Pumpkin Response

Treatment effects on pumpkin injury varied across locations. Overall, injury to pumpkin was greatest at the Painter location (Table 3). At Painter, fluridone, acetochlor EC, acetochlor ME, and halosulfuron were most injurious at 14 DAP, causing 81%, 39%, 34%, and 35% injury to pumpkins, respectively. Injury caused by S-metolachlor and fomesafen was less, ranging from 6% to 14%. Injury caused by all treatments decreased at 28 DAP, but the initial injury trends noted at 14 DAP remained. At Painter, fluridone injured pumpkin 73% at 28 DAP, whereas halosulfuron injured pumpkin 19%, similar to both formulations of acetochlor (9% to 15%). Pumpkin injury in response to fomesafen low rate (LR), fomesafen high rate (HR), ethalfluralin, clomazone, and S-metolachlor ranged from 3% to 6%. Halosulfuron is capable of injuring cucurbit crops, especially on coarse-textured soils with little organic matter (Trader et al. Reference Trader, Wilson and Hines2007), as was the case at Painter. However, halosulfuron injury is normally transitory and does not result in yield loss. Likewise, Trader and others (Reference Trader, Wilson and Hines2008) reported that halosulfuron applied PRE injured squash 42%, but injury was transient and did not result in squash yield loss.

Table 3. Pumpkin injury 14 and 28 d after planting (DAP) in New Jersey and Virginia for herbicide comparison experiment, 2017 and 2018.^{a, b}

^a Means within a column followed by the same letter are not different according to Fisher’s protected LSD test at P = 0.05.

^b Application rates were as listed in Table 2.

^c Pumpkin injury pooled across experiment sites in Painter.

^d Abbreviations: EC, emulsifiable concentrate; HR, high rate; LR, low rate; ME, microencapsulated.

At Virginia Beach, acetochlor EC (53%) injured pumpkin more than acetochlor ME (33%) 14 DAP. Much like acetochlor ME, fluridone caused 26% pumpkin injury. At this location, pumpkin injury caused by all other treatments was minimal (0 to 2%). Pumpkin injury at 28 DAP, like injury at 14 DAP, was greatest in plots treated with acetochlor EC (26%). Again, acetochlor ME (14%) and fluridone (8%) were less injurious than acetochlor EC, and injury by all other treatments was negligible (≤5%). Pumpkin injury at New Jersey was statistically similar across all treatments at 14 and 28 DAP, and injury ranged from 8% to 40% at 14 DAP and 11% to 27% at 28 DAP.

Despite broadcasting over the planted row, S-metolachlor resulted in a maximum pumpkin injury of 16% at 14 DAP and 18% at 28 DAP. Labels for S-metolachlor–containing products require a 30-cm zone of nontreated soil directly over the row or 15 cm to each side of a planted hill or emerged pumpkins to avoid potential stand loss and pumpkin injury (Anonymous 2015; Kuhar et al. Reference Kuhar, Hamilton, VanGessel, Sanchez and Wyenandt2018). Results of this experiment confirm that S-metolachlor is a useful tool for pumpkin weed management with minimal risk for crop injury.

Pumpkin stand in nontreated plots ranged from 6 to 10 plants per 9-m row depending on location (Table 4). At all three sites at Painter, both rates of fomesafen reduced pumpkin stand 20% to 75%. Halosulfuron reduced pumpkin stand compared to nontreated pumpkin at Painter in 2017, whereas fluridone caused complete stand loss at both Painter sites in 2018. The excessive stand loss at Painter in 2018 may have resulted from heavy rainfall shortly after planting. In 2018, Painter received 0.7 cm of rainfall at 2 DAP and 2.6 cm at 3 DAP, compared to 1.5 cm at 4 DAP in 2017; New Jersey received 0.2 cm at 2 DAP. At Virginia Beach (six to seven plants per 9-m row) and New Jersey (four to six plants per 9-m row), no treatments reduced pumpkin stand relative to the nontreated. Moreover, no treatment differences were observed at these locations.

Table 4. Pumpkin stand 14 d after planting (DAP) in New Jersey and Virginia for herbicide comparison experiment, in Virginia and New Jersey, 2017 and 2018.^{a, b}

^a Means within a column followed by the same letter are not different according to Fisher’s protected LSD test at P = 0.05.

^b Application rates were as listed in Table 2.

^c Abbreviations: EC, emulsifiable concentrate; HR, high rate; LR, low rate; ME, microencapsulated.

Weed Control

Ivyleaf morningglory, spurred anoda, and yellow nutsedge were present in four, three, and two sites, respectively. Despite severe pumpkin injury, fluridone controlled ivyleaf morningglory 91% at 28 DAP and 73% at 42 DAP (Table 5). In a previous study, fluridone at 336 g ai ha^–1 controlled pitted morningglory 86% at 12 wk after planting (Hill et al. Reference Hill, Norsworthy, Barber and Gbur2017). No other treatment controlled ivyleaf morningglory >52% at 28 DAP or 33% at 42 DAP. Morningglory species are particularly troublesome in cucurbit crops, and no viable chemical options currently exist (Friesen Reference Friesen1978). Likewise, fluridone controlled spurred anoda 93% at 28 and 42 DAP, whereas clomazone provided 92% control at 28 DAP and 96% at 42 DAP. Comparatively, spurred anoda control by fomesafen, ethalfluralin, halosulfuron, S-metolachlor, and acetochlor was ≤37%. Little information is available on the efficacy of fluridone for spurred anoda control. York (Reference York2018) reported that fluridone + fomesafen controlled spurred anoda ≥80%. Halosulfuron controlled yellow nutsedge 79% at 28 DAP and 63% at 42 DAP. Yellow nutsedge control was similar with halosulfuron, fomesafen HR, S-metolachlor, and acetochlor EC, ranging from 62% to 70% at 28 DAP and 56% to 69% at 42 DAP. Yellow nutsedge is a common weed to vegetable production, and its control has been previously investigated (Trader et al. Reference Trader, Wilson and Hines2008). Halosulfuron, fomesafen, and S-metolachlor have all been reported to provide >70% residual control of yellow nutsedge (Meyers Reference Meyers and Shankle2017; Reed et al. Reference Reed, Boyd and Dittmar2016; Trader et al. Reference Trader, Wilson and Hines2008)

Table 5. Ivyleaf morningglory, spurred anoda, and yellow nutsedge control by herbicides applied PRE for herbicide comparison experiment, 2017 and 2018.^a–c

^a Means within a column followed by the same letter are not different according to Fisher’s protected LSD test at P = 0.05.

^b Application rates were as listed in Table 2.

^c Data pooled across experiment sites where weeds were present. Ivyleaf morningglory control pooled across all experiment sites in NJ and Painter. Spurred anoda control pooled across Painter, 2017 and Painter field 2, 2018. Yellow nutsedge control pooled across experiment sites in New Jersey and Painter field 1, 2018.

^d Abbreviations: EC, emulsifiable concentrate; HR, high rate; LR, low rate; ME, microencapsulated.

Common ragweed, redroot pigweed, and common lambsquarters infested only the New Jersey location. At this location, fluridone completely controlled common ragweed at 28 DAP and provided 83% control at 42 DAP (Table 6). At 28 DAP, fomesafen, halosulfuron, and acetochlor EC controlled common ragweed similarly to fluridone (64% to 100%), whereas all other treatments resulted in 0 to 15% control. Similar trends for common ragweed control 42 DAP were observed with the exception of halosulfuron, which resulted in less control (38%) compared to fluridone. Cahoon and others (Reference Cahoon, Flessner and Hines2017) reported similar common ragweed control by fluridone (93%) and fomesafen (82%) 8 wk after PRE applications to cotton. In contrast to common ragweed control, redroot pigweed was controlled well by S-metolachlor, acetochlor, halosulfuron, and fomesafen; at 28 DAP, 80% to 100% control was observed, whereas at 42 DAP control was 75% to 100%. Fluridone was only marginally effective against redroot pigweed, controlling the weed 63% at 28 DAP and 58% at 42 DAP. Braswell and others (Reference Braswell, Cahoon, York, Jordan and Seagroves2016) reported Palmer amaranth (like redroot pigweed, a member of the Amaranthus genus) control by fluridone to be good (100% control 38 DAT). However, these researchers used 280 g ai ha^–1 fluridone, compared to 168 g ai ha^–1 used in this experiment. Other researchers recommend 336 g ai ha^–1 fluridone to effectively control Palmer amaranth when applied alone (York Reference York2018). Furthermore, for more consistent Palmer amaranth control, labels for fluridone-containing products require a tank-mix partner when fluridone is applied at ≤221 g ai ha^–1 (Anonymous 2018). Rates used in this experiment may explain why fluridone was not more effective at controlling redroot pigweed. Clomazone controlled redroot pigweed poorly and confirms previous reports of inadequate Amaranthus spp. control by clomazone (Brown and Masiunas Reference Brown and Masiunas2002). Fluridone controlled common lambsquarters 99% to 100%. At 28 DAP, clomazone (90%) and halosulfuron (98%) provided control of common lambsquarters similar to fluridone. At 42 DAP, fluridone controlled common lambsquarters 100%, compared to 69% by clomazone and halosulfuron.

Table 6. Weed control 14 and 28 d after planting (DAP) by herbicides applied PRE in New Jersey for herbicide comparison experiment in 2018.^{a, b}

^a Means within a column followed by the same letter are not different according to Fisher’s protected LSD test at P = 0.05.

^b Application rates were as listed in Table 2.

^c Abbreviations: EC, emulsifiable concentrate; ME, microencapsulated.

Yield

Pumpkin yield was not collected at Virginia Beach as a result of severe disease, nor at New Jersey because of lack of rainfall. At Painter in 2017, pumpkin fruit number and total yield ranged from 6,600 to 9,000 fruits ha^–1 and 47,800 to 81,400 kg ha^–1, respectively (Table 7). At this location, fruit number and yield in all plots were similar, regardless of herbicide treatment. At both Painter sites in 2018, fluridone eliminated fruit production. At Painter field 1, plots treated with S-metolachlor produced the greatest number of fruits (10,800 fruits ha^–1) and greatest yield (68,300 kg ha^–1). Pumpkin yield response to acetochlor was variable. Only plots treated with acetochlor ME yielded similarly to S-metolachlor–treated plots. All other plots, including those treated with acetochlor EC, produced fewer fruits (5,000 to 8,000 fruits ha^–1) and lower yield (30,900 to 50,200 kg ha^–1) than plots treated with S-metolachlor. Despite early-season injury and stand loss, Peachey and others (Reference Peachey, Doohan and Koch2012) reported that fomesafen applied at 350 and 700 g ai ha^–1 did not significantly reduce pumpkin yield. This was not the case in this study; fomesafen-treated plots at Painter field 1 in 2018 reduced yield by nearly 50% compared to S-metolachlor, the treatment with the greatest yield. At field 2, all herbicide treatments resulted in 8,400 to 10,000 fruits ha^–1 and 46,300 to 62,700 kg ha^–1, except halosulfuron and fluridone, which produced fewer fruits and lower yield. Comparatively, halosulfuron-treated plots generated only 5,400 fruits ha^–1 and 32,600 kg ha^–1. This is in contrast to Trader and others (Reference Trader, Wilson and Hines2007), who noted that halosulfuron at 27 g ai ha^–1 improved yield compared to nontreated pumpkin. Fomesafen-treated plots had yields similar to S-metolachlor and acetochlor.

Table 7. Pumpkin yield in Virginia for herbicide comparison experiment, 2017 and 2018.^{a, b}

^a Means within a column followed by the same letter are not different according to Fisher’s protected LSD test at P = 0.05.

^b Application rates were as listed in Table 2.

^c Abbreviations: EC, emulsifiable concentrate; ME, microencapsulated.

Crop Response

Pumpkin stand and injury were inversely related with fluridone rate. Nontreated pumpkin averaged seven plants per 9-m row at 14 DAP (Figure 1). The two lowest rates of fluridone had no effect on pumpkin stand. However, fluridone rates at ≥126 g ai ha^–1 reduced pumpkin stand by as many as six plants per 9-m row. Pumpkin injury followed a trend similar to that of pumpkin stand; fluridone rates >126 g ai ha^–1 resulted in 100% injury (Figure 2). Although less injury was observed 28 DAP, the same general trend for pumpkin response to fluridone rate existed.

Figure 1. Pumpkin stand 14 and 28 d after planting (DAP) for fluridone rate experiment in New Jersey and Virginia. Application rates are listed in Table 2. Pumpkin stand was pooled across all experiment sites.

Figure 2. Pumpkin injury 14 and 28 d after planting (DAP) for fluridone rate experiment in New Jersey and Virginia. Application rates are listed in Table 2. Pumpkin injury was pooled across all experiment sites.

Yield

Correspondingly with early-season stand loss and pumpkin injury, pumpkin fruit number and yield generally decreased as fluridone rate increased. Nontreated plots at New Jersey yielded 4,900 fruits ha^–1 (Figure 3). Only plots treated with 84 g ai ha^–1 produced a similar amount of fruit. All other rates of fluridone reduced fruit 1,000 to 4,100 fruits ha^–1 compared to the nontreated. Despite a lesser effect on fruit set, total pumpkin yield in New Jersey was reduced approximately 48% to 100% for fluridone rates >126 g ai ha^–1 (Figure 4). At Painter, nontreated plots resulted in 12,900 fruits ha^–1 and 84,600 kg ha^–1. Only plots treated with 42 g ai ha^–1 fluridone generated a similar number of fruits (12,200 fruits ha^–1) and yield (76,300 kg ha^–1) as did nontreated plots (data not shown). Higher fluridone rates significantly reduced fruit set and total yield. Moreover, fluridone rates of ≥168 g ai ha^–1 caused complete or near-complete fruit loss.

Figure 3. Pumpkin fruits (fruits ha^–1) for fluridone rate experiment in New Jersey and Virginia. Application rates are listed in Table 2. Pumpkin fruit data for Painter were pooled across two experiment sites within location.

Figure 4. Pumpkin yield (kg ha^–1) for fluridone rate experiment in New Jersey and Virginia. Application rates are listed in Table 2. Pumpkin yield data for Painter were pooled across two experiment sites within location.

Surprisingly, average fruit weight (total yield/number of fruit) did not appear to be influenced by fluridone rate. Average fruit weight at New Jersey averaged 5.8 kg fruit^–1 in nontreated plots, whereas fruits in fluridone-treated plots, except for the highest rate, ranged from 3.5 to 6.9 kg fruit^–1 (data not shown). A similar trend was observed in Painter. In treated plots with surviving pumpkins, average fruit weight was 6.2 to 6.4 kg fruit^–1 compared to 6.6 kg fruit^–1 for pumpkins in nontreated plots (data not shown). From this, it can be concluded that fluridone reduces pumpkin yield primarily by interfering with fruit set. This effect is probably a response to severe early-season growth reduction and chlorosis from which pumpkins were unable to recover.

Despite excellent weed control, most notably ivyleaf morningglory, these data indicate unacceptably high pumpkin injury by fluridone. Fluridone rates >42 g ai ha^–1 resulted in significant yield loss. Moreover, fluridone rates that were needed to achieve consistent weed control (>221 g ai ha^–1) also reduced pumpkin yield 75% to 100%. Acetochlor EC and acetochlor ME did not injure pumpkins as much as did fluridone. However, acetochlor caused early-season injury to pumpkin, regardless of formulation. S-metolachlor, also a member of the chloroacetamide family of herbicides and currently labeled for pumpkin, was much safer than acetochlor. In addition, S-metolachlor and acetochlor control a similar spectrum of weeds. For these reasons, like fluridone, acetochlor EC and ME formulations are not recommended for pumpkin production in this region.