCSS) version 5.0 (developed by Jerry Hintze) on plot averages of the character analyzed. Tests of differences Of means were performed using the Newman-Keul's range test (52 level unless otherwise noted). Data sets which exhibited a Poisson distribution (small value count data and percentages) were square root transformed prior to analysis. RESULTS and DISCUSSION Many of the root segments collected on 7 August 1986 produced sprouts, but all 480 root segments failed to root. Therefore, the discussion presented in this section will concern the study established from roots collected on 15 April 1986. F-tests for the clones used in this study are shown in Table 3.3. Rooting was uniformly poor for the Dansville Clone and none of the treatments tested significantly affected rooting (Table 3.4). Apparent differences in the young versus old treatment for the Sandhill clone were accounted for by differences in root diameter. The relatively thin roots of the young treatment produced proportionally fewer roots than the relatively thick roots Of the Old treatment. Although use of root diameter was significant (52 level) for the Kellogg clone, treatment differences were 58 Table 3.3 Analysis of variance tables for the effects of clone, root age, and root orientation on the rooting of black locust. Dansville Source d.f. M.S. F-ratio Significance Block 3 .0577 1.31 ns Age 1 .0425 0.97 ns Orient. 1 .0136 0.31 ns Age by Or. 1 .0010 0.02 ns Error 9 .0440 Total 15 Kellogg Source d.f. M.S. F-ratio Significance Diam. 1 .0967 5.42 * Block 3 .0069 0.38 ns Age 1 .1196 6.71 * Orient. 1 .1867 10.47 * Age by Or. 1 .0002 0.01 ns Error 8 .0178 Total 15 Sandhill Source d.f. M.S. F-ratio Significance Diam. 1 .2360 7.66 * Block 3 .1022 3.32 ns Age 1 .0407 1.32 ns Orient. 1 .0068 0.22 ns Age by Or. 1 .0079 0.26 ns Error 8 .0308 Total 15 * significantly different at the 52 level. us not significantly different at the 52 level. 59 Table 3.4 Effects of age, orientation, and clone on rooting of root sections placed in outdoor nursery beds. CLONE Age Orientation Dansville 2 Kellggg Sandhill ---------- Percent rooted---------- Young Vertical 8 a 1/ 20 be 2/ 15 a Horizontal 10 a 52 c 11 8 Old Vertical 13 a 22 a 56 a Horizontal 17 a 36 ab 57 a 1/ Means followed by the same letter are not significantly different at the 52 level 2/ Actual means are shown. Difference among means tests were performed on means adjusted by diameter when use of diameter as a covariate was shown to be significant. also significant (52 level) for root age and root orientation (Table 3.3). The young and horizontal treatments produced significantly better rooting than the old and vertical treatments (Table 3.4). In a previous study (Swingle 1937), no significant differences in the traits measured could be attributed to root orientation (vertical versus horizontal). Swingle (1937) felt that vertical placement of the roots, with the distal end down, lead to a more balanced plant. The parent root segment was "in line" with shoot and root development and the original root cutting was more easily incorporated iinto the final plant. 60 The results of this study, at least for the Kellogg clone, indicate that horizontal placement of roots can significantly improve rooting percent. Further studies are needed to assess the long term affects Of root orientation on tree form and health. CONCLUSIONS When the three clones used in this study were analyzed together, no significant differences among the treatments could be detected. Only when data for each clone was analyzed independently did significant treatment effects become apparent. This indicates that the treatments are significantly affected by clone. Use of diameter Of the root as a covariate was significant for each clone which had large differences among the treatments. Root diameter Often accounted for large apparent differences in the data. GRAFTING One method of vegetative prOpagation which has great potential value is grafting. One of the uses of grafting techniques is to research the problems and possibilities of controlled pollination systems. Creating successful full- sib crosses in black locust is at present very difficult (Bongarten 1986; Kim 1975). The low success rates reported for controlled hybridizations can be attributed to two main causes, the delicacy of the flowers and the adverse conditions often present in field culture of perennial woody 61 species. Flower morphology is difficult to change, but grafting mature scion-wood onto potted understocks allows pollination in a controlled environment greenhouse, thus eliminating the detrimental effects of poor weather. The use of mature material as scion-wood in grafting Operations can Often lead to earlier abundant flowering than can be achieved with seedling stock (Wright 1976). It could be very advantageous to establish seed orchards with early flowering grafted material, thus reducing the amount of time before seed production commences. If desirable clones are identified, the use Of grafted material in seed orchards would result in large genetic gains over the use of open- pollinated seed prOpagated material. To determine whether grafting is a viable option with black locust, a small grafting trial was initiated. The Objectives of this trial were to determine if grafting can be performed on black locust with a reasonable degree of success, to assess whether the type of graft used has a significant effect on the success rate or resulting vigor of the graft, and to evaluate the influence of clone on graft BUCCGBB 0 MATERIALS and METHODS At the time of the third collection of stem cuttings for the hardwood prOpagation studies (9 March 1987), stem material was also collected for use as scion-wood for a 'small grafting trial. All of the materials used in this trial were from vigorous, mature areas of the plant. 62 Because scion-wood was limited, only three of the four clones collected for the hardwood stem cutting trials were included in the grafting trial. The clones used included one of the two clones from the Dansville State Game Area, the clone from the Sandhill Research Area, and the clone from the Horticulture Department Field Research Area. Two different types of grafts were attempted, a side graft, and a whip and tongue tOp graft. The understocks used in this study consisted of 2-year- Old Open pollinated seedlings. These seedlings had been placed into 2-gallon plastic pots during the summer of 1986. On 19 February 1987, three weeks prior to the initiation of the grafting procedures, the understocks were moved from an Outdoor holding area to a heated greenhouse. The understocks were exposed to this period of warmth so that they would be actively growing when grafted. The tOps Of understocks grafted with the whip and tongue technique were totally removed just before the grafting cuts were made. For the side grafts, the tops of the understock were not removed until approximately three months after the grafting procedure was completed. Scion-wood was prepared by sectioning the long, field collected black locust stems into segments approximately 10 cm in length. Each scion segment had two to three nodes. When making the graft, cuts were made on the understock and scion with a sharp knife, and the cambiums of the scion and understock were matched as well as possible. Since the understock usually had a greater diameter than the scion, it 63 was Often possible to match the cambiums of the scion and the understock on only one side. After the scion and understock had been put together, they were securely wrapped with rubber grafting bands. The graft unions were then covered with stretched parafilm to prevent drying. Grafted plants were held in the greenhouse until 20 May 1987, at which time they were moved Outside to a shade facility. Survival Of the grafts was assessed on 18 September 1987. Also at this time, data were collected on various growth characters, including the total length of all branches produced, the mean thorn length, and the angle of the terminal branch with the vertical. Branch angle and thorniness data were collected so that an assessment of the physiological state of the plant material could be made. RESULTS and DISCUSSION The characteristics of the black locust grafts are shown in Table 3.5. As relatively few grafts were made, statistical analysis of the data was not entirely appropriate. Survival was very good with less than 52 of the grafts failing. Also, of the three clones tested, none showed any signs of incompatibility in the first season following grafting. However, problems with graft incompatibility Often develOp several years after grafting. Completed grafts must be observed over a long time period to fully assess their viability. The vigor of the scions, as measured by their growth, 64 was very good with a mean shoot length of 183 cm. Clone had little effect on growth. All of the scions exhibited traits of mature black locust plants (Table 3.5). The occurrence of mature traits in the grafted scions was especially apparent when growth data for the grafted material was compared with that for 3- year-old seedlings. The grafted material had much shorter thorns (mean thorn length 1.1 mm) and a much more obtuse branch angle (mean branch angle 50°) than did the young seedlings (mean thorn length 8.2 mm, and mean branch angle 10°). The type of graft had little effect on the growth of the scions (Table 3.6). Therefore, selection of one method over the other should be more a matter of personal preference than a decision based on the growth of the plant. Most of the scions used in this study flowered during the spring of 1987, shortly after being grafted. It will be interesting to observe whether the scions continue to flower for subsequent years, or whether they revert to a non- flowering state. Examples of both behaviors have been noted in the literature. Schaffalitzky De Muckadell (1955), working with beech (Fagus sylvatica), found that flower production two years after the time of grafting was abundant 0n scions collected from mature trees, but was nonexistent on scions collected from juvenile trees. However, Sluder (1966) found that tulip pOplar (Liriodendron tulipifera) (scions flowered the first growing season after collection, but did not flower again for several years. 65 Table 3.5 Effects of scion clone on scion growth and morphology. Mean Mean Mean length of thorn branch Clone growth (cm) 1/ length (mm) angle 2/ Dansville (7/7) 3/ 187 1.4 52° Horticulture (5/6) 180 1.0 58° Sandhill (9/9) 181 0.9 42° Seedling 4/ 93 8.2 10° 1/ Values shown are the mean total growth produced by all branches (18 September 1987). 2/ Values shown are the mean angle between the terminal branch the vertical. 3/ Numbers in parentheses indicate successful grafts (18 September 1987) / grafts attempted (12 March 1987). 4/ Values shown indicate the mean of measurements made on the top two branches of ten randomly selected three year old seedlings. All seedlings were tOp pruned 8 July 1987. Values represent regrowth following pruning. 66 Table 3.6 Effect of grafting technique on growth and branch angle of the scion. Mean Mean length of branch Type ofngraft ,ggowth (cm) 1/ angle 2/ Top (8/9) 3/ 190 47° Side (13/13) 179 51° 1/ Values shown are the mean total growth produced by all branches (18 September 1987). 2/ Values shown are the mean angles between the terminal branch the vertical. 3/ Numbers in parentheses indicate successful grafts (18 September 1987) / grafts attempted (12 March 1987). 67 CONCLUSIONS Based upon this study black locust can be easily grafted onto young seedling understock. Vigorous plants can be produced either by top or side grafting techniques. Problems of graft incompatibilities due to clone should be minor, at least for the first year following grafting. Mature locust scions maintain their mature characteristics for at least the first year following grafting. CHAPTER 4 RECOMMENDATIONS The studies described in this paper examined several possibilities for black locust vegetative propagation. Locust was successfully prOpagated by all of the methods evaluated; the rooting of shoot sprouts forced from root sections, direct planting of root segments in the greenhouse and nursery bed, the rooting of hardwood stem cuttings, and grafting. Each technique and type of plant material tested had advantages and disadvantages associated with its use. Therefore, the selection of one method over others should rely mainly on the objectives of the prOpagator and the facilities available. Propagation using roots Differences among the various techniques primarily involved the ease of execution of each technique and the efficiency with which the parent plant material was used. The system which was easiest to execute, planting roots directly in nursery beds, was relatively inefficient in the use of collected prOpagation material. On average, 4, 15 cm root segments were required to produce one plant. The system which required the most amount of care in execution, 68 69 the forcing and rooting of shoots collected from root segments, was the most efficient in the use of collected prOpagation material. On average, 3.7 plants were derived from each 15 cm length of root. The placement of black locust root segments directly in outdoor nursery beds is appropriate if large quantities of roots are available and/or if controlled environment facilities are not available. The forcing and rooting of shoots collected from root segments is apprOpriate when root quantities are limited and/or when controlled environment facilities, such as mist and fog systems, are available. An alternative to either the high or low intensity approaches outlined above is to develop roots and stems on black locust root segments placed in containers in a greenhouse. This system is somewhat inefficient in plant use, producing only one plant per root segment, but a higher proportion of initial root cuttings become usable plants than in the low intensity nursery bed system. On average, 2, 15 cm root sections are required to produce each plant. Roots collected during August sprouted, but produced no usable plants. Roots collected in December and April produced both sprouts and roots. Therefore, regardless of' which root based propagation system is selected, roots should only be collected from dormant plants. Stem propagation DevelOping adventitious roots on hardwood stem material was very successful for some trees but less successful for 70 others. The research presented in this thesis indicates that for all parent trees from which stem material was collected, hardwood stem cuttings collected on or around 19 January rooted best. If stem cuttings are to be collected at some other time of the year, rooting will be least adversely affected if cuttings are collected before mid- January rather than after. Stem cuttings collected from all trees benefited from treatment with IBA at 4000 ppm. Using the procedures outlined in this thesis, no problems should be encountered in black locust grafting operations involving mature stem material. All three trees used as a source of scion material in the grafting procedures produced healthy plants. An additional consideration when propagating black locust from dormant stem material is the desired maturity of the resultant plants. Some uses may require juvenile plants others, mature plants. Root cuttings produced vigorous, juvenile plants, whereas mature plant parts, such as the hardwood stem material used in the direct rooting or grafting trials, produced plants with mature characteristics. Other considerations Any black locust vegetative propagation technique will contain a genetic component which can greatly affect the success of the propagation effort. Different clones have varying capacities to produce adventitious sprouts and roots from roots and stems, leading to differing numbers of __.' 71 sprouts produced per length of root and differing rates of rooting. The genotype of prOpagated plants may also significantly alter the effects of any applied cultural treatments. The genetic aspects of any prOposed black locust vegetative propagation project should be considered if potential problems are to be avoided. LIST OF REFERENCES 72 LIST OF REFERENCES Benson, M.K., and D.E. Schwalbach. 1970. Techniques for rooting aspen root sprouts. Tree Plant. Notes. 21(3):12-14. Bonaminio, V.P. 1983. Comparison of IBA quick-dips with tale for rooting cuttings. Int. Plant PrOp. Soc. 33:565-568. Bongarten, 3.0. 1986. Personal communication. Bonnett, H.T., Jr. and J.G. Torrey. 1965. Chemical control of organ formation on root segments of convolvulus cultured in vitro. Plant Physiol. 40:1228-1236. Brown, A.B. 1935. Cambial activity, root habit and sucker shoot development in two species of poplar. The New Phy. 34(3):164-179. Brown, C.L. and P.P. Kormanik. 1967. Suppressed buds on lateral roots of Liqnidambar snyraciflua. Bot. Caz. 128(3-4):208-211. Cheeke, P.R., M.P. Goeger and G.H. Arscott. 1983. Utilization of black locust (Robinia pseudoacacia) leaf meal by chicks. Nit. Fix. Tree Res. Repts. 1:41. Cheeke, P.R., D.J. Harris and N.M. Patton. 1984. Utilization of black locust leaf meal by rabbits. Nit. Fix. Tree Res. Repts. 2:31. Clark, J.R. 1981. Juvenility and plant propagation. Int. Plant Prop. Soc. 31:449-453. Davies, F.T., Jr. 1983. Influence of juvenility and maturity in propagation. Int. Plant Prop. Soc. 33:559-564. Davis, J.M and D.E. Keathley 1988. Differential response to in vitro bud culture in mature Robinia pseudoacacia L. (black locust). Plant Cell Rpts. In press. Donnelly, J.R. 1970. Individual tree differences confound effects of growth regulators in rooting sugar maple softwood cuttings. USDA For. Serv. Res. Note NE-129, 6p. Northeastern For. Exp. Sta., Upper Darby, Pa. 73 Eliasson, L. 1971a. Adverse effect of shoot growth on root growth in rooted cuttings of aspen. Physio. Plant. 25:268-272. Eliasson, L. 1971b. Growth regulators in POpulus tremula III. Variation of auxin and inhibitor level in roots in relation to root sucker formation. Physiol. Plant. 25:118-121. Eliasson, L. 1971c. Growth regulators in POpulus tremula IV. Apical dominance and suckering in young plants. Physiol. Plant. 25:263-267. Farmer, R.E., Jr. 1962. Aspen root sucker formation and apical dominance. For. Sci. 8(4):403-410. Farmer, R.E., Jr. 1963. Vegetative prOpagation of aspen by greenwood cuttings. J. For. 61:385-386. Fowells, H.A. 1965. Silvics of forest trees of the United States. USDA For. Serv. Ag. Hndbk. No. 271., 762p. Gabriel, W.J., J.W. Marvin and F.H. Taylor. 1961. Rooting greenwood cuttings of sugar maple - effect of clone and medium. USDA For. Serv. Sta. Pap. No. 144., 14p. Northeastern For. Exp. Sta., Upper Darby, Pa. Hanover, J.W. 1985. Maximizing woody-biomass production through genetic selection, hybridization and intensive culture. U.S.D.O.E. Short Rotation Woody CrOps Program subcontract no. 9053 with Oak Ridge National Laboratory. Hanover, J.W. 1986. Maximizing woody-biomass production through genetic selection, hybridization and intensive culture. U.S.D.O.E. Short Rotation Woody CrOps Program subcontract no. 9053 with Oak Ridge National Laboratory. Hartmann, H.I. and D.E. Kester. 1968. Plant propagation: principles and practices. Prentice-Hall, Inc., Englewood Cliffs, N.J. 702p. Hudson, J.P. 1955. Propagation of plants by root cuttings II. Seasonal fluctuation of capacity to regenerate from roots. J. Hort. Sci. 30:242-251. Johnson, W.T. and H.H. Lyon. 1976. Insects that feed on trees and shrubs. Cornell University Press, Ithaca, New York. 464p. Kennedy, J.M., Jr. 1983. Geographic variability in black locust (Robinia pseudoacacia L.). M.Sc. Thesis. The University of Georgii. 66p. 74 Keresztesi, B. 1977. Robinia pseudoacacia: the basis of commercial honey production in Hungary. Bee World. 58(4):144-150. Keresztesi, B. 1980. The black locust. Unasylva. 32(127):23-33. Keresztesi, B. 1983. Breeding and cultivation of black locust, Robinia pseudoacacia, in Hungary. Forest Ecology and Management. 6:217-244. Kim, 0.8. 1975. Studies on the colchitetraploids of Robinia pseudoacacia L. Res. Rep. Inst. Forest Genet., Suwon, Korea. NO. 12. 108p. Kormanik, P.P. and C.L. Brown. 1967. Root buds and the deveIOpment of root suckers in sweetgum. For. Sci. 13(4):338-345. Kramer, P.J. and T.T. Kozlowski. 1979. Physiology of woody plants. Academic Press, Inc., New York, N.Y. 811p. Lee, S.K. 1972. Root and stem cuttings in black locust (Robinia pgeudoacacia L3). Res. Rep. Inst. For. Genet. (9):61-69. Maini, J.S. and K.W. Horton. 1966. Vegetative propagation of Populus app. I. Influence of temperature on formation and initial growth of aspen suckers. Can. J. Bot. 44:1183-1189. Marston, M.E. and P.J. Village. 1972. Regeneration of raspberries from root cuttings in response to physical treatment of the shoots. Hort. Res. 12:177-182. Mebrahtu, T. and J.W. Hanover. 1988. Heritability and expected gain estimates for various traits of black locust in Michigan. Silvae Genetics (submitted). Miller, R.O., P.D. Bloese, and J.W. Hanover. 1987. Black locust: A superior short-rotation intensive culture species for biomass production in the lake states. In: The proceedings of the Institute of Gas Technology/Energy from Biomass and Wastes XI. In press. Moore, G.M. 1984. Mechanisms of hormone action in plants. Int. Plant Prop. Soc. 34:79-90. Otis, G.H. 1931. Michigan trees: a handbook of the native and most important introduced species. The regents of the University of Michigan, Ann Arbor, Michigan. 362 p. 75 Peterson, R.L. 1975. The initiation and development of root buds. in: The develOpment and function of roots, p. 125-161. J.G. Torrey and D.T. Clarkson, eds. Academic Press, New York. Pike, A.V. 1972. PrOpagation by roots. Horticulture 50(5):56-6l. Robinson, J.G. and W.W. Schwabe. 1977. Studies on the regeneration of apple cultivars from root cuttings. II. Carbohydrate and auxin relations. J. Hort. Sci. 52:221-233. Schaffalitzky De Muckadell, M. 1954. Juvenile stages in woody plants. Physiol. Plant. 7:782-796. Schaffalitzky De Muckadell, M. 1955. A develOpment stage in Fagus silvatica characterized by abundant flowering. Physiol. Plant. 8:370-373. Schaffalitzky De Muckadell, M. 1959. Investigations on aging of apical meristems in woody plants and its importance in silviculture. Kandrup and Wunsch's Bogrtykkeri, Copenhagen, Denmark. 145p. Schier, G.A. 1972a. Seasonal variation in sucker production from excised roots of POpulus tremuloides and the role of endogenous auxin. Can. J. For. Res. 3:459-461. Schier, G.A. 1972b. Origin and develOpment of aspen root suckers. Can. J. For. Res. 3:45-53. Schier, G.A. 1974. Vegetative prOpagation of aspen: clonal variation in suckering from root cuttings and in rooting of sucker cuttings. Can. J. For. Res. 4:565-567. Schier, G.A. 1975. Promotion of sucker development on Populus tremuloides root cuttings by an antiauxin. Can. J. For. Res. 5:338-343. Schier, G.A. 1976. Differences among Populus species in ability to form adventitious shoots and roots. Can. J. For. Res. 6:253-261. Schier, G.A. 1978a. Variation in suckering capacity among and within lateral roots of an aspen clone. USDA For. Serv. Gen. Tech. Note INT-241, 6p. Intermt. For. and Range Exp. Sta., Ogden, Utah. Schier, G.A. 1978b. Vegetative prOpagation of rocky mountain aspen. USDA For. Serv. Gen. Tech. Rpt. INT-44, 12p. Intermt. For. and Range Exp. Sta., Ogden, Utah. 76 Schier, G.A. 1981. Physiological research on adventitious shoot development in aspen roots. USDA For. Serv. Gen. Rpt. INT-107, 11p. Intermt. For. and Range Exp. Sta., Ogden, Utah. Schier, G.A. and R.B. Campbell. 1976. Differences among Populus species in ability to form adventitious shoots and roots. Can. J. For. Res. 6:253-261. Schier, G.A. and R.B. Campbell. 1978. Effect of cold storage on develOpment of suckers on aspen root cuttings. USDA For. Serv. Res. Note INT-248, 8p. Intermt. For. and Range Exp. Sta., Ogden, Utah. Schier, G.A. and R.B. Campbell. 1980. Variation among healthy and deteriorating aspen clones. USDA For. Serv. Res. Pap. INT-264, 12p. Intermt. For. and Range Exp. Sta., Ogden, Utah. Scott, T.K. 1972. Auxins and roots. Ann. Rev. Plant Physiol. 23:235-258. Sluder, E.R. 1966. Hand-pollinate yellow-pOplar without climbing. USDA For. Serv. Res. Note SE-54, 7p. Southeastern For. Exp. Sta., Asheville, N.C. Steel, R.G.D., and J.H. Torrie. 1980. Principles and procedures of statistics. McGraw-Hill, Inc., New York, N.Y. 633p. Steiner, W.W. 1966. Black locust clonal selections. USDA Soil Conservation Service Technical Note TSC-PM-UDI. 4p. Sterrett, J.P., W.E. Chappell, and G.M. Shear. 1968. Temperature and annual growth cycle effects on root suckering in black locust. Weed Sci. 16:250-251. Stoutemyer, V.T., J.R. Jester, and F.L. O'Rourke. 1940. PrOpagation of black locust clones by treating hardwood cuttings with growth substances. J. of For. 38:558-563. Swingle, C.F. 1937. Experiments in propagating shipmast locust. J. of For. 35:713-720. Tew, R.K. 1970. Root carbohydrate reserves in vegetative reproduction of aspen. For. Sci. 16(3):318-320. Wright, J.W. 1976. Introduction to forest genetics. Academic Press, Inc., New York, N.Y. 463p. 77 United States Department of Agriculture. 1969. Black locust (Robinia pseudoacacia). Useful trees of the United States. Forest Service Bulletin FS-96 (TDS- 26). 3p. Zuffa, L. 1971. A rapid method for vegetative propagation of aspens and their hybrids. For. Chron. 47:36-39.