The potato vine crusher: a new tool for harvest weed seed control

Abstract Harvest weed seed control (HWSC), an evolving strategy in weed management, is highly effective for the control of a variety of weed species in North American cropping systems. Previous devices for weed seed devitalization at harvest have been limited to tow behind and integrated combine systems. The potato vine crusher (PVC) is a harvester-mounted set of rollers originally designed for crushing and control of Ostrinia nubilalis (Hübner) larvae during potato (Solanum tuberosum L.) harvest. To evaluate the potential of the PVC for HWSC, we conducted stationary testing of spring tension and roller speed settings to maximize devitalization of lambsquarters (Chenopodium album L.), the most problematic weed species in Canadian potato production. In addition, we evaluated the efficacy of the PVC for the devitalization of several pernicious weed species under controlled conditions and during a simulated harvest. Increasing PVC spring tension reduced the devitalization of lambsquarters seed, whereas roller speed had minimal effect. In contrast, maximized spring tension and minimized roller speed reduced lambsquarters emergence (53%) in soil. Hypocotyl and radicle elongation was observed from lambsquarters seed fragments under controlled conditions, potentially contributing to increased control in soil through fatal germination. High levels of seed devitalization (65%–94%) were observed for all species under controlled conditions. During simulated harvest, control of large weed seeds (50%–63%) was observed, whereas smaller seeds were not impacted, signifying the importance of seed size for PVC efficacy. These studies demonstrate the PVC as a promising new tool for HWSC in Canadian potato production systems.


Introduction
The evolution and spread of herbicide-resistant weeds impose significant costs to producers and pressure within the agricultural industry to develop alternative weed management tactics.With reduced chemical options, producers have looked to alternative weed control measures and seed bank management as a means of controlling herbicide-resistant weeds (Norsworthy et al. 2012).Harvest weed seed control (HWSC), which is the collection or devitalization of weed seeds at crop harvest, is one non-chemical tactic producers have looked toward as a mechanism to eliminate herbicideresistant weed seeds from fields.The efficacy of any HWSC approach, however, depends on the propensity of weed species to retain their seeds at crop harvest (Shergill et al. 2020b).For example, many problematic weed species in North American cropping systems, including lambsquarters (Chenopodium album L.), redroot pigweed (Amaranthus retroflexus L.), and volunteer canola (Brassica napus L.), retain greater than 75% of seed up to 4 weeks after physiological maturity of soybean (Glycine max (L.) Merr.), wheat (Triticum aestivum L.), and cotton (Gossypium hirsutum L.) (Beckie et al. 2017;Tidemann et al. 2017b;Schwartz-Lazaro et al. 2021).This high rate of seed re-tention at maturity for many annual weeds enables a large proportion of seeds to be collected at crop harvest and increases the efficacy of HWSC tactics (Tidemann et al. 2017a).
There has been demonstrated success of HWSC in cereal and soybean production systems with various techniques such as chaff carts, bale direct systems, narrow-windrow burning, and chaff lining (Walsh et al. 2018;Norsworthy et al. 2020).Chaff carts and bale direct systems involve the collection of chaff and straw exiting the combine that is often burned or grazed in fields (Norsworthy et al. 2020).In contrast, narrow-windrow burning and chaff lining concentrate chaff and straw in a narrow band which is then burned in field to devitalize weed seeds or left as an undisturbed mulch interfering with seed germination and establishment (Walsh et al. 2018).Narrow-windrow burning is highly effective for the control of numerous weed species including barnyard grass (Echniochloa crus-galli (L.) P. Beauv.) and velvetleaf (Abutilon theophrasti Medik.), yet it has several environmental, safety, and regulatory concerns limiting its use (Norsworthy et al. 2020).In contrast, few studies have documented effective weed kill or reductions in seed bank density through the use of chaff lining (Walsh et al. 2017(Walsh et al. , 2021)).As such, many re-cent HWSC efforts have focused on impact mill systems, such as the integrated Harrington Seed Destructor (iHSD), Seed Terminator, and Redekop Seed Control Unit, that devitalize weed seeds during crop harvest (Schwartz-Lazaro et al. 2017;Tidemann et al. 2017aTidemann et al. , 2020;;Shergill et al. 2020a).These units are integrated systems that, when as compared to previous HWSC methods, eliminate the need for an additional field pass or the use of fire, thus conserving time, fuel, money, and safety.The HSD is an impact mill and a chaff-and strawtransfer system and devitalizes between 97.7% and 99.8% of cleavers (Galium spurium L.), volunteer canola, green foxtail (Setaria viridis (L.) Beauv.), wild oat (Avena fatua L.), and kochia (Kochia scoparia L.) seed during stationary testing (Tidemann et al. 2017a).The efficacy of HWSC tactics, however, varies by weed species and cropping system due to differences in seed retention, seed size, seed number, chaff volume, and chaff type at the time of harvest (Walsh and Powles 2014;Tidemann et al. 2017a;Shergill et al. 2020b).Additionally, current HWSC systems are only viable with grain crops that are combine harvested with a platform header.
Potatoes (Solanum tuberosum L.) are Canada's most valuable horticultural crop, with an annual farm gate value greater than $1.4 billion in 2021 (AAFC 2022).Greater than 38% of Canadian potato acreage is seeded within the Atlantic provinces of New Brunswick and Prince Edward Island (AAFC 2022).Potatoes are relatively poor competitors, with weeds costing Canadian potato producers upwards of $82 million in lost yield potential annually (Ganie et al. 2023).Lambsquarters is the most economically significant weed in potato production systems (Parks et al. 1996) and has evolved resistance to commonly used herbicides in potato production in Atlantic Canada (McKenzie-Gopsill et al. 2020).Retention rates of lambsquarters seed at harvest time are 99% and 90% in swathed and direct-harvested western Canadian spring wheat systems, respectively (Beckie et al. 2017).In addition, 100% control of lambsquarters seed was demonstrated with the integrated HSD during stationary testing (Schwartz-Lazaro et al. 2017).Due to increasing incidence of herbicide resistance and limited available herbicide options, alternative strategies are needed to effectively manage lambsquarters in potato.
Similar to many horticulture crops, potatoes are harvested with specialized equipment.At harvest, potato tubers are dug and separated from the desiccated potato vines which are discarded out of the rear of the harvester.No system for HWSC currently exists in potato production systems; however, a potato vine crusher (PVC) consisting of a roller mill designed to crush stalks and plant biomass during potato harvest was developed by Agriculture and Agri-Food Canada in the early 2000s (AAFC 2020).This system, originally designed for the control of European corn borer (Ostrinia nubilalis Hübner) larvae, consists of brushes which feed plant material into two counter-rotating metal cylindrical rollers during harvest.The PVC is mounted on the rear of a potato harvester and collects all debris, including potato vines and weeds, exiting the main conveyor belt at harvest.Field trials demonstrated the PVC can effectively control European corn borer larvae located within the vines of harvested potato plants (AAFC 2020).This system, however, has not been evaluated for HWSC of common weeds in potatoes.Stationary testing of similar roller mill systems consisting of paired toothed cylindrical rollers reduced downy brome (Bromus tectorum L.) seed germination by 95% and 98% under controlled (Hauhouot-O'Hara et al. 1999) and field conditions (Gossen et al. 1998), respectively.Several studies have shown hammer mills tend to be more effective than roller mills at devitalizing weed seeds due to their mechanism of action (Lyon and Rush 1993;Gossen et al. 1998;Hauhouot-O'Hara et al. 1999).Roller mills crush weed seeds with a single double-sided impact, whereas impact mills apply multiple single-sided impacts resulting in the separation of seed components (Berry et al. 2014(Berry et al. , 2015)).The comparable efficacy of roller mills to impact mills can be achieved through fine-tuning available settings, including the gap between rollers and roller speed to maximize impact force applied to weed seeds (Gossen et al. 1998;Berry et al. 2014Berry et al. , 2015)).
High levels of seed retention of common potato weeds and mechanisms of potato harvest where whole plants are passed through the harvester suggest that HWSC may be an additional integrated weed management tactic for herbicide resistance management.Therefore, the objective of this study was to determine if the PVC could be used to devitalize seeds of various weed species in Canadian potato production systems.Five experiments were conducted to determine the potential of the PVC as an HWSC method in potatoes.The first two experiments were designed to evaluate PVC tension (experiment one) and speed (experiment two) settings to optimize devitalization of lambsquarters seed.We hypothesized that increasing spring tension to narrow the roller gap and decreasing roller speed would increase seed devitalization due to increased pressure and contact time.The results of these experiments led to a novel hypothesis that using the PVC to stimulate lambsquarters seed germination would contribute to seed bank depletion in the field, which was explored in experiment three.Next, we evaluated the efficacy of the PVC for seed devitalization of several common weeds of Canadian potato production under controlled conditions (experiment four) and during a simulated harvest (experiment five) using settings from experiments one, two, and three.

The potato vine crusher
The PVC is a custom-built apparatus that consists of two cylindrical metal rollers (91.5 cm × 17.28 cm, L × D; Fig. 1A) that are attached to the back of a potato harvester below the conveyer belt (Fig. 1B).In this configuration, all plant material exiting the harvester is passed through the PVC.Each roller has seven equally spaced raised metal beads (1.3 mm × 7.8 mm, H × W; Fig. 1) running the length of the roller.Two heavy-duty springs (28.8 cm × 7.48 cm, L × D) attached to the rollers provide tension (Fig. 1).Tension is adjusted with two bolts at either end of the spring.A generator (1.5 HP; 115/230 V) provides power to the rollers.A hydraulic regulator controls roller speed with a locking dial.All experiments were conducted indoors at the Harrington research farm in Harrington PE, Canada (46.3469, −63.1555), from October 2020 to July 2021.3) springs (28.8 cm × 7.48 cm, L × D).Bolts connected to either end of the springs increase spring tension.Each roller is powered by a motor which is both connected to a dial controlling speed (not shown).(B) The PVC mounted on the rear of a potato harvester.

Seed material collection, measurements, and germination testing
Seeds used for all studies were collected from local field populations in 2019.Lambsquarters was chosen as the study species for the optimization of the PVC due to its small seed size, resistance to PSII-inhibiting herbicides (McKenzie-Gopsill et al. 2020), seed retention, and economic importance in potato production.Other species selected included redroot pigweed, volunteer canola, barnyard grass, and yellow foxtail (Setaria glauca (L.) Beauv.) and were chosen to represent problematic species in Canadian potato production and a range of seed sizes.Mature seed was collected, dried on a lab bench, hand-threshed, and stored at 4 • C until use.Seed for all experiments was counted with a benchtop seed counter (Seedburo 801; Seedburo, Des Plaines, IL, USA).The percent accuracy (±5%) of the seed counter was determined on six replicates of 30 seeds for each species prior to counting.Thousand seed weight (TSW) was estimated from eight samples of 1000 seeds of each species.Seed diameter (mm) was estimated as the larger of two measurements (L × W) from 10 seeds per species using digital calipers (Table 1).Initial germi-nation was evaluated on six replications of 30 seeds of each species in 100 mm Petri plates and moist filter paper (P5; Thermo Fisher Scientific, Mississauga, ON, Canada).Seed was surface sterilized with 5% sodium hypochlorite +0.1% Tween (Thermo Fisher Scientific) for 30 min prior to germination testing.Each plate was incubated in the dark at 22 • C for 14 days before counting.Ungerminated seeds were subjected to a pinch test and determined to be viable if the endosperm was white and firm.Initial germination tests were repeated twice.

Experiment one: Optimization of PVC spring tension
To investigate PVC spring tension necessary to maximize the devitalization of lambsquarters, an experiment was conducted with increasing spring tension on the rollers.Spring tension was calibrated by loading each spring onto a mounted steel pole and stacking with known weight.As weight was added, the total length of the spring was measured to calculate displacement.Linear regression was used to generate a spring constant described in eq. 1 (R 2 = 1.00;Fig. S1), where x is equal to the total length of the spring.

Spring tension (
To determine optimal tension settings for devitalization of lambsquarters, three replicates of 1000 seed samples were passed through the PVC at each of the five tension settings (Table 2).As spring tension increased, the space between the rollers decreased and therefore increased the pressure applied to the seeds.Tension settings were selected by measuring maximum possible tension and then increasing spring length in 0.5 cm increments.Spring tension (N) was subsequently calculated using eq. 1.For all tension experiments, roller speed was set to 450 revolutions min −1 (rev min −1 ) measured with a digital tachometer (DT-2234A; Fitnate, Brooklyn, NY, USA).This speed was equivalent to 50% maximum roller speed when spring tension was at 0%.For all tests, samples were slowly poured across the center section of the rollers and allowed to run for an additional 10 s to clear.Compressed air was then passed over the rollers to ensure all sample material had cleared the PVC.Entire samples were collected in a catch tray below the rollers and subjected to germination testing without surface sterilization in 10 cm × 20 cm germination boxes lined with moist filter paper.Germination boxes were kept in the dark at 22 • C for 14 days before counting germinated seedlings followed by a pinch test.Samples were imaged (Infinity3-1; Lumenera, Ottawa, ON, Canada) on a stereomicroscope (SMZ168T-LED; Motic, Richmond, BC, Canada) pre-and post-germination testing at 10× magnification for observation.This experiment was repeated thrice.

Experiment two: Optimization of PVC roller speed
To determine the optimal roller speed for lambsquarters seed devitalization, three replicate 1000 seed samples were passed through the PVC at five speed settings (Table 2).Speed settings were determined with spring tension set to 0.33 N (100% of maximum).This tension was used for all roller speed experiments.Sample processing with the PVC, germination testing, and imaging were similar to those described in experiment one.This experiment was repeated thrice.
Experiment three: Effect of optimized settings on devitalized seed germination in soil The results and observations of experiments one and two, which are later discussed, generated a novel hypothesis that increased germination observed under controlled conditions may result in germination yet failure to emerge when exposed to the fluctuating environmental conditions in soil.To evaluate this hypothesis, 1000 lambsquarters seeds were passed through the PVC using settings that maximized seed germinability and minimized devitalization in Petri plates (0.33 N; 55 rev min −1 ).Sample processing and collection were as previously described with three replications with slight modifications.Samples were collected in the catch tray following processing and spread on a germination tray filled with sterilized Harrington field soil (Orthic Humo-Ferric Podzol with pH 6.6 and 3% OM).Samples were lightly covered with sterilized sand and thoroughly watered.Uncrushed controls were included for statistical comparison.Trays were germinated in a greenhouse at 24/16 • C and 16/8 h day/night cycles for 14 days after which emerged lambsquarters were counted.This experiment was repeated twice.

Experiment four: PVC optimized settings on other weed species
Based on experiments one, two, and three, we evaluated the settings which reduced lambsquarters emergence in soil on other problematic weed species of potato under controlled conditions.Five hundred seeds each of lambsquarters, redroot pigweed, volunteer canola, barnyard grass, and yellow foxtail were passed through the PVC using settings that maximized seed devitalization in experiment three.Each species was processed independently with three replicates per species.Uncrushed controls were included for statistical comparison.Sample processing with the PVC, germination testing, and imaging were as described in experiment one.This experiment was repeated twice.

Experiment five: PVC optimized settings during simulated harvest
To evaluate the PVC during a simulated potato harvest, a mixture of weed seeds consisting of 500 seeds each of lambsquarters, redroot pigweed, volunteer canola, barnyard grass, and yellow foxtail were passed through the PVC with 100 g of dried potato vine biomass using optimized settings (0.33 N; 55 rev min −1 ).Potato biomass was collected from weed-free plots in the summer of 2020 at the Harrington research farm prior to potato desiccation and oven-dried to constant weight at 60 • C until use.This volume of biomass represents an approximate area of 1 m 2 of planted potato row prior to potato vine desiccation.Potato biomass was thoroughly mixed with all weed seeds and processed as described in experiment three in three replications.This experiment was repeated twice.

Statistical analysis
To determine optimized settings in experiments one and two, % viability was calculated as described in eq. 2, where the number of viable seeds in the sample was determined from initial germination testing described above (Table 1; Tidemann et al. 2017a).

% viability =
# of viable seeds after processing # of viable seeds in the sample (2) The percent devitalization for each sample was then calculated by eq.3: Generalized linear mixed effects models were then constructed to evaluate the effects of increasing spring tension and speed on lambsquaters seed devitalization (%) with Proc Glimmix in SAS v9.4 (SAS Institute, Cary, USA) using a beta distribution and logit link function.Spring tension and speed were treated as fixed effects and experimental run as a random effect.The significance of random effects was evaluated with a Wald's test, and data from experimental runs were combined.Least square means from experiments one and two were compared with a Tukey's HSD test and a type I error rate of α = 0.05.To generate curves, linear and quadratic regression analysis was conducted on the least square means using SigmaPlot v.14.0 (Systat Software, Palo Alto, USA) using eqs. 4 and 5: where x is equal to spring tension or roller speed for experiments one and two, respectively, a and b are slopes, and c is the intercept.Model R 2 were compared, and the best fitting model was selected.
To evaluate the effects of optimized PVC settings in experiments three, four, and five, linear mixed effects models were constructed with Proc Mixed in SAS v9.4 (SAS Institute).Fixed effects included treatment for experiment three and species, treatment (crushed or uncrushed), and their interaction for experiments four and five.In all cases, the significance of random effects was evaluated with a Wald's test, and data from experimental runs were combined.Least square means of % devitalization and standard errors were generated for all experiments.Least square means from experiment four were compared with a Tukey's HSD test and a type I error rate of α = 0.05.

Results and discussion
Optimizing the potato vine crusher Lambsquarters seed devitalization was affected by spring tension in experiment one (p = 0.01; Fig. 2A).Increasing tension on the PVC rollers, however, decreased lambsquarters seed devitalization from 47% at 0.10 N (29% of maximum tension) to a low of 27% at 0.33 N (100% of maximum tension) (Fig. 2A).In contrast, the roller speed of the PVC did not affect lambsquarters seed devitalization in experiment two (p = 0.56; Fig. 2B).Therefore, from experiments one and two, we reject our initial hypotheses that increasing spring tension and decreasing roller speed would increase the devitalization of lambsquarters seed through increased pressure and contact time.Increasing spring tension on the PVC reduces the gap between rollers, thus increasing contact and pressure applied to seeds passing through the system which has been shown to be the most important parameter for roller mill crushing efficiency (Fang et al. 1997;Hauhouot-O'Hara et al. 1999).Similarly, the decreasing speed of the rollers increases contact time for weed seed crushing yet had no impact on lambsquarters viability.Hauhouot-O'Hara et  1999).Increased speed differential between the two rollers increases the shear force applied to seeds to improve devitalization efficiency (Fang et al. 1997;Hauhouot-O'Hara et al. 1999).Increased devitalization with greater roller speed differential, however, is dependent on roller gap with minimal response at wide and narrow spacing between rollers (Hauhouot-O' Hara et al. 1999).The speed of the PVC rollers is not independently controlled to allow for differential roller speeds which may explain the limited effect of speed in our study.Further, only the narrowest roller gap width as measured by roller tension was evaluated across speeds in our study and may have masked the effect of speed on lambsquarters devitalization.We observed numerous instances of successful elongation of hypocotyls and radicles from seed fragments of lambsquarters during Petri plate germination tests in experiments one and two (Fig. 3).Many seeds, however, failed to produce both shoots and roots, and we hypothesized these would not emerge in soil.Despite the largest gap between rollers from the lowest spring tension resulting in the highest seed devitalization percentage of lambsquarters, the narrowest gap and highest spring tension had the largest impact on lambsquarters seed viability.Therefore, these settings were chosen to evaluate control in field soil and simulated harvest in subsequent experiments.
There was a significant treatment effect on lambsquarters seedling emergence in soil in experiment three (p < 0.01).Crushing lambsquarters seed with the PVC reduced emergence in soil by 53% relative to the uncrushed control (Table 3), supporting our initial hypotheses and novel third hypothesis that the germination of seed fragments in controlled Table 4. Seed devitalization of several problematic weeds following treatment with a potato vine crusher (PVC) operated with settings that minimized seed devitalization (0.33 N; 55 rev min −1 ) determined in experiments one and two.Species were germinated in Petri plates.Data are presented as % devitalization calculated as 1 − % viability from the initial viability testing.
conditions by the PVC results in seedlings failing to emerge in soil.Similarly, Berry et al. (2014) found that annual ryegrass emergence in soil following roller mill processing was also reduced despite high seed germination of milled seeds on germination paper.Berry et al. (2014) attributed this difference to the harsh environment of field soil where damaged seed did not emerge in soil yet germinated in Petri plates.This contrasts with Schwart-Lazaro et al. ( 2017) who observed similar results between soil assays and Petri plate seed viability tests for a variety of weed species, including lambsquarters, following processing with the integrated HSD while using commercial potting mix.We propose several possible explanations for the discrepancy in our results.First, it is possible the PVC scarifies lambsquarters seeds, damaging the seed coat and thereby breaking primary dormancy and increasing germination percentage under controlled conditions.Lambsquarters seed dormancy is controlled through changes in seed-coat thickness (Karssen 1970) which is dependent on day length during seed ripening of the maternal plant (Bassett and Crompton 1978;Jursik et al. 2003).Studies have shown that the partial or complete removal of the lambsquarters seed coat can increase germination percentage from less than 60% in intact seeds to greater than 70% in seeds with removed seed coats (Martin 1943).Similarly, Currie and Peeper (1988) observed increased germination of slimleaf lambsquarters (Chenopodium leptophyllum Moq.) seed following combine harvesting, whereas hand-harvested seed did not germinate.Currie and Peeper (1988) suggested this was due to seed scarification while passing through the combine.In contrast to our results, Fleischman (1951) observed a 22%, 88%, and 92% reduction in lambsquarters seed germination following 25%, 50%, and 100% removal of the seed coat, respectively, under controlled conditions.These results are potentially explained by comparing our viability testing under controlled conditions (Fig. 2) to experiment three in field soil (Table 3).We observed damaged lambsquarters seeds germinating in controlled conditions, which we suspected would not emerge in soil.This is supported by results in experiment three (Table 3) and past studies (Lyon and Rush 1993;Berry et al. 2014).We acknowledge this may have resulted in double counting (i.e., hypocotyl and radi-Table 5. Effect of the potato vine crusher on weed seedling emergence (% of total) in soil following seed crushing during a simulated harvest.
cal counted as two seeds rather than one) during germination testing and thus artificially inflating seed viability in Petri plate tests, but the identification of individual seeds was not possible as seeds became fragmented after processing with the PVC.When considering our overall results and those of other researchers, we suggest that the PVC scarifies lambsquarters seeds, breaking dormancy and thereby stimulates germination under controlled conditions.This stimulated germination, however, results in a failure to emerge when damaged seeds germinate in soil.Davis et al. (2008) also observed that mechanical injury to the lambsquarters seed coat reduced viability following a 2-month incubation period in the field.Similar to recent efforts to elucidate the effects of the HSD on overwinter survival of weed seeds (Shergill et al. 2020a), future research should consider the long-term effects of the PVC on the lambsquarters seed bank under field conditions.

Devitalization of other weed species
There was a significant effect of PVC treatment on seed devitalization of all species in experiment four (p < 0.01; Table 4).Devitalization of large seeds of barnyardgrass, volunteer canola, and yellow foxtail, as well as the smaller seed of redroot pigweed, was >87% (Table 4).Lambsquarters control was 65% and similar to results from experiment three (Fig. 2).These results demonstrate the PVC to be less effective than impact mill systems (Schwartz-Lazaro et al. 2017;Tidemann et al. 2017a;Shergill et al. 2020a).Further, the control of all weed species was reduced during simulated harvest in experiment five (p = 0.02; Table 5).Control was 59% (p < 0.01), 17% (p < 0.01), and 63% (p = 0.03) for barnyardgrass, volunteer canola, and yellow foxtail, respectively (Table 5).The control of lambsquarters and redroot pigweed was 49% and 51%, respectively, yet did not differ from uncrushed controls.Overall, while still effective, the results of the simulated harvest demonstrate the PVC to have reduced efficacy during harvest relative to controlled conditions and, in particular, when compared to impact mill systems such as the HSD.We did not detect a significant correlation between TSW nor seed size and devitalization (data not shown), though greater devitalization of the large seeds of volunteer canola, barnyard grass, and yellow foxtail relative to the small seeds of lambsquarters and redroot pigweed suggest that there is a seed size effect.This is perhaps due to biomass and debris interfering with contact pressure between weed seeds and the PVC rollers, therefore decreasing overall efficacy.Several authors have suggested that seed size may be of minimal importance in impact mill systems (Schwartz-Lazaro et al. 2017;Tidemann et al. 2017a;Shergill et al. 2020a) possibly due to greater and more efficient damage to seeds in impact mill systems over roller mills (Gossen et al. 1998;Berry et al. 2014).In comparison to impact mills where material is recirculated, maximizing fragmentation of seeds entering the system, material is passed only once through the PVC rollers, potentially increasing the importance of seed size for HWSC.This effect may be exasperated further during conditions of a potato harvest when larger volumes of potato and weed biomass would pass through the PVC.This, however, represents an opportunity to increase the energy transfer efficacy of the PVC through improved engineering such as the addition of more rollers to increase the number of impacts, ensuring consistent roller pressure, adding teeth to the rollers to increase surface area, or modifications to the design to allow for a differential in speed between the counter-rotating rollers.Future studies under field conditions and variable environments and soil types are required to fully elucidate these effects on HWSC efficacy with the PVC and to identify methods for improvement during potato harvest.

Considerations for future HWSC experiments
These results represent the first report of an HWSC device for a horticultural crop and specifically for potato production.In addition, we demonstrate the first report of successful germination of lambsquarters seed fragments.With increased interest in HWSC, another consideration for future studies is the possible contribution of partially crushed seed and seed scarification to control in the field.Our results demonstrate that HWSC is possible.While the PVC is not as effective as impact mill systems it was not specifically designed for devitalization of weed seeds and alterations to the PVC design could dramatically improve its efficacy.Shergill et al. (2020b) suggest that HWSC efficacy of greater than or equal to 20% can be effective for stabilizing and reducing Palmer amaranth (Amaranthus palmerii L.) populations in North American soybean production.While not commercially available, with efficacy greater than 50% on several common and problematic weed species, freely available design plans, and made of relatively low cost components, we have demonstrated that the PVC is a promising integrated weed management tool for HWSC in Canadian potato production.

Fig. 1 .
Fig. 1.The potato vine crusher (PVC) (A) side view of the PVC, (1) paired cylindrical metal rollers (91.5 cm × 17.28 cm, L × D), (2) raised metal beads (1.3 mm × 7.8 mm, H × W), and (3) springs (28.8 cm × 7.48 cm, L × D).Bolts connected to either end of the springs increase spring tension.Each roller is powered by a motor which is both connected to a dial controlling speed (not shown).(B) The PVC mounted on the rear of a potato harvester.

Fig. 2 .
Fig.2.The effects of (A) increasing spring tension (N; kg cm s −2 ) and (B) increasing roller speed (revolutions min −1 ; rev min −1 ) on lambsquarters seed viability, expressed as % devitalization.Values represent least square means ± SEM.Treatments not sharing the same letter are significantly different according to Tukey's HSD (α ≤ 0.05).Data are presented as % devitalization calculated as 1 − % viability from the initial viability testing.

Fig. 3 .
Fig. 3. Examples of successfully germinating lambsquarters seed fragments following processing with the potato vine crusher (PVC) during experiments one and two.All images are at 10× magnification.A scale bar (1 mm) is included for reference.

Table 1 .
Thousand seed weight (TSW), seed size (mm), and initial viability (%) for weed seed lots used to evaluate the efficacy of the potato vine crusher during stationary testing.Values are means ± SE.

Table 2 .
Spring tension (N; kg cm s −2 ) and speed (revolutions min −1 ; rev min −1 ) used in experiments one and two to evaluate the devitalization of lambsquarters seed after passing through the potato vine crusher.Numbers in brackets following spring tension represent % of maximum.Spring tension was set to 0.33 N for measurement of roller speed and numbers in brackets represent % of maximum roller speed at this tension.