SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Weed Control

James Altland Section Editor and Moderator

431 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Preemergence Control of Winter Weeds in Overwintering Ornamental Crops

Caren A. Judge, Joseph C. Neal and Robert E. Wooten Dept. of Horticultural Science, NC State University, Raleigh, NC 27695-7609

Index Words: annual bluegrass, birdseye pearlwort, common chickweed, common groundsel, hairy bittercress, spiny sowthistle, and yellow woodsorrel

Signifi cance to Industry: BroadStar, OH2, and Snapshot TG were compared for control of winter weeds in overwintering nursery crop production. BroadStar, OH2, and Snapshot TG controlled common chickweed, common groundsel, hairy bittercress, and spiny sowthistle. OH2 and Snapshot TG controlled yellow woodsorrel better than Broadstar, although control was greater than 86% with all treatments. Annual bluegrass and birdseye pearlwort were controlled with Snapshot TG; BroadStar and OH2 were less effective on these species. If annual bluegrass or birdseye pearlwort are expected in overwintering structures, Snapshot TG should be used before covering or a dinitroaniline herbicide may need to be included with BroadStar to improve the spectrum of weed control.

Nature of Work: The container nursery industry relies upon broad-spectrum preemergence herbicides and supplemental hand weeding for weed management (1). In the southeastern United States, weed management in container nursery crop production is a year-round endeavor due to the relatively mild winters and overwintering protection of crops. Overwintering structures moderate the temperatures and provide conditions favorable for germination and establishment of winter annual weeds. Weed management options are limited in overwintering structures because no preemergence herbicides are labeled for application within covered structures (3). Preemergence herbicides must be applied several weeks before plants are covered in late autumn. Weeds emerging in containers over the winter must be hand weeded, which is usually feasible only after the overwintering covers are removed in the spring. Few studies have investigated effi cacy of herbicides on weeds in container nursery crops in covered overwintering structures. Effective control of hairy bittercress (Cardamine hirsuta L.) and yellow woodsorrel (Oxalis stricta L.) was obtained from October and February applications of Regal O-O (oxyfl uorfen + oxadiazon) or Gallery (isoxaben) plus Ronstar (oxadiazon) (2). Yet, this trial was conducted outdoors, in large containers, not in a covered overwintering structure. For a more comprehensive evaluation of winter weed control in overwintering ornamental crops, experiments were conducted in 2002 and 2003 to evaluate the effi cacy of commonly used preemergence herbicides on several common winter nursery weeds in overwintering structures.

Broadstar 0.25G (fl umioxazin) at 0.42 kg ai/ha (0.38 lbs ai/A), Scott’s Ornamental Herbicide II 3G (OH 2, oxyfl uorfen + pendimethalin) at 3.4 kg ai/ha (3.0 lbs ai/A), and Snapshot 2.5TG (isoxaben + trifl uralin) at 5.6 kg ai/ha (5.0 lbs ai/A) were applied to 4-L (1 gal) containers using a handheld shaker jar November 7, 2002 and November 25, 2003. All herbicide treatments were compared to nontreated plants. A pine bark plus sand substrate (7:1 v/v) was used and containers were

432 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005 absent of ornamental plants, containing only the weed species. Containers were arranged in a randomized complete block design with fi ve one-container replications of each species in 2002 and eight one-container replications of each species in 2003. About one month after treatments, containers were placed in temporary overwintering structures. Structures were about 2.4 m (8 ft) wide, 10.7 m (35 ft) long, 1.2 m (4 ft) tall (at the center peak), and covered with 6 mil white plastic. The plastic was removed periodically to water plants and evaluate herbicide effi cacy.

In 2002, the weed species evaluated were annual bluegrass (Poa annua L.), common chickweed (Stellaria media L.), common groundsel (Senecio vulgaris L.), hairy bittercress, and yellow woodsorrel. In 2003, the weed species evaluated were annual bluegrass, birdseye pearlwort (Sagina procumbens L.), common chickweed, common groundsel, hairy bittercress, spiny sowthistle (Sonchus asper L.), and yellow woodsorrel. Weed control was visually evaluated compared to nontreated plants through the winter and spring. Final evaluations were recorded approximately fi ve months after herbicide application. Data were subjected to analysis of variance and means were separated using Fisher’s protected LSD at the 5% signifi cance level (4). Because weed control was evaluated relative to nontreated plants, data from nontreated plants were not included in the analysis.

Results and Discussion: In 2002, each herbicide controlled annual bluegrass, common chickweed, common groundsel, hairy bittercress, and yellow woodsorrel for the fi ve month evaluation period (Table 1). In 2003, for the fi ve month evaluation period, common chickweed, common groundsel, hairy bittercress, and spiny sowthistle were equally controlled by each herbicide (Table 1). Yellow woodsorrel control was slightly better with OH2 (98%) and Snapshot TG (100%) than BroadStar (86%). Snapshot TG was most effective on annual bluegrass providing 91% control. BroadStar and OH2 were less effective, providing 17% and 62% control, respectively. Trends for birdseye pearlwort control were similar to annual bluegrass; BroadStar provided 34% control, OH2 provided 72% control, and Snapshot TG provided 97% control.

In general, Snapshot TG provided the best control for the winter weeds evaluated in these studies. Most weeds were equally controlled by BroadStar and OH2 with the exception of annual bluegrass, birdseye pearlwort, and yellow woodsorrel. If the presence of any of these weeds is expected in overwintering structures, Snapshot TG should be used before covering or a dinitroaniline herbicide may need to be included with BroadStar to improve the spectrum of weed control.

Literature Cited: 1. Gilliam, C.H., W.J. Foster, J.L. Adrain, and R.L. Shumack. 1990. A survey of weed control costs and strategies in container production nurseries. J. Environ. Hort. 8:133-135. 2. Mickler, K.D. and J.M. Ruter. 2001. Evaluation of a year long weed control program for container grown ornamentals. Proc. Southern Nursery Assoc. Res. Conf. 46:454-456.

Weed Control Section 433 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

3. Neal, J.C. 1999. Weed control in greenhouses. Horticulture Information Leafl et No. 570. 4. [SAS] Statistical Analysis Systems. 1999. SAS/STAT User’s Guide. Version 8. Cary, NC: Statistical Analysis Systems Institute. 3884 p.

Table 1. Percent control of winter weeds with BroadStar, OH2, and Snapshot TG. The 2002 experiment was evaluated April 14, 2003 and the 2003 experiment was evaluated April 1, 2004.

BroadStar OH2 Snapshot TG 2002 Weed Species ------Percent Control------Annual bluegrass 92 az 100 a 100 a Common chickweed 100 a 100 a 100 a Common groundsel 94 a 100 a 100 a Hairy bittercress 100 a 100 a 100 a Yellow woodsorrel 100 a 100 a 100 a

2003 Weed Species ------Percent Control------Annual bluegrass 17 c 62 b 91 a Birdseye pearlwort 34 c 72 b 97 a Common chickweed 100 a 98 a 96 a Common groundsel 90 a 80 a 79 a Hairy bittercress 95 a 92 a 98 a Spiny sowthistle 90 a 96 a 100 a Yellow woodsorrel 86 b 98 a 100 a z Means within a row followed by the same letter do no signifi cantly differ according to Fisher’s Protected LSD (P ≤ 0.05). Means cannot be compared within a column.

434 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Doveweed, Florida Tasselfl ower and Eclipta Control Under Heavy Rainfall Conditions Using Granular Preemergence Herbicides

Robert H. Stamps and Annette L. Chandler Univ. of Florida, Inst. of Food and Agr. Sci., Dept. of Environ. Hort., Mid- Florida Res. and Edu. Ctr., 2725 S. Binion Rd., Apopka, FL 32703-8504 [email protected]fl .edu

Index words: fl umioxazin, oryzalin, oxyfl uorfen, pendimethalin, prodiamine, trifl uralin, BroadStar, OH2, RegalKade, Rout, Snapshot, Eclipta prostrata, Emilia fosbergii, Murdannia nudifl ora

Signifi cance to Industry: This research shows that even under very heavy rainfall and irrigation conditions, there are several preemergence herbicides that can effectively control eclipta, Florida tasselfl ower and doveweed for at least eight weeks. Flumioxazin was particularly effective in controlling doveweed.

Nature of Work: The introduction and spread of new weeds is a common occurrence in the nursery industry (3). In order to effectively manage these “new” weeds, it is necessary to determine the weed control effi cacy of preemergence herbicides. In addition, irrigation and rainfall amounts (2) and timing (1) can affect herbicide effi cacy. In this study, six commercial granular preemergence herbicides were evaluated for controlling three weeds of increasing signifi cance in ornamental plant production.

The experiment was conducted outdoors at the Univ. of Florida/IFAS’ Mid- Florida Res. and Educ. Ctr. in Apopka, FL. On Aug. 5, 2004, 6-inch [15-cm] diameter plastic pots (Dillen Products) were fi lled with a soilless growing medium composed of 80% aged pine bark and 20% Canadian Sphagnum peat (Fafard). Initial medium pH and EC (dS/m) were 6.86 and 1.52, respectively. After the pots were fi lled, 15 seeds of one of the three weed species were sown in each one: Eclipta prostrata (eclipta), Emilia fosbergii (Florida tasselfl ower) or Murdannia nudifl ora (doveweed) Pots were then irrigated with 1.3 cm [½ in] of water. Each pot was reseeded with another 15 seeds of the same weed species on Sept. 16, six weeks after the initial sowing.

Plots were irrigated daily using overhead sprinklers, except during periods of high rainfall or when there was no electrical power to operate the well pumps. During this three-month long experiment, rainfall + irrigation water totaled 149 cm [58.8 in]. Thirty percent of that total occurred during the fi rst two weeks of the experiment from the heavy rainfall associated with hurricane (Charley). In fact, over 25.4 cm (10 in) of rainfall and irrigation water occurred during the fi rst week after the herbicides were applied.

Herbicide treatments (Table 1) were hand broadcast into each pot on Aug. 6 and irrigated with 1.9 cm [¾ in] of water. There were four replications of each treatment. On Aug. 23, containers were fertilized with a 180 day 18N-2.6P-6.6K

Weed Control Section 435 SNA RESEARCH CONFERENCE - VOL. 50 - 2005 controlled-release fertilizer containing micronutrients (Florikan) applied at a rate of 964 kg N/ha/application [861 lb N lb/acre/appl].

Weed seed germination was monitored after each sowing. Percent weed coverage was determined visually for two weeks after the seeds were fi rst applied and, thereafter, was determined from digital images of each container and using graphics software (Corel). After 84 days from initiation of treatments, the aboveground parts of the weeds were cut from the containers and then dried at 70ºC [158ºF] for one week.

Data were analyzed by analysis of variance and means separations were by Duncan’s new multiple range test (P ≤ 0.05). When necessary to approximate normal distributions, percentage data were transformed using the inverse sine transformation.

Results and Discussion: Weed seed germination (Table 1). Eclipta. Initially, Rout, Snapshot and the BroadStar treatments totally stopped Eclipa seedling development. The other herbicide treatments suppressed but did not eliminate successful plant development. None of the herbicide treatments were 100% effective in controlling Eclipta seed germination after the second sowing. Emilia. The same treatments that controlled Eclipta provided the best control of Emilia, and again, inhibition of successful seed germination was not as great following the second seed sowing. Murdannia. Only Rout and the treatments containing BroadStar provided excellent control of Murdannia following the fi rst seed sowing. Following the second seed sowing, successful germination was limited in all treatments due to competition from the already established weeds in the plots where the herbicides were not effective and by residual herbicide in the other plots, especially in those treated with BroadStar.

Weed coverage percentages (Table(Table 2). Eclipa. All the herbicide treatments provided signifi cant control of Eclipta. Rout, Snapshot and the BroadStar treatments provided better control than Regal O-O and RegalKade. OH2, although not as effective in stopping successful weed seed germination and survival, also provided overall control comparable to the other top products. Emilia. For Emilia, Rout, Snapshot and the BroadStar treatments provided excellent control for eight weeks. After that control declined and by 12 WAT only Snapshot and the BroadStar treatments were still signifi cantly better than the untreated control. Murdannia. Rout provided good control and BroadStar was particularly effective in controlling Murdannia.

Weed dry mass (Table(Table 2). Not surprisinglysurprisingly,, weed dry mass followed a pattern similar to that for weed coverage.

Despite the heavy rainfall throughout this experiment, the preemergence herbicides provided signifi cant weed control. Generally, Rout and the BroadStar treatments provided the greatest control of the three weeds in this test. Snapshot was also effective in controlling Eclipta and Emilia, but not Murdannia. OH2 was as effective in controlling Eclipta as were the abovementioned herbicides. The combination treatment, BroadStar + RegalKade, was no more effective than using BroadStar alone.

436 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Literature Cited: 1. Fain, G. B., K. L. Paridon, and P. M. Hudson. 2004. The effect of cyclic irrigation and herbicide on plant and weed growth in production of Magnolia grandifl ora ‘Alta’. SNA Res. Conf. 49:37–39. 2. Monteiro, A. and I. Moreira. 2004. Reduce rates of residual and post- emergence herbicides for weed control in vineyards. Weed Research 44:117–128. 3. Stamps, R. H. 2003. “New” and emerging weed pests. Soc. of American Florists’ Annu. Conf. on Insect and Dis. Mgt. on Ornamentals 19:61–64.

Weed Control Section 437 SNA RESEARCH CONFERENCE - VOL. 50 - 2005 0.00 a 0.42 a 0.42 a 2.43 a 3.75 a 0.00 a 0.00 a 0.42 a 6.23 a 1.54 a 4.32 ab 4.94 ab 11.25 ab 11.25 17.67 ab 31.54 ab 41.11 ab 41.11 29.88 ab 35.67 ab 47.87 b 22 DASS 32.30 c 12.64 abc 24.94 bc 24.13 bc 24.36 bc ≤ 0.05). P z 2.50 a 3.98 a 0.00 a 0.42 a 0.42 a 2.43 a 3.75 a 0.00 a 0.42 a 0.42 a 3.79 ab 1.54 a 42.88 b 24.36 b 42.88 b 25.09 b 41.15 b 43.26 b 15 DASS 24.13 bc 22.46 bc 29.07 c 11.91 abc 11.91 abc 11.17 18.02 abc Second sowing x 0.43 a 2.50 a 0.43 a 0.00 a 0.00 a 1.67 a 0.42 a 3.75 a 0.83 a 0.00 a 1.67 a 4.94 ab 0.00 a 3.25 ab 8.14 ab 5.95 ab 4.32 ab 3.79 ab 21.52 b 11.12 ab 11.12 10.50 ab 10.05 ab 8 DASS 14.90 b 12.28 b 0.00 a 7.36 b 0.00 a 0.00 a 0.00 a 0.43 a 7.66 b 0.00 a 0.00 a 0.00 a 0.43 a 0.00 a 46.46 c 41.41 c 58.06 c 16.53 b 16.33 b 75.56 d 70.19 d 71.01 d 17.59 b 21.75 b 29.32 bc 31.83 bc 15 DAFS Weed survival from sown seed (%) Weed 0.00 a 2.50 a 0.00 a 0.00 a 0.00 a 9.41 b 0.00 a 0.00 a 9.41 b 0.00 a 0.00 a 1.70 ab 0.00 a 81.59 c 63.37 c 69.20 c 36.24 b 43.84 c 11.44 bc 11.44 10.59 b 55.65 d 17.27 bc 32.83 cd 28.57 cd 11 DAFS 11 First sowing w y d (doveweed) (Eclipta) 0.00 a 4.94 b 3.79 b 0.00 a 0.00 a 0.43 a 0.00 a 0.00 a 0.43 ab 0.00 a 0.00 a 41.56 15.91 c 21.60 c 78.84 c 56.74 c 68.58 c 31.48 b 17.59 b 8 DAFS 11.12 c 11.12 7.36 bc 18.02 c 14.98 c 46.46 d (Florida tasselfl ower) (Florida tasselfl rate (lb a.i./A) Application Eclipta prostrata c weed are not signifi cantly different according to Duncan’s new multiple range test ( according to Duncan’s cantly different c weed are not signifi — 0.375 2.0 + 1.0 2.0 + 1.0 2.0 + 1.0 4.0 + 1.0 1.5 0.375 + 1.5 — 0.375 2.0 + 1.0 2.0 + 1.0 2.0 + 1.0 4.0 + 1.0 1.5 0.375 + 1.5 — 0.375 2.0 + 1.0 2.0 + 1.0 2.0 + 1.0 4.0 + 1.0 1.5 0.375 + 1.5 Murdannia nudifl ora Murdannia nudifl Emilia fosbergii Active ingredient(s) (a.i.) — umioxazin fl uorfen + pendimethalin oxyfl uorfen + oxadiazon oxyfl uorfen + oryzalin oxyfl uralin + isoxaben trifl prodiamine umioxazin + prodiamine fl — umioxazin fl uorfen + pendimethalin oxyfl uorfen + oxadiazon oxyfl uorfen + oryzalin oxyfl uralin + isoxaben trifl prodiamine umioxazin + prodiamine fl — umioxazin fl uorfen + pendimethalin oxyfl uorfen + oxadiazon oxyfl uorfen + oryzalin oxyfl uralin + isoxaben trifl prodiamine umioxazin + prodiamine fl Herbicide effects on weed seed survival from sown the day before herbicide application and again six weeks later. Herbicide effects Trade Name Trade Untreated control BroadStar 0.25G OH2 3G Regal O-O 3G Rout 3G Snapshot 2.5G RegalKade 0.5G BroadStar + RegalKade Untreated control BroadStar 0.25G OH2 3G Regal O-O 3G Rout 3G Snapshot 2.5G RegalKade 0.5G BroadStar + RegalKade Untreated control BroadStar 0.25G OH2 3G Regal O-O 3G Rout 3G Snapshot 2.5G RegalKade 0.5G BroadStar + RegalKade Means followed by the same letter within column for a specifi Percentages were transformed using arcsin transformation, as needed, prior to statistical analysis. rst sowing. DAFS = days after fi DASS = days after second sowing. Table 1. Table z y x w

438 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005 (g) 8.9 bc 3.4 ab 2.9 a 4.4 a 3.2 a 3.4 a 6.2 c 1.2 a 1.2 a 4.2 bc 14.8 c 12.1 c 13.8 c 14.5 c 1.7 ab 1.5 ab 3.1 ab <0.1 a <0.1 a 18.8 b 12.6 b 14.7 b 14.3 b 11.3 d 11.3 Weed top Weed dry weight dry ≤ 0.05). P c a a b b a a ab ab ab 14 24 80 18 a 68 b 20 a 47 a 60 66 50 86 ab 75 91 0.01 a 0.38 a 100 100 b 100 97.44 c 75.07 bc 85.36 bc 33.53 ab 99.40 c 12 WAT 100.00 c b a b b a a b a 15 94 44 23 90 17 3.2 ab 3.1 ab 1.3 a 4.4 ab 0.04 a 0.01 a 100 100 42.1 c 54.5 c 87.6 d 18.2 b 90.62 c 85.36 c 65.92 bc 16.20 ab 83.97 c 10 WAT z 100.00 c 0.1 a 0.1 a 0.2 a 0.7 a 0.2 a 1.0 a 5.6 a 1.3 a 0.6 a 0.0 a 0.0 a 6.5 ab 49.6 c 92.3 d 16.8 b 79.6 b 91.9 b 88.5 b 8 WAT 52.4 bc 43.8 bc 73.6 cd 72.7 cd 100.0 b 100.0 d 4.2 ab 0.0 a 0.0 a 0.0 a 0.0 a 0.0 a 0.0 a 0.0 a 0.7 ab 58.5 c 13.5 b 6 WAT 86.9 c 0.00 a 41.4 b 49.8 b 0.51 a 0.00 a 0.00 a 65.9 bc 28.7 c 67.8 d 20.7 bc 11.8 abc 11.8 14.9 abc Percent weed coverage 2.13 b 0.00 a 0.01 a 0.69 a 0.00 a 0.00 a 0.13 a 0.00 a 4.68 c 0.00 a 1.68 b 3.59 bc 0.04 a 0.00 a 0.00 a 4 WAT 8.96 c 0.00 a 1.84 b 0.00 a 1.56 b 1.93 b 0.00 a 0.88 ab 10.10 d (doveweed) y x (Eclipta) 0.00 a 0.04 a 0.05 a 0.00 a 0.00 a 0.10 a 0.00 a 1.47 b 0.00 a 1.06 b 1.00 b 0.00 a 0.00 a 1.47 b 0.00 a 0.00 a 0.15 b 0.19 b 0.00 a 0.38 b 0.53 c 2 WAT 0.23 bc 0.00 a 0.08 ab (Florida tasselfl ower) (Florida tasselfl rate Eclipta prostrata (lb a.i./A) Application — 0.375 2.0 + 1.0 2.0 + 1.0 2.0 + 1.0 4.0 + 1.0 1.5 0.375 + 1.5 — 0.375 2.0 + 1.0 2.0 + 1.0 2.0 + 1.0 4.0 + 1.0 1.5 0.375 + 1.5 — 0.375 2.0 + 1.0 2.0 + 1.0 2.0 + 1.0 4.0 + 1.0 1.5 0.375 + 1.5 Murdannia nudifl ora Murdannia nudifl Emilia fosbergii Active ingredient(s) (a.i.) — umioxazin fl uorfen + pendimethalin oxyfl uorfen + oxadiazon oxyfl uorfen + oryzalin oxyfl uralin + isoxaben trifl prodiamine umioxazin + prodiamine fl — umioxazin fl uorfen + pendimethalin oxyfl uorfen + oxadiazon oxyfl uorfen + oryzalin oxyfl uralin + isoxaben trifl prodiamine umioxazin + prodiamine fl — umioxazin fl uorfen + pendimethalin oxyfl uorfen + oxadiazon oxyfl uorfen + oryzalin oxyfl uralin + isoxaben trifl prodiamine umioxazin + prodiamine fl Herbicide effects on weed growth from seed sown the day before herbicide application and again six weeks later. Herbicide effects Means within the same column for a specifi c weed followed by the same letter are not signifi cantly different, according to Duncan’s new multiple range test ( according to Duncan’s cantly different, c weed followed by the same letter are not signifi Means within the same column for a specifi Trade Name Trade Untreated control BroadStar 0.25G OH2 3G Regal O-O 3G Rout 3G Snapshot 2.5G RegalKade 0.5G BroadStar + RegalKade Untreated control BroadStar 0.25G OH2 3G Regal O-O 3G Rout 3G Snapshot 2.5G RegalKade 0.5G BroadStar + RegalKade Untreated control BroadStar 0.25G OH2 3G Regal O-O 3G Rout 3G Snapshot 2.5G RegalKade 0.5G BroadStar + RegalKade Percentages were transformed using arcsin transformation, as needed, prior to statistical analysis. = weeks after treatment. WAT Table 2. Table z y x

Weed Control Section 439 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Control of Florida Betony with Herbicides

Mark Czarnota1, Byron Rhodes2 and Tim Murphy3 1University of Georgia, Department of Horticulture, 1109 Experiment Street, Griffi n GA, 30223 2University of Georgia, Cooperative Extension Service, 222 West Jefferson Street, Thomasville, GA 31799-0049 3University of Georgia, Department of Crop and Atmospheric Science, 1109 Experiment Street, Griffi n, GA 30223

Index Words: Rattlesnake weed, Stachys fl oridana, atrazine, carfentrazone, clopyralid, 2,4-D, dicamba, MCPP, dichlobenil, fl uroxypyr, glyphosate, herbicides

Signifi cance to Industry: Several herbicides appeared to do a good job of controlling Florida betony and providing greater than 70% control at 8 or 10 WAT, and these herbicides included: atrazine, dichlobenil, foramsulfuron, glyphosate, metsulfuron, Speedzone (carfentrazone, 2,4-D ester, mecoprop, dicamba), and trifl oxysulfuron. Most of these products are limited in their scope of use as only glyphosate and dichlobenil are the only products with labeling in landscape ornamentals. Dichobenil can be used over-the-top of select ornamentals and glyphosate can be used as a post-directed herbicide in the landscape. All of the other herbicides active on Florida betony are currently labeled as selective turfgrass herbicides.

Nature of Work: Many gardeners of the Southeast are plagued by Florida betony (Stachys fl oridana). With the species name fl oridana, betony is considered by many to be a native invader. However, we are not certain of Florida betony origins. Florida betony or rattlesnake weed is a “winter” perennial, and like most plants in the labiatae family, betony has a square stem with aromatic opposite leaves. Flowers of betony are usually pink and have the classic mint like structure. Unlike its relatives, it has the unique characteristic of producing tubers that look like the rattle of a rattle snake, hence the name. Betony is dormant during the hot, humid summers of the South. In most of Georgia, growth usually begins in early to mid fall and continues until late spring. Betony is a problem in both turf and ornamental beds, yet little control information exists. Most of the information is dated (Houston 1976; Murphy 1991; Norcini et al. 1995; Stamps 1992), and many new herbicides have been or are in the process of being introduced to the turf and ornamental market. A study was designed to determine the effi cacy of several new herbicides as well as herbicides that claim to control betony. In 2004 and 2005, studies were initiated to determine the degree of control provided by a single application of select herbicides. During both years, an area of betony infested turf in Thomasville, Georgia was selected for a study. In 2004, the average number of betony shoots per square foot was 42; and in 2005, the average number of betony shoots per square foot was 40. Identical treatments were applied both years and a treatment list is presented in Table 1. Treatments were arranged in a randomized complete block design with 4 replications. Herbicides were applied with a CO2 backpack sprayer calibrated to deliver 20 GPA. Depending on the year of the test, percent control was visually estimated between 2 and 10 weeks after treatment (WAT). A description of the rating scale is presented in Table 2. All data were subjected to analysis of

440 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005 variance (ANOVA) and means were separated using Fisher’s least signifi cant difference (LSD) test with a signifi cance level of P=0.05.

Results and Discussion: In 2004, acceptable control (> 60%) was not achieved until the 4 WAT rating. Dichlobenil, glyphosate, and metsulfuron were providing at least > 60% control at 4 WAT (Table 3). By 10 WAT, all herbicides but clopyralid were providing signifi cantly better control than the check. Atrazine, dichlobenil, foramsulfuron, glyphosate, metsulfuron, and trifl oxysulfuron were providing greater than 80 % control (Table 3). In 2005, the same trends were seen as acceptable control was not seen until 4 WAT (Table 4). All herbicides but clopyralid and carfentrazone were providing signifi cantly better control than the untreated check. At 4 WAT, atrazine, dichlobenil, foramsulfuron, glyphosate, metsulfuron, Speedzone (carfentrazone, 2,4-D ester, mecoprop, dicamba), and trifl oxysulfuron were all providing greater than 76.7% control (Table 4). This same trend continued into the 8 WAT rating as all treatments but clopyralid, fl uroxypyr, and carfentrazone were providing greater than 70% control.

Literature Cited: 1. Houston, W. 1976. Florida betony control. Fact Sheet # 425. Mississippi State University Cooperative Extension Service. 2 p. 2. Murphy, T. R. 1991. Agronomy fact sheet: Florida betony control in turfgrass and ornamentals. Fact Sheet #L425. University of Georgia. Griffi n. 4 p. 3. Norcini, J. G., J. M. McDowell and J. H. Aldrich. 1995. Control of Florida betony (Stachys fl oridana Shuttlew.) emerging from tubers. Journal of environmental horticulture. 13:89-91. 4. Stamps, R. H. 1992. Prodiamine controlled Florida betony (Stachys fl oridana) in leatherleaf fern (Rumohra adiantiformis). Weed Technology. 6:961-967.

Table 1. List of treatments.

Rate Rate (active # Trade name Active ingredient Formulation (product/ ingredient Acre)* /Acre) 1 Casoron dichlobenil 4 GR 150 lb pr/A 6.0 lb ai/A 2 Monument trifl oxysulfuron 75 WG 0.56 oz pr/A 0.5 lb ai/A 3 Lontrel clopyralid 3 L 1.33 pt pr/A 0.5 lb ai/A 4 Roundup Pro glyphosate 4 L 1.0 qt pr/A 1.0 lb ai/A 5 Atrazine atrazine 4 L 2.0 qt pr/A 2.0 lb ai/A 6 Manor metsulfuron 60 DG 0.5 oz pr/A 0.0188 lb ai/A 7 Revolver foramsulfuron 0.19 L 26.2 oz pr/A 0.039 lb ai/A 8 Spotlight fl uroxypyr 1.5 L 0.67 pt pr/A 0.126 lb ai/A 9 Quicksliver carfentrazone 1.9 L 0.13 pt pr/A 0.031 lb ai/A 10 Speedzone carfentrazone, 2,4- 2.2 L 2.0 qt pr/A 0.405 lb ai/A D Ester, mecoprop, and dicamba 11 Check * lb pr/A = pounds product per Acre; oz pr/A = ounces product per Acre; pt pr/A = pints product per Acre; qt pr/A = quarts product per Acre

Weed Control Section 441 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Table 2. Representations of numeric Florida betony control ratings.

Value Ranges Plant Symptoms 0 No visual injury present. 10-30 Minimal injury to desirable plant. Less than 10% of the plant leaf surface area showing chlorosis and necrosis. A 10 to 30% biomass reduction. 40-70 More noticeable plant injury or stunting. Greater than 50% of the leaf area showing symptoms of chlorosis and/or necrosis. A 40 to 70% biomass reduction. 80-90 Plants severely injured. Most of the leaves and leaf surface showing signs of chlorosis and necrosis. An 80 to 90% biomass reduction. 100 Plant appears dead. No signs of regrowth.

Table 3. 2004 Florida betony (Stachys fl oridana) control ratings 2, 4, 7, and 10 weeks after treatmenta.

Trade Formu- Rate (active # Betony Control name lation ingredient /Acre) 2 WAT 4 WAT 7 WAT 10 WAT 1 Casoron 4 GR 6.0 lb ai/A 23.3 ab 66.7 a 90.0 a 96.7 a 2 Monument 75 WG 0.5 lb ai/A 23.3 ab 46.7 a-d 86.7 a 93.3 a 3 Lontrel 3 L 0.5 lb ai/A 20.0 b 13.3 de 31.7 c 38.3 cd Roundup 4 Pro 4 L 1.0 lb ai/A 40.0 a 60.0 ab 93.3 a 93.3 a 5 Atrazine 4 L 2.0 lb ai/A 13.3 bc 26.7 b-e 43.3 bc 71.7 abc 6 Manor 60 DG 0.0188 lb ai/A 13.3 bc 60.0 ab 88.3 a 83.3 ab 7 Revolver 0.19 L 0.039 lb ai/A 30.0 ab 43.3 a-d 85.0 a 91.7 ab 8 Spotlight 1.5 L 0.126 lb ai/A 20.0 b 40.0 a-d 53.3 bc 68.3 abc 9 Quicksliver 1.9 L 0.031 lb ai/A 16.7 bc 16.7 cde 31.7 c 53.3 bc 10 Speedzone 2.2 L 0.405 lb ai/A 40.0 a 50.0 abc 71.7 ab 60.0 abc 11 Check 0.0 c 0.0 e 0.0 d 0.0 d LSD 18.18 35.34 29.45 39.28 Coeffi cient of 48.92 53.92 28.18 33.82 Variation aAbbreviations: WAT, weeks after treatment; GR, granular; L, liquid; DG, dispersible granular b Means within a column followed by the same letter are not different according to Fisher’s protected LSD at P=0.05.

442 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Table 4. 2005 Florida betony (Stachys fl oridana) control ratings 2, 4, 6, and 8 weeks after treatmenta.

Trade Formu- Rate (active # Betony Control name lation ingredient /Acre) 2 WAT 4 WAT 6 WAT 8 WAT 1 Casoron 4 GR 6.0 lb ai/A 6.7 bc 93.3 ab 98.7 ab 100.0 a 2 Monument 75 WG 0.5 lb ai/A 10.0 ab 100.0 a 100.0 a 100.0 a 3 Lontrel 3 L 0.5 lb ai/A 0.0 d 1.7 de 0.0 e 35.0 de Roundup 4 Pro 4 L 1.0 lb ai/A 5.0 c 90.0 ab 88.3 c 70.0 bc 5 Atrazine 4 L 2.0 lb ai/A 3.3 cd 76.7 c 80.0 d 88.3 ab 6 Manor 60 DG 0.0188 lb ai/A 11.7 a 98.3 a 92.7 bc 100.0 a 7 Revolver 0.19 L 0.039 lb ai/A 6.7 bc 96.7 a 100.0 a 100.0 a 8 Spotlight 1.5 L 0.126 lb ai/A 5.0 c 11.7 d 0.0 e 55.0 cd 9 Quicksliver 1.9 L 0.031 lb ai/A 0.0 d 0.0 e 0.0 e 20.0 ef 10 Speedzone 2.2 L 0.405 lb ai/A 11.7 a 83.3 bc 97.3 ab 88.3 ab 11 Check 0.0 d 0.0 e 0.0 e 0.0 f LSD 26.49 10.87 7.16 29.56 Coeffi cient of 57.66 10.78 7.03 25.23 Variation aAbbreviations: WAT, weeks after treatment; GR, granular; L, liquid; DG, dispersible granular b Means within a column followed by the same letter are not different according to Fisher’s protected LSD at P=0.05.

Weed Control Section 443 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Preemergence Control of Marchantia polymorpha

Adam Newby1, James Altland2, Charles Gilliam1, Donna Fare3 and Glenn Wehtje1 1Department of Horticulture, Auburn University, Auburn, AL 36849 2North Willamette Research and Extension Center, Aurora, OR 97002 3USDA-ARS National Arboretum, McMinnville, TN 37110 [email protected]

Index words: liverwort.

Signifi cance to the Industry: Liverwort is an increasing problem in container- grown ornamental production within the Southeast. Since it is an emerging weed problem in the South, little research exists on preemergence controls

Nature of Work: Liverwort (Marchantia polymorpha) continues to spread throughout the South. Prostrate leaf-like structures of liverwort known as thalli create a mat over media surfaces in containers. Not only is liverwort unsightly, it can impede water and nutrient movement into the root zone (Svenson, 1998). Liverwort propagates sexually by spores and asexually by gemmae. It thrives in low UV light, high fertility and high moisture environments (Svenson, 1997).

Since liverwort is a newly emerging weed problem within the Southeastern United States, growers have limited knowledge about its control. There are promising new products for postemergence liverwort control, but preventing liverwort infestations would be more desirable.

The objective of this study was to evaluated the effectiveness of several commonly used preemergence herbicides for preemergence liverwort control.

Materials and Methods: This study was conducted in the summer of 2004. Full-gallon containers were fi lled with a 6:1 pine bark to sand substrate amended with 14 lb of Polyon 18-6-12, 5 lb of dolomitic lime, and 1.5 lb of Micromax per cubic yard. Herbicide treatments were applied on 6 July 2004. Twelve granular herbicides were applied at the recommended rate: Broadstar (150 lb product/ A), Kansel Plus (100 lb product/A), OH2 (100 lb product/A), Pendulum 2G (200 lb/A), Regal 0-0 (100 lb/A), Regal Kade (200 lb/A), Regal Star (200 lb/A), Ronstar (200 lb/A), Rout (100 lb/A), and Snapshot (200 lb/A). Treatments were applied using a handheld shaker. Each treatment consisted of 4 replications and 4 plants per replication. After treatment, each replication was placed around a container of mature liverwort for inoculation. Treatments and a non-treated control group were arranged in a completely randomized design. There were no potted landscape plants involved in the test. The study was conducted under 47% shade. Overhead irrigation was split into two cycles per day with a total of 0.25” applied. Percent liverwort coverage of the medium surface was recorded 6, 11, and 17 weeks after treatment (WAT). Treatment means were separated with Duncan’s multiple range test (α=0.05).

Results and Discussion: At 6 WAT, all herbicides except Kansel Plus, Pendulum 2G, and Regal Kade provided equally signifi cant control ranging from

444 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

0.3% to 12.4% liverwort coverage as compared to 25.9% liverwort coverage in controls. Percent liverwort coverage in Kansel Plus, Pendulum 2G, and Regal Kade treatments was similar to non-treated controls. At 11 WAT, Broadstar, Regal 0-0, Regal Star, Ronstar, and Rout provided the best control with percent coverage ranging from 2.9% to 28.4%. By 17 WAT, percent liverwort coverage in the Broadstar treatment was numerically lowest at 11.9%. Ronstar provided similar control to Broadstar. All other treatments provided unacceptable control.

This study shows that Broadstar and Ronstar provide superior preemergence control of liverwort compared to the other herbicides tested.

Literature Cited: 1. Svenson, S.E. 1997. Controlling liverworts and moss in nursery production. Comb. Proc. Intl. Plant Prop. Soc. 47:414-422. 2. Svenson, S.E. 1998. Suppression of liverwort growth in containers using irrigation, Mulches, Fertilizers and Herbicides. Proc. Southern Nurs. Res. Conf. 43: 396-398.

Table 1. Preemergence liverwort control with currently available herbicides.

% Liverwort coverage of medium Herbicide lb product/A 6 WATZ 11 WAT 17 WAT Broadstar 150 0.3Y dX 2.9 d 11.9 f Kansel + 100 18.5 ab 57.8 a 79.4 abc OH2 100 6.3 bcd 48.7 abc 79.5 abc Pendulum 2G 200 15.1 abc 60.9 a 86.3 ab Regal 0-0 100 3.6 cd 19.6 d 55.3 de Regal Kade 200 16.3 abc 52.6 ab 74.7 abc Regal Star 200 4.4 cd 28.4 bcd 65.9 cde Ronstar 200 0.3 d 8.7 d 32.5 ef Rout 100 4.1 cd 25.3 cd 72.5 bcd Snapshot 200 12.4 bcd 64.7 a 90 a Non-treated control 25.9 a 59.8 a 73.8 abc ZWeeks after treatment. YPercent coverage of liverwort within container. XDuncan’s Multiple Range Test (α=0.05). Means with same letter are not signifi cantly different.

Weed Control Section 445 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

A Non-Chemical Alternative for Weed Control in Container Nursery Crops

Ben M. Richardson1, Charles H. Gilliam1, Glenn R. Wehtje1, Glenn B. Fain2 1Auburn University, Dept. of Horticulture, Auburn, AL 36849 2USDA-ARS, Southern Horticultural Laboratory, Poplarville, MS 39470 [email protected]

Index Words: Mulch, Preemergence

Signifi cance to Industry: Preemergence herbicides may be impractical for growers that are producing crops in large containers. Container spacing is often such that application of a granular herbicide results in over 50% of the herbicide falling outside the containers. In addition labor involved in hand weeding is expensive therefore growers are seeking alternatives. Pine bark nuggets have been used as mulch in landscape beds for many years to control weeds. This research shows that it can also be used as mulch in the production of nursery crops grown in large containers control weeds.

Nature of Work: Container nursery crops are among the most valuable crops produced in the southeast. Weeds can reduce the value of nursery crops by reducing growth through competitive effects (1) and reduced salability due to consumer demand for weed free crops. Growers primarily use preemergence herbicides for weed control. However demand has increased for nursery crops grown in large containers and preemergence herbicides are often impractical due to the increased spacing between containers. Hand weeding is another option but expensive and labor costs continue to increase (2,3).

Fresh pine bark nugget mulch could be an alternative. It has been used for many years in landscape beds to control weeds and could be an option for use in container crops. Physical properties of pine bark are conducive to weed control. Fresh pine bark is hydrophobic with limited water holding capacity due to the large particle size. Pine bark is well suited to be spread uniformly as a mulch, covering the entire surface of the container while other mulch type products may leave cracks for weeds to establish.

The objective of this study was to evaluate coarse pine bark mulch for weed control in container grown crops and to determine depth of mulch needed for adequate weed control.

These studies were conducted in Auburn, Alabama fall of 2004, and spring 2005. Crapemyrtle (Lagerstroemia indica ‘Acoma’) were transplanted from trade gallon containers into 7 gallon containers. The substrate used was a 6:1 (v:v) aged pine bark: sand amended with 2.3 kg (5lb) of dolimitic lime, 6.4kg (14lb) of Polyon 18-6-12. and 0.68kg (1.5lb) of Micromax. All plants were potted to equal depths, approximately 3 inches below the top of the container. All plants were irrigated twice prior to treatment. Three treatments consisted of broadcasting 25 bittercress (Cardamine hirsuta) seed on the surface of the substrate of each container, then coarse pine bark nugget mulch was hand applied at 0, 1.5 and

446 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

3 inches respectively. The nuggets averaged in size of 2 cm by 6 cm (English units???) fl akes. Two other treatments consisted of fi rst applying mulch at 1.5 or 3 inches, then broadcasting the bittercress seeds on top of the mulch. These procedures were also applied to fi ve more treatments and a granular herbicide (Broadstar 0.25G at 150 lb product/A) was applied after all mulch and seed were present. This study was initiated October 8th 2004 with a total of 10 treatments and 10 single pot reps per treatment.

In a similar study, common gardenia (Gardenia jasminoides) were transplanted from trade gallon containers into 7 gallon containers. On September 30th, 2004 the same treatments were applied to the gardenia except 25 oxalis (Oxalis stricta) seed were used per container instead of bittercress. In both studies, data collected were weed number per container at 30, 60, 90 and 180 days after treatment (DAT) and percent coverage of weeds at 60, 90, 180 DAT. Shoot fresh weight of weeds were taken for each container at 180 DAT. Plants were covered for overwintering from December 15 2004 until March 1 . Growth indices were taken on all gardenia and crapemyrtles at 180 DAT.

Results and Discussion: These studies show that pine bark mulch can provide effective weed control for nursery crops grown in large containers. At 180 DAT oxalis was present in the no mulch, no herbicide treatment which contained 3.9 weeds, averaged 35 % coverage of container surface and averaged a shoot dry weight of 12.9 g per container. All other treatments resulted in minimal oxalis growth. The combination of mulch plus herbicide provided complete oxalis control 180DAT.

At 180 DAT bittercress was growing vigorously in the control containers (no mulch- no herbicide). These containers averaged 8.1 bittercress, 100% coverage of container surface and 59.6 g of bittercress per container. In comparison no herbicide,1.5 inches of mulch treatment with seeding after mulching, averaged 2.6 weeds, 44% coverage of container surface and 33.7 g per container. Both treatments resulted in greater bittercress growth than all other treatments. Similar to the oxalis study, the combination of mulch plus herbicide provided complete bittercress control 180 DAT.

No herbicide or mulch applications affected growth of crapemyrtle and gardenia (Table 2).

Literature Cited: 1. Berchielli-Robertson, D.L., C.H. Gilliam, and D.C. Fare. 1990. Competitive effects of weeds on the growth of container-grown plants. Hortscience 25:77-79. 2. Gilliam, C.H., W.J. Foster, J.L. Adrain, and R.L. Shumack. 1990. A survey of weed control costs and strategies in container production nurseries. J. Environ. Hort. 8:133-135. 3. Judge, C.A., J.C. Neal, and J.B. Weber. 2003. Dose and concentration responses of common nursery weeds to Gallery, Surfl an and Trefl an. J. Environ. Hort. 21:43-45.

Weed Control Section 447 SNA RESEARCH CONFERENCE - VOL. 50 - 2005 0c 0c 0c 0c 0c 3.5c 1.4c 3.8c SDW 59.6a 33.7b 0c 8c 0c 0c 0c 0c 2.5c 1.0c 100a 44.2b %cover 180 DAT 180 DAT 0c 0c 0c 0c 0c 0.2c 0.1c 0.1c 8.1a 2.6b Weed # Weed Bittercress 0b 0b 5b 0b 2b 0b 0b 0b 0b 48a %cover 90 DAT # = 0.05). 0b 0b 0b 0b 0b 0b 0b 9.4a 0.4b 0.1b α Weed Weed 0b 0b 0b 0b 0b 0.9b 1.3b 0.8b 1.1b SDW 12.9a 0b 1b 0b 0b 0b 0b 35a 0.9b 2.5b 2.5b %cover , % cover = percentage of the container surface covered by designated weed species, 180 DAT 180 DAT 0b 0b 0b 0b 0b 3.9a 0.3b 0.4b 0.2b 0.6b Weed # Weed Oxalis 0b 0b 0b 0b 0b 0b 0b Z 0.5b 1.5b 18.5a %cover V 90 DAT 0b 0b 0b 0b 0b 0b 0b 0.2b 0.3b 5.5a Weed # Weed W 0 3 3 0 3 3 1.5 1.5 1.5 1.5 Mulch depth X after after after after before before before before before before Weeds Weeds seeded Y The infl uence of mulch and herbicides on weed control within container nursery crops. The infl No No No No No Yes Yes Yes Yes Yes Applied Herbicide Mulch depth in inches, weed # = number of designated species per container SDW = shoot dry weight (g/ container). Timing of seeding compared to mulching, before= seeds under mulch ,after= on top mulch. Timing Test: Multiple Range (Duncan’s cantly different Means within column followed by the same letter are not signifi DAT= days after treatment. DAT= Application of a preemergence herbicide(Broadstar 0.25G 150 lb product/A). Table 1. Table Z Y X W V

448 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Table 2. The infl uence of mulch and herbicide on growth of container crops.

Growth indexZ 180 DAT Herbicide AppliedY Weeds seededX Mulch depthW Gardenia Crapemyrtle No before 0 59abV 85a No before 1.5 54b 77a No before 3 59ab 80a

No after 1.5 56ab 82a No after 3 56ab 80a

Yes before 0 55ab 80a Yes before 1.5 60a 88a Yes before 3 56ab 79a

Yes after 1.5 54b 75a Yes after 3 58ab 74a ZDAT= days after treatment. YApplication of a preemergence herbicide(Broadstar 0.25G 150 lb product/A). XTiming of seeding compared to mulching, before= seeds under mulch ,after= seeds on top of mulch. WMulch depth in inches, weed # = number of designated weed species per container, % cover = percentage of the container surface covered by designated weed species, SDW = shoot dry weight (g/ container). VMeans within column followed by the same letter are not signifi cantly different (Duncan’s Multiple Range Test: α = 0.05).

Weed Control Section 449 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Container-Grown Ornamental Granular Herbicide Demonstration

L. M. Miller, J. Popenoe, R.H. Stamps, R.C. Miller and D. Lickfeldt Hillsborough County Extension, University of Florida, IFAS, Seffner, FL 33584 [email protected]fl .edu

Index Words: Viburnum odorotissimum, Quercus virginiana Mill., Preemergence weed control

Signifi cance to Industry: Weed control is a continuous challenge for Florida container nurseries despite the numerous chemical control options at their disposal. Many nurseries hand weed container crops to maintain them in marketable condition. This is an increasingly expensive option, especially with the recent rise in Florida’s minimum wage. Each preemergence weed control product available to nurseries has its strengths and weaknesses in terms of weed spectrum, crop safety and worker protection requirements. Each product will perform satisfactorily when applied according to label directions to control target weed species. Failure to adequately control weeds while using granular herbicides in container nurseries may be caused by too much or too little rain/ irrigation, by the presence of weeds that are not controlled by the herbicide(s), by improper rate selection or by inaccurate/non-uniform application (1, 5, 6). This demonstration was designed to show that when applied correctly, most products will control target weed species successfully.

Nature of Work: Two nurseries, one in Hillsborough County and one in Sumter County, were used as test sites for evaluating a number of preemergence herbicides for weed control during containerized landscape ornamental production. Weed pressure, types of weeds, production practices and data collected varied by nursery.

Containers were treated with Snapshot 2.5 TG (200 lbs/acre), Showcase 2.5G (200 lbs/acre), OH2 3G (100 lbs/acre), Rout 3G (100 lbs/acre), BroadStar 0.25G (150 lbs/acre), Dimension 0.15F on a fertilizer carrier (333 lbs/acre), Trefl an 5G (200 lbs/acre in Hillsborough County only), or were untreated controls. In Hillsborough County, one of the untreated groups was hand-weeded on January 7, 2005. Weed control was measured twice in the spring of 2005 and the collected data analyzed using ANOVA (SAS). Differences among treatment means were determined using Duncan’s new multiple range test at P < .05.

Quercus virginiana ‘Cathedral Oak’ P.P. #12015, growing in 5-gallon containers in a pot-in pot production system, were treated in November 2004 at Meyer’s Nurseries in Thonotosassa in Hillsborough County, FL. A randomized block design with four replications of 8 plants each (288 total plants) was used. Herbicides were weighed and applied individually to each pot followed by 0.5 to 0.75 inches of irrigation. Each pot was visually rated for percent weed coverage on January 7 and March 7, 2005. Percentages were transformed using arcsin transformation, as needed, prior to analysis. On March 7, 2005 all weeds were

450 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005 removed and the weeding times recorded. Percent weed control on January 7 was calculated based on zero percent weed control in the untreated controls.

Viburnum odoratissimum growing in 3-gallon containers at Lake Brantley Plant Corporation in Center Hill in Sumter County, FL were treated in December 2004 in a randomized block trial with four replications of 375 plants each (10,500 total plants). Applications were made with a hand-held broadcast spreader to dry foliage on jammed pots, followed by application of 0.5 to 0.75 inches of irrigation water. There were so few weeds in the plots on February 22, 2005 that weeds were counted from each plot as they were hand-weeded, but weeding times and percent weed cover were not taken. On April 28, all weeds were collected, counted, identifi ed, and fresh and dry weights taken. However, the block formation was no longer valid as plants had been moved and set out in different nursery beds in March and no statistics could be applied to the fi nal data.

Results and Discussion: Weed pressure, types of weeds, production practices and data collected varied by nursery. However, in both situations the herbicides performed acceptably in controlling weeds. Weed pressure was much lower at the Center Hill nursery, possibly because it is surrounded by fi elds and pasture and the shrubs quickly fi lled in the pots and shaded out weeds. However, dog fennel, which is more typically a weed found in pastures and row crops, was a signifi cant weed species in these containers. Both nurseries were blessed with the more typical Florida nursery weeds such as hairy bittercress, eclipta, wild carrot and shepherdspurse.

All herbicide treatments reduced weed growth signifi cantly compared to the untreated controls (Tables 1, 2). However Dimension, applied as a fertilizer- based formulation, seems to have encouraged weed growth at the Center Hill site, possibly because surface applied fertilizer has been shown to encourage weed growth (4, 2). While BroadStar appeared to perform best, it was not signifi cantly different from the other treatments.

The time required to hand-weed was cut in half by all treatments except Trefl an compared to the control plots. Hand-weeding once in the mid-season was as effective as herbicide treatments in controlling weeds. The cost of hand-weeding compared to herbicides must be considered. The cost for weeding each pot on 1/7/2005 was 5.625 cents, assuming labor costs of $9.52 per hour (Florida Agricultural Statistics Service, 2005).

Literature Cited: 1. Altland, J. 2003. Weed Control in Container Crops: A Guide to Effective Weed Management through Preventive Measures. Oregon State University Extension Bulletin EM8823. 2. Broschat, T. K. and K.A. Klock-Moore. 2003. Infl uence of fertilizer placement on plant quality, root distribution, and weed growth in container-grown tropical ornamental plants. Hort Tech 13:305–308.

Weed Control Section 451 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

3. Florida Agricultural Statistics Service. 2005. Farm Labor Bulletin. Florida Department of Agriculture and Consumer Services, USDA National Agricultural Statistics Service, University of Florida, Institute of Food and Agricultural Sciences http://www.nass.usda.gov/fl 4. Foster, W. J., J.L. Adrian, and R. L. Shumak. 1990. A survey of weed control costs and strategies in container production nurseries. J. Environ. Hort. 8:133–135. 5. Stamps, R. H. 2003a. Checklist for minimizing weed management failures. Proc. of the Soc. of Amer. Florists’ Annu. Conf. on Insect and Dis. Mgt. on Ornamentals 19:9. 6. Stamps, R. H. 2003b. Effective use of preemergence herbicides for controlling weeds in ornamentals. Proc. of the Soc. of Amer. Florists’ Ann. Conf. on Insect and Dis. Mgt. on Ornamentals 19:100–107.

Table 1. Results of Hillsborough County site.

% Weed % Weed Weeding % Weed coverage control time (min) coverage Treatment Formulation Rate 1/7/2005 1/7/2005 3/7/2005 3/7/2005 Snapshot 2.5 200 8.475 cz 76 ab 3.608 b 61.7 cb lb/acre Showcase 2.5 200 4.100 c 89 a 2.675 b 39.5 cb lb/acre OH2 3 100 9.975 c 66 ab 3.725 b 71.8 cb lb/acre Rout 3 100 7.375 c 67 ab 3.225 b 52.1 cb lb/acre Broadstar 0.25 150 3.500 c 76 ab 2.310 b 25.2 c lb/acre Dimension 0.15 333 4.913 c 85 a 3.158 b 43.0 cb lb/acre Trefl an 5 200 18.063bc 60 ab 5.368 ab 87.5 b lb/acre Handweeded 26.975ab 44 b 3.043 b 49.5 cb 1/7/2005 Untreated 40.088 a 0 c 7.708 a 82.9 a ZValues within each column followed by different letters are signifi cantly different by Duncan’s multiple range test at P < 0.05.

452 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Table 2. Results of Sumter County site.

No. Weed Weed Dry No. Weeds Weeds Fresh Wt. Wt. (g) (per plot of (total of (g) (total of (total of 375 plants) all plots) all plots) all plots) Treatment Formulation Rate 2/22/2005 4/28/2005 4/28/2005 4/28/2005 Snapshot 2.5 200 237.90 20.72 6.000 cz 62 lb/acre Showcase 2.5 200 405.27 39.79 6.000 c 19 lb/acre OH2 3 100 830.34 81.88 12.333 c 38 lb/acre Rout 3 100 611.94 66.32 8.667 c 33 lb/acre Broadstar 0.25 150 89.67 7.98 5.667 c 7 lb/acre Dimension 0.15 333 1746.69 185.21 23.333 b 54 lb/acre Untreated 43.667 a 178 1288.50 132.87 Z Values within each column followed by different letters are signifi cantly different by Duncan’s multiple range test at P < 0.05.

Weed Control Section 453 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Molecular and Morphological Characterization of Cardamine Species

Glenn B. Fain1, James E. Altland2 and Timothy A. Rinehart1 1USDA-ARS Southern Horticultural Laboratory, Poplarville, MS 39470 2North Willamette Research and Extension Center, Oregon State University, Aurora OR 97002 [email protected]

Index Words: Cardamine hirsuta, C. fl exuosa, C. oligosperma, C. scutata, Nursery, Container, Weeds, Hairy Bittercress

Signifi cance to the Industry: This study was conducted to identify and compare species of bittercress (Cardamine) occurring in container crops and fi eld soils. Previously, researchers including the authors of this paper have referred to the bittercress in container nurseries as hairy bittercress (C. hirsuta L.). It is certainly, at least from a research perspective, important to know the species of weeds infesting container nurseries and those being studied in research trials. Weed seed used in research should be the same as those infesting production nurseries. As results of this study indicate, species in containers might not be the same as native or introduced species occurring in the landscape. Another important consideration is that different species might react differently to container nursery management practices.

Nature of Work: Cardamine sp. are problem weeds in container nurseries (3). Proper identifi cation of nursery weeds is important. It has been previously reported that Cardamine occurring in nursery containers is the introduced species C. hirsuta L.(hairy bittercress). It commonly grows as a winter annual throughout the southern, eastern and west coast areas of the US. Some have suggested that Cardamine occurring in northwest nurseries might be the native Cardamine oligosperma Nutt. (little western bittercress). Others have suggested that the introduced species C. fl exuosa With. (woodland bittercress) might also occur in nurseries and greenhouses.

Seed or plants were collected from container plants in Alabama (ALC), Mississippi (three locations MSC1, MSC2, and MSC3), New York (NYC), Oregon (ORC), and Virginia (VAC). Seed were also purchased as C. hirsuta from England (UKH) (Herbiseed). Another seed source was provided from Virginia (VAF) labeled as C. fl exuosa. Plants were also collected from the landscape in Mississippi (MSL) and Oregon (ORL) and from container plants from Florida (FLC). Seed for all but MSL and FLC were sown in a germinating mix and placed in a growth chamber. Stem and leaf tissue from two week old seedlings were used in a genetic analysis with the exception of the ORL plants due to a failure to germinate. DNA was extracted from seed for ORL and stems and leaves for MSL and FLC. Morphological characteristics of each taxon were examined with special attention to stamen number; stem, petiole, and rachis color; leaf shape (especially of young leaves); and stem and leaf pubescence (hairs).

454 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Total genomic DNA was extracted from Cardamine tissue using Plant MiniKit (Qiagen, Valencia, CA) and subjected to PCR amplifi cation of the internal transcribed regions (ITS1 and ITS2) using the primers and methods described by White et al. (4). Amplifi ed DNA was sequenced using BigDye version 3.1 (Applied Biosystems, Hayward, CA). Data were aligned using CLUSTAL and analyzed using UPGMA with 1000 bootstrap replicates.

Results and Discussion: Phylogenetic analysis of the ITS sequence data suggest there were three distinct species of Cardamine in our collection. Based on DNA sequence data, plants from ORL, VAC, UKH, and MSL are C. hirsuta with an average pairwise sequence similarity of 98.2% (Figure 1). Similarly, data for MSC1, MSC2, MSC3, ALC, NYC, and VAF match reference sequences for C. scutata Thunb. (Japanese bittercress, native to Alaska) with an average pairwise sequence similarity of 99.4%. Bootstrap support for these groups is high which also supports our conclusions (data not shown).

The ORC sample produced ITS data that could not be identifi ed based on sequences deposited in GenBank. However, confi dence is high that it is not C. hirsuta or C. scutata. It is likely that ORC is a species not represented in GenBank at this time.

Morphological data support the phylogenetic analysis. ORL, VAC, UKH, and MSL all exhibit pubescent leaves and 100% of the fl owers for all specimens had 4 stamens. By comparison the leaves of all other taxa were glabrous and 100% of the fl owers with 6 stamens. The literature supports pubescent leaves and 4 stamens per fl ower for C. hirsuta (1, 2). While ORC was not positively identifi ed by phylogenetic analysis, morphologically it does appear similar to descriptions of C. oligosperma, and thus could be as speculated.

Other observations were a reddish stem and rachis for VAC, UKH, and MSL. Also observed was a cordate terminal leafl et base for VAC, UKH, and MSL that was not as pronounced on the other samples. In our growth chamber study all Cardamine seedlings were subjected to a 12 hour photoperiod under which VAC and UKH grew prolifi cally but never fl owered. It was not until these plants were subjected to night interrupted lighting did fl owering initiate. This warrants further testing in the effort to determine why C. hirsuta might not be occurring in nurseries in locations where it is prolifi c in the landscape. It should also be noted that this survey did not exhaustively identify bittercress samples throughout all container and landscape areas within each region.

Literature Cited: 1. Al-Shehbaz I.A. 1998. The genera of Arabideae (Cruciferae; Brassicaceae) in the southeastern United States. J. Arnold Arbor. 69:85-166. 2. Marhold, K. 1995. Taxonomy of the genus Cardamine L. (Cruciferae) in the Carpathians and Pannonia. III. Folia Geobot. Phytotax., Praha. 30:397-434. 3. Ryan, G.F. 1977. Multiple herbicide applications for bittercress control in nursery containers. HortScience. 12:158-169.

Weed Control Section 455 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

4. White, T. J., T. Bruns, S. Lee, and J. W. Taylor. 1990. Amplifi cation and direct sequencing of fungal ribosomal RNA genes for phylogenetics. Pp. 315-322 In: PCR Protocols: A Guide to Methods and Applications, eds. Innis, M. A., D. H. Gelfand, J. J. Sninsky, and T. J. White. Academic Press, Inc., New York.

Figure 1. UPGMA methods were used to construct a phylogram from the ITS sequence alignment. C. crassifolia was used to root the tree. Samples appearing on a single vertical line, or branch, are 100% identical. Horizontal branches indicate sequence divergence with longer branch lengths corresponding to an increased number of differences, or divergence, between samples. Samples that group together suggest a shared ancestry.ancestry.

456 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

A New Herbicide for the Nursery Industry

Mark Czarnota1 Darin Lickfeldt2 and Joe Neal3 1University of Georgia, Department of Horticulture, 1109 Experiment Street, Griffi n GA, 30223 2Dow AgroSciences, LLC, 9330 Zionsville Rd., Indianapolis, IN 46268 3North Carolina State University, Raleigh, NC 27695-7609

Index Words: Bittercress, Cardimine hirsuta; crabgrass, Digitaria sanguinalis; eclipta, Eclipta prostrata; common groundsel, Senecio vulgaris; morningglory, Ipomoea spp.; mulberry weed, Fatoua villosa; phyllanthus, Phyllanthus spp.; sowthistle, Sonchus asper; spurge, Chamaesyce spp.; woodsorrel, Oxalis spp. Broadstar, OH2, Ronstar, Showcase, Snapshot, trifl uralin, isoxaben, oxyfl uorfen, pendimethalin, fl umioxazin, oxadiazon, preemergence herbicides

Signifi cance to Industry: From these southeastern US trials, it appears that Showcase™ Specialty Herbicide at 5.0 lb ai/A can provide excellent control of many common weeds in container nursery. This three-way combination of oxyfl uorfen with trifl uralin and isoxaben appears to improve weed control relative to Snapshot™ Specialty Herbicide. On plants with whorled foliage, growers should still choose a herbicide not containing oxyfl ourfen with a long history of plant tolerance, such as Snapshot™ Specialty Herbicide. The addition of Showcase™ Specialty Herbicide to the arsenal of granular herbicides will provide growers with an additional tool to control weeds.

Nature of Work: All container nurseries in the United States continually struggle to control weeds (Derr et al. 1997; Rice 1992). To address this problem, agricultural chemical manufacturers are constantly testing new products and formulations to meet this need. With great success, Dow AgroSciences has been marketing the product Snapshot™ Specialty Herbicide to the landscape and container nursery industry for over 20 years. Snapshot 2.5 TG is a granular herbicide containing 2.0% trifl uralin and 0.5% isoxaben, and is used to control a broad spectrum of broadleaf and grass weeds from seed. To try and increase effi cacy, Dow has incorporated oxyfl uorfen into a preemergence, three-way combination herbicide. The new product was released in 2005 and is being marketed under the trade name Showcase™ Specialty Herbicide. Showcase is a granular herbicide containing 2.0% trifl uralin, 0.25% isoxaben, and 0.25% oxyfl uorfen. Experiments were conducted nationwide in both 2004 and 2005. Trials in the Southeastern U.S. were conducted in both Raleigh, North Carolina and Griffi n, Georgia. Identical treatments were applied at both locations, and a treatment list is presented in Table 1. Treatments were arranged in a randomized complete block design with 4 replications, and each treatment of each replication contained at least 2 subsamples. At the Griffi n location, all one gallon pots were over seeded with a ¼ teaspoon of a weed seed mix immediately before treatments were applied. Weed seed mix contained 1 parts by volumn large crabgrass (Digitaria sanguinalis), morningglory (Ipomoea hederacea and I. purpurea), hairy bittercress (Cardamine hirsuta), spotted spurge (Chamaesyce maculata), mulberryweed (Fatoua villosa), woodsorrel (Oxalis stricta and O. corniculata). Before treatments were applied, all plants in a particular treatment were arranged in a 6ft. x 6ft. square. Pre-weighed herbicides were

Weed Control Section 457 SNA RESEARCH CONFERENCE - VOL. 50 - 2005 then uniformly applied with a shaker jar to the 6ft. x 6ft. area. In NC weeds were seeded to separate pots, with 3 pots of each species per treatment per replicate (12 pots of each species total). In NC, all pots were hand weeded, re-treated, and re-seeded 8 weeks after the initial treatment. Depending on the location, percent control was visually estimated between 1 and 18 weeks after treatment (WAT). A description of the rating scale is presented in Table 2. All data were subjected to analysis of variance (ANOVA) and means were separated using Fisher’s least signifi cant difference (LSD) test with a signifi cance level of P=0.05.

Results and Discussion: In 2004, ornamentals tested at Raleigh, North Carolina and Griffi n, GA are listed in Table 3. Weeds evaluated at both sites are presented in Table 4. At the North Carolina site, ‘Night Beacon’ daylillies were signifi cantly injured by Showcase, OH2, Broadstar and Ronstar, but not by Snapshot TG. The injury observed was “contact type” damage of the new foliage. Foliage emerging after treatment was generally not affected and plants had recovered by 8 weeks after treatment. No other species were injured by any of the treatments at the Raleigh location. In Griffi n, no injury was recoded to either Fothergilla or the Fosteri holly (data not shown).

At the Raleigh location, large crabgrass, oxalis, and spiny sowthistle control was greater than or equal to 70% for the duration of the experiment (up to 10 weeks after treatment) (data not shown). Spotted spurge was controlled by all treatments except Ronstar. Longstalked phyllanthus, and eclipta were more diffi cult to control. Based on data following the 2nd applications, phyllanthus and eclipta were controlled for about 4 weeks by Snapshot TG whereas Showcase, OH2, Broadstar and Ronstar controlled phyllanthus for 10 weeks. Eclipta was not well controlled by Ronstar, or low rates of Showcase, but Broadstar and the high rate of Showcase provided >70% eclipta control 8 weeks after treatment (Tables 5 and 6).

At the Griffi n location, control of morningglory was acceptable until 4 weeks after treatment (WAT). At this rating date both the OH2 and Ronstar were providing signifi cantly less control than the other herbicide treatments. Control continued to decrease with all herbicide treatments during the fi nal 2 ratings, and by 14 WAT none of the herbicide treatments were signifi cantly different from the untreated control (data not shown). With crabgrass, control began to break at 8 WAT. OH2 and the low rate of Showcase and Snapshot were providing less than 70% control. By 14 WAT, only Ronstar and the high rate of Showcase were providing control greater than 70% (data not shown). Hairy bittercress, mulberry weed, woodsorrel, and spotted spurge control didn’t break until 14 WAT (Table 7.). Oxalis and mulberry weed control at 14 WAT was ≤ 73.3% with both the high rate of Showcase and Ronstar. At 14 WAT, control of bittercress was poor with all herbicide treatments. Spurge was completely controlled for the duration of the experiment with the high rate of Showcase, Broadstar, and Ronstar.

Signifi cance to Industry: From these southeastern US trials, it appears that Showcase™ Specialty Herbicide at 200 lb P/A can provide excellent control of many common weeds in containerized ornamentals. This three-way combination of oxyfl uorfen with trifl uralin and isoxaben appears to increase weed control

458 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005 relative to Snapshot™ Specialty Herbicide. On plants with whorled foliage, growers should still choose a herbicide not containing oxyfl ourfen with a long history of plant tolerance, such as Snapshot™ Specialty Herbicide. The addition of Showcase™ Specialty Herbicide to the arsenal of granular herbicides will provide growers with an additional tool to control weeds.

Literature Cited: 1. Derr, J. F., J. C. Neal, L. J. Kuhns, R. J. Smeda, L. A. Weston, C. Elmore, C. A. Wilen, J. Ahrens, A. Senesac and T. Mervosh. 1997. Weed Management in Landscape and Nursery Plantings. In M. E. McGiffen, eds. Weed Management in Horticultural Crops. Alexandria, VA: ASHS. 139. 2. Rice, R. P. 1992. Nursery and Landscape Weed Control Manual. Fresno. Thomson. 290 p.

Table 1. List of treatments.

Rate Rate (active # Trade name Active ingredient Formulation* (product/ ingredient Acre)* /Acre)* 1 Showcase trifl uralin / isoxaben / 2.5 GR 100 lb pr/A 1.0 lb ai/A oxyfl uorfen 2 Showcase trifl uralin / isoxaben / 2.5 GR 150 lb pr/A 3.75 lb ai/A oxyfl uorfen 3 Showcase trifl uralin / isoxaben / 2.5 GR 200 lb pr /A 5.0 lb ai/A oxyfl uorfen 4 Snapshot trifl uralin / isoxaben 2.5 GR 100 lb pr/A 2.5 lb ai/A 5 Snapshot trifl uralin / isoxaben 2.5 GR 150 lb pr/A 3.75 lb ai/A 6 Snapshot trifl uralin / isoxaben 2.5 GR 200 lb pr/A 5.0 lb ai/A 7 OH2 pendimethalin / 3.0 GR 100 lb pr/A 3.0 lb ai/A oxyfl uorfen 8 BroadStar fl umioxazin 0.25 GR 150 lb pr/A 0.375 lb ai/A 9 Ronstar oxadiazon 2.0 GR 150 lb pr/A 3.0 lb ai/A 10 Untreated *lb pr/A = pounds product per Acre; GR = granular formulation.

Weed Control Section 459 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Table 2. Representations of numeric weed control ratings.

Value Ranges Plant Symptoms 0 No visual injury present 10-30 Minimal injury to desirable plant. Less than 10% of the plant leaf surface area showing chlorosis and necrosis. A 10 to 30% biomass reduction. 40-70 More noticeable plant injury or stunting. Greater than 50% of the leaf area showing symptoms of chlorosis and/or necrosis. A 40 to 70% biomass reduction. 80-90 Plants severely injured. Most of the leaves and leaf surface showing signs of chlorosis and necrosis. An 80 to 90% biomass reduction. 100 Plant appears dead. No signs of regrowth.

Table 3. Ornamentals tested in Raleigh, North Carolina and Griffi n, Georgia.

Common Name Scientifi c Name Butterfl y-bush Buddleja davidii ‘Pink Delight’ Boxwood Buxus sinica var. insularis ‘Wintergreen’ Dwarf Fothergilla Fothergilla gardenia* Daylily Hemerocallis spp. ‘Night Beacon’ Shrub Althea Hibiscus syriacus ‘Lucy’ Foster’s Holly Ilex x attenuate ‘Fosteri’* Carrisa Holly Ilex cornuta ‘Carrisa’ Common Crapemyrtle Lagerstroemia indica ‘Natchez’ Waxmyrtle Myrica cerifera Azalea Rhododendron X ‘Formosa’ Doublefi le viburnum Viburnum plicatum var. tomentosum *Plants tested at Griffi n.

Table 4. Weeds evaluated at both the Griffi n, Georgia and Raleigh, North Carolina locations.

Common Name Scientifi c Name Hairy bittercress Cardamine hirsuta Large crabgrass Digitaria sanguinalis Eclipta Eclipta prostrataa Spotted spurge Chamaesyce maculata Mulberryweed Fatoua villosab Morningglory Ipomoea hederacea and I. purpureab Oxalis Oxalis stricta and O. corniculata Longstalked Phyllanthus Phyllanthus tenellusa Spiny sowthistle Sonchus aspera aWeeds evaluated only at Raleigh, North Carolina. bWeed evaluated only at Griffi n, Georgia.

460 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005 24 26 0 d 30 c 45 c 70 b 95 a 22 cd 83 ab 100 a 100 a 100 a 10WAT . a 24 24 0 d 88 a 94 a 95 a 52 b 20 cd 34 bc 100 a 100 a 100 a 8WAT 35 32 0 c 90 a 99 a 39 b 81 a 74 ab 66 ab 100 a 100 a 100 a 6WAT 12 10 0 c 95 a 99 a 68 b 75 b 93 a 100 a 100 a 100 a 100 a Longstalked phyllanthus control 4WAT 20 17 0 c 84 a 90 a 96 a 86 a 64 a 96 a 92 a 100 a 100 a 6WAIT ) control ratings in 2004 at the Raleigh, North Caroline location LSD Variation Coeffi cient of Coeffi Rate (active ingredient /Acre)* 1.0 lb ai/A 3.75 lb ai/A 5.0 lb ai/A 2.5 lb ai/A 3.75 lb ai/A 5.0 lb ai/A 3.0 lb ai/A 0.375 lb ai/A 3.0 lb ai/A Phyllanthus tenellus Formulation 2.5 GR 2.5 GR 2.5 GR 2.5 GR 2.5 GR 2.5 GR 3.0 GR 0.25 GR 2.0 GR Longstalked phyllanthus ( Trade name Trade Showcase Showcase Showcase Snapshot Snapshot Snapshot OH2 BroadStar Ronstar Untreated # 1 2 3 4 5 6 7 8 9 10 Abbreviations: WAIT = weeks after initial treatment; WAT = weeks after 2nd treatment; GR, granular. = weeks after initial treatment; WAT Abbreviations: WAIT according to Fisher’sMeans within a column followed by the same letter are not different protected LSD at P=0.05. Table 5. Table a b

Weed Control Section 461 SNA RESEARCH CONFERENCE - VOL. 50 - 2005 29 82 0 c 62 a 12 cd 41 ab 18 bcd 35 abc 18 bcd 22 bcd 15 bcd 15 bcd 10WAT 33 59 0 c 72 a 35 b 70 a 30 bc 30 bc 20 bc 40 ab 50 ab 40 ab 8WAT 40 53 . 0 d a 89 a 89 a 34 cd 41 bc 45 bc 30 cd 76 ab 6WAT 55 abc 61 abc Eclipta control (%) 26 26 0 d 45 c 87 a 95 a 97 a 60 bc 77 ab 75 ab 81 ab 76 ab 4WAT 28 26 0 c 62 b 92 a 94 a 94 a 60 b 81 ab 78 ab 86 ab 84 ab 6WAIT LSD Variation Coeffi cient of Coeffi Rate (active ingredient /Acre)* 1.0 lb ai/A 3.75 lb ai/A 5.0 lb ai/A 2.5 lb ai/A 3.75 lb ai/A 5.0 lb ai/A 3.0 lb ai/A 0.375 lb ai/A 3.0 lb ai/A ) control ratings in 2004 at the Raleigh, North Caroline location Formulation 2.5 GR 2.5 GR 2.5 GR 2.5 GR 2.5 GR 2.5 GR 3.0 GR 0.25 GR 2.0 GR Eclipta prostrata Eclipta ( Trade name Trade Showcase Showcase Showcase Snapshot Snapshot Snapshot OH2 BroadStar Ronstar Untreated # 1 2 3 4 5 6 7 8 9 10 Abbreviations: WAIT = weeks after initial treatment; WAT = weeks after 2nd treatment; GR, granular. = weeks after initial treatment; WAT Abbreviations: WAIT according to Fisher’sMeans within a column followed by the same letter are not different protected LSD at P=0.05. Table 6. Table a b

462 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005 69 64 0 b 33 ab 67 ab 100 a 67 ab 67 ab 67 ab 33 ab 100 a 100 a 14 WAT and Spurge 9 13 0 c 87 b 100 s 93 ab 100 a 93 ab 100 a 97 ab 100 a 93 ab 8 WAT Oxalis stricta 70 0 c 0 c 101 93 a 13 bc 73 ab 46 abc 53 abc 27 abc 33 abc 67 abc 14 WAT . a 11 14 ), woodsorrel ( 0 c Mulberryweed 93 a 93 a 73 b 73 b 93 a 93 a 87 ab 87 ab 87 ab 8 WAT 58 0 b 0 b 129 63 a 20 ab 36 ab 13 ab 13 ab 33 ab 36 ab 46 ab Fatoua villosa 14 WAT Bittercress 19 16 0 d 63 c 93 a 70 bc 70 bc 83 ab 8 WAT 80 abc 77 abc 80 abc 80 abc ), mulberryweed ( 85 0 a 109 33 a 50 a 80 a 53 a 36 a 33 a 33 a 56 a 76 a 14 WAT ) in 2004 at the Griffi n, Georgia location ) in 2004 at the Griffi Oxalis 17 14 0 d 70 c 93 a 73 bc 73 bc 90 ab 8 WAT 86 abc 83 abc 87 abc 87 abc Cardamine hirsuta Euphorbia maculata LSD Coeffi cient Coeffi of Variation Rate (active ingredient / Acre)* 1.0 lb ai/A 3.75 lb ai/A 5.0 lb ai/A 2.5 lb ai/A 3.75 lb ai/A 5.0 lb ai/A 3.0 lb ai/A 0.375 lb ai/A 3.0 lb ai/A Formulation 2.5 GR 2.5 GR 2.5 GR 2.5 GR 2.5 GR 2.5 GR 3.0 GR 0.25 GR 2.0 GR ), and spotted spurge ( Percent control of hairy bittercress ( Trade Trade name Showcase Showcase Showcase Snapshot Snapshot Snapshot OH2 BroadStar Ronstar Untreated # 1 2 3 4 5 6 7 8 9 10 Means within a column followed by the same letter are not different according to Fisher’sMeans within a column followed by the same letter are not different protected LSD at P=0.05. Abbreviations: WAT, weeks after treatment; GR, granular. Abbreviations: WAT, Table 7. Table O. corniculata º b

Weed Control Section 463 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Oryzalin Movement in One-gallon Containers from Spray and Oryzalin-created Mulch Applications

Luke Case, Hannah Mathers, S. Kent Harrison, Sara Lowe and Alejandra Acuna Dept. Horticulture and Crop Science, The Ohio State University, Columbus, OH, 43212 [email protected]

Index words: herbicide leaching

Signifi cance to the industry: Three to fi ve preemergence herbicide applications are often needed per growing season to maintain acceptable weed control in container nurseries. At The Ohio State University, we are looking at ways to reduce herbicide use via mulches as herbicide carriers. Reducing herbicide use would be environmentally and economically benefi cial to the grower. With mulches as herbicide carriers, the number of herbicide applications could be reduced to one per year. Herbicide-treated mulches also reduce phytotoxicity of sprayable herbicides.

Nature of work: Mathers (3), and Case et al. (1), found that oryzalin-treated pine nuggets extended effi cacy to 130 and 115 DAT (days after treatment), respectively, compared to the over-the-top spray of oryzalin in one-gallon (3.8 L) pots. Fretz et al. (2) also found that herbicide-treated organic mulch provided control comparable to the conventional sprays up to 77 DAT. However, there are no studies showing herbicide movement in containers from herbicide-treated mulch. This study’s objective was to show the movement of the oryzalin in one- gallon pots from applications of an over-the-top spray and oryzalin-treated pine nuggets compared to an untreated control using a bioassay.

Oryzalin-treated mulch and direct sprays were applied at 2.0 lbs ai/ac (2.2 kg ai/ha) to one-gallon pots fi lled with a 7:1 aged pine bark:sand mixture. Mulches were treated by laying them out at one unit-layer thickness on a piece of plastic. One unit-layer thickness represents nuggets of pine bark laying side by side on the plastic with minimal overlap. Mulches were then sprayed in a spray chamber set at 20.2 gallons per acre (188.6 liters/hectare). After the mulches were sprayed, they were allowed to dry for 48 hours before applying to the pots at one unit layer thickness. The study was repeated in time, with trial 1 starting on January 12, 2004 and trial 2 starting on November 15, 2004. Both were conducted in a glass greenhouse at The Ohio State University, Columbus, OH set at 75° F (24° C) day time temperatures and 65° F (18° C) night time temperatures. In trial 1, there were six dates of evaluation: 0, 4, 8, 16, 32, and 64 DAT, and in trial 2, a seventh evaluation was added at 128 DAT. An oat (Avena sativa) bioassay was conducted on three pot levels (0-2 cm, 2-8 cm, and 8-15 cm) and the leachate to determine herbicide presence at the different levels at each evaluation date. The leachate was kept in a silanized glass bottle until the end of the evaluation period. Media from each pot level was thoroughly mixed and a 55.0 gram sample was taken and put into plastic Petri dishes. Five pregerminated oat seeds were placed in each dish. Seeds were allowed to grow

464 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005 for 24 hours; radical tips were marked, and were then measured at 48 hours. The procedure was similar for the leachates, except that the seeds were put in 110 grams of sand fi lled to fi eld capacity with leachate water in the Petri dishes.

Results and Discussion: In trial 1, pots with direct sprays showed more herbicide presence in the top 2 cm than the oryzalin-treated mulch pots at each of the evaluation dates (Fig. 1). In trial 2, results were similar in the 0-2 cm zone up to 32 DAT. At 32 DAT in trial 2, there were very similar amounts of oryzalin between the two oryzalin treatments (Fig. 3), and this lasted until 64 DAT. At 128 DAT, there was more oryzalin present in the oryzalin-treated pine treatments. In both trials, there was a signifi cant increase in herbicide presence in the oryzalin- treated pine nugget pots at the 0-2 cm level from 0 to 4 DAT, suggesting that the mulch does retain the herbicide. Also, results from both trials indicated more herbicide leaching into the 2-8 cm zone with the direct sprays compared to the pots containing oryzalin-treated pine nuggets (Figs. 2 and 4). In trial 2, there was indication of the herbicide getting into the 8-15 cm zone from the direct spray treatment up to 8 DAT (data not shown). There were no signs of herbicide presence in the leachates from any of the treatments from either trial (data not shown).

Literature Cited: 1. Case, L., H.M. Mathers, and N. Tuttle. 2002. Herbicide-treated mulches for ornamental weed control. Proc. Northeastern Weed Sci. Soc. 56:72. 2. Fretz, T.A. and C.W. Dunham. 1971. The incorporation of herbicides into organic mulches for weed control in ornamental plantings. J. Amer. Soc. Hort. Sci. 96(3):280-284. 3. Mathers, H. 2003. Novel methods of weed control in containers. HortTechnology 13:28-31.

Weed Control Section 465 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Figure 1. Radicle lengths of Avena sativa grown in media from the 0-2 cm depth from different treatments across all dates in trial 1.

Figure 2. Radicle lengths of Avena sativa grown in media from the 2-8 cm depth from different treatments across all dates in trial 1.

466 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Figure 3. Radicle lengths of Avena sativa grown in media from the 0-2 cm depth from different treatments across all dates in trial 2.

Figure 4. Radicle lengths of Avena sativa grown in media from the 2-8 cm depth from differentdifferent treatments across all dates in trial 2.

Weed Control Section 467 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Controlling Liverwort (Marchantia polymorpha) Infestations

Adam Newby1, James Altland2, Charles Gilliam1, Glenn Wehtje1 and Donna Fare3 1Department of Horticulture, Auburn University, Auburn, AL 36849 2North Willamette Research and Extension Center, Aurora, OR 97002 3USDA-ARS National Arboretum, McMinnville, TN 37110 [email protected]

Index words: quinoclamine, Diuron, Linuron

Signifi cance to the Industry: Liverwort continues to spread throughout the Southeast as a weed problem in container-grown ornamental crops. The goal of this study to better understand the use of a new product being registered for liverwort control and to continue evaluating alternative controls.

Nature of Work: Liverwort (Marchantia polymorpha) is an ever increasing problem within the Southeast. It is a physiologically primitive plant with no vascular system. Marchantia polymorpha can be identifi ed by prostrate leaf-like structures called thalli that create a mat over media surfaces. They propagate sexually by spores and asexually by gemmae. Thalli can cover the entire media surface of a container and restrict water and nutrient movement into the root zone, as well as reduce the marketability of a crop (Svenson, 1998). Liverwort thrives in low UV light, high fertility, high moisture, and high humidity environments. Propagation houses, shade houses, and other covered structures provide ideal conditions for liverwort (Svenson, 1997).

Liverwort was initially restricted to the northeastern and northwestern United States, but is now found all across the country. This emerging weed problem has necessitated the need for additional research. One herbicide with potential is quinoclamine which has been used as an algaecide in Japan for decades. It is produced as a 25% wettable powder. Quinoclamine has proven to be very effective on postemergence liverwort control, and it is safe on a broad range of ornamental crops (Altland 2003; Newby 2004). The current recommendation for quinoclamine is 2 oz per gallon applied at 2 qt. per 100 ft2 (219 gal/A). Previous research suggests that lower rates and spray volumes provide adequate postemergence control. In 2004, a quinoclamine rate of 1 oz product/gal applied at 1 qt. per 100 ft2 (109 gal/A) provided adequate control (Newby).

The nursery industry in Germany has used diuron for liverwort control (personal communication with Dr. Heinrich Loesing). Diuron is an herbicide widely used in cotton production within the Southeast. It is highly active and inexpensive.

The objective of this study was to evaluate lower rates and spray volumes of quinoclamine than current recommendations and any interaction between rate and volume as well as evaluate the effectiveness of diuron for liverwort control.

468 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Materials and Methods: This study was conducted at Auburn University in the fall of 2004. Liverwort was grown in full-gallon containers consisting of a 6:1 pine bark to sand substrate amended with 14 lb of Polyon 18-6-12, 5 lb of dolomitic lime, and 1.5 lb of Micromax per cubic yard. Postemergence treatments were applied on 4 November 2004 when liverwort covered at least 60% of the container surface. Treatments of quinoclamine were applied in a factorial arrangement consisting of four rates and three spray volumes. Rates of 0.25, 0.5, 1.0 or 2.0 oz product/gal (0.0625, 0.125, 0.25, or 0.5oz ai/gal) were each applied at 27, 54, or 109 gal/A (0.25, 0.5, or 1.0 qt/100 ft2). Two herbicides containing diuron, Diuron 4L and Linuron 4L (Linuron does not contain diuron, it is a completely separate active ingredient), were each applied at 0.5 lb ai/A and 1 lb ai/A at 40 gal/A. All treatments were applied with a CO2 backpack sprayer fi tted with an 8004 fl at fan nozzle at 30 PSI. A non-treated control group was maintained. Treatments were arranged in a completely randomized design with 6 single pot replications. In addition to liverwort, all treatments were also applied to 6 single pot replications of Humata tyermanii (Rabbit foot fern) and Euphorvia pulcherrima (Poinsettia). The study was conducted in a temperature controlled greenhouse under 0.25 inch overhead cyclic irrigation per day split into two cycles. Percent control was recorded at 3, 7, 14, and 28 DAT on a 0 to 100 percent scale where 0 equals no control and 100 equals death of entire liverwort within the container. As a measure of liverwort re-growth, percent of the substrate surface covered with living liverwort was recorded 35 and 70 DAT.

Results and Discussion: Main effects of quinoclamine rate, volume, as well as the interaction thereof were found to be signifi cant according to analysis of variance (p < 0.05). In general, control increased as rate and spray volume increased. Increased spray volume is more likely to improve control at lower rates. At 3 DAT, quinoclamine applied at 0.5, 1.0, and 2.0 oz product/gal at 109 gal/A provided 85 to 99% control (Table 1). Similarly, 1.0 and 2.0 oz product/gal applied at 54 gal/A provided 95% control. Among quinoclamine treatments, higher spray volume provided greater control with lower rates (you already said this above, delete on of these sentences or reword it so you are not reducntand). For example, at 14 DAT quinoclamine applied at 0.25 oz product/gal at 109 gal/ A provided 53% control while the same rate applied at 27 gal/A provided only 28% control. Conversely, lower spray volumes (54 gal/A) can provide adequate control at higher rates (1.0 and 2.0 oz product/A). By 70 DAT, 2.0 oz product/gal applied at 54 and 109 gal/A had the least percent liverwort coverage among the quinoclamine treatments with 40% and 22% coverage. All other quinoclamine treatments had a higher percentage of liverwort coverage. Quinoclamine treatments were compared to Diuron and Linuron treatments as well as the non-treated control group using Duncan’s multiple range test (α = 0.05). Diuron and Linuron treatments were similar to the non-treated control 3 DAT. By 14 DAT, Diuron applied at 1.0 lb ai/A provided similar control to the most effective quinoclamine treatments. Diuron applied at 1.0 lb ai/A provided similar long term control to 2.0 oz product/gal of quinoclamine at 54 and 109 gal/A 70 DAT with only 16% liverwort coverage.

No injury was recorded on Humata tyermanii or Euphorvia pulcherrima at any time throughout the study.

Weed Control Section 469 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

These data show that quinoclamine rate and volume of application infl uence postemergence liverwort control. Rate and spray volume may best be determined by individual growers. Heavy infestations may require 1.0 to 2.0 oz product/gal in 54 to 109 gallons of water per acre. Conversely, if the infestation is lighter, a lower rate (0.5 oz product/gal) may provide adequate control, however the high volume of water (109 gal/A) is required. This study also recognizes Diuron as a promising treatment for liverwort with excellent long term control.

Literature Cited: 1. Altland, J.E., A. Newby, and R. Regan. Determine effi cacy and phytotoxicity of quinoclamine. Comb. Proc. Intl. Plant Prop Soc. 53:383-386. 2. Newby, A., J. Altland, D. Fare, C. Gilliam, and G. Wehtje. 2004. Postemergence Control of Liverwort in Container Production. Proc. Southern Nurs. Res. Conf. 49:396-400. 3. Svenson, S.E. 1997. Controlling liverworts and moss in nursery production. Comb. Proc. Intl. Plant Prop. Soc. 47:414-422. 4. Svenson, S.E. 1998. Suppression of liverwort growth in containers using irrigation, Mulches, Fertilizers and Herbicides. Proc. Southern Nurs. Res. Conf. 43: 396-398.

470 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005 a a a a ab ab a bc bc ab cde de cd e a a a 70 DAT 97 97 87 89 73 79 88 59 54 73 40 22 44 16 94 100 100 % Coverage fg ef de cd cd abc fg ab abc abcd a a bcd abc fgh gf h 4 1 14 DAT 28 36 53 61 66 87 28 94 83 78 97 98 70 86 17 fects of rate and volume as well the interaction thereof % Control =0.05). α Z X ef e cd d cd ab ef ab a bc ab a efg fg efg fg g 7 2 3 DAT 11 25 34 60 56 64 85 27 95 98 76 95 99 18 18 uenced by the main ef

uenced by the main effects of rate and volume as well the interaction thereof uenced by the main effects s multiple range test ( 27 54 27 54 27 54 27 54 40 40 40 40 109 109 109 109 gal/A Volume 0.5 1.0 2.0 0.5 1.0 0.5 1.0 0.25 Rate lb ai/A oz product/gal Y Liverwort control with quinoclamine as infl uenced by rate and volume with selected treatments of Diruon Linuron. Liverwort control with quinoclamine as infl =0.05). α Herbicide Quinoclamine Diuron Linuron Control Means within same column with the letter are similar according to Duncan’ Days after treatment. Analysis of variance the quinoclamine treatments revealed that all responses were infl ( Table 1. Table Z Y X

Weed Control Section 471 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Field Evaluation of Various Herbicide and Mulch Combinations For Ornamental Weed Control

Hannah Mathers and Luke Case The Ohio State University, Columbus, OH 43210 [email protected]

Index Words: Ornamental production, nursery, landscape, preemergent

Signifi cance to the Industry: Weed control is the largest expense facing the nursery and landscape industries that nationally exceed $10 billion and $625 billion annually (Hall et al., 2005). In an industry where aesthetics determine profi tability, zero-tolerance of weeds is often adopted. Data indicates the integration of two tactics of weed control, mulch + preemergent herbicides can produce a positive interaction, offering a promising alternative pest management system and simplifying and enhancing the safety and effectiveness of applications through the utilization of an integrated pest management approach.

Nature of Work: Oliveira et al. (2000) found that the controlled release of herbicides using lignin as the matrix offered a promising alternative technology for weed control. Knight et al. (2001) found that the application of preemergent herbicides onto organic mulches reduced herbicide leaching by 35-74% compared with bare soil preemergent herbicide applications. This research project included two experiments and three objectives: 1) determine the effi cacy and duration of weed control of 10 herbicide-mulch combinations; 2) assess the phytotoxicity of the 10 herbicide-mulch combinations on two ornamental plants; and 3) determine effi cacy and phytotoxicity of three application methods for each herbicide-mulch combinations.

The two experiments conducted were effi cacy (experiment 1) and phytotoxicity (experiment 2). Both experiments were started on May 1, 2004, ended April 15, 2005, and are being repeated in 2005 at the Ohio State University Waterman Farm, Columbus, OH. The plots in experiment one contain no crop plants. Evaluations of effi cacy were conducted at 30, 60, 90, 120 and 350 DAT using dry weights and visual ratings from 1 X 1 ft sections in the 3 X 3 ft (0.9 m) plots. Effi cacy ratings were on a scale of 0 (no control) to 10 (complete control) and ≥ 7 (commercially acceptable). In experiment two, dogwood shrubs and crabapple tree liners were evaluated. A visual rating score of 1 (no injury) to 10 (complete kill) and ≤ 3(commercially acceptable) will be used for the shoots. The herbicide treated mulches and herbicide-mulch application methods were compared to sprays of the fi ve chemicals applied directly to the surfaces of the plots, the two untreated mulches applied to the plots and a weedy check (no herbicide, no mulch). Mulches were applied untreated, over the top of soil surfaces sprayed with the different herbicides. Mulches were also applied untreated to untreated soil surfaces and then sprayed with the different herbicides in the fi eld.

The fi ve chemicals applied were oryzalin, Surfl an (AS) (aqueous solution) 2 (ai)lb/acre, fl umioxazin, (SureGuard WDG) 0.34 (ai) lb/acre, acetochlor (Harness) 2.5 lbs ai/ac, dichlobenil (Casoron CS) 4 (ai) lb/acre and a combination

472 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005 of oryzalin and fl umioxazin. Two bark types were evaluated, pine nuggets and shredded hardwood. Pretreated bark mulch treatments were prepared by placing the mulches on a sheet of plastic, as a single layer thick and sprayed over the top with the different herbicide treatments and allowed to dry for 48 h. Treated barks when dry and untreated mulches were applied directly to evaluation plots in varying amounts according to the mulch thickness. The mulches were applied as close as possible to a single layer.

Results and Discussion: Effi cacy ratings and dry weights showed signifi cant difference with treatment and date. Only dry weights had signifi cant treatment X date interactions. Twenty of 38 treatments gave effi cacy rating of ≥ 7, pooled over all evaluation dates (Fig.1). Only one was a direct spray, Surfl an + SureGuard (7.6). Three were pretreated mulches, Surfl an + SureGuard (8.2), Harness (7.8) and Surfl an (7.4) treated pine (Fig.1). None of the pretreated hardwood barks provided ratings of ≥ 7. Eight of the 20 were treatments with the herbicides applied under the bark. Seven of the eight provided ratings of ≥ 8, Surfl an + SureGuard under pine (9.1), Casoron under pine (8.9), Surfl an under pine (8.7), Harness under pine (8.0), Surfl an + SureGuard under hardwood (8.0), SureGuard under hardwood (8.0) and SureGuard under pine (8.0) (Fig.1). Eight of 20 were treatments with the herbicides applied over the bark with fi ve providing ratings of ≥ 8, SureGuard over pine (9.1), Casoron over pine (9.0), Harness over pine (8.3), Surfl an over pine (8.3), Casoron over hardwood bark (8.0). The untreated pine (3.5) and untreated hardwood (1.5) provided signifi cantly better effi cacy than the control (0.15); however, these three treatments were three of the fi ve least effi cacious treatments in the trial (Fig.1). At 350 DAT, four treatments were still providing above commercially acceptable weed control, none were direct sprays, none involved hardwood bark and one was a pretreated mulch, Surfl an + SureGuard pretreated pine (7.3) (Fig.2). The other three treatments were Casoron over pine (8.2), Surfl an + SureGuard over pine (7.6), Casoron under pine (7.4) (Fig.2).

Literature Cited: 1. Hall, C., Hodges, A., Haydu, J. 2005. Economic Impacts of the Green Industry in the United States. Draft (in review). 2. Knight, P.R, C.H. Gilliam, S.L. File and D. Reynolds. 2001. Mulches reduce herbicide loss in the landscape. Proc. South. Nurs. Assn. Res. Conf. 46:461- 463. 3. Oliveira, S.C., Pereira, F.M., Ferraz, A., Silva, F.T. and Goncalves, A.R. 2000. Mathematical modeling of controlled-release systems of herbicides using lignins as matrices. Applied Biochem. And Biotech. Vol. 84-86:595-615.

Weed Control Section 473 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Fig.1. Effi cacy rated score data for herbicide treated mulch experiment pooled over 30, 60, 90 and 120 days after treatment (DAT). The abbreviations OV, U, P and Hdwd mean over, under, pine and hardwood bark, respectively. Different letters signify least signifi cant difference (LSD) P=0.05.

Fig.2. Effi cacy rated score data for herbicide treated mulch experiment 350 days after treatment (DAT). The abbreviations OV, U, P and Hdwd mean over, under, pine and hardwood bark, respectively. Different letters signify least signifi cant difference (LSD) P=0.05.

474 Weed Control Section SNA RESEARCH CONFERENCE - VOL. 50 - 2005

Effect of Compost Type on Bermudagrass (Cynodon dactylon) Invasion

Derald A. Harp1, Kevin Ong2, John Sloan2 and Kristen L. McDowell1 1Texas A&M-Commerce, Department of Agricultural Sciences, Commerce, Texas 75482 2Texas A&M University – Dallas, Dallas, Texas 75252

Index Words: dairy compost, poultry litter, yard waste compost, grassy weeds

Signifi cance to Industry: Bermudagrass is an important and diffi cult to control grassy weed in beds throughout the southern United States. As we evaluate various composts and compost blends for use in gardens, it becomes important to know if the spread of bermudagrass may be exacerbated by compost type.

Nature of Work: A study was conducted on the campus of Texas A&M University – Commerce in Commerce, Texas to evaluate various composts for ornamental rose growth. As a part of this study, the beds were evaluated concerning the reestablishment of bermudagrass following incorporation of 4 different compost blends. The experiment was arranged in 4 blocks with 5 treatments in each block and 5 reps per treatment.

At initiation of the study, bermudagrass was mechanically removed from each bed. No herbicides were used either before or during this study. All planting areas were then tilled to a depth of 15 cm (6”). Four treatments then received one of the four following composts: 1) poultry litter compost (PLC), 2) yard waste compost (YWC), 3) dairy compost (DC), and 4) 1:1 mixture of dairy and poultry litter composts (MC). Composts were tilled in, and roses planted. No mulch was applied to the planting area. All composts were analyzed for %N prior to incorporation. Plots were analyzed post-plant for variations in pH and EC.

Plots were visually graded once every 30 days for bermudagrass coverage. Plots were scored on a scale of 1 – 10, with each point representing an estimated 10 percent of area covered. Plots were scored until 100% coverage. Statistical analysis was conducted using an ANOVA (SigmaStat, Systat, Inc.). Means were separated using Duncan’s.

Results and Discussion: Chemical analysis of plots and composts showed only minor variation among the treatments (Table 1). The amount of nitrogen was similar in all composts. As expected, those composts derived from animal wastes were slightly higher in %N, with PLC being the highest at 2.9%. EC and pH did not differ among the treatment plots or blocks.

At day 30, PLC and YWC had signifi cantly higher bermudagrass invasion than other treatments with roughly 70% coverage in all treatment plots (Table 2). Dairy compost had more bermudagrass than the control with approximately 40% coverage. MC did not differ from the control.

Weed Control Section 475 SNA RESEARCH CONFERENCE - VOL. 50 - 2005

By day 60, the PLC plots were 100% covered, signifi cantly faster than any other treatment (Table 2). This can be explained by the slightly higher N concentration in the compost. Percent coverage in the control group was signifi cantly lower than other treatments with approximately 67% coverage. DC, MC, and YWC treatments had approximately 80% coverage.

By day 90, all treatments had reached at least 83% coverage (Table 2). Treatment C remained signifi cantly lower than other treatments. YWC matched PLC at 100% coverage. Bermudagrass in MC and DC treatments exceeded 90%. Full coverage in all treatments was reached at day 120 of the study.

The type of compost can affect the invasion of bermudagrass. Because bermudagrass responds quickly to increases in N, it is likely to invade adjacent gardens as incorporated composts increase soil fertility. This further emphasizes the need for barriers and other weed control measures in new beds and gardens.

Table 1. Chemical analysis of treatment plots. %N analysis of composts was conducted pre-incorporation.

Compost Blend %N pH EC (μS/cm) Control (None) N/A 7.12 152 Yard Waste Compost (YWC) 1.1 7.15 141 Poultry Litter Compost (PLC) 2.9 7.05 164 Dairy Compost (DC) 2.0 7.12 136 Mixed Compost (MC) 2.2 7.02 134

Table 2. Bermudagrass invasion scores for each compost blend. Each point represents 10% coverage.

Compost Blend Day 30 Day 60 Day 90 Day 120 Control (C) 3.3a 6.7a 8.3a 10.0a Dairy Compost (DC) 4.7b 8.0b 9.0b 10.0a Mixed Compost (MC) 4.3ab 7.7b 9.3b 10.0a Poultry Litter Compost (PLC) 7.7c 10.0c 10.0c 10.0a Yard Waste Compost (YWC) 6.7c 8.3b 10.0c 10.0a

476 Weed Control Section