Impact of pasture herbicides on the seedling growth response of three tall fescue varieties

Chemical application is a viable and popular method for weed control, but the effectiveness of various herbicides depends on the target weed. The objectives of this study were to determine the appropriate herbicide for three tall fescue varieties that either do or do not contain endophytic fungi. Plants were split into six herbicide treatments and an untreated control for 7, 14, 21, and 28 days in a randomized complete block design experiment with four replicates. Herbicides evaluated were 2,4-D, aminopyralid, 2,4-D plus dicamba, metsulfuron plus chlorsulfuron, aminocyclopyrachlor, and metsulfuron plus aminopyralid. In all three tall fescue varieties, the application of 2,4-D, aminopyralid, 2,4-D plus dicamba, or aminocyclopyrachlor failed to reduce phytotoxicity, height, tiller number, and dry matter when compared with the untreated control. However, metsulfuron-containing treatments reduced all variables in every variety. We thus conclude that herbicides containing metsulfuron are the most phytotoxic to tall fescue.


INTRODUCTION
Tall fescue is the predominant cool-season pasture species of the eastern United States, and over 90% of the tall fescue pastures in the United States are believed to be infected with fungal endophytes (Bacon and Siegel, 1988).One example is cultivar KY 31, a dominant forage grass for cattle in Kentucky and the surrounding states, which frequently contains the endophytic fungus Neotyphodium coenophialum.The suitability of tall fescue as a pasture grass is diminished when infected with endophytes (Smith, 1989), and infected fescue plants are considered weeds.Thus, the development of tall fescue varieties without N. coenophialum, or other endophytic fungi, is of interest for improving livestock performance.
Unfortunately, endophyte-infected fescues differ in their physiology and thus, may also differ in their response to herbicides.For instance, endophyte infection decreased the performance of perennial ryegrass in intra-and interspecific competition (Marks et al., 1991).In contrast, other studies found that endophyte infection significantly enhanced ryegrass shoot and root growth, because the endophyte-infected plants tended to accumulate more biomass and soluble sugar in the sheath and root than endophyte-free plants (Ren et al., 2009).Furthermore, endophyte infection appears to cause physiological changes that improve growth and survival under stress, compared with uninfected plants (Malinowsky and *Corresponding author.E-mail: pedromoraes@utfpr.edu.br. Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License Belesky , 2000).However, while a few recent studies (Aiken et al., 2012;Sather et al., 2013) have evaluated the effectiveness of various herbicides on tall fescue, no research is available that examines the herbicide phytotoxicity on infected versus non-infected fescues.As new herbicides, such as metsulfuron, are introduced to the pasture herbicide market, the need for relevant data becomes increasingly necessary.We took advantage of the new tall fescue varieties that were being developed at the University of Kentucky to evaluate the effectiveness of several common herbicides (including the new metsulfuron mixtures) on fescue with and without endophytic fungal infection.Our objective was to evaluate how fescue variety, fungal infection, and length of application time will affect phytotoxicity.As a result, we aim to determine the selective herbicide for use in different fescues.

Herbicides
Plants were divided into six herbicide treatments and an untreated control for 7, 14, 21, and 28 days in a randomized complete block design experiment.Each condition had four replicates.Seeds of all plants listed were planted in plastic pots (10 cm diameter) filled with 800 ml of sand/soil mixture using local soil (Maury-Kentucky bluegrass silt loam complex [fine, mixed, active, mesic Typic Paleudalfs]).The mixture (pH = 6.8) contained 7% sand, 75% silt, 18% clay, and 2.5% organic matter.After germination and thinning, two plants per pot were retained, pots were filled with approximately 2 cm water, and plants were grown in a greenhouse.Herbicides were applied to grasses approximately 5 cm high (3 to 4 leaves).
Herbicides evaluated are listed in Table 1 and the rates of herbicide application were those normally used for weed control (Table 1).
Herbicides were applied using a track sprayer chamber fitted with a Teejet 8002 EVs flat fan-tip nozzle.A single pass at 3 km/h was used to deliver 262 L/ha.The delivery rate was verified gravimetrically prior to the applications.The spray nozzle was positioned 20 cm from the surface of the highest fully expanded leaf.Treatments were administered to all plants in a particular group simultaneously.Immediately after spraying, the plants were left in the track sprayer, and the leaves were dried with an exhaust fan.Treated plants were then removed from the track sprayer and isolated for another 30 min to ensure dry leaf surfaces before their return to the greenhouse bench.

Measurements of plant characteristics
Visual injury, height and number of tillers were determined 7, 14, 21, and 28 days after treatment (DAT), and grass biomass was determined at 28 DAT.The percentage phytotoxicity was estimated visually, rated on a scale of 0 (no injury symptoms, as defined by healthy plants) to 100 (no green tissue: complete death of the plant) (Frans and Talbert, 1977).Plant height was determined by measuring the base of the plant to the ground next to the most extended leaf with a graduated ruler.All tillers present in plants were counted and recorded, regardless of size.At 28 DAT, the aerial parts of plants were cut at the soil surface, separated by variety, counted, placed into paper sacks, and dried in a forced-air oven at 70°C for 72 h.

Statistical analysis
All data were tested using analysis of variance (ANOVA) and individual means compared using a Tukey's test.Assumptions of normality and homoscedasticity were checked.Significance was set at p < 0.05.

RESULTS AND DISCUSSION
The results indicated that all varieties exhibited evidence of injury (1 to 20%) from all treatments at 7 DAT; however, only treatments containing metsulfuron resulted in injury (63.75 to 93.33%) at 28 DAT (Table 2).Compared with infected KY 31, the new endophyteinfected fescue varieties (KYFA 9821, KYFA 9301) experienced less injury following herbicide treatment: in response to metsulfuron plus chlorsulfuron and metsulfuron plus aminopyralid herbicides, average differences in injury were 23 and 17%, respectively.This outcome demonstrates that endophytic fungal infection influences increased tolerance to herbicides in the new fescue varieties compared with KY 31 (Table 2).In addition, within the new varieties treated with metsulfuroncontaining herbicides, phytotoxicity was 20% higher in uninfected fescue than in infected fescue.
Besides generating necrosis and tissue death,  metsulfuron-based herbicides also caused a decrease in height, number of tillers, and dry matter weight.In comparison, visual damage caused by herbicides containing 2,4-D was relegated only to the stunted formation of tillers and spiral leaves.Our results suggest that metsulfuron and metsulfuron mixtures are the most effective herbicides of infected fescue, and are generally in line with previous research indicating that metsulfuron plus aminopyralid treatment suppresses seed head emergence of the fescue and reduces the severity of toxicosis in toxic endophyte-infected tall fescue pastures, an effect slightly intensified by early application (Aiken et al., 2012).
In terms of height specifically, treatments containing metsulfuron resulted in shorter plants of all varieties compared with control, except for KY31, which was not reduced by any treatment 28 DAT.While metsulfuron plus aminopyralid affected all new varieties, metsulfuron plus chlorsulfuron caused stunted growth only in AR584infected KYFA 9821 and KYFA 9301, which experienced an 18-68% height reduction (Table 3).The effect of metsulfuron plus chlorsulfuron found in this study is corroborated to a lesser degree by a previous study (Sather et al., 2013) that showed that tall fescue height decreased by 21 to 38% when the herbicide was applied at the vegetative stage.The far greater height reduction we observed may be because KYFA 9821 is a taller variety.Moreover, chlorsulfuron foliar applications appear to severely inhibit tall fescue growth (Maloy and Christians, 1986).Next, in response to metsulfuron-containing treatments, tiller numbers per plant varied but was reduced overall (50 to 88%) in all new varieties 28 DAT (Table 4).The presence of the endophyte AR584 did not alter the response of KYFA 9821 or KYFA 9301 to the herbicides evaluated.However, EFKYFA 9821 (uninfected by AR584) and KY 31 (infected by N. coenophialum) decreased in tiller number compared with the control  when metsulfuron plus aminopyralid herbicide was applied.
Similar to our study, Aiken et al. (2012) reported that an application of metsulfuron plus aminopyralid reduced the number of endophyte-infected tall fescue reproductive tillers.Such results have led others to suggest that tiller number is useful as an indicator of vigor or the persistence of grass in the pasture (Teixeira et al., 2013).Thus, characterization of individual tillers is a relatively simple method for determining the effects of various management techniques, in order to find the most efficient means of weed control.
Finally, metsulfuron-containing treatments reduced dry matter in all varieties by 28 DAT.Non-metsulfuron treatments resulted in higher but similar amounts of dry matter from all varieties (Table 5).In another study (James et al., 1999), the herbicide metsulfuron-methyl reduced the production of dry matter in pastures at different times of harvest.Previous research found that metsulfuron-containing herbicide treatment boosts steer weight gain, despite the fact that the suppression of seedhead production reduced forage mass but did not affect ergovaline concentrations in leaf blades and sheaths (Aiken et al., 2012).However, cattle selectively graze tillers and metsulfuron application significantly reduced the density of reproductive tillers with seeds and stems containing high ergot alkaloid concentrations (Aiken et al., 2012).Thus, chemical suppression of tall fescue seedheads appears to be efficient at reducing the severity of fescue toxicosis.The considerable (by 40 to 63%) tiller-reducing effects of metsulfuron herbicides have also been corroborated elsewhere (Sather et al., 2013).
In summary, across all varieties of tall fescue examined, we found no reduction in phytotoxicity, height, tiller number, or dry matter following treatment at both application times with 2,4-D, aminopyralid, 2,4-D plus dicamba, or aminocyclopyrachlor when compared with the untreated control.However, all metsulfuron-containing treatments resulted in a reduction of all variables in every variety.These results are consistent with earlier findings (Teixeira et al., 2013), indicating that the metsulfuron application to fescues could be used as an alternative control measure to reduce the population of the tall fescue and broadleaf species.Future studies could potentially examine the effects of herbicide treatment on long-term fescue recovery (for example, the season posttreatment) (Goatley et al., 1990), which should further improve treatment methods against endophytic toxicosis.

Conclusion
In this study, we found that herbicides containing metsulfuron were the most phytotoxic to tall fescue for all variables measured, whereas other herbicides had no effect on measured plant variables after 28 days.
the same letter were not significantly different as determined by Tukey's test (p ≤ 0.05).
the same letter were not significantly different as determined by Tukey's test p ≤ 0.05.

Table 1 .
Herbicide treatments and application rates evaluated in experiment.