Research Report Series No. 7 RESEARCH REPORT 1991 ~Fi iv Alabama Agricultural Experiment Station Auburn University Lowell T. Frobish, Director Auburn University, Alabama September 1991 1~ ~U~ i'l '40 CONTENTS Page GREENHOUSE CROPS. ................................................ 3 Hypoestes phyllostachya Responds to Growth Regulators.......... 3 Response of Two Forcing Azalea Cultivars to Bonzi and B-Nine Applications ................................ 4 Postproduction and Marketing of Nandina domestica as a Potted Foliage Plant . .............. I....................... 6 Evaluation of Growth Regulators on Lisianthus as Potted Plants and for Cut Flowers ......................................... 7 Uniconazole Suppresses Bypass Shoot Development and Alters Flowering of Two Forcing Azalea Cultivars...................... 7 Alice du Pont Mandevilla Responds to Sumagic................... 9 Growth Inhibition of Dianthus Species..........................10 Consumer Preferences for Annual Bedding Plant Containers........ 12 Hardy Chrysanthemum Trial .................................... 13 Profile of the Perennial Plant Industry............................14 Influences of Subirrigation on Postproduction Longevity of Poinsettias ................................................ 15 Effects of Sumagic on Seed-Propagated Physostegia virginiana Alba .................................. 16 Branching of Vinca minor Increased by Growth Regulators........ 17 W OODY ORNAMENTALS............................................... 19 Fireblight Susceptibility of Ornamental Pears in Southern Conditions ....................................... 19 Growth of Five Shade Trees as Influenced by Trickle Irrigation Based on Net Evaporation .................................... 20 Chemically Induced Branching of Woody Landscape Plants ......... 21 Plant Response to Planting Method and Media....................23 Paclobutrazol Inhibits Growth of Woody Ornamental Plants....... 25 INSECT, DISEASE, AND WEED CONTROL ................................... 28 Herbicide Effects on Rooting and Root Growth ................... 28 Landscape Fabrics Suppress Growth of Weed Species.............. 29 Weed Control in Field-Grown Holly .............................. 31 Jumping Tree Bugs Are Important Natural Enemies of Obscure Scale Infesting Landscape Plantings of Pin Oak......... 33 Evaluation of Labeled Miticides for Control of Southern Red Mite on Azaleas ................................. 33 Chemical Control of Powdery Mildew on Miniature Roses......... 34 Effect of Nova Fungicide on Entomosporium Leaf Spot on Photinia .................................................. 35 Evaluation of Stirrup M as a Tank-Mix Addition to Miticides for Control of Spider Mites on Roses......................... 36 FIRST PRINTING 3M, SEPTEMBER 1991 Information contained herein is available to all without regard to race, color, sex, or national origin. GREENHOUSE CROPS Hypoestes phyllostachya Responds to Growth Regulators James T. Foley and Gary J. Keever PINK POLKA DOT PLANT, Hypoestes phyllostachya Bak., though usually considered an indoor plant, has the potential to be a popular addition to the herbaceous garden border. The plant has the uncommon charac- teristic of dark green foliage mottled with bright pink spots. Hypoestes forms a loosely branched mound 18 to 24 inches in height and is easily grown from seed. It prefers moist, well-drained soil and full sun. It may also grow in partial shade, producing a more open habit. Pinching is suggested when growing Hypoestes to decrease height and encourage branching. Despite pinching, however, the plant has an unkempt appear- ance that reduces the overall effect. By reducing plant height, a more compact habit may be achieved, con- centrating color and thereby increasing the foliage effect. Chemical growth retardants are commonly applied to bedding plants to produce more compact plants and extend the duration of attractiveness for sales. Reducing height may make Hypoestes more attractive to con- sumers as a pot crop or bedding plant. Paclobutrazol (Bonzi ? ) is effective in height control of chrysanthe- mum, geranium, and various other flowering annuals. Chlormequat chloride (Cycocel?) has growth retarding activity on dianthus, geranium, and many ornamental plants. The objective of this experiment was to improve plant quality of Hypoestes by suppressing shoot elon- gation with use of chemical growth retardants. Seeds of H. phyllostachya Pink Splash were sown on December 11, 1987, in flats of Pro-Mix BX? drenched with benomyl fungicide and placed under intermittent mist (5 seconds per 10 minutes) in a double polyeth- ylene greenhouse with minimum day/night tempera- tures of 70?F/60 0 F. Once cotyledons had fully emerged, seedlings were removed from mist on January 8, 1988, and 11 days later were transplanted to 4-inch pots of identical growth medium. Foliar spray treatments were applied just before runoff on February 23, 1988, and consisted of one or two applications of paclobutrazol at 25, 50, or 100 mg per liter and one or two appli- cations of chlormequat chloride at 3,500 mg per liter; second applications were made 2 weeks after initial treatment. An untreated control was included for com- parison. Buffer-X? was added as a surfactant at 0.2 percent by volume to spray solutions. Plants were about 1.6 inches tall with about three axillary shoots per plant when treated. The pruned treatment was pinched to one set of true leaves at a height of about 1.2 inches just prior to growth regulator application. Plant height was measured at 2-week intervals beginning 2 weeks after initial application until experiment termination at week 12. Plants were fertilized weekly with 200 mg per liter N from 20-10-20 Peters Peatlite Special?. Plant quality was based on plant size and shape relative to pot size and foliage color. One and two applications of 3,500 mg per liter chlormequat chloride provided the greatest suppression of growth throughout the experiment when compared to the control, table 1. Plants treated with chlormequat chloride exhibited darker green foliage compared to the control, an effect observed in other plants when this chemical was applied. These high-quality plants were also compact and consistently mounded, making them potentially more attractive to consumers. Four weeks after a single application of each rate of paclo- butrazol, plant height was suppressed compared to the TABLE 1. PLANT HEIGHT OF HYPOESTES PHYLLOSTACHYA AS AFFECTED BY GROWTH SUPPRESSANTS AT ONE AND TWO APPLICATIONS Plant height 1 , by weeks after treatment Treatment 2 4 6 8 12 cm cm cm cm cm Paclobutrazol- 1 application 25 p .p .m . ............................................................................................... 4.6 6.1 11.1 19.8 46 .3 50 p.p.m . .............................................................................................. . .4.9 5.6 8.7 15.0 36.9 100 p.p.m ............................................................................................. 4.7ns2 5.1ns 7.0q 10.5q 28.11 Paclobutrazol-2 applications 25 p .p .m . ............................................................................................... 5.5 7.1 13.6 23.1 47 .9 50 p .p .m ................................................................................................ 5.1 5.6 13.3 3 1.6 3 1.6 100 p.p.m .............................................................................................. 4.6ns 5.4ns 7.1q 11.2q 30.9c Chlormequat chloride-1 application 3,500 p.p.m . .......................................................................................... 4 .7 5.0 6.5 8.8 22.6 Chlormequat chloride-2 applications 3,500 p.p .m . .......................................................................................... 4 .1 4.4 5.7 7.6 22.0 Control .............................................................. ........................................ 6.2 8.8 18.6 29.0 53.3 LSD 3 ............................................ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 1.9 4.3 5.7 5.1 'Measurements in the tables are given in centimeters (2.5 cm = 1 inch). 21 = linear; q = quadratic; c = cubic; ns = not significant; control included in regression. 3 Mean separation within columns by a protected Fisher's least significance test, P = 0.05; LSD used for comparisons among growth retardants. control and height decreased as rate increased with either one or two applications of paclobutrazol. Leaf color of paclobutrazol-treated plants did not vary from the control and plants were loosely branched with a sprawling habit. Due to this loose branching habit, paclobutrazol-treated plants were of lower quality com- pared to the chlormequat chloride treatment. One or two applications of chlormequat chloride at 3,500 mg per liter acts to enhance the pink spots against the green background of the leaves, while reducing plant height. This results in a low, mounded plant and concentrates foliage color, making Hypoestes more effective in mass plantings. Foley is Graduate Student and Keever is Associate Professor of Horticulture. Response of Two Forcing Azalea Cultivars to Bonzi and B-Nine Applications Gary J. Keever FLORIST AZALEAS respond to long photoperiods, warm temperatures, ample moisture, and nutrients with rapid vegetative growth. This rapid growth necessitates fre- quent pruning to produce well-branched plants before flower buds initiate and develop. Short days and growth retardants hasten flower initiation and bud develop- ment. Daminozide (B-Nine*) and chlormequat (Cyco- cel?) are the principal growth retardants applied to florist azaleas. Complete flower bud development requires chilling, followed by warm temperatures dur- ing forcing. An excessive period from final pruning to the start of forcing usually increases the development of vege- tative axillary shoots (bypass shoots) during forcing. These. shoots often must be removed by the grower or florist before marketing. Bypass shoots are a particular problem on cultivars forced in January or later. In previous research, paclobutrazol (Bonzi?), which is not labeled for florist azaleas, effectively suppressed bypass shoot development of forcing azaleas. The objective of this test was to determine if the plant growth retardant currently labeled for florist azaleas, daminozide, or the retardant paclobutrazol could control bypass shoots if applied just prior to or immediately after chilling. Uniform liners of Dorothy Gish White (mid-season, white) and Roadrunner (late-season, red, which readily forms bypass shoots) florist azaleas were potted into 6- inch containers of amended peat moss:shavings (3:2, by volume) growth medium March 1987. Plants were placed in a double polyethylene greenhouse in a com- mercial nursery in Semmes, Alabama, and maintained according to common commercial practices. Plants were pruned in July and sprayed the following day with 3,627 p.p.m. dikegulac (Atrinal?) to increase lateral branch- ing. Six and 8 weeks later, plants were sprayed with 3,000 p.p.m. daminozide to suppress vegetative growth and hasten flower bud formation. Plants were held at 40*F minimum temperature in a polyethylene green- house until February 17, 1988, when they were trans- ferred to a double polyethylene greenhouse at Auburn University and maintained at the same minimum tem- perature until treatments were applied. TABLE 2. GROWTH REGULATOR TYPE, TIME OF APPLICATION, AND RATE EFFECTS ON BYPAss SHOOT AND FLOWER DEVELOPMENT OF DOROTHY GISH WHITE AZALEA Bypass shoots Days to Bloom TreatmentNumber ength full bloom 2 diameter 3 cm No. cm Paclobutrazol Applied 1 day before cooling: 50 p.p.m....................... 4.2 1.5 25.4 5.8 100 p.p.m.................... 5.2 1.4 27.0 6.0 150 p.p.m.................... 3.6 1.2 27.0 5.8 Applied 1 day after cooling: 50 p.p.m ......................... 5.7 2.7 24.7 5.9 100 p.p.m ....................... 6.0 2.6 5.4 5.8 150 p.p.m ........................ 5.4 1.8 28.8 5.9 Daminozide Applied 1 day before cooling: 3,000 p.p.m ................ 9.1 3.4 25.0 5.9 Applied 1 day after cooling: 3,000 p.p.m.................... 7.5 3.4 24.0 5.9 Control ................................................................. ... ..... 7.7 3.5 22.3 6.3 Significance of contrasts 4 Before vs. after ............................................................ ns ns ns Paclobutrazol vs. daminozide ........................................ ns Paclobutrazol vs. control ....................................... Dam inozide vs. control ................................................ ns ns Significance of paclobutrazol rate 5 ................................... . 1 1 1 c Time x rate interaction ................................................. ns ns ns ns 'Mean length of three longest bypass shots on each plant. 2 Days to full bloom beginning when plants were moved from cooler to greenhouse. 3Mean of three randomly selected blooms per plant. 4 ns = not significant;* = significant at 5 percent level. 5Control included in regression analysis; 1 = linear, c = cubic. [4] Treatments consisted of foliar sprays of two growth retardants applied either 1 day before cooling or 1 day after removing plants from the cooler. A single paclo- butrazol spray of 50, 100, or 150 p.p.m. or a single daminozide spray of 3,000 p.p.m. was applied. A non- treated control was included for comparison. Pre-cool- ing treatments were applied on February 23, 1988, and all plants were subsequently cooled in the dark at 38*F for 4 weeks. Plants were removed from the cooler on March 24 and placed under shade (47 percent light exclusion) in a double polyethylene greenhouse at 68*F minimum night temperature. On March 25, post-cooling treatments of the same rates of paclobutrazol and dam- inozide were applied. After 3 days, shade cloth was removed and plants were forced in full sun. Time until flowering was determined from the time plants were removed from the cooler until 75 percent of flowers were fully open. At this time, flower number and diameter (three randomly selected blooms per plant) and bypass shoot number and length (mean length of the three longest bypass shoots on each plant) were determined. Bypass shoot number and length decreased as paclo- butrazol rate increased for both cultivars, tables 2 and 3. Applying paclobutrazol after cooling was as effective in controlling bypass shoot development as applying paclobutrazol before cooling, with the exception of shorter bypass shoots on Dorothy Gish White treated before cooling. Plants treated with paclobutrazol devel- oped fewer bypass shoots than did daminozide-treated plants or control plants, which developed similar num- bers of bypass shoots. Paclobutrazol suppressed bypass shoot length of both cultivars compared to the control and of Dorothy Gish White, but not Roadrunner, com- pared to daminozide. Daminozide was more effective in suppressing bypass shoot length than the control on Roadrunner but not on Dorothy Gish White. Days to flower for both cultivars increased with increasing rates of paclobutrazol, from 22 days to 29 days for Dorothy Gish White and from 33 days to 41 days for Roadrunner. The time paclobutrazol was applied relative to cooling did not influence days to flower. Paclobutrazol delayed flowering compared to daminoz- ide and the control, and daminozide delayed flowering of Dorothy Gish White compared to the control. Bloom diameter either decreased (Dorothy Gish White) or was not affected (Roadrunner) by increasing rates of paclo- butrazol. The decrease with Dorothy Gish White was from 2.5 to 2.3 inches. Daminozide application also resulted in smaller blooms on Dorothy Gish White compared to the control. Neither time of application (before or after cooling) nor growth regulator (paclo- butrazol vs. daminozide) affected bloom diameter of either cultivar. Bypass shoot development is more extensive when there is excessive time from final pinch to forcing, particularly on certain cultivars forced in January or later for the florist trade. Paclobutrazol applications of 50, 100, or 150 p.p.m. suppress bypass shoot devel- opment, while minimally affecting bloom size. How- ever, flowering is likely to be delayed. Applications made just prior to or immediately after cooling were equally effective in controlling bypass shoot develop- ment. Daminozide is less effective than paclobutrazol in controlling bypass shoot development. Paclobutrazol may provide the grower of forcing azaleas an additional tool for controlling bypass shoots on late season cul- tivars. Keever is Associate Professor of Horticulture. TABLE 3. GROWTH REGULATOR TYPE, TIME OF APPLICATION, AND RATE EFFECTS ON BYPASS SHOOT AND FLOWER DEVELOPMENT OF ROADRUNNER AZALEA Treatment Bypass shoots Days to Bloom Number Length 1 full bloom 2 diameter 3 cm No. cm Paclobutrazol Applied 1 day before cooling: 50 p.p.m ...................... 11.4 5.6 35.7 5.8 100 p.p.m.................... 12.3 5.7 38.6 5.8 150 p.p.m.................... 12.5 5.8 37.5 5.9 Applied 1 day after cooling: 50 p.p.m ......................... 14.5 5.9 41.0 5.9 100 p.p.m........................ 10.3 5.9 37.6 5.8 150 p.p.m........................ 12.3 6.0 40.7 5.9 Daminozide Applied 1 day before cooling: 3,000 p.p.m ................. 15.3 5.4 32.4 5.9 Applied 1 day after cooling: 3,000 p.p.m .................... 18.7 5.3 33.0 5.7 Control ................................................................... .. ..... 17.3 6.9 32.5 5.8 Significance of contrasts 4 Before vs. after ............................................................ ns ns ns ns Paclobutrazol vs. daminozide ........................................ ns ' ns Paclobutrazol vs. control .............................................. " * " ns Dam inozide vs. control ................................................. ns ns ns Significance of paclobutrazol rate 5 . . . . .. . . . . . . .. .. ... ... . . .. .. q q q ns Time x rate interaction .................................................. ns ns ns ns 'Mean length of three longest bypass shoots on each plant. 2 Days to full bloom beginning when plants were moved from cooler to greenhouse. 3 Mean of three randomly selected blooms per plant. 4 ns = not significant; * = significant at 5 percent level. 5 Control included in regression analysis; q = quadratic; ns = not significant. [5] Postproduction and Marketing of Nandina domestica as a Potted Foliage Plant Bridget K. Behe, C. Fred Deneke, and Gary J. Keever PRODUCTION of foliage plants has stabilized in the past 10 years, indicating that the market has peaked in sales for many varieties currently on the market. New or improved products are often introduced to "restart" the life cycle for many products, generating more sales and more profits. For foliage plants, the introduction of new varieties may stimulate sales and profits for plant retailers. According to an industry expert, a high-volume supermarket floral department can expect to sell 6 to 12 four-inch foliage plants each week. Nandina domestica, or nandina, is a frequently planted, woody landscape shrub. The cultivars Harbour Dwarf and San Gabriel have shown promise as interior landscape plants and have potential to be sold through retail outlets, particularly supermarket floral depart- ments. The objectives of this research were to determine the postproduction performance and the market poten- tial for these two cultivars of nandina as potted foliage plants. To study postproduction performance, tissue-cul- tured plugs of Harbour Dwarf and San Gabriel nandina were potted on June 1, 1990, in 6-inch containers using a medium of 7 pine bark: 1 sand (by volume) amended per cubic yard with 5 pounds dolomitic lime, 1.5 pounds Micromax?, and 14 pounds of Osmocote? 18-6-12. Plants were grown in Mobile, Alabama, under either 30, 47, or 62 percent shade. On October 23, plants were moved to a simulated consumer environ- ment to evaluate postproduction performance. The post- production environment consisted of an insulated room with fluorescent lamps maintained at an average tem- perature of 70 0 F. Both cultivars of nandina performed well as interior foliage plants for at least 4 months in the simulated consumer environment. There were trends for increased plant width of Harbour Dwarf and increased height of San Gabriel with increasing shade. To determine the market potential and consumer perception of nandina, plugs of Harbour Dwarf and San Gabriel were planted in a medium as previously de- scribed. Plants were grown for 12 to 16 weeks in 4 2-inch azalea pots. A care tag was added to the pot to aid customers in the identification and care of the plant. A self-addressed, stamped postcard survey form was attached to the pot, to be completed and returned by the purchaser or gift recipient. Test plants were sold in two Birmingham, one Au- burn, and one Opelika, Alabama, and two Columbus, Ohio, supermarket floral departments. Ten plants of each variety were tested in separate 4-week studies at each location. Harbour Dwarf plants were evaluated from September 25 to October 23, 1990. San Gabriel plants were evaluated from October 23 to November 23, 1990. A total of 60 Harbour Dwarf and 60 San Gabriel plants was distributed to six stores in an 8- TABLE 4. PERCENTAGES OF SURVEY RESPONDENTS RATINGS Six FACTORS CONCERNING THEIR PURCHASES OF NANDINA DOMESTICA Factor Respondents' rating 1 1 2 3 4 5 Pct. Pct. Pct. Pct. Pct. Newness ........... 60 10 5 14 11 Problem free ........ 50 22 10 9 9 Use inside ............ 47 6 22 11 14 Price .................... 38 3 18 12 29 Smallness ........... 35 24 19 11 11 Form .............. 29 26 19 10 16 'Rated from most important (1) to least important (5), as an average for Harbour Dwarf and San Gabriel. week period. Plants were test marketed at a price of $2.99, comparable to the price of other foliage plants in the same size pot. Sales of all 4-inch foliage plants averaged 12.2 plants per store per week. Sales of the Nandina cultivars av- eraged 1.9 plants per store per week. Thus, the new nandina foliage plants captured 16 percent of the mar- ket for 4-inch foliage plants. Forty of 120 postcards were returned (33 percent response rate). Of those responding, 90 percent were female and 10 percent were male. Twenty-six percent had completed a high school education or less, while 44 percent had completed a 4-year college degree or more. The average years of education completed was 14.7. The median age was 49 years. No other demo- graphic information was collected. A majority of con- sumers, 78 percent, purchased only one nandina at the time of the purchase. Most of the plants purchased were for personal use, 90 percent, while 10 percent were purchased for use as gifts. Consumers were asked to rate the importance of six characteristics pertaining to the foliage plants: the new- ness of the plant, the ability to use the plant inside the home, its pest and disease resistance, the small size of the plant, price, and the form of the plant. Attributes were rated as most important by a different percentage of respondents: newness, 60 percent; problem free, 50 percent; use inside, 47 percent; price, 38 percent; small size, 35 percent; and form, 29 percent. Price was rated unimportant by the most respondents (29 percent), while the form of the plant (16 percent) and use inside (14 percent) were rated as unimportant by the most respondents. Thus, when prices of other foliage plants are equal, the newness of the plant was identified as an important factor to a majority of the purchasers. The consumer perception reported here, combined with the sales data, indicate that the N. domestica cultivars Harbour Dwarf and San Gabriel have good potential in the marketplace as interior foliage plants. Their post-production evaluation indicated that they should perform well in an interior environment, and the market information indicates that consumers ap- preciate them for their newness and are willing to purchase them at a price similar to other foliage plants of the same size. Behe is Assistant Professor, Deneke is Assistant Professor, and Keever is Associate Professor of Horticulture. [6] Evaluation of Growth Regulators on Lisianthus as Potted Plants and for Cut Flowers C. Fred Deneke and Gary J. Keever LISIANTHUS (Eustoma grandiflorum) is a herbaceous plant native to the Central United States that produces flowers over 3 inches in diameter in colors of blue, white, or pink. It is grown both for cut flowers, for which long stems are desirable, and for use as a flow- ering potted plant, for which compact plant growth is desirable. Therefore, different chemical growth regu- lators are needed to produce lisianthus for the two uses. The objectives of this research were to evaluate growth retardants for restricting excessive vegetative growth of lisianthus as a potted plant and gibberellic acid for promoting stem elongation for cut flower uses. Plugs of lisianthus Yodel Blue were potted in 6-inch azalea pots using Pro-mix BX? on March 14, 1990. Plants were pinched to three nodes on April 3. For an evaluation of growth retardants, plants were treated on April 17 with foliar sprays of uniconazole (Sumagic?) or B-Nine?, or a drench of A-Rest?. Additional plants received a foliar spray of gibberellic acid on April 19 to evaluate growth stimulation. Gibberellic acid was reapplied on April 26 and May 3. Plants were fertilized weekly with 300 p.p.m. N from Peters? 20-10-20. The highest rate of A-Rest produced the shortest flowering plants, followed by the highest rate of Su- magic, table 5. Lower rates of Sumagic, both rates of B-Nine, and the low rate of A-Rest produced plants 15 to 20 inches tall, which were similar to the heights of nontreated plants. Time to flowering was increased by 11 days for the highest rates of Sumagic and A-Rest. Flower diameter and number of flower buds were not influenced by any of the treatments. Lisianthus Yodel Blue can be grown as a flowering potted plant by using either a spray application of 25 p.p.m. Sumagic or a drench of 1.0 mg active ingredient (a.i.) A-Rest per pot. These treatments were judged to produce plants of appropriate height for a flowering potted plant. TABLE 5. EFFECTS OF THREE GROWTH RETARDANTS ON PLANT HEIGHT AT FLOWERING AND TIME TO FLOWERING FOR LISIANTHUS Plant Days to Treatment height flower In. No. 5 . .............................. 17.6 ? 2.0' 46 ? 3 10 pp m ............................ 15.2 ? 3.6 45 ? 2 15 p.p.m ............................ 14.8 ? 4.2 50 + 3 25 p.p.m .................... 12.8 ? 4.1 50 ? 2 2,500 p.p.m ....................... 17.7 ? 3.2 44 ? 3 5,000 p.p.m. ..................... 16.0 ? 2.9 44 ? 2 A-Rest 0.5 mg a.i./pot .................. 16.0 ? 3.4 44 ? 1 1.0 mg a.i./pot .................. 10.9 + 3.3 50 ? 3 Control ................ ............. 19.5 + 2.0 39 ? 2 'Mean ? standard deviation. TABLE 6. EFFECTS OF THREE GIBBERELLIC ACID APPLICATIONS ON PLANT HEIGHT AT FLOWERING AND TIME TO FLOWERING FOR LISIANTHUS Rate, Plant Days to p.p.m. height flower In. No. 0 .............................. 21.1 ? 2.51 42 ? 21 25 ........................... .. 23.5 ?2.3 37 ?2 50 ............................. 23.6 ? 2.5 36 ? 2 75 .......................... 24.4 ? 3.1 37 ? 3 100 .................... 24.9 ? 1.7 36 ? 2 125 .................... 25.6 + 2.1 41 + 2 150 .......................... 22.5 ? 2.2 36 ? 2 'Mean ? standard deviation. Three applications of 25 to 125 p.p.m. gibberellic acid increased plant height by only 2 to 4 inches, table 6. The highest rate of gibberellic acid resulted in plant heights similar to nontreated plants. Time to flowering decreased slightly with the gibberellic acid-treated plants. Flower diameter and number of flower buds were similar for all treatment rates. Gibberellic acid appears to have limited usefulness in increasing stem lengths of lisianthus Yodel Blue for use as cut flowers. Deneke is Assistant Professor and Keever is Associate Professor of Horticulture. Uniconazole Suppresses Bypass Shoot Development and Alters Flowering of Two Forcing Azalea Cultivars Gary J. Keever and William J. Foster GROWTH RETARDANTS are commonly applied to forc- ing azaleas to suppress internode elongation, promote flower bud initiation and development, and inhibit the growth of vegetative shoots that develop basipetally to flower buds (bypass shoots). Daminozide (B-Nine?) and chlormequat chloride (Cycocel?) are the principal growth retardants applied to forcing azaleas, both of which have undesirable side effects: delayed flowering, smaller flower size, and greater bypass shoot develop- ment for daminozide and delayed flowering and smaller plant size for chlormequat chloride. Triazole inhibitors, a relatively new group of plant bioregulants represented by paclobutrazol (Bonzi?) and uniconazole (Sumagic?), have growth retardant activity on a wide range of crop species. Paclobutrazol sprays of 150 and 200 p.p.m. controlled bypass shoot devel- opment, increased flower number and size, and de- creased forcing time of forcing azaleas when compared to daminozide. The objective of this study was to com- pare the effects of uniconazole, a less active triazole growth retardant, on bypass shoot development and flowering of forcing azaleas with daminozide and pa- clobutrazol. [71 Uniform liners of Redwings and Gloria azaleas were potted December 16, 1988, into 7 -inch azalea pots of amended 3 sphagnum peatmoss:2 softwood shavings (by volume) growth medium. Plants were placed in a double polyethylene greenhouse in a commercial azalea nursery in Semmes, Alabama, and maintained according to common commercial practices. Plants were sheared on March 22 and June 25, 1989; 1 day after each shearing, plants were sprayed with dikegulac to increase lateral branching. Plants were transferred to an outdoor shade structure July 20. The following treatments were applied on September 15: single uniconazole sprays of 0, 5, 10, 15, 25, 50, 75, 100, 150, and 200 p.p.m., single paclobutrazol sprays of 150 or 200 p.p.m., and a daminozide spray of 3,000 p.p.m. repeated 1 week later. Plants were cooled in the dark at 39*F for 6 weeks beginning November 1. Plants were removed from the cooler on December 13 and forced into flower in an unshaded double polyethylene greenhouse. Time until flowering was determined from the time plants were removed from the cooler until flowers were fully open. At that time, flower number and diameter (three ran- domly selected flowers per plant) and bypass shoot number and length (mean length of the three longest bypass shoots of each plant) were determined. Bypass shoot number and length of Gloria and Red- wings decreased with increasing concentrations of un- iconazole, tables 7 and 8. At concentrations above 50 p.p.m., essentially no bypass shoots developed on either cultivar. Plants treated with paclobutrazol sprays of 150 or 200 p.p.m. developed bypass shoot numbers and lengths similar to uniconazole sprays of 10 p.p.m. or less. Gloria sprayed with daminozide exhibited more bypass shoots than plants sprayed with uniconazole or paclobutrazol, or the control plants. Bypass shoot num- bers and lengths of Redwings treated with daminozide were similar to those of paclobutrazol-treated plants and plants sprayed with the lower concentrations of uniconazole. Days to flower for both cultivars increased with in- creasing concentrations of uniconazole. The delay in flowering was most pronounced when plants were sprayed with uniconazole concentrations of 50 p.p.m. or higher. Comparisons of days to flower among growth retardants varied with cultivar. Days to flower was sim- ilar when Redwings were sprayed with daminozide, paclobutrazol, or 25 p.p.m. or lower uniconazole. Days to flower of Gloria was similar for plants sprayed with daminozide or 150 p.p.m. or lower uniconazole, but greater than that of plants treated with 150 p.p.m paclobutrazol. Flower number increased (Gloria) or was minimally affected (Redwings) by lower concentrations of uni- conazole when compared to the control; flower numbers of plants treated with uniconazole decreased dramati- cally at concentrations higher than 75 p.p.m. Paclo- butrazol or daminozide sprays resulted in similar flower numbers to uniconazole sprays of 75 p.p.m. or lower, but more flowers than on plants treated with the three highest concentrations of uniconazole. Flower diameter increased slightly (Redwings) or was not affected (Gloria) by increasing concentrations of uniconazole. Flower diameters of plants treated with TABLE 7. PLANT GROWTH RETARDANT EFFECTS ON BYPASS SHOOT AND FLOWER DEVELOPMENT OF GLORIA AZALEA Bypass shoots Days to Flowers Treatment Number Length 1 open Number Diameter 3 flowers 2 cm No. cm Uniconazole 5 p.p.m ................................................... 2.1 3.0 41 359 6.7 10 p. .m . .................................................. 3.1 1.8 40 307 7.0 15 p.p.m ................................................... 2.0 1.8 42 323 6.9 25 p.p.m .................................................. .. 7 1.4 44 336 6.7 50 p.p.m ................................................... 1.3 1.7 46 333 7.2 75 p.p.m .................................................. .. 0 - 47 347 6.6 100 p.p.m ................................................ .. 0 - 49 283 7.2 150 p.p.m ................................................ .. 0 - 49 215 7.1 200 p.p.m ................................................ .. 0 - 55 142 7.0 Significance of rate 4 .. . .. .......... . . . . . . . . . .. .. ... . .. ... . q q c c ns Paclobutrazol 150 p.p.m ................................................ 1.0 3.2 39 332 6.8 200 p.p.m ............................................... 1.9 2.8 41 338 7.1 Daminozide 3,000 p.p.m. ............................................ 7.4 3.6 45 312 7.1 Control ..................................... 2.0 4.9 37 315 6.8 LSD ............................................... ......... 3.5 1.6 4.6 59 .6 'Mean length of three longest bypass shoots on each plant. 2 Days to full bloom beginning when plants removed from cooler to greenhouse. 3Mean of three randomly selected blooms per plant. 4 Significance of regression analysis at P= 0.05: q= quadratic, c=cubic, ns = nonsignificant. SMean separation within columns by a protected Fisher's least significance test, P= .05; LSD used for comparisons among growth retardants. 6 F value not significant. [8] TABLE 8. PLANT GROWTH RETARDANT EFFECTS ON BYPASS SHOOT AND FLOWER DEVELOPMENT OF REDWINGS AZALEA Bypass shoots Days to Flowers Treatment Number Length' open Number Diameter 3 flowers 2 cm No. cm Uniconazole 5 p.p.m. ...................................... 3.0 2.5 44 274 7.3 10 p.p.m. ................................................ 2.8 2.8 45 275 7.4 15 p.p.m. ..................................... 2.5 1.7 46 268 7.2 25 p.p.m. ............................................... 2.1 1.6 48 268 7.1 50 p.p.m. ...................................... 1.8 1.1 53 273 7.2 75 p.p.m. ...................................... .5 1.0 56 249 7.5 100 p.p.m. ................................................. 0 - 60 215 7.5 150 p.p.m. ................................................. 0 - 60 158 7.1 200 p.p.m. ................................................. 0 - 63 148 7.3 Significance of rate 4 . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. c q c c Paclobutrazol 150 p.p.m. ...................................... 3.6 2.8 44 311 7.3 200 p.p.m ............................................... 4.5 2.7 46 260 6.9 Daminozide 3,000 p.p.m .......................... ................ . 4.4 3.1 46 260 7.3 Control ....................................................... 5.5 3.5 46 267 6.9 LSD 5 ................................................... .......... 3.4 1.1 3.9 46 0.4 'Mean length of three longest bypass shoots on each plant. 2 Days to full bloom beginning when plants removed from cooler to greenhouse. 3Mean of three randomly selected blooms per plant. 4 Significance of regression analysis at P=0.05: c=cubic, q-quadratic, 1= linear. nMean separation within columns by a protected Fisher's least significance test, P= .05; LSD used for comparisons among growth retardants. 6 F value not significant. paclobutrazol or daminozide were similar to each other and to those of uniconazole-treated plants. Reductions in bypass shoot number and length of Redwings occurred with uniconazole concentrations as low as 5 p.p.m.; concentrations of 25 p.p.m. and 10 p.p.m. were required to achieve a similar reduction in bypass shoot number and length, respectively, of Gloria. Flowering was minimally delayed with the lower con- centrations of uniconazole, but was delayed a week or longer with concentrations of 50 p.p.m. or higher. Flowering, as indicated by flower number and diameter, was enhanced with lower concentrations of unicona- zole, but flower number was greatly reduced with con- centrations above 75 p.p.m. Uniconazole sprays were much more effective in controlling bypass shoots than paclobutrazol sprays of similar concentrations or than two daminozide sprays of 3,000 p.p.m. Daminozide was the least effective retardant in suppressing bypass shoot development of Gloria, but similar in effect to paclobutrazol and the lower concentrations of uniconazole when applied to Redwings. Alice du Pont Mandevilla Responds to Sumagic Gary J. Keever and C. Fred Deneke THE GENUS Mandevilla consists of more than 100 species of tropical and subtropical twining vines and shrubs. Alice du Pont, the most widely available cultivar of Mandevilla, is grown for greenhouse use or as a horticultural annual in temperate areas, photo 1 (page 18). The cultivar blooms over a long season and is useful for arbors, trellises, or other supports about which the stems can twine. The vigorousness of Alice du Pont creates production problems for the grower since consumers are interested in manageable plants in flower. To produce flowering plants, growers frequently contend with excess vege- tative growth that twines around other plants and struc- tures. Sumagic?, an experimental growth regulator, has effectively suppressed growth of bedding and pot plants. This research investigated the effects of single or mul- tiple applications of Sumagic on vegetative growth and flowering of Alice du Pont Mandevilla. In two experiments conducted in 1989, rooted cut- tings of Alice du Pont were potted into # 1 containers of amended 7 pine bark: 1 sand. Plants were pruned to two nodes before applying Sumagic and were fertilized weekly with 300 p.p.m. N from 20-10-20. Plant height [9] Keever is Associate Professor of Horticulture and Foster is former Superintendent of the Ornamental Horticulture Substation. FIG. 1. Height of Alice du Pont mandevilla in response to multiple Sumagic applications reapplied as needed. was measured weekly until plants were in full flower, when flower diameter and days to flower were deter- mined. In preliminary work, a single application of 30 p.p.m. or higher Sumagic retarded growth excessively for at least 6 weeks, after which plants began to exhibit normal growth. All rates of Sumagic induced leaf cup- ping, delayed flowering, and reduced bloom size. In the first experiment, one or two applications of 0, 5, 10, 15, or 20 p.p.m. Sumagic were applied. The second application was applied to all treatments (rates) when plants treated with 10 or 15 p.p.m. began to regrow normally. In the second experiment, applications of Sumagic from 0 to 20 p.p.m. in 2.5-p.p.m. increments were repeated as necessary when plants within a treat- ment (rate) resumed a normal growth pattern. In the first experiment, single applications of 5, 10, 15, or 20 p.p.m. Sumagic did not provide acceptable control of internode elongation. With two applications of Sumagic, 5 p.p.m. was inadequate, 10 and 15 p.p.m. were acceptable, and 20 p.p.m. was excessive in con- trolling shoot elongation. In the second experiment, multiple applications of all tested rates of Sumagic effectively suppressed elongation, figure 1. As the con- centration of Sumagic increased, the interval between applications increased from 28.5 days with 2.5 p.p.m. to 39.5 days with 20 p.p.m. Multiple applications of Sumagic from 2.5 to 20 p.p.m. reapplied when shoots begin to elongate is an effective means of controlling excessive vegetative growth of Mandevilla and may provide growers with an additional management tool in the production of this flowering horticultural annual. Keever is Associate Professor and Deneke is Assistant Professor of Horticulture. Growth Inhibition of Dianthus Species James T. Foley and Gary J. Keever WHEN USED as bedding plants or pot crops, carnation, Dianthus caryophyllus, and garden pinks, D. chinensis, are treated as annuals, blooming the first year from seed. Annual carnation produces large double blooms tradi- tionally used as cut flowers, while garden pinks bloom in loose clusters of single flowers and are more com- monly used as bedding plants; both species provide a showy display of blooms in shades of red, pink, or white. When these plants are offered as pot crops, they should be proportional in height for 4-inch pots. Growth retardants or pruning may be needed to meet size spec- ifications. Paclobutrazol (Bonzi?) suppresses stem elongation of many plant species including chrysanthemum, ge- ranium, and a wide variety of other floriculture crops. Chlormequat chloride (Cycocel?), labeled for use on poinsettia, effectively inhibited shoot elongation of D. chinensis and D. barbatus cultivars and Reiger begonia when applied at 3,000 p.p.m. and ivy geranium at 1,500 p.p.m. Ancymidol (A-Rest?) has effectively retarded the growth of English ivy with the application of 0.125- mg drenches and chrysanthemum with foliar sprays of 62 p.p.m. Seeds of Knight Hybrid Scarlet annual carnation and Queen of Hearts garden pinks were sown in 36-cell TABLE 9. ANNUAL CARNATION PLANT HEIGHTS IN RESPONSE TO FOUR PLANT GROWTH REGULATORS Plant height, by weeks after treatment applied 2 4 7 10 14 cm cm cm cm cm Paclobutrazol 100 p.p.m ................................................ . . 7.3 10.2 13.9 16.3 17.0 200 p.p.m ................................................ . 6.311 8.81 11.6 14.0q 14.4q 100 p.p.m. + pruning ..... ................ 5.4 7.8 12.9 17.6 18.7 Chlormequat chloride 3,500 p.p.m . ................. . ........................... 8.1 10.6 17.3 21.9 24.7 Ancymidol 100 p.p.m . ................................................ 9.8 15.5 28.3 31.1 33.4 Pruned ........................................................ 6.6 10.8 20.3 30.3 33.1 Control ......................................................... 8.7 14.3 26.1 31.1 31.6 LSD 2 .................... ........................................ 1.4 1.9 2.4 3.6 3.7 Significance of regression analysis at P= 0.05, l= linear; q=quadratic; control, but not plant growth regulator + pruning treatment, included in regression. 2 Mean separation within columns by a Fisher's least significance test, P=0.05; LSD used for comparisons among growth regulators. [101 flats of Pro-Mix BX? and placed in a double polyethylene greenhouse under intermittent mist (10 seconds/5 min- utes) until cotyledons had fully emerged. Minimum day/night temperatures in the greenhouse were 70*F/ 60?F. Seedlings of both Dianthus species were trans- planted to 4-inch pots on January 29, 1988, 17 days prior to treatment with foliar sprays. Treatments were applied just prior to runoff to annual carnation and garden pinks when plants were about 3.1 inches and 4.7 inches tall, respectively. Treatments consisted of paclobutrazol at 100 or 200 p.p.m.; an- cymidol at 100 p.p.m. (applied to carnations only); two applications of chlormequat chloride at 3,500 p.p.m. applied 3 weeks apart; and an untreated control. The surfactant Buffer-X? was added at 0.2 percent to spray solutions. There were also two treatments which included pruning of plants above the fourth node to a height of about 1.6 inches. These treatments consisted of pruning alone or pruning with an application of paclobutrazol at 100 p.p.m. on annual carnation or 200 p.p.m. on garden pinks. Plants were fertilized weekly with 200 p.p.m. N from 20-10-20 Peters Peatlite Spe- cial.? Height from the growth medium surface to the veg- etative shoot apex of annual carnation was determined 2, 4, 7, 10, and 14 weeks after treating; heights of garden pinks were measured 2, 4, and 8 weeks after treating. Terminal data consisted of numbers of axillary shoots longer than 0.4 inch, number of flowers and flower buds, and shoot dry weight. Days to first fully open flower from date of seeding were also taken as plants flowered. Beginning at week 4 and continuing through the study (week 14), the height of annual carnation treated with paclobutrazol decreased as rate increased, table 9 and photo 2 (page 18). Fewer axillary shoots developed on paclobutrazol-treated plants than on control plants, table 10. Neither days to flower, bloom number, nor flower bud number was affected by paclobutrazol ap- plication. Shoot dry weight decreased at the higher paclobutrazol rate. Plant height was suppressed with chlormequat chloride application beginning at week 4 and continued throughout the study; however, paclo- butrazol was more effective in suppressing growth dur- ing the same time period. There were no other differences between growth regulator treatments and the control. The pruned treatment grew to a height equal to that of the control by week 10 and produced a greater number of axillary shoots; however, flowering was de- layed about 11 days. Numbers of blooms, flower buds, or shoot dry weight did not differ from the control when plants were pruned. When pruning was combined with 100 p.p.m. of paclobutrazol, plant height was suppressed compared to the pruned treatment begin- ning at week 4 and continuing throughout the study. Axillary shoot numbers and dry weight decreased when pruning was combined with paclobutrazol compared to the pruned treatment; flower number was decreased while bud number and days to flower did not differ from the pruned treatment. Plants treated with ancy- midol did not differ from the control in any of the observations measured. At the higher paclobutrazol rates, plant height of garden pinks was suppressed beginning 2 weeks after spray application until the study ended at week 8, table 11 and photo 3 (page 18). Secondary and tertiary axillary shoots and shoot dry weight were suppressed with paclobutrazol application; primary axillary shoots were suppressed at the 200 p.p.m. rate. Total number of flowers and days to flower did not differ from the control when plants were treated with paclobutrazol. When pruning was combined with paclobutrazol at 200 p.p.m., plant height was suppressed compared to the pruned treatment beginning at week 4; however, plant height did not differ from the unpruned 200 p.p.m. paclobutrazol treatment. Fewer secondary and tertiary axillary shoots developed on plants pruned and treated with 200 p.p.m. paclobutrazol as compared to pruned plants. Flower number and shoot dry weight of plants pruned and treated with paclobutrazol were less than with the pruned treatment, while days to flower did not differ. TABLE 10. ANNUAL CARNATION AXILLARY SHOOT, FLOWER, AND FLOWER BUD NUMBER AND DRY WEIGHT IN RESPONSE TO THREE PLANT GROWTH REGULATORS, 14 WEEKS AFTER TREATMENT Treatment Axillary Flowers F er Shoot Days toshoots Flowersbuds dry flower weight fwe No. No. No. Grams No. Paclobutrazol 100 p.p.m ............................................... 13.0 9.7 13.7 9.0 143.4 200 p.p.m . ................................................ 14.0q 2 9.3ns 16.9ns 7.9q 145.3ns 100 p.p.m. + pruning .... ................. 18.0 5.3 19.4 8.8 156.4 Chlormequat chloride 3,500 p.p.m.............................................. 13.1 13.1 12.0 10.5 140.1 Ancymidol 100 p.p.m ................................................. 15.1 16.4 12.6 13.5 139.0 Pruned ........................................................ 23.1 12.4 15.1 14.2 153.0 Control ........................................................ 16.3 13.4 11.4 14.0 141.6 LSD 3 .................... ........................................ 2.2 4.3 5.6 1.6 6.4 'Days to flower = number of days from date of seeding to anthesis of first flower. 2 Significance of regression analysis at P=0.05: q=quadratic; ns=nonsignificant; control, but not plant growth regulator + pruning treatment, included in regression. 3 Mean separation within columns by a Fisher's least significance test, P=0.05; LSD used for comparisons among growth regulators. [11] TABLE 11. PLANT HEIGHTS OF GARDEN PINKS IN RESPONSE TO TWO PLANT GROWTH REGULATORS, PRUNING, OR COMBINATIONS Plant height, by weeks after Treatment treatment 2 4 8 cm cm cm Paclobutrazol (1 application) 100 p.p.m. ................................... 6.7 10.6 19.0 200 p.p.m. ................................... 6.2q' 7.7q 13.0q 200 p.p.m. + pruning .............. 4.6 7.7 12.8 Chlormequat chloride (2 applications) 3,500 p.p.m. ................................ 9.1 23.7 32.0 Pruned ..................................... 6.3 14.3 31.5 Control ..................................... 12.3 26.2 34.6. LSD 2 ............. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.8 3.3 3.4 'Significance of regression analysis at P= 0.05: q= quadratic; control, but not plant growth regulator + pruning treatment, included in regression analysis. 2 Mean separation within columns by a Fisher's least significance test, P=0.05; LSD used for comparisons among growth regulators. Chlormequat chloride was ineffective in height con- trol. Shoot dry weight and number of tertiary axillary shoots decreased with the application of chlormequat chloride when compared to the control; no other ob- servations were significantly different. When plants were pruned, heights were less than the control through week 4, while shoot dry weight, flower number, and primary and tertiary axillary shoots were less compared to the control. Days to flower was delayed about 12 days as a result of pruning. Growth regulator application did not induce any symptoms of phytotoxicity, such as chlorosis, on either species. Both species displayed darker green foliage as a result of paclobutrazol application, which may add to plant quality. Experimentation on each species was terminated after all plants had bloomed, 8 weeks after treating of garden pinks and 14 weeks after treating of annual carnation; effects of paclobutrazol application persisted until experimentation of both species had been terminated. Two applications of chlormequat TABLE 12. FLOWER NUMBER, DAYS TO FLOWER, AND SHOOT DRY WEIGHT OF GARDEN PINKS IN RESPONSE TO Two PLANT GROWTH REGULATORS, PRUNING, OR COMBINATIONS Days Shoot Treatment Flowers to dry flower weight No. No. Grams Paclobutrazol (1 treatment) 100 p.p.m.. ..................... 46.2 97 5.6 200 p.p.m.. ..................... 44.712 98ns 4.9q 200 p.p.m. + pruning ................. 12.6 110 3.7 Chlormequat chloride (2 applications) 3,500 p.p.m ................................. 45.5 99 7.9 Pruned ............................................ 27.9 108 8.3 Control ........................... 52.5 96 10.4 'Days to flower = number of days from date of seeding to opening of first flower. 2 Significance of regression analysis at P=0.05: l=linear; q--quadratic; ns= nonsignificant; control, but not plant growth regulator + pruning treatment, included in regression analysis. 3 Mean separation within columns by a Fisher's least significance test, P=0.05; LSD used for comparisons among growth regulators. chloride inhibited shoot elongation of annual carnation, though not as effectively as paclobutrazol. Garden pinks may require at least three applications of chlormequat chloride at 2-week intervals to suppress shoot elon- gation; two applications at 3-week intervals were inef- fective. Foley is Graduate Student and Keever is Associate Professor of Horticulture. Consumer Preferences for Annual Bedding Plant Containers Bridget K. Behe, Ginger Purvis, and Charles H. Gilliam PACKAGING of many products, which includes the container itself, influences consumer purchases. For ornamental horticultural products, this influence has not yet been defined. Some garden center managers reported that customers want larger, more mature plants and are willing to pay the extra costs involved. Con- sumers like these larger bedding plants because they provide "instant color" to their home landscapes. The objective of this study was to determine consumer pref- erences for pansy plants in three types of containers: 6-plant cell packs (black), 6-inch plastic pots (black), and a plastic grow bag (white), photo 4 (page 18). Uniform pansy plants (Viola X wittrockiana) were grown by a commercial producer using standard cul- tural practices. These plants were then containerized in each type pot or bag and placed in a greenhouse for 2 or 4 weeks until used in the study. Two 4-packs were used for the cell packs, eight plants were used in the bag, and two 6-inch pots (three plants each) were used. Seventy-four consumers participated in a personal interview and written questionnaire study at two garden center locations on two Saturdays in Montgomery, Ala- bama, in October and November, 1989. Consumers were evaluated for characteristics of age, education, household size, gender, and income. They were also asked their perceptions of certain plant and container characteristics. Finally, they were asked to select one group of pansy plants they would most prefer if making a purchase that day. No prices were indicated on the different containers. Respondents by age group were 25-34 (17 percent), 35-49 (36 percent), and 50-79 (47 percent). Seventy- five percent of the respondents had completed some college or had earned a college degree. Median per capita income was $10,416, while average per capita income was $9,767. Fifty-one percent of the respond- ents stated that their households contained only two people and the mean number of persons in each house- hold was 2.4. Thirty-seven percent of the respondents were male and 63 percent were female. Respondents were asked to rank the importance of several plant and package characteristics. Consumer ratings were evaluated using a 9-point Likert Scale, table 13. Ninety percent of the respondents indicated that [121 TABLE 13. CONSUMER PERCEPTION RATING OF EIGHT PLANT CHARACTERISTICS Rated most Rated Rated least Characteristic important neutral important 1-4 5 6-9 Pct. Pct. Pct. Health and appearance of plant .......... 90 1 8 Shape or form of the plant .............. 86 6 9 Color of the flowers on the plant ........ 80 9 10 Color of the leaves on the plant .......... 78 6 16 Size of the plant .............................. 53 25 21 Price of the plant ................................ 49 26 26 Care and planting instructions ......... 49 12 42 Type of container plant comes in ....... 18 7 75 the health of the plant was an important attribute. Conversely, the type of container in which the plant was sold was given an important rating by only 18 percent of the respondents. Consumers were asked to indicate their attitudes or perceptions about buying plants on a 5-point Likert scale. Only 14 percent of the respondents strongly agreed or agreed with the statement, "I usually grow plants in the container I buy them in." Eighty-nine percent agreed or strongly agreed that they could tell the difference between a healthy and an unhealthy plant. Respondents were then asked to choose which pansy on display they would purchase if making a purchase for themselves that day. Of those responding, 51 percent chose the plants in the bag, 28 percent chose the plants in the cell packs, and 19 percent chose the plants in the 6-inch pots. Price did not appear to be a concern to these con- sumers, yet purchasing a high quality plant was. Health is one measure of plant quality, and it is one charac- teristic respondents believed they were able to recog- nize. Plants in the grow-bag may have appeared to grow larger than plants in the cell pack or pots. Consumers may have equated larger sized or more colorful plants with higher quality plants. The plants may have grown larger because the grow bags held more soil, retained more moisture, or provided some nutrition. Results of this study offer helpful information for marketing bedding plants. Larger plants or those with a more colorful display are preferred by consumers. Optimum container size for greenhouse production may not be the optimum size container in which to market plants. Annual plants could be transplanted into larger containers prior to retail sales, which could result in more visual appeal. While many plants would still be purchased in traditional cell pack containers, suggesting uses to the consumer may stimulate additional sales. Displaying mature plants in large containers, or a variety of containers, may suggest additional uses. The container does not appear to play a significant role in the consumer purchase decision directly. How- ever, it may influence the consumers' perception of quality, most likely by influencing the growth and de- velopment of the plant itself. Therefore, the container selection should be given further consideration by pro- ducers and retailers of annual bedding plants. Behe is Assistant Professor, Purvis is Graduate Student, and Gilliam is Professor of Horticulture. Hardy Chrysanthemum Trial Bridget K. Behe, C. Fred Deneke, Art McDow, and Dan Land HARDY CHRYSANTHEMUMS, ones which should over- winter outdoors from year to year, are popular flow- ering plants for fall gardens in Alabama. Hardy chry- santhemums are available in a wide range of colors and forms, some of which bloom earlier or longer than others. To evaluate new varieties, a study was initiated to determine the date the first flower opened and the date of peak flowering for new chrysanthe- mum cultivars. Uniform unrooted cuttings of 19 cultivars of chry- santhemums were treated with a basal dip of Hor- modin #2, and directly stuck into trade gallon containers on June 28, 1990. The medium contained 4:1:1 (by volume) of composted bark, sand, and peat moss. Incorporated into the medium were 1.5 pounds dolomitic limestone, 10 ounces superphosphate, 10 ounces gypsum, and 2 ounces Micromax? minor ele- ments per cubic yard. Cuttings were placed under mist for 3 days. Fifty plants of each cultivar were planted in ground beds on August 26 and visually evaluated weekly until October 31. Plants were fer- tilized with a liquid feed of 200 p.p.m. nitrogen using a 20-10-20 formulation on July 9, August 31, and September 28. Results of the trial are shown in table 14. The color of the petals is shown followed by the form of the flower. Daisy-form flowers have two colors indicated: the color of the ray flowers (petals) followed by the color of the disc flowers (center). The flowering mechanism in chrysanthemums is regulated by the photoperiod; thus, date when the first flower opened is important in determining when the floral display will begin. Date of first flowering was recorded as the date of peak flowering, when a majority of blooms TABLE 14. EVALUATION OF 19 HARDY CHRYSANTHEMUM VARIETIES ON COLOR, FORM, DATE OF FIRST FLOWER, AND DATE OF PEAK FLOWER Variety Color 1 Form 2 Flowering First Peak Allure .................... Y/Y Daisy 9/19 9/30 Bravo .................... Red Cushion 9/19 10/12 Dark Grenadine ...... Red Cushion 9/19 10/12 Debonair ............... Pink Cushion 9/3 10/7 Donna ................... Y/Y Spoon/Daisy 9/8 9/30 Grace .............. B/Y Daisy 8/31 10/11 Illusion ............... W/Y Quilled/Daisy 9/15 9/30 Jessica ................. Y Cushion 9/7 9/23 Legend ............. Y Pompon 9/18 10/12 Naomi ................. P&W/Y Daisy 9/27 10/12 Red Remarkable ..... R Cushion 10/1 10/14 Sandy .................. B/B Daisy 9/28 10/12 Sarah ............... B Cushion 10/4 10/21 Sundoro .............. P Cushion 10/1 10/21 Volunteer ............ W Cushion 9/16 9/19 'Flower colors: Y=yellow, R=red, P=pink or lavender, W=white, and B=bronze. 2 Forms were daisy, cushion, and pompon. [13] on each plant were fully open. Notes were made on bloom fading and susceptibility to Botrytis blight (grey mold) on the blooms. The white varieties included Hekla, Illusion, To- lima, and Volunteer. Of these, Hekla was the first to flower, on August 31, a desirable characteristic in garden performance, and also the first to reach peak flowering, on September 30. This variety did, how- ever, grow taller than the other white varieties eval- uated. The bloom color of Hekla discolored early, turning purple soon after opening. Volunteer ap- peared to be highly susceptible to Botrytis blight in this trial. The red varieties included Bravo, Dark Grenadine, Minnigopher, and Red Remarkable. All of the red cultivars except Red Remarkable showed their first flower on September 19. Red Remarkable showed first flower on October 1, nearly 2 weeks later. Minni- gopher reached peak flowering before the other three red varieties on October 7. Bravo retained its petals quite well in the evaluations. The yellow varieties included Allure, Donna, Jes- sica, Legend, and Target. Jessica was the first of the yellow varieties to flower, on September 7, and one of the first of all variety colors to flower. Jessica was also the first to reach peak flowering, while Legend and Target were the last yellow varieties to reach peak flowering, on October 12. Donna appeared to be more susceptible to Botrytis grey mold on the blooms, and the blooms faded quickly after opening. Three pink varieties were evaluated: Debonair, Na- omi, and Sundoro. Debonair was the first pink to flower, on September 3. It appeared to be one of the top performers in the evaluation as it faded slowly and appeared to be resistant to Botrytis blight. Naomi had an interesting color pattern, bright pink ray petals with white near the bottom and yellow center disc flowers. Naomi reached first flower on September 27 and peaked on October 12. Naomi also seemed less resistant to Botrytis blight, and the petals appeared to fade more quickly than other colors. The bronze varieties included Grace, Sandy, and Sarah. Grace flowered first of the bronze varieties, on August 31, while the last bronze variety to flower was Sarah on October 4. Grace, photo 5 (page 18), per- formed well early in the study and kept a high rating into the final week of the study. Overall, Grace performed the best of all 19 vari- eties. It flowered very early, only 1 week after being planted in the ground beds, the blooms faded little, and it appeared to be resistant to Botrytis blight. Other top performing varieties included Hekla, a white petal and yellow center daisy form, Bravo, a red cushion, and Legend, a yellow pompon form chrysanthemum. Behe is Assistant Professor, Deneke is Assistant Professor, McDow is undergraduate student, and Land is Greenhouse Supervisor of Horticulture. Profile of the Perennial Plant Industry Bridget K. Behe and Lisa Beckett ONE OF THE FASTEST growing categories for sales of horticultural products recently has been herbaceous perennials. Businesses which market perennials often handle a tremendous variety of species from which their clients may choose. Little information is available to current industry members and to firms considering entering this lucrative market. What little information is available on the size of the perennial industry is found in the U.S. Agricultural Census of Horticultural Specialties publication. It contains information on only the wholesale value of a wide variety of plants, in- cluding annuals and herbaceous perennials. The census information is further limited by its publication once every 10 years. The annual USDA Floriculture Crops Summary shows the wholesale value of selected florist crops and bedding plants in 28 states, but excludes perennials. Thus, reliable information on important characteristics of the perennial plant industry is lacking. In an effort to provide perennial plant growers and marketers with additional information, research by the Alabama Agricultural Experiment Station, in coopera- tion with the Perennial Plant Association (PPA), inves- tigated characteristics of the perennial plant industry in the United States. PPA members who grew perennials in 1989 were used as the sample for the study. In July, 1990,439 surveys were mailed to perennial businesses across the country. Two surveys were mailed to each business to encourage a response. A total of 146 surveys was returned, for a 33 percent return rate. Characteristics of Perennial Plant Businesses Many members of the industry think of perennial plants as newcomers, and that perhaps the businesses which sell these plants are newcomers as well. The perennial plant firms responding to this study were not newcomers, however, having been in business from 1 to 75 years, or an average of 16 years. The legal form of businesses can add another facet of information to the perennial plant industry. Forty- seven percent of the businesses were sole proprietor- ships, 12 percent were partnerships, 39 percent were corporations, and 2 percent had adopted some other business form. This meant that almost half of the firms were owned and operated by one person. Over half of the businesses (57 percent) had two or fewer full-time year-round employees and 76 percent had one or more seasonal employees. The total sales for each company in 1989 was re- quested. This total included wholesale and retail sales of items which may have included more products than strictly perennial plants (many businesses sold annuals, woody plants, hard goods, ilorist crops, and fertilizers). Thirty-four percent of the businesses had total revenues of $50,000 or less in 1989, while 14 percent had revenues of $1 million or more. These results showed [14] that many of the firms selling perennial plants are relatively small. Market Area Respondents to the survey came from 31 states. Many indicated that over half of the plants they marketed went to buyers in the state where their business was located. In fact, 78 percent of the plants were sold in the state where the business was located; 21 percent were sold out of the state and 1 percent were sold out of the country. Responses to a question about method of sale re- vealed that telephone sales accounted for an average of 25 percent of revenues, mail sales 14 percent, and walk-in or drive-in sales 42 percent of total sales. Methods of Propagation Results indicated that, on average, 30 percent of plants propagated by the businesses are propagated from seed, 14 percent are propagated from plugs, 27 percent are from division of plants or root cuttings, and 29 percent are propagated by other vegetative methods, including stem cuttings and tissue culture. These results show that over half of the perennial plants produced in the United States are propagated vegetatively, from cuttings or division. This means that stock plants must be maintained, or a source of cuttings located. Vege- tative propagation is also more expensive than seed propagation, increasing the costs of production for per- ennial plants. Implications The size of the perennial industry is estimated to be between $66 and $150 million, although the 1990 Census of Horticultural Specialties has not yet been published. The information examined thus far shows that perennials are a sizeable component of the horti- cultural industry. Perennial plant growers and marketers are most likely sole proprietors who operate businesses which sell more than one type of plant (i.e., perennials and annuals, woody plants, or florist crops). Most per- ennials are propagated vegetatively, adding consider- ably to the cost of production for these plants. These relatively small businesses utilize a number of methods to sell their plants, but the majority are sold when the customer visits the operation. Perennial growers are not large businesses with the financial resources to obtain sophisticated marketing studies; rather, they rely on government or other re- searchers to compile information for them. The few studies published thus far support the widely held notions that perennials have earned a substantial portion of the horticulture industry and have grown markedly in the past 10 years. The results of this study should provide those business managers with useful informa- tion to help them formulate marketing strategies for the next decade. Behe is Assistant Professor and Beckett is Graduate Student of Horticulture. Influences of Subirrigation on Postproduction Longevity of Poinsettias C. Fred Deneke, Bridget K. Behe, and John Olive INCREASING CONCERN for Alabama's water quality has focused much attention on businesses that use ni- trate fertilizers and pesticides. Ebb and flow production technology is a closed subirrigation system which pre- vents water used in irrigating greenhouse crops from being routinely discharged into the environment. This technology, developed in the early 1950's, has been only recently adopted in Europe but only infrequently adopted by American producers. In an ebb and flow system, plants are grown in a trough. Water and fertilizer from an enclosed tank are pumped into the trough where the plants are watered from below. Water that is not absorbed by the media after a few minutes drains back into the tank for recirculation. One concern expressed by many growers and re- searchers is that subirrigation may reduce the post- production longevity of plants. The objective of this study was to compare the postproduction longevity of poinsettias grown with traditional irrigation and subir- rigation. On September 12, 1990, rooted cuttings of Pink Peppermint, Supjibi, V-10 Amy, and V-14 Glory were stuck in 6-inch azalea pots using a 3:1 milled pine bark:peat moss medium amended per cubic yard with 6 pounds dolomitic lime, 2 pounds gypsum, and 1.5 pounds Micromax?. Plants were pinched to four nodes on September 25. A drench of Dexon?, Benlate?, or Truban? was applied four times. Plants were watered with either drip irrigation or subirrigation. Plants on drip irrigation were fertilized about twice per week beginning September 19 with 300 p.p.m. N from 20-10-20. This fertilization rate was gradually reduced to 200 p.p.m. by November 20. Similarly, the conductivity of the irrigation water of subirrigated plants was monitored so that fertilization rates were gradually reduced from 300 to 200 p.p.m. TABLE 15. EFFECTS OF IRRIGATION METHOD ON POSTPRODUCTION LONGEVITY OF FOUR POINSETTIA CULTIVARS 6 WEEKS AFTER PLACEMENT IN A SIMULATED CONSUMER ENVIRONMENT Cultivar Irrigation Plant Leaf Bract Flower method grade 1 drop drop drop Pct. Pct. Pct. V-10 Amy drip 2.0a 2 19a 24b 77a sub- 1.0b 4b 55a 76a V-14 Glory drip 2.6a 6a 4a 52b sub- 2.0b 3a 6a 59a Pink drip 2.8a 7b 3b 86a Peppermint sub- 1.8b 24a 10a 91a Subjibi drip 2.0a 20a 14b 72a sub- 1.0b 14a 50a 65a 'Rated from 5 (excellent, no discoloration and no bract or leaf drop) to 1 (poor, faded bracts and leaves, or leaf and bract drop). 2 Means followed by the same letter within cultivars are not significantly different (Duncan's multiple range test, 5 percent level). [15] N. Plants in both irrigation methods were irrigated with water only after November 20. On November 29, plants were moved to a simulated consumer environment to evaluate postproduction quality. After 3 and 6 weeks in the postproduction environment, data were taken on plant grade and leaf, bract, and flower (cyathium) drop. There were no differences in plant grade among cultivars and treatments when plants were placed in the postproduction environment. After 3 weeks, plant grade significantly declined and leaf drop significantly increased only for subirrigated V-10 Amy as compared to the drip irrigation treatment. After 6 weeks, plant grade for all cultivars and treatments had declined; this decline was greater in subirrigated plants, table 15. Similarly, the numbers of leaves and bracts dropped were greater in subirrigated plants. Flower abscission of subirrigated V-14 Glory was greater than with drip irrigated plants. Regardless of the irrigation method, V-14 Glory and Pink Peppermint had the highest plant grades. Also, these two cultivars retained their bract coloration. V-10 Amy dropped bracts sooner than the other cultivars; other researchers have noted that this cultivar is subject to severe leaf and flower abscission. Supjibi, because of its large brightly colored bracts, was the most striking cultivar when plants were moved to the postproduction environment and also 3 weeks later. However, after 6 weeks in the postproduction environment, extensive leaf and bract drop and bract discoloration greatly de- tracted from its appearance. The differences in postproduction longevity observed in this research were surprising. These experiments will be replicated in 1991 to verify these results. Also, additional data will be taken on media and root system characteristics to possibly explain any differences in postproduction longevity. Deneke is Assistant Professor and Behe is Assistant Professor of Horticulture; Olive is Superintendent of the Ornamental Horticulture Substation. Effects of Sumagic on Seed- Propagated Physostegia virginiana Alba C. Fred Deneke, Patricia F. Thomas, and Gary J. Keever IN THE LAST several years, herbaceous perennials have increased in popularity. The market for perennials could be expanded by using them in the home for several weeks before planting in the landscape. However, many herbaceous perennials are too tall for use as potted plants, and they also have the problem that many do not come true from seed. Physostegia virginiana (obedient plant or false- dragon head) is a herbaceous perennial that was judged to have potential for use as a potted plant. A member of the mint family, P. virginiana is native to the Eastern United States and is cultivated in wildflower gardens and for cut flowers. White or pink florets are arranged on a densely flowered raceme in summer on plants 2 to 4 feet tall. The florets are unusual in that they can be repositioned on the raceme, which is beneficial in cut flower arrangements and interesting to flower en- thusiasts. Plants have few disease or insect problems but can be invasive in the landscape. Floricultural crops are often treated with plant growth retardants to control height. Uniconazole (Sumagic?), which is a member of the triazole family of growth retardants like Bonzi?, has not yet been registered for use on horticultural crops. No plant growth regulators have been extensively used on herbaceous perennials. The objective of this research was to evaluate drench applications of Sumagic? in restricting vegetative growth in P. virginiana Alba. Seeds of P. virginiana Alba were sown on January 12, 1990, transplanted into cell packs on January 30, and repotted into 5-inch pots on February 19. A soilless medium of 7 pine bark: 1 sand amended per cubic yard with 6 pounds dolomitic limestone, 2 pounds super- phosphate, 1.5 lb. of Micromax?, and 6 pounds Os- mocote? 14-14-14 was used. Plants were maintained in a greenhouse with a minimum night temperature of 70*F. Night-break incandescent lighting from 10:00 p.m. to 2:00 a.m. was used since long-day photoperiods enhance flowering. On March 6, Sumagic was applied as a drench of 0, 0.75, 1.50, 2.25, 3.00, or 3.75 mg active ingredient (a.i.) per pot. Drench applications of Sumagic effectively con- trolled excessive vegetative growth, table 16. Plant height decreased linearly as drench rate increased. Time to flower was not affected by application rate. Most plants flowered within 14 weeks of treatment, and no plants exhibited phytotoxicity symptoms. Increasing drench rates decreased the number of lateral shoots, and thereby decreased the number of lateral inflores- cences. The highest application rate of 3.75 mg a.i. per plant was judged to result in plant heights proportional to the 5-inch container size; however, flowering quality was reduced. A drench application of 2.25 mg a.i. per plant was judged to be the best treatment for controlling plant height without adversely affecting flowering qual- ity. There was noticeable variation within all treatments TABLE 16. RESULTS OF DRENCH RATES OF SUMAGIC ON PHYSOSTEGIA VIRGINIANA ALBA, TAKEN WHEN THE FIRST FLORET OPENED ON THE INFLORESCENCE Drench rate, Flowering Days to Plant Lateral mg/pot flower height shoots Pct. No. In. No. 0 ............................ 100 69 28 6 2.25 ....................... 100 68 21 2 3.00 ....................... 80 73 19 1 3.75 ....................... 90 73 13 1 Significance 2 .............. ns 1"" 1' 'Percentage of plants to flower within 14 weeks after treatment. 2 ns = not significant; 1 = linear; "** and = - significant at 0.001 and 0.05 levels, respectively. [16] for time to flowering and plant height. A problem with producing P. virginiana from seed, as well as many other species of herbaceous perennials, is the large variation in a number of plant characteristics assumed to be genetically influenced. Deneke is Assistant Professor, Thomas is former Graduate Student, and Keever is Associate Professor of Horticulture. Branching of Vinca minor Increased by Growth Regulators James T. Foley and Gary J. Keever THE COMMON VINING groundcover Vinca minor is characterized by long runners and little lateral shoot development. Vinca is typically sold by the number of runners on the plant. Crops are potted several plants per pot or pruned repeatedly during production to produce well-branched, high-quality plants. The first approach requires more plant material than using one plant per pot, while the latter is labor intensive. A study was instituted to determine the effectiveness of several plant growth regulators in inducing lateral budbreak and elongation of V. minor. Cuttings of Vinca were stuck in 36-cell flats of Pro- Mix BX? medium and placed in a double polyethylene greenhouse under intermittent mist (5 seconds/10 min- utes) for 11 weeks. Rooted cuttings were transferred to 4-inch pots and pruned to three nodes each; all side and basal shoots were removed. The following foliar spray treatments were then applied to the plants: Prom- alin? at 125, 250, and 500 p.p.m.; Atrinal? at 1,000, 2,000, and 3,000 p.p.m.; BA at 62.5, 125, and 250 p.p.m.; and Accel? at 62.5, 125, and 250 p.p.m. Sprays were applied just prior to runoff in a volume of about 2.0 ml per plant. Buffer-X* was added as a surfactant at 0.2 percent to BA, Promalin, and Accel. Plants were fertilized weekly with 200 p.p.m. N from Peters Peatlite Special? 20-10-20. A pruned control was included for comparison. Promalin treatments up to 250 p.p.m. increased run- ner number and runner length. Atrinal delayed runner production through week 4, but numbers increased thereafter and runner length decreased as Atrinal rate TABLE 17. PRIMARY RUNNER NUMBER AND LENGTH OF VINCA MINOR Final Final Treatment runner runner number 1 length In. Promalin 250 p.p.m. ....... ................ 4.2 14 500 p.p.m. ....... ................ 5.1 12 1,000 p.p.m. .... ................. 7.1 12 Atrinal 250 p.p.m. ....... ............... 3.3 14 500 p.p.m. ...... ............... 2.3 14 1,000 p.p.m. .... ................. 3.0 14 Control ...... ............ ......... 2.5 15 1 Runners produced from a 3-node cutting. increased. BA did not increase runner number, and increased rates of BA tended to increase runner length. Accel did not increase runner number, and increased rates of Accel decreased runner length. Based on results of the first experiment, a second test was initiated to determine appropriate rates of the most effective branching compounds from the initial test. Foliar spray treatments consisted of Promalin and Atrinal at 250, 500, and 1,000 p.p.m. Treatments were applied a second time without pruning 6 weeks after the initial application. Data were taken at 2-week in- tervals and included primary runner number and length, secondary runner number (from primary runners), and basal runner number (from the growth medium). Final data included a node count of primary runners and a measure of the three longest secondary runners. The results of the second study showed that Promalin increased primary runner number from the three-node cutting, increased secondary and total runner number, and decreased primary runner length. Secondary runner length was found to increase as rate of Promalin in- creased. Atrinal did not promote runner production or runner elongation. Treatment with Promalin resulted in light to severe chlorosis, especially at higher rates. Plants eventually developed normal foliar color and the majority of sec- ondary runners survived. These results indicate that two applications of Promalin at 500 or 1,000 p.p.m. at about 6-week intervals effectively induce lateral branch- ing of runners of V. minor. Foley is Graduate Student and Keever is Associate Professor of Horticulture. [17] PHOTO 7. Effect of (left to right) 0, 500, and 1,000 p.p.m. BA, and pruning on " L - - ibranching of Helleri holly. PHOTO 2. Carnations treated with (left to right) 100, 200, and 1 t 200 p.p.m. paclobutrazol + pinch and untreated control. -S . .- A PW ,1- I PHOTO 3. Garden pinks treated with (left to right) , o and untreated control. PHOTO 8. Infestation of ob- scure scale, Melanaspis Ob- scura (comstock), on pin oak. PHOTO 1. Alice du Pontmandevilla is a popular greenhouse or horticultural annual crop. PHOTO 4. Containers for annual bedding plants used in study: grow-bag, cell packs, and round pots. PHOTO 9. An adult jumping tree bug, Lidopus heidemanni Gibson, probing beneath the cover of an adult obscure scale. .4M ~, (f~f was the top performer was the top performer PHOTO 6. Fireblight damage on Autumn Blaze callery pear. ulr* ~~;" - ~,,r ROLI~P* WOODY ORNAMENTALS Fireblight Susceptibility of Ornamental Pears in Southern Conditions Donna C. Fare, Charles H. Gilliam, and Harry G. Ponder RELEASE of Bradford pear by the USDA in the early 1960's has led to extensive plantings along streets, highways, and urban landscapes in many Middle Atlantic and Southeastern States. Seasonal contributions of spring flowering, summer shade, and fall leaf color have led to Bradford pear being ranked among the "Ten Most Recommended Trees" in several states. However, ma- turing Bradford trees have exhibited robust growth, causing overcrowded conditions in some landscapes. Also, trunk and canopy splitting of older trees has been reported in the Southeast. Splitting is often reported as storm damage, but is likely caused by unnoticed split- ting prior to storms caused by acute branch angles. These problems plus the success of Bradford have led to the release of 12-15 other selections of Pyrus cal- leryana. Since 1980, P. calleryana and nine ornamental pear cultivars have been evaluated for tolerance to fireblight (Erwinia amylovora), a bacterial leaf and stem blight. All trees were planted as bare root whips in a grid spacing of 30 feet. During May 1988, 1989, and 1990, fireblight injury was assessed by counting the number of infected shoot tips per tree. Also, three infected shoots per tree were chosen at random to measure the length of the fireblight dieback. At this stage of infec- tion, shoots were darkened as if scorched by fire. Meas- urements were taken to the base of the dieback, though some fireblight cankers were observed below the die- back on what appeared to be healthy branches. No selective pruning or preventative (or corrective) spray program was employed to reduce fireblight pressure. Fireblight development is enhanced by warm, moist Infected shoots per tree, number 250 - 225 - 200 " 175 - 150 - 125 - 100 75 50 i 25 Aris. Aut.EB N 1988 0 1989 0 1990 31. Brad. Cap. CI.Sel. Earr. Faur. Red. Whit. P.cal. weather conditions, particularly during flowering. Blight development can occur at temperatures between 65 and 95*, but the 75-81F range is most favorable. Rain is critical to the development and spread of fireblight. Warm, cloudy weather following a rain, as often occurs in the Southeast, promotes the growth and spread of causal bacterium. Other environmental factors, such as frost, high winds, and hail, create wounds through which the bacteria can enter the plant. Weather con- ditions in 1988 and 1989 in central Alabama mimic the favorable weather description for disease devel- opment. In 1990, flowering occurred about 1 month earlier than in previous years, due to unseasonably warm temperatures and less rainfall as compared to the pre- vious years; consequently, disease severity was greater in 1988 and 1989 than 1990. Prior to the spring of 1988, the incidence of fireblight infection was insignificant. In 1988, Aristocrat and Red- spire trees had 112 and 40 fireblight damaged shoots, respectively. P. calleryana and other cultivars had fewer than 12 diseased shoots per tree, figure 2. The number of infected shoots of Aristocrat (208) and Redspire (70) was higher in 1989 than in previous years. These data concur with a report from Kentucky that Aristocrat and Redspire were more susceptible to fireblight than Brad- ford, Capital, Fauriei, and Whitehouse. Their data were based on a survey of four sites where cultivars were growing in proximity to each other. In 1988, the average length of dieback per infected shoot was greatest on Redspire (19 inches), Autumn Blaze (17 inches), Aristocrat (15 inches), and Earlyred (14 inches), figure 3. In 1989, Autumn Blaze was se- verely infected with fireblight (209 infected shoots per tree averaging 78 inches dieback per infected shoot), photo 6 (page 18). Autumn Blaze trees died during the winter of 1989-90, following 2 years of intense fire- blight-related damage. Bradford had the least dieback per infected shoot, with the dieback averaging 3, 8, and 5 inches in 1988, 1989, and 1990, respectively. Dieback per infected shoot, cm 225 - 1988 200 0 1989 20 - 0 1990 175 - 150 - 125 - 100 - 75- Aris. Aut.BI. Brad. Cap. CI.Sel. Earr. Faur. Red. Whit. P.cal. FIG. 3. Length of dieback per infected shoot. [19] FIG. 2. Number of infected shoots per tree. The selection of callery pear for use in southern landscapes will be strongly dependent on varietal re- action to fireblight. Fireblight from plant species in the native habitat in the South will maintain constant pres- sure in cultivated landscapes, therefore the success of any cultivar may be largely determined by fireblight tolerance. Callery pears most susceptible to fireblight in this study were Autumn Blaze, Aristocrat, Redspire, and Earlyred. The least susceptible cultivars were Brad- ford, Capital, Fauriei, and Whitehouse. Fare is Research Associate, Gilliam is Professor, and Ponder. is Professor of Horticulture. Growth of Five Shade Trees as Influenced by Trickle Irrigation Based on Net Evaporation Donald J. Eakes, Charles H. Gilliam, Harry G. Ponder, William B. Webster, and Clyde E. Evans PREVIOUS RESEARCH has shown trickle irrigation to enhance growth of field-grown nursery stock. Most re- search to date has been conducted to evaluate irrigation rates delivering a fixed amount of water on a daily basis during the growing season regardless of environmental conditions. In the Southeastern United States, summers are characterized by prolonged drought periods with occasionally widely scattered, afternoon thunderstorms, causing soil moisture to vary from day to day. Since trickle irrigation is generally considered as the daily maintenance of adequate soil moisture to at least 25 percent of the root system to prevent moisture stress, water requirements will vary on a daily basis. A fixed irrigation rate may cause overwatering on cloudy, over- cast days and may not provide adequate moisture for optimum growth during drought periods. The purpose of this experiment was to evaluate the effects of trickle irrigation rates based on net evaporation from a class A pan on five field-grown shade tree species. On March 29, 1-year-old liners of river birch, flow- ering dogwood, green ash, Bradford pear, and pin oak were planted in a Decatur silt loam soil at the Tennessee Valley Substation, Belle Mina, Alabama. All plants were 18- to 24-inch bareroot liners, except Bradford pear had been potted in #1 containers after rooting the previous season. Rows were 6 feet apart, and plants were spaced 4 feet within rows. Four irrigation rates based on 0, 25, 50, and 100 percent replacement of net evaporation from a class A pan were evaluated. The five shade tree species were measured for height and caliper at 6-week intervals. Trickle irrigation was installed on May 13. One Rain Bird EM-J10@ series 1-gallon-per-hour pressure com- pensating emitter was inserted into the 1/2-inch black polyethylene pipe adjacent to individual plants receiv- ing 25 percent and 50 percent replacement of net evaporation. Irrigation was supplied to plants receiving 100 percent replacement of net evaporation with Rain Bird EM-J20 2-gallon-per-hour pressure compensating emitters. Emitters were placed within 6 inches of the plant's base. Rainfall and net evaporation from a class A evapo- ration pan were recorded daily throughout the exper- iment. Plants were irrigated from May to October, when net evaporation reached or exceeded 1/2 inch. Irri- gation prior to May was not necessary due to rainfall distribution and lower daily temperatures. Daily water replacement was based on each plant occupying a space of 1.5 square feet. Results varied according to species and irrigation rate. Trickle irrigation on flowering dogwood resulted in greater caliper and height growth when compared to nonirrigated trees. Irrigated flowering dogwood had greater caliper by the end of the first growing season, and the differences continued through the second year, table 18. Treatments of 50 percent and 100 percent replacement of the net evaporation produced the great- est height. Both caliper and height of green ash re- sponded in a quadratic manner to irrigation rates. Caliper of river birch responded in a linear and quadratic fash- ion to increasing irrigation rates, table 18. Maximum caliper size occurred with application of 50 and 100 percent replacement of net evaporation. Height of river birch was not affected by irrigation. Pin oak and Bradford pear growth was not influenced by irrigation. While these data do not demonstrate a positive re- sponse for all five tree species to trickle irrigation, they do support the concept that trickle irrigation is not TABLE 18. EFFECTS OF TRICKLE IRRIGATION ON CALIPER AND HEIGHT OF FIVE WOODY TREE SPECIES AFTER Two GROWING SEASONS Irrigation rate, percent Dogwood Green ash River birch Pin oak Bradford pear replacement of net evaporation Caliper Height Caliper Height Caliper Height Caliper Height Caliper Height cm cm cm cm cm cm cm cm cm cm 0 ...................................................... 2.2 134.1 3.8 244 4.4 261 3.6 169 3.3 237 25 .................................................... 2.5 141.7 3.9 252 4.1 250 3.6 177 3.4 231 50 .................................................... 2.7 150.4 4.1 258 4.6 242 3.5 170 3.5 231 100 ................................................... 2.7 147.8 4.2 264 4.8 257 3.9 178 3.5 226 Significance Irrigation rate-linear ......................... 1 ns ns ns * ns ns ns ns ns -quadratic .................... .. ? ns ns ns ns ns 'Nonsignificant (ns) or significant at the 5 percent () or 1 percent (**) level. [20] economically beneficial for all field-grown nursery stock. This may be particularly true for silty loam soils which have a high water holding capacity and typically pro- duce excellent nursery stock. Eakes is Assistant Professor, Gilliam is Professor, and Ponder is Professor of Horticulture; Webster is Superintendent of the Tennessee Valley Substation; Evans is Professor of Agronomy and Soils. Chemically Induced Branching of Woody Landscape Plants Gary J. Keever and William J. Foster REPEATED PRUNING of many woody landscape species is a labor-intensive practice required to produce well- branched, marketable materials. Shoot tip pruning removes the source of apical dominance, a process controlled by a balance between auxin and cytokinin levels, and stimulates lateral bud development. Exog- enously applied cytokinins, including BA, promote axil- lary bud growth and branching of woody plants. Synthetic cytokinins may induce only a partial release from apical dominance; further growth of released axillary buds requires treatment with a synthetic auxin or gibberellin. Two experiments were conducted to determine if axil- lary growth of several woody landscape species could be promoted by exogenous cytokinin and gibberellin application. Species used in the tests typically require multiple prunings during production for the develop- ment of a well-branched, compact plant. In experiment 1, uniform liners of Helleri holly, Stoke's Dwarf holly, Fraser photinia, Indian hawthorn, Formosa azalea, and cleyera were grown in 3-inch- square containers in a glass greenhouse. Growth medium was amended milled pine bark-peat moss (4:1, by vol- ume). On March 7, 1986, a single spray of BA at 0, 125, 250, 500, and 1,000 p.p.m. was applied to the foliage of all species. Foliar sprays included 0.2 percent (by volume) Buffer X?, an ionic su factant. A pruned control in which plants were cut 2 inches above the growth medium also was included. On April 29, 1986, newly developed lateral shoots longer than 0.4 inch were counted. Due to a minimal response to BA, the experiment was terminated in June 1986 for all species except Helleri and Stoke's Dwarf hollies. On June 9, 1986, Helleri and Stoke's Dwarf hollies were repotted into #1 containers of amended 100 percent milled pine bark. Plants were moved from the greenhouse to an outdoor production area in full sun, and treatments were reapplied. Growth indices [(height + width + width)/3] were measured on November 18, 1986. On January 28, 1987, lateral shoot length (five longest shoots per plant), plant height, and relative root rating (1 = no roots on rootball surface; 2-5 = 25, 50, 75, and 100 percent coverage of rootball, respectively) were determined. Because several species were not responsive to the BA rates applied, a second experiment was conducted using higher BA rates and several rates of Promalin. Promalin is a mixture of equal parts by weight of BA and GA 4 + 7 . Species evaluated included Fraser photinia, cleyera, Formosa azalea, Indian hawthorn, and Harbour Dwarf nandina (Harbour Dwarf nandina has little or no lateral branching and does not readily respond to mechanical pruning). Uniform liners were grown in 3- inch containers of the same amended liner growth medium as in the first experiment. On November 4, 1987, BA treatments of 0, 1,000, 1,500, 2,000, and 2,500 p.p.m. and Promalin? treatments of 0, 2,000, 3,000, 4,000, and 5,000 p.p.m. were applied. A pruned control was included for comparison. Plants were main- tained in a glass greenhouse. On January 26, 1988, plant height and axillary bud- break were determined, and on February 8, treatments were reapplied. Plant height and axillary budbreak data were collected on April 1 from all species except Indian hawthorn, which was collected on May 13. On May 24, apical and subapical single-node cuttings were taken from Harbour Dwarf nandina to determine treatment effects on cuttage; cuttage production limits the avail- ability of Harbour Dwarf nandina. Experiment 1 Induction of axillary budbreak in response to BA application was species-dependent, table 19. Budbreak of cleyera, Formosa azalea, and Indian hawthorn was marginally or not affected by BA, whereas hand pruning induced equal or greater branching than all rates of BA. Axillary budbreak of Helleri and Stoke's Dwarf hollies and Fraser photinia increased with increasing rates of BA and resulted in a greater number of new shoots at 1,000 p.p.m. than did hand pruning, photo 7 (page 18). Six months after repotting and a second BA appli- cation, little visual response was evident in the relative root rating for Helleri and Stoke's Dwarf hollies. Growth indices, shoot length, and plant height measurements TABLE 19. EFFECT OF BA FOLIAR SPRAYS AND HAND PRUNING ON AXILLARY BUDBREAK OF FIVE WOODY NURSERY CROPS, EXPERIMENT 1 Average number of new shoots/plant Treatment Formosa Helleri Fraser Stoke's Cleyera azalea holly photinia Dwarf holly No. No. No. No. No. BA' 125 p.p.m . 6.3*2 11.2 11.5' 1.5 39.4 250 p.p.m. 6.8 10.6" 14.7 1.5 45.2 1,000 p.p.m... 5.9' 9.9' 42.1' 7.3' 63.4' Control ......... 6.7 9.9' 6.8' 1.7' 29.7 Hand pruned ...... 9.5 13.7 19.1 2.5 35.4 Significance 3 ...... q c c c 1 dApplied as a foliar spray with 0.2 percent Buffer X, a surfactant, 2 Dunnett's test for least significant difference; means within a col- umn followed by an asterisk differ from the mean of the pruned treatment, 5 percent level. 3q = quadratic; c = cubic regression response; 1 = linear; 1 percent level. [21] TABLE 20. GROWTH INDEX, SHOOT LENGTH, AND PLANT HEIGHT OF HELLERI AND STOKES'S DWARF HOLLIES TREATED WITH BA, EXPERIMENT 1 Growth index' Shoot length 2 Plant height Treatment Helleri Stoke's Helleri Stoke's Helleri Stoke's Dwarf Dwarf Dwarf cm cm cm cm cm cm BA 125 p.p.m... 28.8 27.6'3 6.9 7.3 19.9 25.5' 250 p.p.m... 26.9 29.2' 4.7' 7.3 19.4 25.2* 500 p.p.m... 29.8 27.2 6.1 7.9 19.3 26.5* 1,000 p.p.m. 27.2 24.6 5.4 6.9 18.7 22.4 Control ........... 30.4 26.9 7.6 7.7 20.6 26.5 Hand pruned ... 27.1 25.5 7.4 6.9 18.2 22.2 Significance 4 .... ns c c q ns c 'Growth index = (height + width + width)/3. 2 Shoot length = mean of 5 longest shoots/plant. 3Dunnett's test for least significant difference; means within a col- umn followed by an asterisk differ from the mean of the pruned treatment, 5 percent level. 4 ns = not significant; c = cubic regression response; q = quadratic; 1 percent level. tended to decrease with increasing BA rates, table 20. BA application, especially at the higher rates, resulted in dense and compact plants of Helleri and Stoke's Dwarf hollies. Phytotoxicity was not observed on any species. Experiment 2 Axillary shoot number taken when treatments were first applied did not differ among treatments, so sub- sequent budbreak data were not adjusted. As in the first TABLE 21. EFFECTS OF BA AND PROMALIN ON AXILLARY BUDBREAK AND HEIGHT OF FORMOSA AZALEA, EXPERIMENT 2 Axillary budbreak Plant height Treatment Jan. 26, Apr. 1, Jan. 26, Apr. 1, 1988 1988 1988 1988 No. No. cm cm BA 1,000 p.p.m .......... 5.5*1 8.1* 29.0' 49.7' 1,500 p.p.m .......... 4.9* 8.7* 31.9' 53.3* 2,000 p.p.m .......... 5.1* 14.1 28.9* 51.7' 2,500 p.p.m .......... 6.1 13.6 27.1* 42.8* Promalin 2,000 p.p.m......... 5.5' 13.7 33.9' 46.4* 3,000 p.p.m......... 5.9' 15.9 28.6' 41.7' 4,000 p.p.m .......... 5.5* 16.2 28.0' 37.6' 5,000 p.p.m .......... 6.0' 22.1' 29.5* 38.4' Pruned ...................... 7.7 15.3 16.0 14.5 Unpruned ................. 6.9 9.1' 30.7* 49.3* Comparison BA vs. Promalin 2 .... ns " ns " Sign of rate 3 BA ......................... q q q q Promalin ............... ns 1 c 1 'Dunnett's test for least significant difference; means within a col- umn followed by an asterisk differ from the mean of the pruned treatment, 5 percent level. 2 ns = not significant or significant at the 1 percent (") level. q = quadratic; ns = not significant; 1 = linear; c = cubic regression response; 1 percent level; unpruned control included in regression analysis. [22] experiment, species responded differently to growth regulator application. Neither axillary budbreak nor plant height of cleyera was affected by BA or Promalin foliar sprays. Axillary budbreak of Formosa azalea, determined in January 1988, decreased at the lower BA rates before increasing at the highest rate, table 21. By April, bud- break was positively correlated with BA rate, increasing from 9.1 breaks per plant for the unpruned control to 14.1 breaks per plant receiving 2,000 p.p.m. BA. Bud- break was not affected by Promalin in January, but by April 1988, budbreak increased linearly with increasing Promalin rate, from 9.1 breaks on the unpruned control to 22.1 breaks on plants treated with 5,000 p.p.m. Pruning induced greater budbreak than all plant growth regulator (PGR) treatments except 2,500 p.p.m BA in January and the lower two BA rates in April. Only the highest Promalin rate promoted greater budbreak than pruning. There was a decrease in plant height with increasing plant growth regulator rate; this trend was particularly evident in April. Indian hawthorn responded minimally to both plant growth regulators. With increasing BA and Promalin rates there was a slight increase in budbreak noted in January 1988, however this response was not evident in May. Budbreak of pruned plants was significantly less in January than budbreak of plants treated with three rates of BA and Promalin and less in May than that of plants treated with all rates of both materials. Budbreak of pruned plants also was less than that of unpruned plants in May. Axillary budbreak of Fraser photinia was promoted by BA and Promalin on both sampling dates, table 22. TABLE 22. EFFECTS OF BA AND PROMALIN ON AXILLARY BUDBREAK AND HEIGHT OF FRASER PHOTINIA, EXPERIMENT 2 Axillary budbreak Plant height Treatment Jan. 26, Apr. 1, Jan. 26, Apr. 1, 1988 1988 1988 1988 No. No. cm cm BA 1,000 p.p.m .......... 3.8 11.5*1 22.3' 33.7' 1,500 p.p.m .......... 4.7 15.3' 25.7* 40.7' 2,000 p.p.m .......... 4.9 18.1' 22.1* 32.2* 2,500 p.p.m .......... 6.9* 23.8* 24.5* 36.0' Promalin 2,000 p.p.m .......... 4.7 20.7* 29.3* 42.6* 3,000 p.p.m .......... 5.9' 24.7' 29.3' 41.5* 4,000 p.p.m .......... 6.3* 24.6* 30.3' 40.9* 5,000 p.p.m .......... 9.6* 28.7* 36.9' 43.5' Pruned...................... 3.6 3.7 14.5 12.8 Unpruned .................. 1.8* 11.0* 23.5' 38.1' Comparison BA vs. Promalin 2 .... " Sign of rate 3 BA ......................... q 1 ns ns Promalin ............... c 1 c ns 'Dunnett's test for least significant difference; means within a col- umn followed by an asterisk differ from the mean of the pruned treatment, 5 percent level. 2 ns = not significant or significant at the 1 percent (") level. 3q = quadratic; ns = not significant; 1 = linear; c = cubic regression response; 1 percent level; unpruned control included in regression analysis. In January, budbreak of BA-treated plants increased from 1.8 for the unpruned treatment to 6.9 breaks with 2,500 p.p.m. BA. In April, the increase was from 11.0 to 23.8 breaks per plant. Budbreak increased to 9.6 in January and to 28.7 in April with the highest Promalin rate. Budbreak of pruned plants was twice that of unpruned plants in January but less in April. Pruning was less effective in inducing axillary budbreak than 2,500 p.p.m. BA and the three highest Promalin rates in January and all rates of both in April. Plant height, measured in January, was promoted by Promalin, but was not evident in April. BA and Promalin increased axillary budbreak of Har- bour Dwarf nandina compared to unpruned plants, table 23. The BA-induced increase was from 1.1 breaks for unpruned plants to 3.9 breaks with 2,000 p.p.m. in January and from 1.3 to 7.4 breaks with 2,000 p.p.m. in April; Promalin promoted budbreak to 3.3 breaks in January and 5.5 breaks in April with 5,000 p.p.m. Pruning induced more budbreaks compared to the unpruned control and was more effective than the two lowest Promalin rates in January but less effective than the three highest BA rates and the highest Promalin rate in April. Plant height of Promalin-treated plants was greater than height of unpruned plants on both sampling dates. TABLE 23. EFFECTS OF BA AND PROMALIN ON AXILLARY BUDBREAK AND HEIGHT OF HARBOUR DWARF NANDINA, EXPERIMENT 2 Axillary budbreak Plant height Treatment Jan. 26, Apr. 1, Jan. 26, Apr. 1, 1988 1988 1988 1988 No. No. cm cm BA 1,000 p.p.m. ........ 2.0 3.6 12.6 16.9 1,500 p.p.m. ........ 3.9 5.8 ' 11.4 17.3 2,000 p.p.m. ........ 4.1 7.4' 13.1 18.6 2,500 p.p.m. ........ 3.8 7.4* 13.3 18.2 Promalin 2,000 p.p.m ........ 1.6* 3.5 18.0 27.2* 3,000 p.p.m......... 1.3' 2.4 18.5* 26.9' 4,000 p.p.m ......... 1.9 3.5 22.4' 36.8' 5,000 p.p.m ........... 3.3 5.5' 21.6' 43.3' Pruned ..................... 3.4 2.9 11.3 15.7 Unpruned ................. 1.1' 1.3 12.1' 16.8 Comparison BA vs. Promalin 2 .... Sign of rate 3 BA ... ................ c 1 ns ns Promalin .................. 1 c 1 1 'Dunnett's test for least significant difference; means within a col- umn followed by an asterisk differ from the mean of the pruned 2 ns - not significant or significant at the 1 percent (") level. 3 c = cubic regression response; 1 percent level; 1 = linear; ns = not significant; unpruned control included in regression analysis. More single-node apical and subapical cuttings were taken from Promalin-treated Harbour Dwarf nandina and more apical cuttings were taken from BA-treated plants than from unpruned plants, table 24. This increase represents an important source of propagation material for the grower from a cultivar that does not naturally branch or form multi-node shoots. TABLE 24. NUMBER OF CUTTINGS TAKEN FROM HARBOUR DWARF NANDINA SPRAYED WITH BA OR PROMALIN, EXPERIMENT 2 Treatment Cuttagel Apical Subapical Total No. No. No. BA 1,000 p.p.m. .......... 1.6 0.0 1.6 1,500 p.p.m ............. 3.0 .0 3.0 2,000 p.p.m. ........... 3.4 .0 3.4 2,500 p.p.m. ........... 4.8 .4 5.2 Promalin 2,000 p.p.m. ........... 1.8 2.2 4.0 3,000 p.p.m. ........... 2.0 1.2 3.2 4,000 p.p.m. ........... 2.2 2.4 4.6 5,000 p.p.m. ........... 3.2 4.6 7.8 Pruned ....... ........... 1.4 .0 1.4 Unpruned ........... ...... 1.0 .0 1.0 'Means of five single-plant replicates. BA promoted axillary budbreak of Fraser photinia, Harbour Dwarf nandina, Helleri holly, Stoke's Dwarf holly, and Formosa azalea, while budbreak of Harbour Dwarf nandina, Fraser photinia, and Formosa azalea was stimulated by Promalin. These responses were species and rate-dependent, with higher rates generally induc- ing more lateral budbreak than lower rates. Branching of cleyera and Indian hawthorn was not influenced by BA or Promalin application. BA and Promalin generally induced equivalent or greater numbers of axillary shoots compared to mechanical pruning. Plant height of all species and root rating, growth index, and shoot length of Helleri and Stoke's Dwarf hollies were minimally influenced by PGR application. Keever is Associate Professor of Horticulture; Foster is former Superintendent of the Ornamental Horticulture Substation. Plant Response to Planting Method and Media Gary J. Keever and Gary S. Cobb CONTROLLED-RELEASE granular fertilizers are typi- cally applied either uniformly incorporated into con- tainer growth media or surface-applied as a topdressing. Incorporation has proven successful in a wide range of applications, but uniform blending is essential and sub- sequent storage for more than a week is not recom- mended due to the potential release of fertilizer salts. Longer storage necessitates leaching prior to planting to avoid phytotoxicity, but this wastes fertilizer and could result in undesirable pollution of the surrounding area. Intermittent drying of surface-applied fertilizer slows release due to a lack of continuous moisture, and fertilizer may be lost if the container is overturned or rapidly flooded. Placement of the fertilizer directly under the liner at transplanting (dibbling) is a third method that pre- sents no storage, mixing, drying, or spilling problems. In some cases, dibbling has resulted in more growth or superior-quality plants compared to incorporation. [23] In other studies, either no benefit or a negative response to dibbling was observed. A possible explanation for the poor results with dibbling is that compaction of the growth medium at planting restricts root growth. The objective of this research was to compare dibbling with excavating (removal of a core to form the planting hole without compaction of the growth); media effects on planting method were also evaluated. Aged pine bark was mixed with a sandy loam soil in 4 ratios (1 pine bark:1 soil, 4:1, 9:1, 1:0, by volume). Media were amended by preplant incorporation of 6 pounds dolomitic limestone, 2 pounds gypsum, and 1.5 pounds Micromax? per cubic yard. Osmocote? 17-17-12 at 0.63 ounce per #1 container was placed under each liner at planting. Bulk densities, particle size distributions, water holding capacities, and air porosities were obtained for each growth medium. Uniform liners of Compacta holly and Hino Crimson azalea were transplanted into # 1 containers of the four growth media on March 28, 1984. Two planting meth- ods were compared: (1) compression to form the plant- ing hole (dibbling), and (2) removal of a core to form the planting hole (excavating). Compressed hole and core were similar in size to a linear rootball. Plants were grown in full sun. Plants within media treatments were irrigated as needed by overhead impact sprinklers. After 7 months, growth index [(height + width 1 + width 2 )/3], shoot dry weight, and relative root density were determined. In a second experiment, liners of Helleri holly and Trouper azalea were transplanted on April 12, 1985, into #1 containers, by either dibbling or excavating. Growth media included four media commonly used in the Southeastern United States: 100 percent milled pine bark; pine bark:sandy loam soil (4:1, by volume); pine bark:peat moss (3:1, by volume); and peat moss:softwood shavings (1:1, by volume). Media were amended as in the first experiment, and 0.67 ounce of Osmocote 17- 7-12 was placed under each holly and azalea liner, prior to transplanting. Plants were grown in full sun and watered as needed by overhead irrigation. After 7 months, growth index, shoot dry weight, and relative root den- sity were determined. Experiment 1 Physical properties varied greatly among the four media. For example, the higher the percent bark in the pine bark:soil media the higher the percentage of coarse material and the higher the air porosity. Conversely, the higher the percent soil in the media the greater the fraction of fine material, and the greater the bulk density. Water-holding capacity was not greatly influ- enced by media; however, irrigation frequency was increased 23.4, 29.8, and 44.7 percent with 4:1, 9:1, and 1:0 pine bark:sandy loam soil media, respectively, compared to the 1 bark: 1 soil growth medium. Planting method influenced shoot growth and root density of both holly and azalea, tables 25 and 26. Shoot dry weight of holly and azalea averaged 0.17 and 0.13 ounce, respectively, more when media were excavated at transplanting compared to dibbled. Root densities TABLE 25. EFFECTS OF PLANTING METHOD AND GROWTH MEDIA ON GROWTH OF COMPACTA HOLLY 7 MONTHS AFTER TRANSPLANTING Comparison Growth Shoot Relative Comparison index' dry root weight density 2 cm Grams Method Dibble ........ ........................ 35.7a 33.3b 3.2b Excavate ...... ....................... 36.1a 3 38.1a 3.5a Pine bark:sandy loam ratio 1:1 ............................................ 33.7 34.5 2.3 4:1 ......................................... 36.6 35.5 3.7 9:1 ................ ............... 37.0 38.3 3.6 1:0 ......................................... 36.4 34.7 3.8 Significance 4 ......... . . . . . . . ... .. q c c Method x ratio of pine bark: sandy loa . .. . .. ... ..... . . . . . . .. ... ns ns ns 'Growth index = (height + width, + width2)/3, in centimeters. 2 Relative root density: 1 few surface roots on root ball; 3 moderate root density over entire rootball; 5 = dense matting over entire rootball. 3Mean separation within columns by Duncan's multiple range test, 5 percent level. 4 Quadratic (q) or cubic (c) regression response significant at 5 percent (') or 1 percent (') level. sPlant method x media interaction not significant (ns). TABLE 26. EFFECTS OF PLANTING METHOD AND GROWTH MEDIA ON GROWTH OF HINO CRIMSON AZALEA 7 MONTHS AFTER TRANSPLANTING Growth Shoot Relative Comparison index' dry root weight density 2 cm Grams Method Dibble ..................................... 31.1a 39.9b 3.2b Excavate ...... ....................... 31.8a 3 43.7a 3.4a Pine bark:sandy loam ratio 1:1 ......................... ............. 28.6 38.9 2.3 4:1 ......................... .............. 30.9 38.9 3.2 9:1 ............................................ 33.3 44.0 3.9 1:0 .................... .................. 33.0 45.5 4.0 Significance 4 ........ . . . . . . . . .. ... c q c Method x ratio of pine bark: sandy loam s ..... ............... ns ns 'Growth index = (height + width, + width 2 )/3, in centimeters. 2 Relative root density: 1 = few surface roots on root ball; 3 = moderate root density over entire rootball; 5 = dense matting over entire rootball. 3Mean separation within columns by Duncan's multiple range test, 5 percent level. 4 Cubic (c) or quadratic (q) regression response significant at 5 percent (*) or 1 percent (") level. 5Plant method x media interaction not significant (ns). also were greater with excavating. Growth indices of the two species were not influenced by planting method. Growth index and relative root density of holly in- creased as the percentage of sandy loam soil in the media decreased from 50 percent (1:1) to 20 percent (4:1); there were only minor differences in these mea- surements among the three media with lower percent- ages of soil. Shoot dry weight was similar among treatments, except holly growth was greater in the 9:1 growth medium. Growth index, shoot dry weight, and root density of azalea increased with increasing percentages of pine bark up to 90 percent; there was little change in mea- surements as the percentage of pine bark increased [24] TABLE 27. EFFECTS OF PLANTING METHOD AND GROWTH MEDIA ON GROWTH OF HELLERI HOLLY 7 MONTHS AFTER TRANSPLANTING Growth media Growth index' Shoot dry weight Relative root density 2 Dibble Excavate Dibble Excavate Dibble Excavate cm cm Grams Grams Pine park (100 percent) ................................................... 38.9a(b) 3 39.3a(b) 30.4b(c) 34.7a(b) 2.2ab(ab) 2.4a(ab) Pine bark-sandy loam (4:1) ......... ......................... ... 355b(c) 41.5a(ab) 25.4b(c) 41.Oa(a) 2.0b(b) 2.5a(a) Pine bark-peat moss (3:1) .................................................... 42.5a(a) 43.2a(a) 39.1a(b) 42.3a(a) 2.4a(a) 2.4a(ab) Peat moss-shavings (1:1) ...................................................... 44.6a(a) 43.1a(a) 49.2a(a) 45.5a(a) 2.2a(ab) 2.2a(b) 'Growth index = (height + width, + width 2 )/3, in centimeters. 2 Relative root density: 1 = few surface roots on rootball; 3 = moderate root density over entire rootball; 5 = dense matting over entire rootball. 3Mean separation within planting method made using LSD at 5 percent level; mean separation within growth media ( ) by Duncan's multiple range test, 5 percent level. All planting method x media interactions were significant. from 90 to 100 percent. Root density was greater with excavation than with dibbling in the 4:1 and 9:1 media, but similar for plants in the other two media trans- planted by the two methods. Experiment 2 Excavation resulted in a higher growth index than dibbling in the pine bark-sandy loam growth medium and a greater shoot dry weight and relative root density in 100 percent pine bark and pine bark-sandy loam media, table 27. Planting method did not affect mea- surements of plants grown in other media. Growth index and shoot dry weight of dibbled hollies were greater when plants were grown in media con- taining peat moss compared to media without peat moss. Plants of the excavation planting method had the lowest growth index and shoot dry weight when grown in 100 percent pine bark, possibly because of less water re- tention compared to the other media. Root density of dibbled hollies was greater in a pine bark medium when peat moss was a component compared to sandy loam soil. Excavation resulted in less root growth in a peat moss-shavings medium than in the pine bark-sandy loam medium. Neither growth index, shoot dry weight, nor relative root density of azalea was influenced by planting method. Growth index and shoot dry weight were greater in peat-based media than in media not containing peat moss. Root density of plants grown in pine bark + sandy loam soil or peat moss was greater than root density in 100 percent pine bark or peat moss-shavings medium. Shoot and root growth of two species in two exper- iments were either greater or not influenced when the planting hole was excavated rather than dibbled. This response was media-dependent, occurring in media with a range of pine bark:sandy loam ratios but not in peat- based media. Alterations in the physical properties of the media during formation of the planting hole or differences in moisture-holding capacity of excavated and dibbled media may explain growth differences. Placement of controlled-release fertilizer directly un- der the liner at transplanting is an effective method of fertilization that avoids media storage and mixing prob- lems. However, plant growth may be adversely affected if the planting hole is formed by compression (dibbling) rather than removal of a core (excavating). This effect is more likely to occur in pine bark-based media that do not contain peat than in peat-based media. Keever is Associate Professor of Horticulture; Cobb is former Su- perintendent of the Ornamental Horticulture Substation. Paclobutrazol Inhibits Growth of Woody Landscape Plants Gary J. Keever, William J. Foster, and James C. Stephenson MECHANICAL PRUNING to control excessive vegeta- tive growth and improve plant form is a major expense in the production and maintenance of some woody landscape plants. Over time, numerous compounds have been used to retard woody plant growth, but most remain uneconomical or cause undesirable side effects. Currently, chemical growth inhibitors are being actively evaluated by the electric utility industry, which spends an estimated $800 million per year on tree trimming. Paclobutrazol, registered as Clipper?, is an inhibitor of gibberellin biosynthesis and is used to suppress regrowth of pruned trees along utility rights-of-way. Clipper is labeled for trunk injection of trees, which avoids chemical contact with nontarget plants and reduces environmental residues. Paclobutrazol also is labeled as Bonzi? for use on poinsettias, bedding plants, chrysanthemums, geraniums, and potted freesias. Little published research is available on the potential uses of paclobutrazol in the production of woody landscape plants. The objectives of this research were to evaluate the magnitude and duration of growth inhibition by media- and foliar-applied paclobutrazol for eight woody landscape species. Uniform liners of eight species were potted March 27, 1986, in an amended 100 percent milled pine bark growth medium. Plant species and container sizes included euonymus, dwarf Burford holly, compacta holly, juniper, Hino Crimson azalea, photinia, and For- mosa azalea in #1 gallon containers and privet in #3 containers. Plants were placed outdoors in full sun or under 47 percent shade (euonymus and azaleas) and maintained following typical nursery cultural practices. OnJuly 23, 1986, the following treatments were applied: paclobutrazol sprays of 0, 250, 500, 1,000, and 2,000 p.p.m. and drenches of 6.3, 25, and 100 mg active [25] TABLE 28. GROWTH INDICES OF EIGHT WOODY LANDSCAPE SPECIES DRENCHED OR SPRAYED WITH PACLOBUTRAZOL, NOVEMBER 19, 1986 (17 WAT) Index' , by species Treatment (a.i.) Copac Dwarf Formosa Hino Euonymus Co acta Burford Juniper Privet Photinia Formosa Crison holly azalea Drench 6.3 mg/pot .................................. 33.8 42.4 24.0 44.5 76.0 50.8 49.8 38.8 25.0 mg/pot ................................ 32.8 39.7 25.6 42.9 75.2 47.0 53.8 37.7 100.0 mg/pot .............................. 30.4 38.5 25.8 36.6 68.5 43.5 49.9 38.8 Control ............................... 44.3 45.5 28.1 52.2 100.5 78.2 62.6 41.8 Significance 2 ............ .. ... .. .. . . . . . . . .. ... ... c q c c" c" c" c Spray 250 p.p.m. .............................. 45.1 46.9 25.0 47.5 91.3 63.3 56.7 38.5 500 p.p.m ................................... 45.0 41.5 24.1 50.0 89.5 56.0 55.3 38.7 1,000 p.p.m ................................... 41.5 42.3 24.7 45.9 82.6 55.4 54.1 39.4 2,000 p.p.m .................................. 40.7 40.3 27.1 44.5 74.9 50.7 53.4 38.5 Control .......................................... 44.3 45.5 28.1 52.2 100.5 78.2 50.7 41.8 Significance .................................... c " 1" q" 1" 1" c" q" c* Drench ............................................... 32.3*" 40.2' 25.1 41.3* 73.2' 47.1 51.1' 38.4 Spray ................................................. 43.1 42.8 25.2ns 47.0 84.6 56.4 54.9 38.8ns 'Growth index = (height + width, + width2)/3 in centimeters, where width, = width at the widest point and width 2 = width at a right angle to width,. 2 Significant at the 5 percent (*) or 1 percent (") level; c = cubic; I = linear, q = quadratic. Zero rate included in regression analysis. 3Significant (*) or not significant (ns) at the 5 percent level; zero rate not included in mean determination ingredient (ai) in a volume of 8.5 ounces per container. (A dosage of 100 mg = 0.0035 ounce.) On November 19, 1986, 17 weeks after treatment (WAT), growth indices and foliar color ratings were determined for all species. On March 27, 1987 (35 WAT), flowers on five plants of Formosa azalea were counted and the diameter of 10 flowers each on four plants of Hino Crimson azalea was measured. Growth indices were taken again on June 19 (48 WAT) and December 1, 1987 (71 WAT); foliar color was rated on December 9, 1987 (72 WAT), and fruit of dwarf Burford holly was counted on January 18, 1988 (78 WAT). On March 8, 1988 (84 WAT), root systems of all species were rated for density, foliage color was rated, and shoots were severed at the container medium surface for dry weight determination. By November 19, 1986 (17 WAT), significant differ- ences in growth indices had occurred in response to both concentration and application method, table 28. In general, growth indices of all species decreased with increasing drench and spray concentrations. Drench treatments were more active than spray treatments for six species, while two species (dwarf Burford holly and Hino Crimson azalea) responded similarly, regardless of the application method. Similar trends in growth indices to those on November 19 were observed on June 19, 1987 (48 WAT), and December 1, 1987 (71 WAT). Growth indices for all species decreased with increasing paclobutrazol concentrations except for spray- treated euonymus (both dates) and juniper (June date only). Drenches again were more effective than sprays in suppressing growth indices for all species on both sampling dates. For most species, foliar color ratings taken November 19, 1986 (17 WAT), were not influenced by treatment. By December 9, 1987 (72 WAT), foliar color ratings of most species were influenced by paclobutrazol, Foliar color ratings for six of eight species drenched and four of eight species sprayed with paclobutrazol improved with increasing concentration, while foliar color ratings of drench-treated dwarf Burford holly and Hino Crimson azalea were greatest at the low concentrations. Flower number for Formosa azalea increased dra- matically in response to paclobutrazol, with as much as a 360 percent and 238 percent increase in flower number with sprays and drenches, respectively, com- pared to nontreated plants, table 29. Flower diameter of Hino Crimson azaleas drenched or sprayed was reduced as much as 46 and 11 percent, respectively, compared with the control. Flowering of drench-treated plants of both cultivars was delayed about 3 weeks. TABLE 29. FLOWERING OR FRUITING OF THREE WOODY LANDSCAPE SPECIES DRENCHED OR SPRAYED WITH PACLOBUTRAZOL Formosa Hino Dwarf Burford Treatment (a.i.) Formosa Crimson flower holly flowers diameter 2 fruit 3 No. cm No. Drench 6.3 mg/pot - 2.9 239.6 25.0 mg/pot .......... 189.1 2.1 73.0 100.0 mg/pot ........ 153.0 1.9 1.5 Control .................. 79.2 3.5 40.2 Significance ........... q" q c" Spray 250 p.p.m .............. 119.2 3.3 40.2 500 p.p.m............. 238.6 3.2 208.4 1,000 p.p.m .......... 287.6 3.1 159.5 2,000 p.p.m. ......... 274.4 3.3 272.5 Control .................. 79.2 3.5 275.0 Significance ............ q q' q Drench ...................... 171.1** 5 2.3**" 104.7 Spray ......................... 230.0 3.2 228.9 'Means of five single-plant replicates, March 27, 1987. Due to a delay in flowering, data on drench-treated plants were taken April 20, 1987. Data for the 6.25 mg a.i. drench not available. 2 Means of 10 flowers per plant, 4 single-plant replicates, March 27, 1987. Due to a delay in flowering, data on drench-related plants were taken April 20, 1987. 3Means of 10 single-plant replicates, January 18, 1988. 4Significant at the 1 percent level; q = quadratic, c cubic. Zero rate included in regression analysis. 5Significant at the 1 percent (") level. [26] Fruit number of dwarf Burford holly increased with increasing spray concentrations compared to nontreated plants, while drench-treated plants increased in fruit number at the lowest concentration and decreased to essentially zero at the highest concentration. Terminal data collected on March 8, 1988 (84 WAT), were root density and shoot dry weight. Roots of most species covered the entire rootball surface and were densely matted, regardless of treatment. However, dif- ferences among treatments were evident with the two azalea species and dwarf Burford holly. Root density of Formosa azaleas treated with the highest drench con- centration was less than plants in other treatments, whereas root density of Hino Crimson azalea decreased with increasing drench concentration. Root coverage of sprayed and control plants was similar for both azalea cultivars. Root coverage of dwarf Burford holly was highest for nontreated plants and decreased with increasing paclobutrazol concentration, with drenches suppressing root growth more than sprays. Twenty months after paclobutrazol was applied, shoot dry weight was suppressed on all species drenched and on six of eight species sprayed, table 30. Only dry weight of spray-treated dwarf Burford holly and juniper was not affected by treatment. Growth retardation was greater for seven of eight species drenched compared to sprayed. In addition to quantitative differences among treat- ments, visual or aesthetic changes were observed with paclobutrazol-treated plants. Generally, plants respond- ed to increasing drench and spray concentration by producing shorter internodes and smaller leaves. Axil- lary buds began to develop on several species, but these buds seldom elongated more than 0.8-1.6 inches. Foli- age of some species was darker green when treated with paclobutrazol; with other species, treatment had no effect on foliar color. Relatively high concentrations of paclobutrazol generally do not cause phytotoxicity. However, in this test several species exhibited phyto- toxicity symptoms in response to the highest paclo- butrazol concentrations; for example, foliage of dwarf Burford holly developed tip and marginal chlorosis while new foliage of photinia and the two azalea cul- tivars curled downward. As a result of excess internode suppression, other species developed dense clusters of leaves closely adpressed along the stems. Paclobutrazol is a powerful inhibitor of internode elongation. Growth inhibition was detected as early as 4 months after paclobutrazol was applied and persisted for at least 20 months. Generally, the magnitude and duration of growth suppression was greater when paclo- butrazol was applied as a drench than as a spray, as exemplified by the growth indices, table 28, and shoot dry weight, table 30. Foliar color ratings generally increased when paclobutrazol was applied as a drench; response to sprays varied among species. Paclobutrazol promoted flowering and fruiting, table 29, of several species. Paclobutrazol is an effective growth retardant on a wide range of woody landscape plants when applied as either a drench or spray. This may offer growers an additional management tool; for example, its use offers the ability to retard growth during a depressed market or avoid transplanting. Due to the magnitude and per- sistence of growth suppression, drench applications during production are probably not practical and spray concentrations should be carefully chosen. Drench and spray application methods have potential for the land- scape industry, however established plants may respond differently than container-grown plants to paclobutrazol concentration and application method due to differ- ences in growth medium or other factors. Sensitivity to paclobutrazol varied greatly among species, and appro- priate concentrations are likely to be highly species- dependent; hence, paclobutrazol should first be tested on a small group of plants before committing to large scale application. Keever is Associate Professor of Horticulture; Foster is former Superintendent and Stephenson is Associate Superintendent of the Ornamental Horticulture Substation. TABLE 30. SHOOT DRY WEIGHT OF EIGHT WOODY LANDSCAPE SPECIES DRENCHED OR SPRAYED WITH PACLOBUTRAZOL, MARCH 8, 1988 (84 WAT) Dry weight, by species Treatment (a.i.) Cmat Dwarf Formosa Hino Treatment (a.i.) Euonymus Compacta Burford Juniper Privet Photinia Forosa Crimson holly holly ip azalea Grams Grams Grams Grams Grams Grams Grams Grams Drench 6.3 mg/pot .................................... 52.2 82.6 45.0 161.9 172.2 111.4 129.6 67.9 25.0 mg/pot .................................. 47.9 81.0 26.0 182.4 181.1 95.2 116.7 46.2 100.0 mg/pot................................ 37.0 63.9 21.7 112.9 144.5 43.9 32.4 27.3 Control ........................................... 83.8 119.2 63.4 171.2 277.9 164.0 242.6 102.6 Significance 2 ............ .. .. ... .. ... .. .. .. ... .. . c" c* q" q c" c" c" c" Spray 250 p.p.m ...................................... 77.2 127.9 65.8 164.3 232.3 162.5 248.3 96.3 500 p.p.m ....................................... 82.3 106.7 70.6 151.5 235.5 128.0 255.8 74.8 1,000 p.p.m ................................... 78.7 102.5 57.0 166.1 182.8 110.6 213.9 69.4 2,000 p.p.m ................................... 67.3 88.4 61.3 1,710.6 200.6 164.0 174.1 64.2 Control.. ........................... 83.8 119.2 63.4 171.2 277.9 92.9 242.6 102.6 Significance ..................................... l" l" ns ns q'" 1" c" q" Drench .............................................. 45.7 75.8' 30.9' 152.4 165.9* 83.5' 128.4 47.1" Spray ................................................. 76.4 106.4 63.7 163.1ns 212.8 123.5 223.0 76.2 'Significant or not significant (ns) at the 5 percent (*) or 1 percent (") level; q = quadratic, c = cubic; I = linear. Zero rate included in regression analysis. 2 Significant (') or not significant (ns) at the 5 percent level; zero rate not included in mean determination. [27] INSECT, DISEASE, AND WEED CONTROL Herbicide Effects on Rooting and Root Growth Mack Thetford, Charles H. Gilliam, and D. Joseph Eakes MANY GROWERS propagate cuttings by placing the cuttings in small containers (rose pots) under mist in greenhouses or outdoor groundbeds. Weed control in these areas is a recurring problem currently addressed by hand weeding. Use of herbicides to control weeds in these areas would be beneficial; however, information is limited on how herbicides affect rooting and sub- sequent root growth of woody cuttings. Previous work has demonstrated that suppressed rooting and lower root quality occurred when herbicides were broadcast over the top of cuttings during propagation. Defoliation of some species was also observed. In commercial production, cuttings are stuck directly into individual pots, pots are filled with media, placed in flats, and the flats are moved to the propagation house 1-2 days prior to sticking the cuttings. During this time, the pots are watered to thoroughly wet the medium. Preemergence application of herbicides to the pots before sticking the cuttings may avoid the direct her- bicide injury previously reported. The objective of this study was to determine if pre-propagation application of preemergence applied herbicides would affect root- ing and subsequent root growth of selected woody plants. Experiment 1 was initiated in November 1986, when 3-inch-square pots were filled with a pinebark and sand medium (1:1, by volume) amended per cubic yard with 6 pounds Osmocote? 18-6-12, 3 pounds dolomitic lime- stone, and 1.5 pounds Micromax?. Treatments included five herbicides and a nontreated control. The following four herbicides were applied at 3.0 pounds a.i. per acre: Rout 3G?, OH-2 3G, Surflan 4AS?, and Prowl 4L?. Ronstar 2G? was applied at 4.0 pounds per acre. These herbicides were selected for evaluation because (1) Rout, OH-2, and Ronstar are the three major herbicides used in container production, (2) Surflan is one of the most widely used herbicides in ornamental crops, and (3) Prowl has recently received an ornamental label. Granular formulations were applied with a hand-held shaker; Surflan and Prowl were applied with a CO 2 sprayer at 29 p.s.i. with 20 gallons of water per acre. All herbicides were applied to the medium prior to sticking cuttings. The base of each cutting was dipped in a 5,000 p.p.m. K-IBA solution prior to sticking and placed under intermittent mist (2.5 seconds per 5 min- utes from 8 a.m. to 5 p.m.) in a 50 percent shade glass house. Two species, Ilex X attenuata Fosteri (Foster holly) and Juniperus borizontalis Wiltoni (Blue Rug juniper) were used. Data collected included rooting percentage, root number, and root rating scale: 1 = distorted roots < 0.4 inch, 2 = 1-4 roots 0.4-2.0 inches long, 3 = 5-10 roots 0.4-2.0 inches long, 4 = 10- 15 roots 0.4-2.0 inches long, 5 = 16+ roots 0.4-2.0 inches long. Experiment 2 was initiated in June 1987. Species used were: Abelia X grandiflora (abelia), Buxus mi- crophylla Koreana (Korean boxwood), and Ilex crenata Compacta (Compacta holly). Procedures and treatments were identical to those in experiment 1. Rooting data were collected 8 weeks later (August), and growth indices and root density ratings were measured in De- cember 1987 [4 months after potting (4 MAP)], March 1988 (7 MAP), and September 1988 (13 MAP). Cuttings were maintained in the greenhouse until May 1988 in the original 3-inch pots when they were potted into 1-gallon containers using the medium previously de- scribed, except that the Osmocote rate was 14 pounds per cubic yard, and moved to an outdoor container- growing area. The root density rating scale was: 1 = no roots visible on outer edge of rootball, 3 = root de- velopment to outer edge of rootball, and 5 = total root coverage of rootball. Due to a lack of roots on the outer edge of the rootball in September 1988, root fresh weights were collected on the Japanese holly. In experiment 1, rooting percentage, root number, and root ratings of Blue Rug juniper were not affected by any herbicide treatment. In contrast, rooting per- centage (3 percent vs. 36 percent) and root ratings (1.0 vs. 2.3) of Foster holly were severely suppressed with Surflan compared to nontreated controls. None of the other herbicide treatments affected Foster holly rooting percentage or root quality. In experiment 2, Korean boxwood rooting percent- age and root ratings 8 weeks after sticking were sup- pressed with Surflan compared to all other treatments, table 31. Growth index measurements indicated shoot growth suppression continued up to 14 MAP with Sur- flan, compared with the nontreated control. Korean boxwood root growth suppression also continued throughout the test with Surflan. These data support previous findings that reduced root growth of Japanese holly liners occurred with Surflan-treated plants. There were no measurable effects with the other treatments until 13 MAP, when root density ratings were suppressed with Rout, OH-2, and Prowl. These data indicate po- tential long-term effects from using products containing dinitroaniline herbicides (Rout and OH-2) in the prop- agation of woody plants. Both Rout and OH-2 are com- monly used in commercial production of container nursery crops and producers may consider their use in propagation since these products would be available at the nursery. Ronstar and nontreated plants were sim- [28] TABLE 31. ROOT AND SHOOT GROWTH RESPONSE OF KOREAN BOXWOOD PROPAGATED IN HERBICIDE-TREATED MEDIUM, EXPERIMENT 2 Results, by months after potting Treatment/rate Rooting 1 Root Growth index 3 Root density per acre rating 2 rating 4 4 7 14 4 14 Pct. Rout, 3.0 lb. .................... 82a 2.7a 10.0abc 20.2ab 19.5abc 4.0a 2.3bc OH-2, 3.0 lb .................... 92a 2.6a 10.0abc 20.6ab 20.0ab 4.0a 2.6b Ronstar, 4.0 lb. ................... 92a 2.8a 10.2abc 20.8ab 21.4a 3.9ab 2.8ab Surflan, 3.0 lb. .................. 75b 1.2b 9.2c 17.4b 17.6c 3.7b 2.0c Prowl, 3.0 lb ..................... 83a 2.7a 10.9ab 21.9a 21.6a 4.1a 2.5bc Nontreated ......................... 90a 3.1a 11.3a 20.1a 20.5ab 4.2a 3.2a 'Rooting percentage measured August 1987, 8 weeks after sticking. 2 Root rating scale: 1 = distorted roots < 1 cm, 2 = 1-4 roots 1-5 cm, 3 = 5-10 roots 1-5 cm, 4 = 10-15 roots 1-5 cm, 5 = 16+ roots 1- 5 cm; August 10, 1987. 3Growth index: (height + width at widest point + width perpendicular to widest point)/3. 4 Root density rating scale: 1 = no roots visible on outer edge of rootball, 3 = root development to outer edge of rootball, 5 = total root coverage of rootball. 'Mean separation within columns by Duncan's multiple range test (5 percent level). ilar throughout the test with respect to all data col- lected. Compacta holly rooting percentage was suppressed with Surflan and initial root ratings were lower with Surflan and Rout. Root density ratings 4 MAP showed that all treatments resulted in less root growth, com- pared with nontreated plants. At the end of the study (13 MAP), root fresh weights were less with Surflan than with the nontreated control (57.6 vs. 94.9 grams); shoot fresh weight was not affected by Surflan appli- cation. All other treatments had similar root and shoot fresh weight, compared with the nontreated plants. Initial abelia root ratings were lower with Rout and Surflan, while rooting percentage was lower with OH-2; however, at 4 MAP, root and shoot growth were similar among all treatments. This study demonstrates several important consid- erations when using preemergence-applied herbicides in propagation of woody plants. First, plant response varies with herbicide application. In experiment 1, Blue Rug juniper rooting percentage was not affected by any preemergence herbicide application, while in experi- ment 2, both Korean boxwood and Compacta holly rooting percentage was less with Surflan. The results also demonstrate that herbicide appli- cation may inhibit root growth for varying lengths of time depending on the plant species. In experiment 2, abelia root growth was suppressed initially with Rout and Surflan; however, by 4 MAP, treatment effects on abelia root growth were not significant. With Korean boxwood, root suppression continued for 13 MAP. Din- itroanaline herbicides (Surflan and Prowl) were in- volved in all cases of extended suppression of root growth. Since root inhibition of annual crops is a major characteristic of dinitroanaline herbicides and in con- sideration of previous work with woody plants, strong evidence is provided against the use of Surflan or Prowl and their combination products during propagation. Thetford is former Graduate Student, Gilliam is Professor, and Eakes is Assistant Professor of Horticulture. Landscape Fabrics Suppress Growth of Weed Species Chris A. Martin, Charles H. Gilliam, and Harry G. Ponder MULCHING LANDSCAPE planting beds may enhance aesthetic value, reduce soil temperature fluctuations, and increase resistance to weed pressure. Laying plastic underneath an organic mulch is a common landscape installation practice. Previous work evaluating mulch- ing with pine bark over black polyethylene concluded that the presence of black polyethylene made little difference in controlling weed growth, while winter kill of landscape plants was increased by 60 percent compared with bark mulch alone. An alternative to black polyethylene is polypropylene fabrics now on the market. Two types of fabrics available are the woven and nonwoven polypropylene polymers. Polymers of nonwoven fabrics are spun-bound or meshed, while polymers of woven fabrics are generally of thicker diameter and tightly woven. Most polypro- pylene polymers are readily oxidized when exposed to ultraviolet light and are limited for use as an undercover supplemental mulch under a surface component such as pine bark. However, some fabrics, such as the Dewitt Pro 5 Weed Barrier? (woven), are surface-coated with carbon black, conferring a degree of resistance to deg- radation from ultraviolet light and therefore may be used singly as a mulching material. Unlike polyethylene sheets, polypropylene landscape fabrics are permeable to water. This is an advantage in facilitating improved soil aeration and reducing root growth at the plastic-soil interface. However, the prin- cipal purpose of mulching is to suppress weeds. This study was designed to compare the effectiveness of several polypropylene landscape fabrics in suppressing the emergence and growth of selected weed species. Two separate 30-day experiments were conducted in an unshaded double polycovered greenhouse with a daily temperature range of 66* to 88*F. Rectangular plastic flats 11 X 20.5 inches were filled with pine [29] TABLE 32. AVERAGE COUNT (CNT) AND SHOOT DRY WEIGHT (SDW) PER FLAT OF WEEDS PENETRATING up THROUGH Six LANDSCAPE MATS 30 DAYS AFTER SOWING, EXPERIMENT 1 Yellow nutsedge Bermudagrass Johnsongrass Pigweed Sicklepod mriglor Mat type___ __ _ __ _ _ __ CNT SDW CNT SDW CNT SDW CNT SDW CNT SDW CNT SDW No. Grams No. Grams No. Grams No. Grams No. Grams No. Grams Dewitt Landscape Fabric .. 1.3ab' 1.5a 0.Oa 0.Oa 0.8a 0.7a 0.Oa O.Oa 0.Oa 0.Oa 0.Oa 0.Oa Geoscape Landscape Fabric -3a .1a 7.Oab 3-7b 1.7a 11.4ab .Oa .Oa .Oa .Oa .Oa .Oa Amoco Rit-a-Weed ........... 2.5bc .2a 11-3b 7-3b 7-5a 40.7c 5.3ab 7-5b .Oa .Oa .Oa .Oa Phillips Fibers Duon ........ 1.3ab .4a 13.8b 5.2b 8.Oa 42.6c 12-5b 9.6b .Oa .Oa .Oa .Oa Weedblock Fabric............ 3.8c 9.8b 92.3c 22-3c 5.3a 31.7bc 110-7c 24.7c .Oa .Oa .Oa .Oa Control....................... 11.8d 44.5c 126.1c .25.9c 2 1.0b 70-3d 139.Oc 27.2c 3 3-3b 18-7b 15-5b 13.2b 'Mean separation within columns by Duncan's multiple range, 5 percent level. bark:sand:sandy loam (1: 1: 1, by volume) amended with 2 pounds of Osmocote? 18-6-12 per cubic yard. Two weed species were sown separately on op posite halves of each flat, covered with vermiculite, and irrigated. After 2 days, flats were covered with polypropylene fabric with overlapping fabric tucked underneath the flats. A 2-inch-deep mulch of landscape bark nuggets was placed on top of all flats to simulate typical land- scape practices. At the end of 30 days, weeds emerging through the fabric were counted. Shoots were subse- quently clipped at the fabric surface, dried, and weighed. In experiment 1, the following six weed species were used: pigweed (Aranthus sp.); sicklepod (Cassia obtusifolia); bermudagrass (Cynodon dactylon); yel- low nutsedge (Cyperus esculentus); small flower mor- ningglory (Jacquemontia tamrnfolia); and johnsongrass (Sorghum halepense). Approximately 0 .3 tablespoon of weed seed were sown per half flat for all species, except for yellow nutsedge where 15 tubers were planted in each flat. The five polypropylene landscape fabrics used to cover the flats included: Dewitt Weed Barrier? woven; Amoco Rit-a-Weed? heavy-meshed nonwoven; Phillips Fiber Duon? 2.5-ounce meshed nonwoven; Geoscape Landscape Fabric? meshed nonwoven; and Weedblock Fabric? perforated-polyethylene nonwoven. The control treatment consisted of flats without fabric coverings. In experiment 2, the same weed species were used, except sicklepod and smallflower morningglory were omitted. These two species were completely inhibited by all fabrics in experiment 1. Weed seeds were sown or planted as in experiment 1. Polypropylene fabrics used included five fabrics listed previously plus three additional fabrics: Weed Barrier MatO woven; Dupont Typar 307? spunbound nonwoven; and Typar 312? spun-bound nonwoven. The control treatment consisted of flats without mat coverings. In experiment 1, shoot emergence of sicklepoctand smallflower morningglory, both species with broadleaf cotyledons, was completely suppressed by all landscape fabrics, table 32. Shoot emergence of pigweed was completely suppressed by the Dewitt and Geoscape Landscape fabrics, while Amoco Rit-a-Weed and Phillips Fibers Duon provided partial suppression, compared with the control.- Shoot emergence of bermudagrass was completely suppressed by Dewitt, whereas bermuda- grass shoots emerged through the Weedblock fabric in numbers equivalent to the control. All other treatments provided similar control of bermudagrass shoot growth. Johnsongrass shoot emergence was suppressed by all treatments, but the greatest suppression was achieved with the Dewitt and Geoscape fabrics. Johnsongrass growth after emergence through all fabrics was altered in that aerial adventitious roots developed immediately above the fabric surface, while not developing the caliper of the portion of stem which initially emerged through the landscape fabric. These weeds were easily rouged by hand. All landscape fabrics partially suppressed the emerg- ence of yellow nutsedge compared to the control; how- ever, yellow nutsedge dry weights were greater with TABLE 33. AVERAGE COUNT (CNT) AND SHOOT DRY WEIGHT (SDW) PER FLAT OF WEEDS PENETRATING up THROUGH NINE LANDSCAPE MATS 30 DAYS AFTER SOWING, EXPERIMENT 2 Mat type Yellow nutsedge Pigweed Bermudagrass Johnsongrass Ma yeCNT SDW CNT SDW CNT SDW CNT SDW No. Grams No. Grams No. Grams No. Grams Dupont Typar 307 .......... 0.Oa'0.Oa .Oa 2.5a 2.9ab 2.9ab 0.0a O.Oa Dupont Typar 312.............. Oa .Oa .3a .8a .9a .9a .Oa .Oa Weed Barrier Mat...................3ab .4ab Oa Oa .Oa Oa 3a3a Dewitt Pro 5 ................... .Oab 3.6ab 1.9a 1.5a 3.Oab 3.Oab 2.3a 5.8ab Geoscape Landscape Fabric .... .ab 4.6b .Oa 9-3a 4.4b 4.4b .3a 4.3ab Amoco Rit-a-Weed ........... 2.5b 4.Ob 5.ob 17.5a 5.2b 5.2b 7.Ob 9.7bc Phillips Duon Fiber..........33b 3.6ab 6.6b 12.Oa 5.lb 5.lb 65b 13.4cd Weedblock Fabric ............1.8ab 3.8ab 12.c 72.6b 10.3c 10.3c 8.Ob 10.4bc Control.......................... 8.3c 23-1c 13.2c 11 2.8c 12.8c 12.8c 12.8c 19-7d 'Mean separation within columns by Duncan's multiple range test, 5 percent level. [301 the Weedblock Fabric compared to the other landscape fabrics and less than the control, table 32. The elasticity of the Weedblock fabric, a polyethylene derivative, appeared greater than the other polypropylene fabrics and may have facilitated yellow nutsedge penetration. In experiment 2, shoot emergence of yellow nutsedge and johnsongrass was completely suppressed by the spun-bound nonwoven fabrics from Dupont, table 33, although etiolated growth of yellow nutsedge was ob- served underneath the Dupont landscape fabrics. All treatments provided partial control of yellow nutsedge compared with the control. Johnsongrass dry weights with Phillips Fiber Duon were similar to the control; however, the number of weeds emerging was reduced, suggesting that weeds that penetrated the fabric were able to grow to a larger size. Pigweed was completely suppressed by Typar 307, Geoscape Landscape Fabric, and the Weed Barrier Mat, and growth was inhibited by Typar 312 and Dewitt Pro 5, table 33. Bermudagrass was best controlled by the Dupont Typar fabrics, the Weed Barrier Mat, and Dewitt Pro 5, while bermudagrass growth with the Weedblock Fabric was similar to the control. Use of landscape fabrics by American Woven Fabrics, Dewitt Co., Dupont Corp., and Innovative Geotextile resulted in the best overall suppression of emergence and growth inhibition of the weed species tested. Re- sults indicate that weed suppression may not be cor- related to polypropylene polymer type; however, weed suppression using spun-bound nonwoven fabrics was superior to meshed nonwoven fabrics. This research would also tend to agree with earlier findings that some hand weeding and herbicide application may be nec- essary when landscape fabrics are used. Martin is former Graduate Student, Gilliam is Professor, and Ponder is Professor of Horticulture. Weed Control In Field-Grown Holly Glenn R. Wehtje and Charles H. Gilliam WEED CONTROL in field-grown nursery crops and landscape beds generally requires the application of herbicide combinations, as well as multiple applica- tions of these combinations during a growing season. In Maryland, herbicides applied twice per season to field-grown azaleas controlled greater than 99 percent of the weeds at the end of the growing season. In contrast, a single application provided only 63-77 per- cent weed control. In Illinois, Roundup? applied pre- plant, followed by Princep? alone or in combination with either Surflan?, Enide?, Lasso?, or Devrinol?, re- sulted in season-long weed control in field-grown nurs- ery crops. In Connecticut, November applications of Princep alone at 1.5 to 2.5 pounds a.i. per acre or at 1.0 pound per acre when in combination with other herbicides provided over 95 percent control of weeds as rated in late May of the following year. Information on the use of herbicides in nursery crop production in the Southeastern United States is limited. Compared to northern locations, achieving satisfactory weed control is more difficult in the Southeast because the growing season is longer and the warmer soil tem- perature may hasten herbicide dissipation. Previous work in Alabama evaluated several herbicide combinations applied twice annually and reported that Surflan + Princep at 2.0 + 0.75 pounds a.i. per acre provided acceptable weed control in field-grown photinia and boxwood. The first objective of this research was to examine how time of Surflan + Princep application affected weed control in field-grown flex. Goal? has been demonstrated to provide excellent weed control in selected woody crops. The 2 percent granular formulation has been evaluated extensively in container production and reported to be safe to a wide range of woody plants. Comparison of the 2E and 2G formulations has shown the 2E formulation to be sig- nificantly more injurious to woody plants. Other work has reported injury from application of Goal 2E to 3 of 13 plant species, however three Ilex crenata cultivars were not injured. Previous work has showed mid-sum- mer application of Goal to be noninjurious when ap- plied over the top of four field-grown woody plants. This lack of injury was attributed to the semi-dormant state that plants enter during an environmentally stress- ful period of the year. The second objective was to evaluate multiple applications of Goal on three species of field-grown Ilex for efficacy and weed control. Two experiments with separate objectives were con- ducted simultaneously. The first examined the effect of herbicide application timing, and the second evaluated the performance of various herbicide combinations. Three Ilex cultivars were chosen for these experi- ments because of large acreage plantings in north Ala- bama. Uniform liners of: I. x meserveae China Girl, I. x aquipernyi San Jose, and I. aquifolium x cornuta Nellie R. Stevens were planted on March 12, 1985, in a Hartsells fine sandy loam soil at the Sand Mountain Substation, Crossville, Alabama. The research area was uniformly infested with large crabgrass (Digitaria san- quinalis), entire leaf morningglory (Ipomoea heder- acea), and prickly sida (Sida soinosa). Herbicides were applied with a tractor-mounted, compressed-air boom-type sprayer operating at 32 p.s.i. and delivering 15 gallons of water per acre. Initial herbicide treatments were applied over the top 2 weeks after the March 28 planting. Treatments were reapplied according to the pertinent treatment schedule. Data collected included crop injury (0 = no effect, 100 = death) and percentage weed control (0 = no control, 100 = total control); hand hoeing times were determined in mid-July just prior to the second appli- cation and in early October at the end of the growing season. Plots were uniformly weed-free prior to the second application. Growth indices [(height + width + width)/3] were taken in early October each year. Both experiments had two control treatments. The first was hand weeded in July just prior to the second herbicide application, and again at the end of the grow- [31] TABLE 34. WEED CONTROL AS INFLUENCED BY TIMING OF SURFLAN + PRINCEP' (2.0 + 0.75 POUNDS PER ACRE) TO THREE FIELD-GROWN ILEX, EXPERIMENT 1 Broadleaf weed control Grass weed control Timeofn 1985 1986 1987 1985 1986 1987 application 7/18 9/19 7/15 10/7 7/1 9/20 7/18 9/19 7/15 10/7 7/1 9/30 Pct. Pct. Pct. Pct. Pct. Pct. Pct. Pct. Pct. Pct. Pct. Pct. March ............................. 78ab 2 78bc 80ab 88abc 100a 95ab 74a 71b 100a 100a 98a 100a Mar.-July ................ 76ab 90a 84ab 97ab 100a 100a 80a 96a 100a 100a 100a 100a Mar.-July-Nov ................. 84a 88a 93a 99a 100a 99a 84a 97a 100a 100a 100a 100a Mar.-Nov ........................ 70ab 68bc 93a 78c 100a 95ab 90a 81b 100a 100a 96a 100a July-Nov ........................ Oc 85ab 60bc 95ab 100a 100a Oc 90a 85a 99a 100a 100a Hand weeded ............. 63b 64c 49c 84bc 36b 90b 20b 70b 24b 58b 88b 30b Nonweeded ................... Oc Od Od Od Oc Oc Oc Oc Oc Oc Oc Oc 'Rates were 2.0 and 0.75 pounds per acre. 2 Means within columns followed by the same letter are not significantly different at the 5 percent level as determined by Duncan's multiple range test. ing season in early October. The second control was hand weeded only at the end of the growing season. Experiment 1 Surflan? + Princep? at 2.0 + 0.75 pounds a.i. per acre were applie applied with five treatment schedules: March; March + July; March + July + November; March + November; and July + November, table 34. In 1985, maximum season-long broadleaf weed con- trol was obtained with all treatments that contained a minimum of two Surflan + Princep applications that included the March + July treatments. An additional herbicide application in November (i.e., a total of three applications) offered no further improvement in broad- leaf weed control. Other treatments that had two ap- plications (i.e., March + November orJuly + November) also provided maximum control at one or more of the rating periods. However, only the March + July appli- cation obtained maximum control for the duration of the growing season. Broadleaf weed control in 1986 and 1987 exhibited a similar trend to that which was established in 1985, table 34. However, treatments that were marginal in performance in 1985 tended to be more effective in 1986 and 1987. This may be attributable to increased crop competition and residual herbicide effects. These data also show that during the first year in production, when crop competition is minimal, maximum herbicide inputs are necessary. Furthermore, the data indicate that during the second and subsequent years, herbicide inputs may be decreased. Grass control followed a similar pattern to that ob- served with broadleaf control. In 1985, two applications of Surflan + Princep resulted in maximum grass control. A third application of Surflan + Princep in November did not improve control when compared to the March + July application. In all years, all herbicide treatments were effective in providing the maximum level of grass control. Hoeing times were generally reflective of weed con- trol ratings, table 35. Application of Surflan + Princep in March + July resulted in minimum hoeing times. Hoeing time, as averaged over all treatments, decreased each year. This trend probably reflects increased crop competition and residual herbicide activity. Maximum growth indices of China Girl holly oc- curred when Surflan at 2.0 pounds and Princep at 0.75 pound per acre were applied in March. Applying Surflan + Princep in July + November resulted in reduced growth indices throughout the study. These data dem- onstrate the importance of early season weed control in reducing weed crop competition. Similar plant size among March-based treatments during the third year tends to confirm that if adequate weed control practices are maintained, fewer herbicide applications are nec- essary in the later crop years. Experiment 2 Six different herbicide combinations were evaluated under a common schedule (March + July), table 35. Across all years, all herbicide combinations provided equivalent maximum levels of both broadleaf and grass weed control. No differences between individual treat- ments were detected. The only exception was Goal + Princep (low rate), in which grass control at the first TABLE 35. TIME REQUIRED TO REMOVE REMAINING WEEDS FROM PLOTS AT THE END OF THE GROWING SEASON, EARLY OCTOBER Time required/plot, by year Treatment 1985 1986 1987 Min. Min. Min. Exp. 1: Timing of application' March ................. .......................... 7.8b 2 2.2bc 0.6 Mar.-July .. . ....................... 2.9c .4c .2c Mar.-July-Nov ................................. 2.6c .2c .2c Mar.-Nov ........................................ 6.5b 3.Obc .7c July-Nov . ....................................... 4.1bc .9c .3c Hand weeded ................................ 10.7ab 7.7b 5.2b Nonweeded .................................. 29.1a 38.2a 11.1a Exp. 2: Herbicide combinations Surflan + Princep (2.0 + 0.75 lb.)... 2.9c 0.4c .2c Surflan + Princep (3.0 + 1.0 lb.) ..... 1.1c .lc .2c Goal + Princep (1.0 + 0.75 lb.) ....... 1.5c .4c .3c Goal + Princep (2.0 + 0.75 lb.) ...... 1.6c .0c .3c Surflan + Goal (2.0 + 1.0 lb.) ......... 2.5c .0c .3c Surflan + Goal (2.0 + 1.0 lb.) ......... 1.3c .lc .2c Handweeded ... ...................... 10.7b 7.7c 5.2b Nonweeded ....................................... 29.1a 38.2a 11.1a 'Surflan + Princep, 2.0 + 0.75 pounds/acre. 2 Means within columns followed by the same letter or letters are not significantly different at the 5 percent level as determined by Duncan's multiple range test. 3 Applied March-July. [32] rating in 1985 was less than the other treatments. This was expected since both of these herbicides are pri- marily active against broadleaf species. No crop injury resulted from the mid-season appli- cations of Goal, which agrees with earlier results. lex typically exhibits episodic growth patterns character- ized by periodic stem elongation. Less than ideal con- ditions (nonirrigated field conditions during the summer months) may reduce shoot growth until satisfactory conditions prevail. Elongated stems harden off and re- main in that condition until favorable conditions occur. The application of Goal-containing herbicide combi- nations during this summer dormant period (July) caused no injury over the 3-year test to the three field-grown Ilex. Other woody crops have responded similarly in the Southeast. Data from both experiments indicate that application timing is of greater importance than herbicide selection. Provided the combination contains both a broadleaf and grass-active herbicide and rates are sufficient, acceptable weed control can be obtained. The critical factor was time of application. Nope of the herbicides evaluated had sufficient residual activity to provide control for the entire season with a single application. An initial application in March served to eliminate much of the weed competition during spring growth. A second ap- plication extended control through the remainder of the season. These data also suggest that chemical weed control needs are greatest during the first year following planting. Wehtje is Associate Professor of Agronomy and Soils; Gilliam is Professor of Horticulture. Jumping Tree Bugs Are Important Natural Enemies of Obscure Scale Infesting Landscape Plantings of Pin Oak Michael L. Williams and Harlan J. Hendricks THE PIN OAK, Quercus palustris, is one of the most widely used landscape trees on home grounds, parks, golf courses, and street borders in Alabama. Obscure scale, Melanaspis obscura, is the primary insect of pin oak, causing rapid decline in plantings in the South, photo 8 (page 18). A survey of the Auburn University campus, where pin oak is one of the most common shade trees, revealed that more than 90 percent of 305 pin oaks were infested with obscure scale. Obscure scale infestations seldom kill the tree, but can cause extensive dieback of branches and make the tree more susceptible to secondary infestation by other insects and diseases. Pesticide sprays applied to kill the obscure scale have generally provedl unsuccessful and may have a detrimental effect on ' its natural enemies as well. Biologically, there are several predators and parasites that have an important impact on obscure scale pop- ulations. In a study of the bionomics of obscure scale and its natural enemies on the Auburn Campus during 1985-88, 21 species of insects and mites were observed preying on or parasitizing obscure scale. The most important predators were four species of predaceous plant bugs in the family Miridae. Included were the jumping tree bugs Corticoris signatus (Heidemann), Lidopus heidemanni Gibson, photo 9 (page 18), and Myiomma cixiiforme (uhler). Jumping tree bugs are predaceous plant bugs that are difficult to detect be- cause of their small size, cryptic coloration, and secre- tive habits. These three species were the most frequently encountered predators and were observed on all areas of infested trees sampled, particularly in cracks and crevices of the trunk. Immature stages were more com- mon on the undersides of branch bases. Both adults and immature stages of all three species were observed feeding on obscure scale. The active searching behavior, aggressive feeding habits in the laboratory, and large populations of these jumping tree bugs associated with obscure scale in this study suggest that these predators play an important role in regulating obscure scale populations. Feeding observations and those conducted in a similar study in Pennsylvania strongly suggest that jumping tree bugs may well be specialized scale feeders. Knowing and understanding the natural enemy complex of a pest species are of utmost importance in managing a pest population, particularly when chemical controls are ineffective. Williams is Associate Professor and Hendricks is former Graduate Student of Entomology. Evaluation of Labeled Miticides for Control of Southern Red Mite on Azaleas Michael L. Williams, James C. Stephenson, and Gary L. Miller A PERSISTENT nursery pest of azaleas is the southern red mite, Oligonychus ilicius (McGregor). Although this mite can be found throughout the year, greater numbers are found during cooler weather when dam- aging populations can build up quickly. Mite popula- tions often fluctuate dramatically on their host. This fluctuation may be due in part to the host's poor foliage condition during the cooler part of the year. This test was initiated as a single curative treatment to determine the efficacy of many currently labeled miticides. Eight pesticides were selected for application on mite-infested White Gumpo azaleas. Plants were grown in trade gallon plastic containers with amended pine bark-peat moss medium. Miticides were applied to up- per and lower leaf surfaces using a CO 2 pressurized test plot sprayer at 35 p.s.i. No spray adjuvant was included in the test. Four single-plant replicates were [33] TABLE 36. CONTROL OF SOUTHERN RED MITE ON RHODODENDRON X WHITE GUMPO Mean no. of mites/sample, Treatment, rate/100 gal. 1 by days after treatment 2 3 7 21 Talstar 10WP?, 1 lb. ................... 7.8a 1.3a Vendex 50WP?, 1 lb. ................ 12.5a 1.0a Avid 0.15 EC?, 4 fl. oz. .............. 39.5b 21.Obc Morestan 25WP?, 1 lb ............... 10.3a .0a Pentac Aquaflow 4FO, 16 fi. oz ..... 5.a .3a Mavrik 2F?, 10 fi. oz ................... Oa .Oa Omite 30W?, 1 lb. ..................... 8.5a 1.3a Safer's Ins. Soape, 128 fi. oz. ........ 12.0a 8.5ab Untreated check ......................... 48.5b 31.Oc 'Treatments applied January 18, 1989. Dry bulb 58.5?F, wet bulb 560F. 2 Evaluation January 25 and February 8, 1989. 3 Values followed by the same letter are not significantly different at the 5 percent level by Duncan's multiple range test. randomized and held in an open, unheated glass green- house for the duration of the study. Evaluations were conducted 7 and 21 days after treatment. Seven days after treatment, all materials except Avid? provided significant control when compared to the un- treated check, table 36. Even 21 days after treatment, excellent control was observed with most miticides when compared with the untreated check. As with the 7-day evaluation, Avid did not provide good control. Most of the labeled miticides evaluated in this test provided adequate control of southern red mite pop- ulations. Because this mite is often found on the un- derside of foliage, it is most important to achieve full coverage of the leaf surface. It is also good policy to rotate various compounds in a spray program to reduce chances for pesticide resistance. Williams is Associate Professor of Entomology; Stephenson is As- sociate Superintendent of the Ornamental Horticulture Substation; Miller is Research Associate of Entomology. Chemical Control of Powdery Mildew on Miniature Roses Austin K. Hagan and John Olive POWDERY MILDEW, caused by the fungus Sphaero- theca pannosa, is a common disease on field- and greenhouse-grown roses. In Alabama, this disease is particularly troublesome on greenhouse-grown minia- ture roses. Although the damage is largely cosmetic, powdery mildew-damaged miniature roses often appear unsightly and may be unmarketable. In this study, the efficacy of Spotless fungicide for the control of powdery mildew on miniature rose was compared with that of several registered fungicides. Red miniature rose (Rosa spp.) liners were planted in a pine bark:peat moss medium (3:1, by volume) amended with 14 pounds of 17-7-12 Osmocote?, 6 pounds of dolomitic limestone, 2 pounds of gypsum, and 1.5 pounds of Micromax? per cubic yard of media. Plants were maintained in the greenhouse and watered daily using drip irrigation. Disease incidence, as indi- cated by an estimate of the percentage of diseased shoot tips, and canopy height and width were recorded on January 3. The growth index was calculated as follows: (height + width 1 + width 2)/3. Widths were meas- ured perpendicular to each other. In 1989, the fungicide Spotless 25W at 0.2, 0.3, and 0.4 pound per 100 gallons of water was compared with Benlate 50DF? at 0.25 pound per 100 gallons of water, Bayleton 25W? at 0.25 pound per 100 gallons of water, and Rubigan A.S. 1 .OE at 8 fluid ounces per 100 gallons of water for the control of powdery mildew on miniature roses. All rates of Spotless were applied with and with- out 1 pint of Agrioil AG-98? crop oil concentrate sur- factant. Treatments were applied with a CO 2 pressurized sprayer every 2 weeks from October 4 to December 12. TABLE 37. POWDERY MILDEW CONTROL OF MINIATRE ROSES FROM DIFFERENT FUNGICIDE TREATMENTS, 1989 Treatment, rate/10 gal. Growth Disease Treatment, rate/00 gal. index incidence Pct. Spotless 25W 0.2 lb ....... ........................... 35.2 35.5 0.3 lb. ....................................... 31.6 4.0 0.4 lb. ....................................... 30.5 2.0 Spotless 25W + Ag-98 (0.5% volume) 0.2 lb ....................................... 19 .4 .0 0.3 lb ....... ........................... 22.4 .0 0.4 lb ....... ........................... 18.4 .0 Benlate 50W 0.25 lb .... ......................... 31.2 28.1 Bayleton 25W 0.025 lb. ........ .... ............ 28.7 37.9 Rubigan A.S. 1.0 E 8 ft. oz. ......... ................ 33.0 3.6 Non-sprayed control ..................... 33.4 60.9 'Growth index = (height + width 1 + width 2)/3. All fungicides reduced disease incidence, but con- siderable differences in controlling powdery mildew on rose were seen among the fungicide treatments, table 37. Spotless and Rubigan generally gave the best disease control. Only the lowest rate of Spotless applied without the AG-98 surfactant failed to provide good disease control. Addition of AG-98 improved the effectiveness of the lowest rate of Spotless against powdery mildew, but not the two higher rates of the same fungicide. Benlate and Bayleton reduced disease incidence, but were not as effective in controlling powdery mildew as Rubigan and the two higher rates of Spotless. Although some improvement in powdery mildew control was obtained by adding AG-98 surfactant to the lowest rate of Spotless, 40-50 percent reductions in shoot growth were associated with the use of this tank- mix combination. Brown necrotic spots were also seen on the leaves of the roses sprayed with all three Spotless [34] + Agrioil AG-98 tank-mix combinations. Surprisingly, the surfactant-free spray mixes of Spotless had no effect on plant growth, nor did any of the other fungicides screened. Hagan is Associate Professor of Plant Pathology; Olive is Superin- tendent of the Ornamental Horticulture Substation. Effect of Nova Fungicide on Entomosporium Leaf Spot on Photinia Austin K. Hagan, John Olive, and William J. Foster ENTOMOSPORIUM LEAF SPOT is a destructive disease on red tip photinia in production and landscape settings across Alabama. Control of existing disease outbreaks often depends on an intensive fungicide spray program. Previous Alabama Agricultural Experiment Station trials have clearly shown that the registered fungicides Da- conil*, Zyban*, and Triforine? will give good control of this disease. The objective of this trial was to de- termine whether the experimental fungicide Nova would provide better control of Entomosporium leaf spot on photinia than registered fungicides. Healthy red tip photinia (Photinia fraseri Birming- ham) liners were potted in a pine bark medium amended with 12 pounds of 17-7-12 Osmocote?, 6 pounds of limestone, 2 pounds of gypsum, and 1.5 pounds of Micromax? per cubic yard in trade gallon containers. The plants were watered daily with overhead impact sprinklers. Inoculum pressure was maintained by plac- ing diseased photinia within the blocks of fungicide- treated plants. Nova 40W* fungicide was applied from May 9 to July 28 at the rates listed in table 38 at 1-, 2-, and 4-week intervals. Weekly applications of label rates of Daconil 2787 4.17F and Triforine 1.6E were also included. A Penetrator 30 spray adjuvant was added to all Nova tank-mixes. Disease incidence was evaluated using the Horsfall and Barratt rating system. Plant di- mensions were recorded on July 31. The growth index was calculated using the formula (height + width 1 + width 2)/3. All rates of Nova applied on a 1- and 2-week spray schedule gave excellent control of Entomosporium leaf spot, table 38. No spotting of the leaves was seen across all application rates at these spray schedules, except for plants treated every 2 weeks with the lowest rate of Nova. Disease incidence increased sharply as interval between applications of all rates of Nova was lengthened from 2 to 4 weeks. Applied weekly, all rates of Nova gave disease control similar to that obtained with Da- conil 2787. Weekly applications of the 12-fluid-ounce TABLE 38. IMPACT OF SPRAY SCHEDULE AND APPLICATION RATE ON THE CONTROL OF ENTOMOSPORIUM LEAF SPOT AND GROWTH OF PHOTINIA Fungicide rate/100 Disease Growth gal. water incidence' index 2 cm Nova 40W 2.5 oz. Sprayed weekly .................... 1.0 56.4 Sprayed every 2 weeks .............. 1.8 59.3 Sprayed every 4 weeks .............. 5.3 60.4 Nova 40W 5.0 oz. Sprayed weekly ......................... 1.0 49.6 Sprayed every 2 weeks ............... 1.0 50.7 Sprayed every 4 weeks..............3. 60.4 Nova 40W 10.0 oz. Sprayed weekly .......................... 1.0 40.1 Sprayed every 2 weeks.............. 1.0 62.4 Sprayed every 4 weeks .............. 4.1 54.9 Daconil 2787 4.1F 2 pt. Sprayed weekly .................... 1.3 57.8 Triforine 1.6E 12 fi. oz. Sprayed weekly .................... 4.1 57.0 Nonsprayed control .................. 7.1 58.8 'Disease incidence was measured on a scale of 1-12 (1 = no disease, 2 = 0-3 percent, 3 = 3-6 percent, 4 = 6-12 percent, 5 = 12-25 percent, 6 = 25-50 percent, 7 = 50-75 percent, 8 = 75-87 percent, 9 = 87-94 percent, and 12 = 100 percent of leaves diseased). 2 Growth index = (height + width 1 + width 2)/3. rate of Triforine failed to control Entomosporium leaf spot. Plant growth reductions were seen on photinia treated with the two higher rates of Nova applied on a 1- and/ or 2-week spray schedule. The 10-ounce rate applied weekly had the greatest impact on plant growth. The lowest rate of Nova, Daconil 2787, and Triforine did not have an adverse impact on plant growth. Nova fungicide across a range of application rates provided excellent control of Entomosporium leaf spot on photinia. Applied on a 1- or 2-week spray schedule, little if any disease was seen on the Nova-treated pho- tinia. The level of disease control with Nova was com- parable to that obtained with Daconil 2787 and superior to that from Triforine. Previous results have shown that Nova may prove more effective in controlling Ento- mosporium leaf spot under heavy disease pressure than Daconil 2787. Unfortunately, reductions in shoot growth were seen on photinia treated with high rates of Nova on short spray schedules. Similar reductions were not recorded in the previous Nova spray trial. Further work must be done to identify application rates of this fun- gicide that will give good disease control without ap- preciable reductions of plant growth. Hagan is Associate Professor of Plant Pathology; Olive is Superin- ticulture Substation. [35] Evaluation of Stirrup M as a Tank- Mix Addition to Miticides for Control of Spider Mites on Roses James C. Stephenson MITES CONTINUE to be a concern for many nursery- men in Alabama. The ineffectiveness of some com- pounds, plus problems of spray coverage in dense plant canopies, close plant spacing, and environmental con- ditions, complicates control measures. There are several mites which are economic pests in Alabama nurseries. The twospotted spider mite, Tetranycbus urticae Koch, continues to be one of the most serious and widespread. Damaging populations can build rapidly and under fa- vorable conditions take as- little as 5 days to develop from egg to adult. One recent approach to offset some spray coverage problems is to bring the pest into contact with a toxicant by the addition of a pheromone. A pheromone is a substrate produced by one individual that induces a specific reaction by others of the same species. It is a means of communication. In this case, Stirrup M?, a sex pheromone, reportedly results in mites moving around within the plant canopy out of the more shel- tered areas. This improves the chance and length of time a mite contacts the miticide. This new approach has been reported to be successful in the Western United States on fruit trees and in the desert Southwest on cotton. This study was initiated as a preliminary eval- uation of Stirrup M as a tank-mix addition to miticides for use on ornamentals. Mite-infested miniature roses growing in a glass greenhouse were selected for treatment. Good spray coverage on this plant is difficult due to its waxy leaf surface. Treatments were applied to run-off using a hand- held compressed air sprayer to upper leaf surfaces. No spray adjuvant was included in the first trial. Chevron X-77? at the rate of 8 fluid ounces per 100 gallons of water was added in the second trial. TABLE 39. EVALUATION OF STIRRUP M FOR SPIDER MITE CONTROL ON ROSES Mean number of mites/sample 2 , Treatment, rate/i00 gal. by days after treatment 7 21 Pentac Aquaflow?, 16 fl. oz. .......... 10a 8a Pentac Aquaflow + Stirrup M, 16 fl. oz. + 2 fl. oz ................. 38a 18ab Morestan 4F 8 f. oz .................... 5 ab 63bc Morestan 4F + Stirrup M, 8 fl. oz. + 2 ft. oz . ....................... 23a 4 abc Vendex 50 WP?, 1 lb .................. 13a 35ab Vendex 50 WP? + Stirrup M, 1 lb. + 2 fl. oz . ....................... 13a 41abc Stirrup M, 2 fl. oz ..................... 134c 87c Untreated check ...................... 91bc 66bc 'Treatments applied March 5, 1990. 2 Mean separation within columns by Duncan's multiple range test, P=0.05. In the first trial, there was a high degree of variability within treatments and an apparent lack of control. This prompted the addition of a spray adjuvant to insure better coverage of the leaf surfaces in the second trial. In the second trial, table 39, good control was ob- served with most miticide treatments compared to the untreated check 7 days after treatment. In this test, there was no advantage gained by adding Stirrup M to the spray tank. In a final evaluation 21 days after treat- ment, the same conclusion concerning Stirrup M was reached. In summary, these two trials indicate that no clear advantage is gained by adding Stirrup M to the spray tank for control of the twospotted spider mite on min- iature roses. However, there are other factors which affect pheromone use, such as rate, specificity, and environmental conditions, that must be addressed be- fore a definite conclusion can be reached. No incom- patibility or phytotoxicity was observed. Stephenson is Associate Superintendent of the Ornamental Horti- culture Substation.