Nitrogen Response and Soil Microbial Activity in Potato Cropping Systems As Affected by Fumigation

executive summary

Fumigation is commonly used to control soil-borne pathogens in potato fields. Its shortterm benefits include improved disease control and healthier root systems, which may decrease nutrient input requirements. However, fumigation also eliminates beneficial soil organisms, which may depress the capacity of the soil microbial community to control pathogens and facilitate nutrient cycling. The goal of our research was to determine the effects of fumigation and
fumigation source on N response and disease prevalence in Russet Burbank potatoes. The objectives of this study were to: 1) determine the effects of Vapam and Chloropicrin fumigation on potato response to N fertilizer, and 2) characterize the effect of fumigation on soil microbial activity and nitrogen transformations.

The two-year study was conducted at the Sand Plain Research Farm in Becker, Minnesota, on a Hubbard loamy sand soil in 2016 and 2017. Treatments were applied in a split-plot randomized complete block design with four blocks. Whole plots received either Chloropicrin, Vapam, or no fumigant. Each whole plot was split into five subplots, each receiving N at one of five total rates (including 40 lbs·ac-1 N as DAP at planting): 40 lbs·ac-1, 120 lbs·ac-1, 180 lbs·ac-1, 240 lbs·ac-1, and 300 lbs·ac-1. Fumigation treatments were applied in October 2015 and November 2016 at the rates of 70 gal·ac-1 Vapam and 100 lbs·ac-1 Chloropicrin. Nitrogen treatments as Environmentally Smart Nitrogen (ESN: 44-0-0; Agrium, Inc.) were applied at emergence. Plant responses to treatment were evaluated in terms of tuber yield and N uptake. Soil responses were measured in terms of microbial respiration (CO2 produced in 24 hours, a measure of microbial activity), NH4-N concentration, and NO3-N concentration.

Fumigation significantly decreased the concentration of Verticillium propagules in the soil in both years. It also decreased the severity of Verticillium wilt observed in late summer. The severity of Verticillium wilt also decreased with increasing N application rate in each year, suggesting that greater plant vigor in the fumigated plots partially explains the diminished wilt severity observed in these plots.

In both 2016 and 2017, total and marketable yields were lower in the non-fumigated plots than plots receiving either fumigant. In 2017, but not in 2016, plots receiving Vapam had higher total and marketable yields than plots receiving Chloropicrin, averaged across N rates. In both years, yield increased with increasing N rate, regardless of fumigation treatment, but yield did not change significantly across N rates of 180 to 300 lbs·ac-1 N. Based on a quadratic N response
function, fumigation decreased the estimated N application rate at which peak yield was achieved, from 251 lbs·ac-1 N for the non-fumigated plots to 229 lbs·ac-1 N for the Chloropicrin-treated plots and 224 lbs·ac-1 N for the Vapam-treated plots, based on combined data from both years.

While fumigated and non-fumigated plots had similar rates of microbial respiration prior to fumigation, the non-fumigated plots had significantly higher respiration rates at planting and mid-season than the plots receiving either fumigant in both years. The effect of fumigation on microbial respiration was no longer significant by October in 2016, but in 2017, the non-fumigated plots continued to have significantly higher microbial respiration rates than the plots treated with Vapam, with the Chloropicrin-treated plots intermediate. Nitrogen application rate was related to microbial respiration rate in both years (though only weakly in 2017), but the effect was inconsistent between years and difficult to interpret in each.

In each year, the fumigated plots, especially those treated with Chloropicrin, had high soil NH4-N and low soil NO3-N relative to the non-fumigated plots, indicating that fumigation interferes with nitrification. Soil NH4-N and NO3-N concentrations increased with the application rate of N in July of each year, but not at other times, indicating that the effect of N fertilization on soil mineral N concentrations did not persist beyond harvest.

Nitrogen uptake was greater in the fumigated plots than the non-fumigated plots in both 2016 and 2017. This difference was due to a difference in biomass production, especially tuber yield, rather than a difference in tissue N concentrations. Nitrogen uptake also increased with N application rate. Averaged across both years and relative to the non-fumigated plots, fumigation increased N uptake at the recommended N application rate of 240 lbs·ac-1 N by 45 lbs·ac-1 N in
Chloropicrin-treated plots and 52 lbs·ac-1 N in Vapam-treated plots.

Overall, while fumigation suppressed microbial soil activity and inhibited nitrification, it was also effective in controlling Verticillium wilt, and it was presumably effective against other pathogens as well. The net result was that fumigated plots had higher yields, achieved peak yield at lower application rates of N, and took up more N at a given application rate than non-fumigated plots in both years of the study. Higher uptake of N with fumigation will reduce soluble or residual N in the soil as well as the potential for nitrate leaching. A major long-term goal in soil science is to obtain the benefits of fumigation, in terms of pathogen control and yield, without the negative impacts on the non-pathogenic components of the soil microbial community.

summary

Fumigation is commonly used by potato growers to control soil-borne pathogens. Its short-term benefits include improved disease control and healthier root systems, which may decrease nutrient input requirements. However, fumigation also eliminates beneficial soil organisms, which may depress the soil community’s capacity for pathogen control and nutrient cycling. The goal of our research was to determine the effects of fumigation and fumigation source on N response and disease prevalence in Russet Burbank potatoes. We applied treatments in a split-plot randomized complete block design with four blocks. Whole plots received either Chloropicrin, Vapam, or no fumigant, and each whole plot was split into subplots, each receiving N at one of five total rates (including 40 lbs·ac-1 N as DAP at planting): 40 lbs·ac-1, 120 lbs·ac-1, 180 lbs·ac-1, 240 lbs·ac-1, and 300 lbs·ac-1. Fumigation treatments were applied in October and November 2016, and N treatments as Environmentally Smart Nitrogen (ESN: 44-0-0; Agrium, Inc.) were applied at emergence, on May 11, 2017. Soil 24-hour CO2 production, NH4-N, and NO3-N were determined for six-inch soil samples collected before fumigation in 2016 and before planting, after emergence N application, and after harvest in 2017. Leaflet chlorophyll contents (SPAD readings) and petiole NO3-N concentrations were measured at five times between hilling and harvest. The severity of Verticillium wilt was assessed from late July until September 12, nine days before harvest. Tuber yield, size, and quality (including the prevalence of scab) and vine and tuber N uptake were determined after harvest. Tuber sugar concentrations and French fry color were measured at harvest and after three months storage. Total and marketable yields were higher in the fumigated plots than in the non-fumigated plots, and yields were higher with Vapam than Chloropicrin. Yields  increased with N rate regardless of fumigation. The percentage of yield represented by tubers weighing over six or ten ounces was higher and less responsive to N rate in fumigated plots than in non-fumigated plots, suggesting that fumigation may decrease N requirements for tuber bulking. Plots receiving Chloropicrin had a lower prevalence of scab than those receiving Vapam or no fumigant. Tuber specific gravity was higher in plots receiving Vapam than in non-fumigated plots, with Chloropicrin-treated plots intermediate. Soil respiration was lower in the fumigated plots than in the non-fumigated control plots after the fumigation treatments were applied, though this effect diminished over time. The fumigated plots, especially those treated with Chloropicrin, had high soil NH4-N and low NO3-N relative to the non-fumigated plots, indicating that fumigation may interfere with nitrification. Leaflet chlorophyll content increased with N application rate and was slightly higher in the plots receiving either fumigant than in the non-fumigated plots. Petiole NO3-N concentrations also increased with N application rate, but while the fumigated plots had higher petiole NO3-N concentrations initially, there was no significant effect of fumigation treatment on petiole NO3-N concentration overall. Vine and tuber N concentrations increased with N application rate. Vine, tuber, and total N uptake also increased with N application rate, and N uptake was also higher in the treatments receiving either fumigant than the control plots. The non-fumigated control treatment had greater Verticillium wilt severity than the treatment receiving Chloropicrin, which had greater severity than the treatment receiving Vapam. Tuber glucose concentration decreased with increasing N application rate, but the effect of fumigation treatment on tuber glucose was ambiguous. The bud ends of tubers from the non-fumigated control treatment produced slightly darker French fries than those from the treatment receiving Vapam. Overall, we found that fumigation increased marketable yield at all N rates tested and decreased N requirements for tuber bulking, increased N uptake and leaf chlorophyll content, decreased Verticillium wilt severity, and decreased soil microbial activity, including nitrification, during the growing season.

background

Fumigation of potato fields to control pathogens has well-known short-term benefits. Most directly, fumigation decreases disease incidence. An apparent consequence of this is that potato plants in fumigated soil have healthier root systems, which may result in a decreased requirement for nutrient inputs. However, a major drawback of soil fumigation is that it eliminates beneficial soil organisms in addition to the pathogens. The benefits such organisms provide include pathogen control and nutrient cycling activities. Consequently, once a field is fumigated, additional applications of fumigant are required to control pathogens each time potatoes are planted in the field and nutrient cycling may be disrupted during and beyond the years when fumigant is applied. The objectives of this study were to: 1) determine the effects of Vapam and Chloropicrin fumigation on potato response to N fertilizer, and 2) characterize the effect of fumigation on soil microbial activity and nitrogen transformations.

methods

Study design
The study was conducted at the Sand Plain Research Farm in Becker, Minnesota, on a Hubbard loamy sand soil. The previous crop was soybeans. The plots had been cropped to potatoes in a 3 to 4 year rotation for the previous 25 years, without fumigation.

Fumigation treatments were arranged in a randomized complete block design with four blocks and three fumigation treatments. The fumigation treatments were: no fumigation, with cultivation on October 24, 2016; cultivation on October 18, 2016, followed by fumigation with Chloropicrin on October 19, 2016 at 100 lbs·ac-1 applied in strips; and cultivation on October 24, 2016, followed by fumigation with Vapam at 70 gal·ac-1 injected at 6” and 10” on November 10.

Five N fertilization treatments were arranged as randomized 20 X 21-foot subplots within each fumigation plot, giving the study a split-plot randomized complete block design. All subplots received 40 lbs·ac-1 N as DAP at planting, plus 0, 80, 140, 200, or 260 lbs·ac-1 N as ESN at emergence, depending on the assigned N treatment.

A summary of the treatments is presented in Table 1.

Soil sampling
To assess initial soil characteristics in the study field, soil samples to a depth of six inches were collected on April 13 and sent to Agvise Laboratories (Benson, MN) to be analyzed for Bray P2O5; NH4OAc-extractable K2O, Ca, and Mg; Ca(H2PO2)2 / Ba-extractable SO4-S; hot-water-extractable B; DTPA-extractable Cu, Fe, Mn, and Zn; soil water pH; and LOI soil organic matter content. Results are presented in Table 2.

To evaluate responses of soil mineral N concentrations to treatment application, soil samples to a depth of six inches were collected from each plot on October 17, 2016, and April 13, July 5, and October 13, 2017. The samples were dried at 95°C for 48 hours, ground, and subsamples were extracted with 2N KCl. The extracts were analyzed for NH4-N and NO3-N concentrations using a Wescan nitrogen analyzer.

The soil microbial respiration rate was determined for a 40-g subsample of each sample using Solvita Soil CO2 Burst Test kits (Woods End Laboratories), which measure the amount of CO2 a wetted sample emits in a 24-hour period. The sample was placed a 150-mL plastic beaker inside a glass jar and wetted to achieve 50% water-filled pore space. A CO2-detecting gel on a plastic paddle was placed inside the jar but outside the beaker, and the jar was sealed with a plastic lid with a CO2-proof rubber gasket. The jars were incubated at 20°C for exactly 24 hours. The CO2-detecting gel was immediately
analyzed with a Solvita Digital Color Reader to measure the CO2 concentration in the jar in ppm. A duplicate of one subsample, as well as a standard, were run with each set of Solvita tests to ensure accuracy.

Planting and N treatments
The subplots were planted with Russet Burbank whole “B” seed potatoes on April 25, 2017, with one-foot spacing within rows and three-foot spacing between rows. Each subplot was seven rows wide. In each subplot, the fourth and fifth rows from the irrigation alley were designated as harvest rows. In these two rows, the first and last seed potato in each subplot was replaced with a Chieftain cut “A” seed potato to identify the boundaries between subplots during harvest. Each adjacent pair of whole plots was surrounded by a buffer strip of Russet Burbank potato plants five feet wide on the ends and three feet (one row) wide along the sides. At row opening, 40 lbs·ac-1 N, 103 lbs·ac-1 P2O5, 182 lbs·ac-1 K2O, 41 lbs·ac-1 S, 21 lbs·ac-1 Mg, 1.1 lb·ac-1 Zn, and 0.6 lbs·ac-1 B were banded in as a blend of DAP (18-46-0), MOP (0-0-60), SulPoMag (0-0-21.5-21S-10.5Mg), BluMin (17.5% S, 35.5% Zn), and Granubor (14.3% B). Environmentally Smart Nitrogen (ESN; 44-0-0; Agrium, Inc.) was hand-broadcast on subplots per the assigned N treatments shortly after shoot emergence, on May 11, and then hilled in.

Plant stand, leaflet chlorophyll content, and petiole NO3-N
For each plot, plant stand in the harvest rows was recorded on June 8. The number of stems per plant for ten plants in the harvest rows was recorded on June 13. On 5 days throughout the summer, chlorophyll content in the terminal leaflet of the fourth leaf from the tip of 20 shoots per plot was recorded with a Konica Minolta SPAD-502 chlorophyll meter, generating a single average SPAD reading for each plot. SPAD readings were taken on June 15 and 27, July 11 and 25, and August 8 (i.e., 35, 47, 61, 75, and 89 days after the emergence fertilizer was applied). On the same days that SPAD readings were collected, the petiole of the fourth leaf from the tip was collected from each of 20 shoots per plot.

Harvest, tuber quality, and tuber sugars and fry color
Tubers were harvested on September 21 (149 days after planting) and sorted by size and USDA grade. Representative 25-tuber samples were evaluated for hollow heart, brown center, dry matter content, and specific gravity. Representative subsamples from each plot were sent to USDA-ARS (East Grand Forks, MN) to determine the sucrose and glucose concentrations of the stem and bud ends of the tubers on a fresh-weight basis. Samples from the stem and bud ends were French-fried by USDA on October 10 and 11, and their reflectance was determined using a Photovolt reflectometer. Measurements of tuber sugar concentrations and reflectance were repeated three months later.

Data analysis
The data were analyzed with SAS 9.4m3® software (copyright 2015, SAS Institute, Inc.) using the MIXED procedure. For each dependent variable related to tuber yield or quality or plant N uptake, fumigation treatment, N treatment, their interaction, and block were treated as fixed effects, while the interaction between block and fumigation treatment (the factor differentiating whole plots) was treated as a random effect. For repeatedly measured dependent variables, including soil respiration, soil NO3-N, leaflet chlorophyll content, petiole NO3-N concentration, and tuber sugar concentrations, dependent variables were analyzed as functions of sampling time, fumigation treatment, N treatment, their interactions, and block as fixed effects, block*fumigation treatment as a random effect, plot as the subject variable, and sampling time as the repeated-measures variable. An autoregressive (AR[1]) covariance matrix structure was used for chlorophyll content, petiole NO3-N concentration, and tuber sugars, while a compound symmetrical (CS) structure was used for soil respiration and mineral N measurements, which were separated by significant events (the application of fumigation and N treatments and harvest) that were expected to minimize the effect of autocorrelation between sampling
dates. In all models, denominator degrees of freedom were estimated using the Kenward-Roger approach (the KENWARDROGER option in SAS). Marginal means for dependent variables were determined using the LSMEANS statement, and post-hoc pairwise comparisons (alpha = 0.10) were conducted using the DIFF option. Pairwise comparisons are only presented where the significance (Pvalue) of fumigation, N treatment, or their interaction in the model is less than 0.10.

results and discussion

Tuber yield, size, and grade
Tuber yield, size, and grade results are presented in Table 3. Total and marketable yield were related to both fumigation treatment and nitrogen application rate. The treatments receiving Vapam had higher yields than those receiving Chloropicrin, which had higher yields than the non-fumigated treatments, averaged across N application rates. Yields increased with increasing N application rate, especially between 40 and 180 lbs·ac-1 N total, averaged across fumigation treatments. The nonfumigated plots showed a stronger response to N rate than the plots receiving either fumigant, but the effect of the interaction between fumigation treatment and N application rate was not quite significant for either total or marketable yield.

The percentage of yield represented by tubers weighing over six ounces was lower in the nonfumigated plots than in those receiving Chloropicrin or Vapam, and a parallel but non-significant difference was observed for the percentage of yield represented by tubers over ten ounces. For both tuber-size thresholds, the percentage of yield in large tubers increased as the application rate of N increased, especially between 40 and 180 lbs·ac-1 N total. The non-fumigated control plots showed a much stronger response of the percentage of yield in tubers over six or ten ounces to N rate between 40 and 120 lbs·ac-1 N applied in total than did the plots receiving Chloropicrin or Vapam. As a result, the effect of the interaction between fumigation treatment and N rate was significant for both variables.

Tuber quality
Tuber quality results are presented in Table 4. The prevalence of hollow heart and brown center was higher in the subplots receiving 180 lbs·ac-1 N at total than those receiving other rates. The cause of this result is unclear.

The prevalence of scab was lower in the plots receiving Chloropicrin than in the nonfumigated plots or the plots receiving Vapam. The effect of the interaction between fumigation treatment and N application rate was significant at α = 0.10, but this appears to be a reflection of the sporadic occurrence of scab, which was absent from 22 of 45 subplots, but present in up to 32% of tubers in others. 12 of the 23 subplots with scab were in the non-fumigated plots, versus 5 in Chloropicrin-treated plots and 7 in Vapam-treated plots, suggesting that both fumigants have some suppressive effect on scab.

Fumigation affected tuber specific gravity, with tubers from Vapam-treated plots having higher specific gravity than those from non-fumigated plots. Plots treated with Chloropicrin produced tubers with specific gravity intermediate between the non-fumigated and Vapam plots.

Soil respiration
The results of 24-hour CO2 burst tests (a measure of soil microbial activity) are presented in Table 5. Fumigation treatment and the fumigation*date interaction were significantly related to soil CO2 production. The non-fumigated control treatment had a higher rate of soil CO2 production, averaged across N treatments, than either fumigated treatment in April and July 2017, but not in October 2016 (before the fumigation treatments were applied). The control treatment had a higher rate of soil CO2 production than the Vapam treatment in October 2017, with the CO2 production rate of the Chloropicrin treatment intermediate between the two and not significantly different from either.

The effect of the interaction between nitrogen treatment and date on soil CO2 production was also significant. Three of the five nitrogen treatments showed decreases in CO2 production between October 2016 and April 2017 and between July and October 2017, with increases in production between April and July 2017. The other two treatments did not follow this pattern. The treatment receiving no N at emergence had a steady decrease in soil CO2 production across all four sampling times, and the treatment receiving 240 lbs·ac-1 N total had higher soil CO2 production in October 2017 than in July of that year. The treatment receiving no N at emergence probably showed decreasing respiration throughout the study because N availability limited microbial activity. The high average CO2 production in October 2017 of subplots receiving 240 lbs·ac-1 N total is difficult to explain.

Soil NH4-N and NO3-N
Soil NH4-N and NO3-N concentration results are presented in Table 6. Soil NH4-N concentrations did not vary with treatment on October 17, 2016, before treatments were applied, nor on October 13, 2017, after harvest. Plots receiving Chloropicrin had higher soil NH4-N concentrations than the non-fumigated control plots on both April 13 and July 5, 2017. Plots receiving Vapam had soil NH4-N concentrations greater than the non-fumigated control plots but less than the plots
fumigated with Chloropicrin on April 13, after fumigation but before N treatments applied at hilling. By July 5 (55 days after application of N treatments and hilling), the difference in soil NH4-N concentration between the Vapam-treated plots and the non-fumigated plots was no longer significant.

Soil NO3-N concentrations were unrelated to treatment on October 17, 2016, before treatments were applied. In all three samples taken after fumigation treatments were applied, the plots receiving Chloropicrin had lower soil NO3-N concentrations than the non-fumigated plots. On July 5, 2017, the plots treated with Vapam had soil NO3-N concentrations higher than the Chloropicrin-treated plots, but lower than the non-fumigated plots.

The high soil NH4-N concentrations and low NO3-N concentrations observed in the fumigated plots, particularly the Chloropicrin-treated plots, suggest that fumigation had an inhibitory effect on nitrification. Nitrification is mediated by microbes, and the inhibitory effect of fumigants on this process is presumably a result of the negative effect of fumigants on microbial activity. We observed similar effects on soil NH4-N, but not NO3-N, in 2016.

Soil NH4-N concentration was only related to N application rate on July 5, 2017, after the N treatments were applied but before harvest. The subplots receiving no N at emergence had lower soil NH4- N concentrations than those receiving between 140 and 260 lbs·ac-1 N, with the subplots receiving 80 lbs·ac-1 N intermediate. In contrast, soil NO3-N increased with N application rate in both July and October 2017, though the relationship was stronger in July.

Plant stand and leaflet chlorophyll content
Results for plant stand and leaflet chlorophyll content are presented in Table 7. The number of stems per plant 33 days after the emergence fertilizer was applied was unrelated to treatment. However, plant stand was related to both fumigation treatment and the interaction between fumigation treatment and N treatment. The plots receiving Vapam had higher stand than those receiving Chloropicrin or no fumigant. The interaction effect is a result of the plots receiving Chloropicrin having higher stand than the non-fumigated plots among the subplots receiving 180 lbs·ac-1 N total; the non-fumigated plots had higher or equal stand to the plots receiving Chloropicrin at all other N application rates.

Chlorophyll content (based on readings from a Konica Minolta SPAD-502 chlorophyll meter) increased with N application rate on all five sampling dates. Chlorophyll content generally declined over time, while the response of chlorophyll content to N rate grew stronger over time. The nonfumigated control plots had slightly lower chlorophyll content than the plots receiving Chloropicrin or Vapam on each sampling date, resulting in a weak overall effect of fumigation treatment on chlorophyll content. There was a significant effect of the interaction between N treatment and sampling date.
Chlorophyll content declined more rapidly over time in treatments receiving less N.

Petiole NO3-N concentration
Results for petiole NO3-N concentration are presented in Table 8. Petiole NO3-N concentration generally decreased over time. Season-average petiole NO3-N concentration increased with the application rate of N, but was not related to fumigation treatment or the interaction between fumigation treatment and N application rate.

While all N treatment showed rapid declines in petiole NO3-N concentration at some point during the season, this decline occurred later in the season in treatments receiving higher rates of N. This resulted in a highly significant effect of the interaction between N application rate and sampling date on petiole NO3-N concentration.

The effect of the interaction between fumigation treatment and sampling date was also significant. The treatments receiving no fumigation had a significantly lower mean petiole NO3-N concentration (averaged across N application rates) on the first sampling date (June 15) than the treatments receiving either fumigant. On the third sampling date (July 11), the treatments receiving Vapam had a significantly higher mean petiole NO3-N concentration than the treatments receiving either Chloropicrin or no fumigant. There was no significant difference in petiole NO3-N concentration
between any two fumigation treatments on any of the other three sampling dates.

Tissue N concentration and N uptake
Results for tissue N concentration and N uptake are presented in Table 9. Tissue N concentration increased with the application rate of N in both vines and tubers but did not respond to fumigation treatment. Tissue N uptake in both vines and tubers also increased with the application rate of N. In addition, the treatments fumigated with Chloropicrin or Vapam took up more N into both vines and tubers than the non-fumigated control plots. The results for total N uptake paralleled those for vines and tubers separately.

Verticillium wilt development
Results for the development of Verticillium wilt between July 31 and September 12 are presented in Table 10. The severity of Verticillium wilt increased between late July and mid-September, as expected. The non-fumigated control treatment had greater Verticillium wilt severity than either fumigated treatment on all four sampling dates, even though both fumigated treatments had severity close to 90% by September 12. The non-fumigated control treatment therefore had a greater relative area under the disease progression curve (RAUDPC) than either fumigated treatment. The difference in Verticillium severity between the two fumigated treatments was less pronounced, but the plots receiving Chloropicrin had greater severity on August 22 and 31, as well as a greater RAUDPC, than the plots receiving Vapam.

N application rate also affected Verticillium severity, with disease severity declining as N application rate increased.

Tuber sugars and French fry color
Tuber sugar and French fry reflectance results are presented in Tables 11 (glucose), 12 (sucrose), and 13 (reflectance). In both stem end and bud end tissues, glucose concentrations, averaged across sampling times, generally decreased as the application rate of N increased. Fumigation treatment was not related to glucose concentration in either end of the tuber, but the effect of the interaction between fumigation treatment and N application rate on stem-end glucose concentration was significant in both ends. In either tissue, this interaction effect is difficult to interpret.

In both the stem end and the bud end, sucrose concentrations generally decreased as the application rate of N increased. The average stem-end sucrose concentration at harvest of treatments fumigated with Chloropicrin was relatively low, resulting in a significant effect of the interaction between storage time and fumigation treatment. Similarly, sucrose concentration in the bud end was significantly related to the interaction between storage time and fumigation treatment because the mean sucrose concentration of non-fumigated treatments at harvest was relatively high. The effect of the
interaction between fumigation treatment and N application rate on bud-end sucrose concentration was also significant. This effect is difficult to interpret.

French fries made from the stem ends of tubers receiving 40 lbs·ac-1 N in total were darker (i.e., had lower reflectance scores) than those made from tubers receiving N at higher rates. French fries made from the bud ends of tubers from the non-fumigated treatments were darker than fries made from Vapam-fumigated treatments, with the treatments fumigated with Chloropicrin intermediate.

conclusions

Fumigation increased tuber yield and size relative to the non-fumigated control plots. While tuber yield and size increased with increasing N rate, fumigation lowered the N requirement for bulking. Fumigation with Vapam improved plant stand relative to the non-fumigated control, while fumigation with Chloropicrin did not, raising the possibility, that Vapam is more effective at controlling some pathogen that reduces stand. The two fumigated treatments had slightly higher leaflet
chlorophyll contents than the non-fumigated control, as well as higher petiole NO3-N concentrations early in the season and higher N uptake into both vines and tubers, indicating that fumigation improves the ability of plants to acquire N from the soil. Fumigation had no clear effect on tuber sugar content, though it slightly improved the color of French fries made from the bud end of the tuber. Overall, fumigation treatment appeared to affect soil N cycling processes and overall microbial activity negatively, but plants in fumigated plots had higher tuber yields, larger tubers, and greater ability to take up N than those in non-fumigated plots.