Plant materials.

Seeds of selected four weed species including florida beggarweed, ivyleaf morningglory, johnsongrass, and spanishneedles were collected from a citrus grove near Lake Alfred, FL. These weed species were chosen because they are problem weeds in Florida citrus, their variability in susceptibility to glyphosate, and their apparent differences in leaf surface characteristics. Experiments were conducted under greenhouse conditions at the Citrus Research and Education Center, University of Florida, Lake Alfred. Six to eight seeds of each weed species were planted in a 15 × 10-cm plastic pots containing commercial potting mix (Conrad Fafard, Agawam, MA). The pots were kept in a greenhouse, maintained at 25/16 °C (± 0.5 °C) day/night temperature, 70% (± 5%) day/night relative humidity (RH), and a light intensity of 250 μmol·m⁻²·s⁻¹ photosynthetically active radiation (PAR). Plants of each weed species used in all studies were watered regularly and fertilized once (2 weeks after planting) with 20N–8.7P–16.6K fertilizer (Tracite; Helena Chemical Co., Collierville, TN) to promote optimum plant growth.

Herbicide and surfactant.

Glyphosate (Roundup Original; Monsanto, St. Louis) was used in all experiments. The adjuvant Ortho X-77 (Chevron Chemical Co., Richmond, CA), a mixture of alkylaryl polyoxyethylene glycols, was used as NIS and added to glyphosate at 0.25% v/v. Glyphosate at 6 and 12 oz/acre applied alone and with NIS (0.25% v/v) was evaluated for a ¹⁴C-glyphosate absorption and translocation study. In a radiolabel study, ¹⁴C-glyphosate isopropylamine salt with 23.87 curies per mole specific activity was used. The solubility of glyphosate isopropylamine salt was 900 g·L⁻¹ at pH 7, water temperature of 25 °C, and octanol–water partition coefficient (K_ow) of 0.0006–0.0017 (Vencill, 2002).

Efficacy study.

To examine the effects of different rates of glyphosate with and without NIS on control of four weed species, glyphosate was applied at 3, 6, 12, and 24 oz/acre alone or with NIS (Ortho X-77 at 0.25 v/v) using a chamber track sprayer fitted with a flat fan spray nozzle (Teejet 8002; Spraying System Co., Wheaton, IL) delivering 20 gal/acre at 30 psi. Deionized water was used as a carrier for all spray solution. All species were thinned to four plants per pot. Glyphosate treatments were applied to 15-cm-tall (three to four leaves) ivyleaf morningglory, 12-cm-tall (four leaves) florida beggarweed, 10-cm-tall (four leaves) spanishneedles, and 12- to 15-cm-tall (three to four leaves) johnsongrass. Percent control was estimated visually at 14 d after treatment (DAT) based on a scale of 0% to 100%, where 0% indicates no damage and 100% indicates complete plant death (Frans et al., 1986). Aboveground biomass from each pot was harvested by cutting the plants from the base at 14 DAT, and fresh weight was recorded. Harvested samples were oven dried at 40 °C for 10 d, and dry weight of each weed species (weed biomass) was recorded. Percent biomass reduction over untreated control was calculated using the following equation: percent biomass reduction = [(a – b)/a] × 100, where a is biomass of untreated plants and b is biomass of the treated plants.

Absorption and translocation.

One week before applying ¹⁴C treatments, plants were transferred from greenhouse to a growth chamber and maintained at 25/16 °C (± 0.5 °C) day/night temperature, 55%/70% (± 5%) day/night RH, and 200 μmol·m⁻²·s⁻¹ PAR. For each plant, the leaf to be treated with ¹⁴C was carefully covered with aluminum foil, before over the top application of glyphosate at 6 or 12 oz/acre with or without NIS (Ortho X-77 at 0.25% v/v) using an air pressured chamber track sprayer as mentioned in the efficacy study. After removing the aluminum foil, five 2-μL droplets (10 μL contained 300 to 335 Bq) of ¹⁴C-glyphosate were applied to a discrete area of the adaxial leaf surface in the median part of the third fully expanded leaf at four leaves stage using a 10-μL microsyringe. The quantity of ¹⁴C applied to leaves was calculated by dispensing a similar number of droplets directly into 5 mL scintillation cocktail in a 7-mL polyvinyl vial. At 72 h after the ¹⁴C application, plants were dissected into two sections (treated leaf and remaining tissues). The excised treated leaf was washed twice with a mixture of 4 mL water and ethanol (1:1 v/v) to recover unabsorbed ¹⁴C-glyphosate and then rinsed twice with 3 mL ethanol solution. A 200 μL subsample from each washing was dispensed to 7-mL polyvinyl vial containing 6 mL scintillation liquid and afterward radio-assayed using a liquid scintillation counter. Plant samples were oven dried at 50 °C for 48 h and were combusted using a Biological Oxidizer OX 500 (R.J. Harvey Instrument Corp., Hillsdale, NJ). Evolved ¹⁴CO₂ was trapped in Carbon-14 Cocktail (R.J. Harvey Instrument Corp.) flushed in 20-mL glass vials. The trapped ¹⁴CO₂ was quantified by liquid scintillation spectrometry to determine distribution pattern of ¹⁴C-glyphosate. Foliar absorption (percent applied) was defined as ¹⁴C not recovered by washing the treated leaves and was calculated using the following equation: percent foliar absorption = (¹⁴C in all plant tissues/¹⁴C applied) × 100.

Translocation of ¹⁴C-glyphosate was expressed as percent of applied as well as a percent of the total quantity absorbed. The recovery efficiency of ¹⁴C-glyphosate was calculated by including all the oxidized radio assayed fractions and ¹⁴C-glyphosate present in washing. Total recovery of ¹⁴C-glyphosate was ≥94%, and there was no difference in recovery among the treatments.

Chloroform extract.

Plants of the four weed species were grown in the greenhouse under similar conditions to those in the glyphosate efficacy study. At the three to four leaves stage of each weed species, 10 to 15 leaves were harvested and leaf area was determined using a portable leaf area meter. Harvested leaves were immersed in 30 mL chloroform for 20 s at room temperature in preweighed vials. The surface wax was extracted according to the procedure by Bukovac et al. (1979). Because of the difficulty in determining if only epicuticular waxes were extracted, all extracted wax was referred as “chloroform extract” (Westwood et al., 1997). Each treatment was replicated four times, and the experiment was repeated three times.

Statistical analysis.

All the studies were conducted in a randomized complete block design with four replications and were repeated. There was a nonsignificant experiment-by-treatment interaction; therefore, data were pooled and represented the mean of the repeated experiments. The data from efficacy study (percent control and biomass reduction) and ¹⁴C absorption study were subjected to two-way factorial analyses with glyphosate treatments and weed species as two factors. All data were subjected to analysis of variance and means were separated using Fisher's protected least significant difference test (P = 0.05) in SAS (version 9.2; SAS Institute, Cary, NC). The data of percent weed control and biomass reduction were arcsine transformed before analysis; however, nontransformed percentages are presented with mean separation based on transformed data.

Glyphosate efficacy.

Glyphosate treatments had variable effects on the tested weed species, and there was a significant interaction between weed species and glyphosate application rates; but the addition of NIS did not improve control of any species (Table 1). Compared with untreated control, glyphosate applied alone or with NIS at any rate provided significant control of florida beggarweed, spanishneedles, and johnsongrass, indicating that these species were not glyphosate tolerant. Similar results were obtained in a study that evaluated the response of four weed species including barnyardgrass (Echinochloa crus-galli), hemp sesbania (Sesbania exaltata), pitted morningglory (Ipomoea lacunose), and prickly sida (Sida spinosa) to glyphosate with or without NIS (Latron AG-98; Rohm and Hass, Philadelphia) (Norsworthy et al., 2001). Florida beggarweed was the most sensitive and recorded 99% control even at the lowest rate (3 oz/acre), and there was no difference among treatments (Table 1). The application of glyphosate at 3 oz/acre resulted in 90% and 84% control of spanishneedles and johnsongrass, respectively. Increasing glyphosate rates from 6 to 24 oz/acre provided 98% to 100% control of these two weed species (Table 1).

Table 1.Effects of application rates of glyphosate with and without nonionic surfactant (NIS) on control of four weed species.

View Fullscreen

Ivyleaf morningglory was the most tolerant to glyphosate among the four weed species. For example, ivyleaf morningglory was not controlled when glyphosate applied alone or with NIS at 3 oz/acre (Table 1). Increasing glyphosate rate to 24 oz/acre resulted in ≤25% control. Similar to this result, Shaw and Arnold (2002) reported 32% control of pitted morningglory with 12 oz/acre glyphosate. In contrast to this, a previous study reported >90% control of pitted morningglory (Koger and Reddy, 2005); however, glyphosate was applied at almost double rate (39 oz/acre) than the maximum rate used in this study.

Percent biomass reduction in all the tested weed species followed almost identical trend to percent control in response to glyphosate treatments (Table 2). The effect of glyphosate application rates on weed biomass was significant; however, there was no difference because of addition of NIS. For example, ivyleaf morningglory showed the lowest biomass reduction compared with other weed species and recorded <10% and <65% biomass reduction at 3 and 24 oz/acre glyphosate, respectively (Table 2). Biomass reduction was 78% to 84% in florida beggarweed, 64% to 75% in spanishneedles, and 76% to 88% in johnsongrass with the application of 3 oz/acre glyphosate (± NIS), and the biomass reduction was higher with increasing application rates (Table 2). Similar to this result, a study reported 24% and 69% biomass reduction in pitted morningglory when glyphosate was applied at 240 and 1680 g·ha⁻¹ (3.43 and 23.98 oz/acre), respectively (Norsworthy et al., 2001). Field experiments conducted by Blackshaw and Harker (2002) reported that glyphosate applied at 445 g·ha⁻¹ (6.35 oz/acre) provided significant biomass reduction in green foxtail (Setaria viridis), wild oat (Avena fatua), redroot pigweed (Amaranthus retroflexus), and wild mustard (Sinapis arvensis) without injury to glyphosate-resistant wheat (Triticum aestivum).

Table 2.Effects of application rates of glyphosate with and without nonionic surfactant (NIS) on biomass reduction of four weed species.

View Fullscreen

Overall results of glyphosate efficacy study suggested that species sensitivity to glyphosate was ranked as florida beggarweed > spanishneedles > johnsongrass > ivyleaf morningglory, averaged over use of NIS and glyphosate rate.

Absorption and translocation.

Differences in absorption and translocation of ¹⁴C-glyphosate among four weed species were evaluated. The results suggested that ivyleaf morningglory absorbed the lowest ¹⁴C-glyphosate (63% of applied) at the 6 oz/acre compared with other weed species (Table 3); however, at the highest rate (12 oz/acre), it was comparable with spanishneedles and johnsongrass. The glyphosate efficacy data suggested that ivyleaf morningglory was the most tolerant (Table 1) and had the lowest biomass reduction compared with other species (Table 2). However, it has to be noticed that tolerance level is relative and, in this case, glyphosate still reduced the biomass of ivyleaf morningglory in the range of 3% to 63% (Table 2). Similar results were reported by Starke and Oliver (1998) and Norsworthy et al. (2001) showing minimal glyphosate absorption by pitted morningglory.

Table 3.Effects of application rates of glyphosate with and without nonionic surfactant (NIS) on carbon-14-labeled glyphosate (¹⁴C-glyphosate) absorption in four weed species.

View Fullscreen

The effect of the addition of NIS on ¹⁴C-glyphosate absorption was species dependent. For example, the addition of NIS improved the absorption of ¹⁴C-glyphosate in ivyleaf morningglory and spanishneedles; however, there was no effect of NIS on the absorption of ¹⁴C-glyphosate in florida beggarweed at any rate of glyphosate and in johnsongrass at lower rate (Table 3). A study by Sherrick et al. (1986) revealed that absorption was increased 2- to 3-fold in field bindweed (Convolvulus arvensis) with addition of tallow amine (MON 0818, Monsanto) compared with glyphosate applied alone. In a similar study, Norsworthy et al. (2001) reported that the addition of NIS to glyphosate did not affect ¹⁴C-glyphosate absorption in prickly sida and pitted morningglory, whereas absorption was reduced by the addition of NIS in hemp sesbania and barnyardgrass.

Florida beggarweed showed the highest ¹⁴C-glyphosate absorption (>85%), and it did not vary with application rates or addition of NIS (Table 3). This was consistent with the results of the glyphosate efficacy study, which indicated that florida beggarweed was the most sensitive (Table 1). Except in florida beggarweed, ¹⁴C-glyphosate absorption increased with increasing glyphosate rates. A study by Koger and Reddy (2005) suggested that glyphosate absorption and translocation was more affected by application rate than the degree of plant exposure to glyphosate in pitted morningglory.

The results of the ¹⁴C-glyphosate-translocation in the tested weed species revealed that johnsongrass showed the highest ¹⁴C-glyphosate translocation (52% to 58%) regardless of application rate and inclusion of NIS. In other species, ¹⁴C-glyphosate-translocation was variable [spanishneedles (32% to 52%), florida beggarweed (34% to 51%), and ivyleaf morningglory (16% to 39%)] based on rate and NIS (Table 4). For example, the ¹⁴C-glyphosate translocation was increased with increasing rate of glyphosate from 6 to 12 oz/acre in ivyleaf morningglory and spanishneedles. Addition of NIS to glyphosate further improved the translocation of ¹⁴C-glyphosate in ivyleaf morningglory and florida beggarweed. Usually similar trend was observed when translocation was calculated as percent of absorbed (Table 5). A study by Westwood et al. (1997) reported that neither RH nor uptake and gross translocation processes accounted for differential glyphosate sensitivity in different field bindweed biotypes. The tolerance of ivyleaf morningglory to glyphosate observed in this study could be due to several factors including elevated levels of the target enzyme EPSPS, glyphosate metabolism, or sequestration away from the site of action. Most weed species do not metabolize glyphosate enough to avoid its toxic effects; however, in field horsetail (Equisetum arvensis), it has been reported that low activity was due to ability to metabolize the herbicide (Marshall et al., 1987).

Table 4.Effects of application rates of glyphosate with and without nonionic surfactant (NIS) on carbon-14-labeled glyphosate (¹⁴C-glyphosate) translocation in four weed species.

View Fullscreen

Table 5.Effects of application rates of glyphosate with and without nonionic surfactant (NIS) on carbon-14-labeled glyphosate (¹⁴C-glyphosate) translocation in four weed species.

View Fullscreen

Leaf wax content.

The four weed species differed in leaf wax content and showed variable relationships between ¹⁴C-glyphosate absorption and leaf wax content (data not shown). Ivyleaf morningglory that had the lowest leaf wax content (10.8 μg·cm⁻² of leaf surface) showed relatively lower ¹⁴C-glyphosate absorption and translocation (percent applied and percent absorbed) compared with other weed species. However, spanishneedles had lower wax content (15.7 μg·cm⁻²) compared with florida beggarweed (24.7 μg·cm⁻²) and johnsongrass (34.6 μg·cm⁻²), but it showed the same ¹⁴C-glyphosate absorption as johnsongrass and the same ¹⁴C-glyphosate translocation as florida beggarweed (Tables 3–5). Similarly, Norsworthy et al. (2001) reported that glyphosate absorption was not correlated with the quantity of chloroform leaf wax or leaf wettability in tested weed species.

As reported in previous study (Elmore et al., 1998), this study has shown that leaf wax content was less effective in the penetration of the herbicide through the cuticle. Lipoidal nature of the leaf surface minimizes the loss of water from the plant and may act as a barrier in the absorption of herbicides (Holloway, 1993). Compared with spanishneedles, johnsongrass had higher leaf wax content; however, the ¹⁴C-glyphosate absorption was the same in both the species at all the rates of glyphosate (Table 3). Harr et al. (1991) performed the qualitative analysis of leaf wax in johnsongrass and reported that 93% of the total leaf waxes to be of polar nature which absorbed water soluble chemicals easily. In this study, ivyleaf morningglory and spanishneedles had lower leaf wax contents (data not shown); however, their percent nonpolar and polar waxes might be different, which might have affected the absorption of glyphosate. Therefore, further research is required for qualitative wax analysis in florida beggarweed, spanishneedles, and ivyleaf morningglory. Results of this study will contribute to the limited understanding of differential glyphosate tolerance in these weed species and may aid in developing alternative management options.

Combining agronomic practices with judicious herbicide use can lead to successful weed management and reduce the selection pressure for herbicide resistant weeds (Beckie et al., 2011). Many studies reported that cropping system diversity is the long-term solution for proactive herbicide-resistant weed management in annual crops (Harker et al., 2005; Owen, 2001); however, this option is less effective in perennial crops such as citrus. Therefore, integrated weed management practices, which might include sanitation, mulching, soil solarization, preplant fumigation, use of herbicides with different mode of action, tank mix of herbicides, and monitoring of weeds for herbicide-tolerance is required for effective management of weeds in Florida citrus.

Units

View Fullscreen