The objective of this study was to evaluate the performance of three experimental 'NB' herbicide formulations against violets, a 'hard-to-control' broadleaf weed species in turfgrass. The trial was conducted in an established area of 'common' Kentucky bluegrass at the Ames Public Cemetery on South Dakota Street in Ames, IA. The experimental design was a split plot with herbicide as the main plot treatment and number of applications as the subplot factor. Individual plot size was 5 x 10 ft with three replications. The 5 x 10 ft main plots were split into 2 - 5 x 5 ft subplots. Herbicide treatments and number of applications were randomly assigned to main and subplots, respectively. The experimental area was laid out so that each 5 x 10 ft plot had approximately 30-35% violet cover pre-treatment.

Three experimental formulations from PBI Gordon - EH1381, EH1382, and EH1383 were screened in single and sequential applications (21 days apart) with Trimec Classic, Millenium Ultra, and an untreated control. Initial applications were made on June 7 to the main 5 x 10 ft plot. Twenty one days later on June 28, the materials were re-applied at the same rate to one 5 x 5 ft subplot for each main plot. The treatments were applied using a carbon dioxide backpack sprayer equipped with TeeJet #8006 nozzles at a spray pressure of 30-40 psi. The formulations were mixed with water to a volume equivalent to 3 gal/1000 ft².

Evaluations of violet damage were made from June 11 through July 18 (Table 1). Damage was assessed using a scale from 9 to 1 with 9 = no damage, 8 = slight discoloration, leaf cupping, and/or stem curling, 7 = more uniform symptoms from #8, 6 = all symptoms of #7 plus leaf mottling, 5 = all symptoms of #6 plus browning on leaves, 4 = all symptoms of #5 plus some dead leaves, 3 = more uniform symptoms of #4 with more dead leaves, 2 = all symptoms of #3 plus some dead plants, and 1 = all violets dead within the plot. After July 18, all remaining violets were assumed to have survived treatment and violet populations were assessed for each subplot. The number of violets per plot was counted on July 24, August 3, August 9, August 14, August 24 and August 31 (Table 2). The study was monitored for possible bluegrass phytotoxicity and turf quality for the duration of the study. No phytotoxicity was detected and turf quality was similar for all plots.

Data were analyzed using the Statistical Analysis System (SAS Institute Inc., 1989-1996) and the Analysis of Variance (ANOVA) procedure. Treatment effects on weed populations and visual quality were tested using Fisher's Least Significant Difference (LSD) test. In addition, orthogonal contrasts were conducted using the General Linear Model (GLM) procedure to test for differences between individual treatments (Tables 3).

All herbicides caused significant levels of damage to violets from June 11 through July 11 (Table 1). The three 'EH' formulations provided higher levels of damage than Trimec Classic or Millenium Ultra on June 11, four days after the initial treatments were applied. By June 18, all treated violets expressed similar damage. On July 18, the violets in the untreated turf were beginning to succumb to the high temperatures and dry conditions so the damage levels caused by the treatments were not significant. However, there were differences in damage levels recorded between subplots (P > F = 0.0377). Those violets receiving the sequential treatment on June 28 recorded higher levels of damage than those treated only once on June 7.

There were no statistical reductions in the number of violets in the treated turf as compared with the untreated turf (Table 2). There were numerical differences between the violet counts among the main and subplot treatments but they were not statistically significant. By August 3, violets in the untreated areas were dying because of the high temperatures and low rainfall. As a result, treatment effects are hard to interpret from the data (Table 3). Mean counts show that there were numerical reductions as compared with the untreated controls but the differences are not statistically significant.

Table 1. Violet damage¹ in turfgrass treated for the 2001 PBI Gordon Postemergence violet study.