Peanut (Arachis hypogaea L.) has several unique features that contribute to challenging weed management. Peanut cultivars grown in the United States require a fairly long growing season (140 to 160 d), depending on cultivar and geographical region (Henning et al., 1982; Wilcut et al., 1995). Peanut also has a prostrate growth habit, a relatively shallow canopy, and is slow to shade inter-rows allowing weeds to be more competitive (Walker et al., 1989; Wilcut et al., 1995). Consequently, soil-applied herbicides may not provide season-long control and mid-to-late season weed pressure can occur. Additionally, peanut fruit develops underground on pegs originating from branches that grow along the soil surface. This prostrate growth habit and pattern of fruit development restricts cultivation to an early season control option (Brecke and Colvin, 1991; Wilcut et al., 1995). With conventional row spacing (91 to 102 cm), complete ground cover may not be attained until 8 to 10 wk after planting. In some areas of the United States peanut growing region, complete canopy closure may never occur.

Paraquat is one of the most frequently used postemergence (POST) herbicides in Southeastern United States peanut production systems but is seldom used in the Southwestern production region because crop injury may occur, reducing yield and grade characteristics (Knauft et al., 1990; Wilcut and Swann, 1990). The addition of bentazon to paraquat is a common practice to reduce peanut injury, although it may be either antagonistic or synergistic on weed control depending on the weed species and herbicide rate (Wehtje et al., 1992) and often does not improve peanut yield despite the reduction in crop injury (Wehtje et al., 1986, 1992; Wilcut et al., 1989). Due to the low price of paraquat, growers in the southwest continue to express an interest in its use (authors' personal observations). However, concerns have been expressed about peanut cultivar response to paraquat applications. Evaluations of cultivar response to herbicide treatments containing paraquat have been studied (Knauft et al., 1990; Wehtje et al., 1991; Wilcut and Swann, 1990).

Flumioxazin is a N-phenyl phthalamide soil-applied herbicide that received a federal label in the United States for use in peanut in 2001 (Grichar et al., 2004). Flumioxazin inhibits the enzyme protoporphyrinogen oxidase (Anderson et al., 1994; Senseman, 2007; Yoshida et al., 1991). In Georgia, flumioxazin applied preemergence (PRE) was shown to control morningglory spp. (Ipomoea spp.), prickly sida (Sida spinosa L.), and Florida beggarweed (Desmodium tortuosum (Sw.) DC) (Wilcut, 1997) while in Texas, pitted morningglory [Ipomoea lacunosa L.] were controlled greater than 75% (Grichar and Colburn, 1996). Flumioxazin has been reported to cause peanut injury especially when the application is delayed until peanut emergence (Johnson et al., 2006; Jordan et al., 2009; Tredaway-Ducar et al., 2009). When applied soon after peanut planting (1 to 2 d), Grichar et al. (2004) reported that flumioxazin plus metolachlor combinations, under cool, wet conditions resulted in peanut stunting which was evident throughout the growing season. They attributed this to increased uptake of flumioxazin and metolachlor with the heavy rainfall and the slowed metabolism of these herbicides as a result of cool temperatures (Yoshida et al., 1991). Askew et al. (1999) reported that flumioxazin at 0.07 and 0.11 kg ai/ha injured peanut 45 and 62%, respectively, when evaluated 2 wks after peanut planting. Peanut stunting of greater than 60% was followed by as much as 35% leaflet discoloration, which was characterized as necrotic spots on foliage. Scott et al. (2001) reported that flumioxazin treated peanuts were injured 10% when evaluated 3 wks after planting. However, injury was transient and was not appartent 6 wk after planting. Flumioxazin enters plants mainly by shoot and root uptake, and plant injury can be avoided via rapid metabolism (Yoshida et al., 1991; Anderson et al., 1994).

Smellmelon is becoming more of a problem in south Texas peanut production fields and has become a problem in several crops along the Texas Gulf coast (author's personal observation). The range of smellmelon stretches from Georgia to the southern part of California and as for north as Arkansas (SWSS, 1999). Smellmelon can be a problem at peanut harvest as the melon can become broken apart when run through the combine and increase drying time because of the high moisture content of the melon itself (author's personal observation). In IMI-tolerant corn (Zea mays L.), Thompson et al. (2005) reported that imazapic at 0.07 and 0.14 kg ai/ha applied either PRE, early POST, or late POST controlled smellmelon greater than 90%. Tingle and Chandler (2004) reported that smellmelon control was at least 93% with low-, medium-, and high-input herbicide systems. In cotton (Gossypium hirsutum L.), glyphosate systems have provided effective smellmelon control (Livingston et al., 2004; Livingston, 2006). Tingle et al. (2000) reported when smellmelon was allowed to compete with cotton for at least 6 wks, yield was reduced 7% compared to the weed-free check but when smellmelon was allowed to compete for 10 to 12 wks, cotton yield was reduced 22 and 27%, respectively.

Some growers in the southwest have questioned the feasibility of applying flumioxazin and paraquat at planting or soon after peanut emergence to control weeds that are not normally controlled with either herbicide alone and response of peanut cultivars to these herbicides and application timing. The product label states that PRE applications of flumioxazin must be made within 2 d after planting and prior to peanut emergence. Application after the peanut have begun to crack, or are emerged, will result in severe crop injury (Anonymous, 2009). Although off-label with respect to flumioxazin, the objective of this research was to evaluate the effect of flumioxazin and paraquat alone and in combination applied PRE or after peanut cracking for smellmelon (Cucumis melo L. var. Dudaim Naud.) control and peanut response.

Smellmelon control and peanut tolerance studies were conducted in south Texas near Yoakum and in the Texas High Plains near Lamesa and Halfway with runner and Spanish market-types during the 2009 through 2011 growing seasons. The studies in south Texas in 2009 and 2010 were conducted in areas with high populations of smellmelon while the study in 2011 and at the High Plains locations in 2009 and 2010 were conducted under weed-free conditions. The soils at Yoakum were a Denhawken sandy clay loam (fine, smectitic, hyperthermic, Vertic Haplustepts, 1.6% organic matter, pH 7.6); at Lamesa the soils were a Amarillo fine sandy loam (fine-loamy, mixed, superactive, thermic Aridic Paleustalf, 0.4% organic matter, pH 7.8); and at Halfway the soils were a Acuff sandy clay loam (fine-loamy, mixed, superactive thermic Aridic Paleustolls) with < 0.8% organic matter and pH 7.7. Planting date, application dates, and other variables for each study are given in Table 1.

Treatments consisted of a factorial arrangement of two application timings (PRE and 7 days after peanut cracking [DAC]) and 8 herbicide treatments. Herbicide treatments included flumioxazin alone at 0.07 and 0.11 kg ai/ha, paraquat alone at 0.14 and 0.28 kg ai/ha, flumioxazin at 0.07 kg/ha plus paraquat at 0.14 or 0.28 kg/ha, and flumioxazin at 0.11 kg/ha plus paraquat at 0.14 or 0.28 kg/ha. An untreated check was also included for comparison.

Plot size was two rows (97 cm apart) by 9.5 m. at Yoakum and four rows (101 cm apart) by 7.3 m. at Lamesa and Halfway and the two middle rows per plot were sprayed and the other rows were left untreated to serve as buffers. All PRE applications of herbicide treatments including paraquat and all 7 DAC treatments included a non-ionic surfactant, Induce® [non-ionic surfactant, a blend of alkylarylpolyoxyalkane ether, free fatty acids, and isopropyl (90%), and water and formulation acids (10%); Helena Chemical Co., 225 Schilling Boulevard, Suite 300, Collierville, TN 38017] in south Texas or R-11® (90% akylphenol ethoxylate, butyl alcohol dimethylpolysiloxane; Wilbur-Ellis Company, P.O. Box 16458, Fresno, CA 93755) in the High Plains area at the rate of 0.25% v/v. Herbicides were applied in water using a CO₂- pressurized backpack sprayer with TeeJet® 11002 DG (Spraying Systems Company, P.O. Box 7900, North Avenue, Wheaton, IL 60188) nozzles calibrated to deliver 190 L/ha at 180 kPa at Yoakum and TurboTee® 110015 nozzles calibrated to deliver 140 L/ha at 207 kPa at the Lubbock location.

Runner market-type peanut varieties ‘Tamrun OL02’ (Simpson et al., 2006) in 2009, ‘Florida 07’ (Gorbet and Tillman, 2009) in 2010 and ‘Tamrun OL07 (Baring et al., 2006) in 2011 were planted at Yoakum while at the High Plains locations, Tamrun OL02 was planted in 2009 and FlavorRunner 458 (Beasley and Baldwin 2009) was planted in 2010 at Lamesa. At the Halfway location the Spanish peanut OLin (Simpson et al., 2003) was planted.

Supplemental irrigation was supplied as needed at both locations. Traditional production practices were used to maintain peanut growth, development, and yield. In the weed-free studies, all plots received a dinitroaniline herbicide applied preplant incorporated and were cultivated and hand-weeded throughout the growing season to maintain weed-free conditions. Clethodim at 0.18 kg ai/ha + crop oil concentrate (COC) was applied POST to control annual grass escapes at the south Texas location. No insecticides were needed at any location in any year.

Peanut stunting was evaluated approximately 60 d after peanut planting with the runner peanuts cultivars in south Texas and the High Plains area, while the Spanish cultivars were rated prior to peanut digging. Peanut stunting were based on a scale of 0 (no peanut stunting) to 100 (peanut death). Peanut yield was determined by inverting the pods based on maturity of untreated control plots, air-drying in the field for 6 to 10 d, and harvesting individual plots with a combine. Yield samples were cleaned and adjusted to 10% moisture. For grades, a 200-g pod sample from each plot was obtained and grades determined following procedures described by the Federal-State Inspection Service (USDA, 1993).

Data for percentage of peanut injury and stunting were transformed to the arcsine square root prior to analysis; however, nontransformed means are presented because arscine transformation did not affect interpretation of the data. Data were subjected to ANOVA and analyzed using SAS PROC MIXED (SAS 2002). Treatment means were separated using Fisher's Protected LSD at P ≤ 0.05. The untreated check was used for peanut yield and grade calculation comparison and a visual comparison for peanut injury and was only included in yield and grade analysis.