Annual and Perennial Alleyway Cover Crops Vary in Their Effects on Pratylenchus penetrans in Pacific Northwest Red Raspberry (Rubus idaeus)

Evaluation of alleyway cover crops in an established raspberry field: The experiment was conducted from the Fall of 2014 to the Fall of 2016 (24 mon) in an established commercial ‘Meeker’ red raspberry field in Lynden, WA. The entire experimental area was 0.35 ha and comprised 17 rows of raised beds (approximately 25 cm tall, 45 cm wide) with 3 m of spacing between beds (alleyways). Before establishing this experiment, alleyways had been routinely cultivated throughout the year to control weeds, and no ground cover had been planted in the alleyways. The soil in this field is a Lynden sandy loam, with a pH of 6.4, 3.94% organic matter, a cation exchange capacity of 10 meq/100 g, 89.4 ppm of nitrate, 200.9 mg/kg of phosphorus, and 227.1 mg/kg of potassium (Brookside Laboratories, Inc., New Bremen, OH). The mean P. penetrans population density in raspberry roots in this field was 4,320 ± 798 P. penetrans/g of root. The raspberry crop was managed by the commercial grower, including irrigation and fertilization, throughout the duration of this experiment and recommended conventional practices for the region were followed (Pacific Northwest Extension, 2007).

Experimental treatments were arranged in a completely randomized design and replicated four times. Each treatment plot consisted of 9.1 m of raised bed with 1.2 m of alleyway on each side of a raised bed; the total area of each plot was approximately 28 m². A buffer area 9 m long was established between each treatment plot in the same row. Untreated beds and alleyways were adjacent to each treatment plot to avoid any undesired interaction between treatments.

Eight cover crop species were planted in the alleyways during the fall of each year (Table 1). Cover crops included in the study were selected because seeds are readily available to growers in the region and because they are suitable for the environment. Each year, a weedy mow control (Mow) and a cultivated, bare soil control (Till) were also established in the alleyways. These treatments were first planted in the Fall of 2014 and replanted in the same plots in the Fall of 2015. The exception was the perennial ryegrass mix of 51.25% Lolium hybridum ‘Tetralite’ and 48.24% L. perenne ‘Kentaur’ (grass mix 1) which was seeded only once at the beginning of the experiment and was maintained over the duration of the study.

For cover crop establishment in year 1, immediately before seeding, all alleyways were cultivated to a depth of 15 cm. Cover crop treatments were seeded on 1 October 2014 using a compact drill (3P500V; Land Pride, Salina, KS). All plots were seeded with approximately a 15 cm wide band left unseeded adjacent to the raised raspberry beds on either side. This area is where the tires of machinery and equipment pass throughout the season. Both ryegrass mixes were seeded at 28 kg/ha, per the supplier’s instructions. The remaining cover crop treatments were seeded at 112 kg/ha. Cover crop establishment, stand, and coverage were visually monitored throughout the experiment.

All of the cover crop treatments, perennial and annual, and the Mow treatment were mowed for the first time on 27 April 2015. Till plots were cultivated twice in the first year of the experiment, on 9 June and 17 August 2015. On 22 September 2015, the annual cover crop treatments and the Mow and Till treatments were terminated and alleyways were subsoiled and cultivated by the grower. The perennial treatment, grass mix 1, was left intact for the second year of the experiment. The raspberry floricanes were pruned and the canes were flail-mowed and then incorporated into the alleyways by the grower on 5 October 2015. In year 2, cover crops were seeded at the same rates as in year 1 on 8 October 2015 using a custom-made double-disc planter with a cone seeder that seeded seven rows with 15 cm spacing. The cover crop and Mow treatments were mowed once in year 2 on 13 May 2016. Till plots were cultivated on 20 May 2016 and again on 27 July. Treatments were not irrigated or fertilized throughout the duration of the experiment. The experiment was terminated on 15 August 2016.

Population densities of P. penetrans in alleyway treatments and adjacent raspberry beds were determined in the Spring and Fall of 2015 (year 1) and 2016 (year 2). To determine P. penetrans population densities in the soil, four cores, 2.5 cm diam. and 20 cm deep, were collected and combined from each side of the bed within 15 cm of the raspberry crown and from both alleyways adjacent to the bed; soil cores from the bed were kept separate from alleyway soil cores. To determine P. penetrans population densities in roots, roots from the raspberry crop were collected from three randomly selected raspberry plants in the beds, and roots from the cover crop treatments were collected from four locations in the alleyways using a square-blade shovel (15 cm³ core). All samples were placed in a cooler and transported to the laboratory for P. penetrans extraction and quantification. Mixed stages of P. penetrans were extracted from a 50 g subsample of soil using the Baermann funnel extraction method (Ayoub, 1980). Pratylenchus penetrans were collected after 5 d. For P. penetrans extraction from roots, roots ≤2 mm in diameter were preferentially selected, rinsed free of soil, and placed under intermittent mist for 5 d (Ayoub, 1980). Extracted roots were oven-dried for 1 wk at 70°C and then weighed. Pratylenchus penetrans collected from soil and roots were identified and counted using a stereoscope at ×40 magnification. The nematodes were identified as P. penetrans based on morphology as well as the presence of males, a diagnostic trait for this species (Castillo and Vovlas, 2007). Population densities are expressed as number of P. penetrans/100 g of soil or number of P. penetrans/g of root.

Fruit was mechanically harvested from late June to mid-July in 2015 and early June to early July in 2016 by the grower. To estimate yield, a modified version of the yield estimation methodology developed by Daubeny (1986) was conducted on the raspberry crop in June of each year. Three raspberry plants per treatment plot were randomly selected and the total cane number from each plant were counted and averaged. The number of laterals on two canes from each of the selected plants were counted and averaged. Then, the fruit, including buds, flowers, and green fruit on two laterals in five different fruiting zones in the raspberry canopy were counted and averaged. These means were multiplied along with the number of plants/ha, average berry weight (described below), and a yield loss correction factor which took into account potential yield loss due to mechanical harvesting. Early-, mid-, and late-season ripe raspberry fruit collection occurred on 26 June, 7 July, and 16 July in year 1 and on 14 June, 21 June, and 5 July 2016 in year 2, respectively. At each sampling date, 30 fruit were randomly selected from each treatment plot, weighed, and frozen for future total soluble solids (TSS) analysis. To perform TSS analysis, the fruit from each treatment plot and time point was crushed in a sample mesh bag (Agdia^®, Inc., Elkhart, IN) and the juice was strained out into a test tube. Three drops of juice were placed on a digital refractometer (Palm Abbe digital refractometer, Model #PA201; MISCO, Solon, OH) for each measurement. Juice from each treatment plot and time point was analyzed three times, the value of each was recorded, and the mean of the three values was calculated.

All data were subjected to statistical analysis using Statistical Analysis System software (Version 9.3; SAS Institute Inc., Cary, NC). Alpha was set at 0.05 for all data. All raspberry root and soil data were subjected to a Dunnett’s t-test, comparing all other treatments to Till. Raspberry root and soil data were compared across all four sampling dates. The cover crop soil data within each sampling date were analyzed using a Dunnett’s t-test. The cover crop root data were analyzed using an analysis of variance with Tukey as the post hoc test, instead of a Dunnett’s t-test, because there was no root data for the Till control (no roots were present). Annual cover crops were only compared within year (spring and fall sampling dates). The perennial treatments, Till and grass mix 1, were compared across all four sampling dates. To compare across sampling dates, repeated measures were used for both raspberry plant and cover crop data. Much of the nematode data had unequal variance. When this occurred, means were transformed by log (x + 10) and reanalyzed. All data are presented with original means, even when transformations were performed.

Greenhouse evaluation of cover crops as hosts for P. penetrans: Host assay experiments were conducted at the WSU-NWREC, Mount Vernon, WA, to determine the host suitability of the cover crops used in the field trial (Table 1) as well as B. juncea ‘Pacific Gold’ and S. alba ‘Ida Gold’ for P. penetrans. ‘Meeker’ raspberry was included in the experiments as a positive control. The assays were conducted in a greenhouse and treatments were arranged in a completely randomized design on a bench; treatments were replicated six times. Lynnwood sandy loam soil, collected from a raspberry field in Lynden, WA, was used for all assays. The soil was sieved through a 4-mm mesh sieve and homogenized using a cement mixer. Six subsamples of 50 g of soil were collected and P. penetrans population densities determined as described earlier; no P. penetrans were found in the soil. Two pots sizes were used in the assays; 656 ml (6.4 cm diam. and 25.4 cm height; D40H, Stuewe & Sons, Inc., Tangent, OR) and 2.6 liter (14 × 14 × 15 cm; McConkey & Co., Sumner, WA) pots.

In the first experiment (host assays 1 and 2), all the cover crop species used in the field trial and the two mustard cover crops were seeded into the 656 ml pots (Table 1). The cover crops were overseeded and then thinned to one plant before inoculation. A single tissue culture ‘Meeker’ raspberry (Northwest Plant Co., Ferndale, WA) was planted per pot. In host assay 1, plants were inoculated 14 d after seeding. To obtain inoculum, raspberry roots were collected from the field trial as described earlier and P. penetrans were collected under intermittent mist for 5 d as described earlier. In host assay 1, the 656 ml pots were inoculated with 626 P. penetrans in 3.5 ml of water, corresponding to approximately 2 P. penetrans/g of soil. Two holes, 5-cm deep, were made on either side of the plant and the P. penetrans-water suspension was added to the root zone. The pots were kept in the greenhouse for the duration of the assay at 21 to 24°C with supplemental lighting to provide a 16-hr photoperiod. Pots were watered once daily and fertilized once weekly with 20N–20P–20K dissolvable fertilizer (Jack’s Classic All Purpose; JR Peters Inc., Allentown, PA). The mustards were terminated 59 d after inoculation because of bolting. The remaining plants were terminated 72 d after inoculation. At assay termination, shoots were cut at soil level and discarded. The remaining contents of the pots were then separated into roots and soil for P. penetrans extraction as described earlier. Host assay 2 was conducted using the same methods as described for host assay 1. Inoculation of cover crops occurred 22 d after seeding. Pots were inoculated with 918 P. penetrans in 3.5 ml of water. This inoculation density corresponded to 2 P. penetrans/g of soil. The ‘Ida Gold’ and ‘Pacific Gold’ mustards were terminated 24 d after inoculation because of bolting. The rest of the plants were terminated 69 d after inoculation. Data collection for assay 2 was as described for assay 1.

In the second experiment (host assays 3 and 4), the grass mixes 1 and 2 were seeded at a rate of 1 g of seed/pot in the 2.6 liter pots and were not thinned. A single tissue culture ‘Meeker’ raspberry was planted per pot. In host assay 3, the soil in the pots was inoculated with 1,969 P. penetrans in 11 ml of water per pot, corresponding to 1 P. penetrans/g of soil 14 d after seeding. Four holes were made in the soil around the plant and the P. penetrans-water suspension was added to the root zone. Plants were terminated 72 d after inoculation. Host assay 4 was similar to host assay 3 with a few exceptions. Inoculation of cover crops occurred 22 d after seeding with 3,231 P. penetrans in 12.4 ml per pot. This inoculation density corresponded to 2 P. penetrans/g of soil. The plants were terminated 69 d after inoculation. Greenhouse conditions, plant maintenance, and data collection for host assays 3 and 4 was as described for host assay 1.

The reproductive factor (RF) value was calculated for each pot by dividing the final population of P. penetrans recovered from roots and soil by the initial inoculation population of P. penetrans (Oostenbrink, 1966). The results from the host assays were different, therefore, they were analyzed separately. All data were subjected to statistical analysis using statistical analysis system. Alpha was set at 0.05 for all data. Data were initially subjected to analysis of variance with Tukey as the post hoc test. Much of the data had unequal variance. When this occurred, means were transformed using log (x + 10) and reanalyzed. Certain data that did not meet all the assumptions of analysis of variance were analyzed using a Kruskal–Wallis nonparametric test. All data are presented with original means, even when transformations were performed.

Evaluation of alleyway cover crops in an established raspberry field: In both years of the study, cover crops germinated and emerged by 1 November. Stand and alleyway coverage were consistent and persisted into the spring. Annual cover crops began to senesce in late July, whereas grass mixes 1 and 2 remained green and continued to grow. In the first year of evaluating the cover crops in the field, although there was obvious numeric variability among the treatments, there were no significant differences for P. penetrans per gram of cover crop root (Table 2) in the Spring of 2015, the first spring after seeding. The ‘Norwest’ wheat and cereal rye treatments had P. penetrans densities 20 times greater than ‘Rosalyn’ wheat, which was the annual treatment with the lowest density. No P. penetrans were found in grass mix 1 roots; it was the only alleyway treatment to not have P. penetrans recovered from its roots. In the alleyways, the ‘Nora’ oat treatment had the highest density of P. penetrans per 100 g of soil. Population densities of P. penetrans in ‘Nora’ oat (116 ± 40 P. penetrans/100 g of soil) were significantly different than in Till (2 ± 2 P. penetrans/100 g of soil) at this sampling date and approximately twice as high as the treatment with the next highest density, grass mix 2 (61 ± 40 P. penetrans/100 g of soil). In the Fall of 2015, almost 1 yr after seeding, while there were no statistically significant differences among treatments, grass mix 1 continued to support low P. penetrans per gram of root, with only grass mix 2 having fewer P. penetrans per gram of root (Table 2). There were also no significant differences in P. penetrans population densities in alleyway soil comparing all other treatments to Till. There were 0 P. penetrans/100 g soil in grass mix 1, ‘Trical 103BB’ triticale, and ‘TriMark 099’ triticale alleyway soil samples. The Till treatment had the fourth highest average density of P. penetrans in soil (16 ± 6 P. penetrans/100 g of soil), 1 yr after treatment establishment.

In the second year of evaluating the annual cover crops in the field, in the Spring of 2016 (the first spring after the second seeding), ‘Trical 103BB’ triticale, ‘TAM 606’ oat, and ‘Rosalyn’ wheat supported low P. penetrans densities per gram of root (Table 2). During this year, grass mix 1 was considered a perennial cover crop treatment because it was maintained (no tillage) throughout the duration of the experiment. In year 2, grass mix 1 continued to support low densities of P. penetrans in roots, and the mean density was not significantly different from the annual cover crops or industry standard Till (Table 2). In the corresponding alleyway soil samples from this sampling date, there were no significant differences between Till (5 ± 1 P. penetrans/100 g of soil) and all other treatments (0 to 14 P. penetrans/100 g of soil); P. penetrans population densities were low in all treatments (data not shown). In the Fall of 2016, 1 yr after the second seeding, cereal rye had the highest P. penetrans density per gram of root and was nearly twice the density of the treatment with the next largest density, ‘TAM 606’ oat (Table 2). The grass mix 1 treatment had the lowest P. penetrans population density. Population densities of P. penetrans in alleyway soil were not significantly different between Till (55 ± 55 P. penetrans/100 g of soil) and all other treatments (3 to 88 P. penetrans/100 g of soil). The perennial cover crop, grass mix 1, was the only treatment that consistently supported extremely low densities of P. penetrans in its roots and surrounding soil throughout the 2 yr of the study.

There was a treatment by season interaction for P. penetrans population densities per gram of root for ‘Rosalyn’ wheat in the first year of the study (Table 2), with significantly higher P. penetrans densities in the fall compared with the spring; those differences did not persist into the second year. In the alleyway soil, ‘Nora’ oat had significantly higher densities of P. penetrans (116 ± 40 P. penetrans/100 g of soil) in the Spring of 2015 compared with all other seasons (3 to 42 P. penetrans/100 g of soil). The ‘TriMark 099’ triticale treatment had significantly higher P. penetrans densities in the Fall of 2016 (9 ± 5 P. penetrans/100 g of soil) compared with all other sampling dates (0 P. penetrans/100 g of soil). Across sampling dates in the perennial cover crop grass mix 1, the P. penetrans population density in roots was significantly higher in the Spring of 2016 compared with all other sampling dates (Table 2).

In the Spring of 2015, raspberry plants grown in raised beds adjacent to alleyways seeded with cover crops or residential weeds (Mow) did not have significantly higher P. penetrans population densities compared with raspberry plants grown adjacent to cultivated, bare soil (Till; Table 3). There were also no significant differences between Till and all other treatments for P. penetrans population densities in bed soil surrounding roots (2 to 203 P. penetrans/100 g of soil). This trend continued in both raspberry roots and soil for the duration of the study; there were no significant differences between Till and all other treatments at any sampling date (Table 3).

There was also a treatment by season interaction in the P. penetrans population densities of raspberry roots growing adjacent to Mow, cereal rye, grass mix 1, ‘TAM 606’ oat, ‘Trical 103BB’ triticale, and ‘Rosalyn’ wheat alleyways (Table 3). In general, population densities in the Fall of 2015 were significantly higher compared to P. penetrans population densities in the other seasons for all of the previously mentioned treatments. In addition, there was a trend across treatments that the P. penetrans population densities in raspberry roots in all treatments were lower at the end of year 2 (Fall 2016) than at the end of year 1 (Fall 2015). A treatment by season interaction was also observed in raised bed soil adjacent to cereal rye, ‘Nora’ oat, ‘TAM 606’ oat, and ‘TriMark 099’ triticale alleyways. Similar to observations for root P. penetrans populations, P. penetrans soil population densities in the raised beds generally declined over the course of the experiment with densities being the highest in the Fall of 2015 (128 to 355 P. penetrans/100 g of soil) and lowest in the Fall of 2016 (14 to 438 P. penetrans/100 g of soil). The exception to this was Till and ‘Norwest 553’ wheat. Densities of P. penetrans for both treatments were higher in the Fall of 2016 (438 ± 320 and 270 ± 95 P. penetrans/100 g of soil, respectively) than the Fall of 2015 (273 ± 23 and 165 ± 60 P. penetrans/100 g of soil, respectively), but differences were not significant.

There were no significant differences in estimated fruit yield in either year between Till and all other treatments (data not shown). Yield from raspberry plants growing adjacent to industry standard Till alleyways did not have the highest mean yield in either year (23.6 ± 3.0 t/ha in 2015, 19.5 ± 4.4 t/ha in 2016). Raspberry grown adjacent to ‘Rosalyn’ wheat alleyways had the highest mean yield in 2015 (23.7 ± 1.9 t/ha), whereas raspberry in ‘TAM 606’ oat plots had the highest mean yield in 2016 (23.4 ± 1.5 t/ha). There were also no significant differences in measured TSS in either summer of the study (data not shown). In 2015, fruit collected from raspberry adjacent to ‘TriMark 099’ triticale alleyways had the highest TSS (11.1 ± 0.2 °Brix). Fruit collected from raspberry adjacent to grass mix 1 alleyways had the highest TSS in 2016 (12.0 ± 0.2 °Brix). The TSS in 2016 was significantly higher across all treatments compared with 2015 (data not shown).

Greenhouse evaluation of cover crops as hosts for P. penetrans: Data from the host assay experiments are presented separately. In host assay 1, ‘Pacific Gold’ mustard had the highest P. penetrans per gram of root and also the highest RF value (Table 4). However, the number of P. penetrans per gram of root for ‘Pacific Gold’ mustard was not significantly greater than that of ‘Ida Gold’ mustard or either oat cultivar. The RF value for ‘Pacific Gold’ mustard was significantly higher than all other treatments with the exception of cereal rye and ‘Nora’ oat. The cover crops ‘TriMark 099’ triticale and ‘Norwest’ wheat had RF values less than 1.0, but were not significantly different than those of ‘TAM 606’ oat, ‘Trical 103BB’ triticale, ‘Rosalyn’ wheat, and the raspberry control. The cover crops ‘Nora’ oat, ‘TAM 606’ oat, ‘Pacific Gold’ mustard, and ‘Ida Gold’ mustard all had significantly higher P. penetrans densities per gram of root compared with the raspberry control and cereal rye. ‘Nora’ oat and ‘Pacific Gold’ mustard had significantly higher RF values than the raspberry control.

In host assay 2, all treatments performed similarly to host assay 1 when comparing their host suitability relative to each other, with the exception of ‘TriMark 099’ triticale. The ‘TriMark 099’ triticale cover crop had the lowest mean P. penetrans per gram of root in host assay 1, but had the 5th highest P. penetrans density in host assay 2. The RF value for ‘TriMark 099’ triticale also reflected this change. In host assay 1, it had the lowest RF value, but in host assay 2 it had the 4th highest RF value. Similar to host assay 1, ‘Norwest’ wheat had a low mean P. penetrans per gram of root (Table 4) in host assay 2, but it was only significantly lower than ‘Nora’ oat, ‘TAM 606’ oat, and ‘Ida Gold’ mustard. The ‘Norwest’ wheat cover crop was the only cover crop to have a RF value less than 1.0 in host assay 2 (Table 4), and the RF value was significantly lower than that of cereal rye, ‘Nora’ oat, and ‘TAM 606’ oat. The only cover crops to have significantly higher densities of P. penetrans per gram of root than the raspberry control were ‘Nora’ and ‘TAM 606’ oats, whereas the cereal rye and ‘TAM 606’ oat cover crops were the only cover crops to have significantly higher RF values compared with the raspberry control.

In host assay 3, grass mix 1 and 2 had significantly lower densities of P. penetrans/g of root and RF values compared with the raspberry control. Both of these cover crops had RF values < 1.0 (Table 5). In host assay 4, grass mix 2 performed similarly to host assay 3, with the lowest mean P. penetrans per gram of root and RF value (Table 5) and was significantly lower than that of grass mix 1 and the raspberry control. The grass mix 2 was the only species to have an RF value less than 1.0.

The cover crop mean root weights rankings relative to each other were consistent across the assays. In general, the root weights were larger in host assay 2 than in host assay 1 (Table 4). The ‘Pacific Gold’ and ‘Ida Gold’ mustards had the smallest root weights in both assays. The ‘Nora’ and ‘TAM 606’ oat cover crops had the 7th and 8th, respectively, largest root systems in host assay 1 and were tied for 7th in host assay 2. Raspberry, ‘Norwest’ wheat, and ‘Rosalyn’ wheat were ranked in the middle in both assays. The cover crops cereal rye and ‘TriMark 099’ triticale had heavy root weights in both assays. The only exception was ‘Trical 103BB’ triticale which was ranked 2nd for the largest root system in host assay 1, but had a much smaller root weight in host assay 2.

The current industry practice in PNW raspberry production is to maintain bare, cultivated soil in the alleyways between raspberry beds. This practice, referred to as Till in this study, did not perform better than any of the cover crop treatments. The presence of annual and perennial alleyway cover crops did not increase P. penetrans population densities in the alleyways or in adjacent raspberry plants compared with the industry standard. This was evident when comparing population densities of P. penetrans in raspberry roots (2,555 to 4,941 P. penetrans/g root when averaged across treatments) with those in cover crop roots (108 to 293 P. penetrans/g root when averaged across cover crops). These results are similar to what was previously observed by Vrain et al. (1996). In this study, cover crops that were poor hosts for P. penetrans did not affect P. penetrans population densities in adjacent ‘Willamette’ raspberry roots and soil, positively or negatively. The lack of significant results in our study confirms this as well in ‘Meeker’ raspberry, which is one of the most widely planted raspberry cultivars in the PNW (Moore and Daubeny, 1993; Pacific Northwest Extension, 2007). Low population densities of P. penetrans in alleyway soil were also observed, both in soil with cover crops growing (9 to 26 P. penetrans/100 g of soil) and bare cultivated soil (2 to 55 P. penetrans/100 g of soil). Similar population densities in soil were observed in a study comparing broadcast fumigation with bed fumigation in the raspberry production system (Walters et al., 2017); population densities of P. penetrans ranged from 3 to 174 P. penetrans/100 g of soil at three different field sites.

Raspberry was a superior host relative to the alleyway cover crops in the field environment, whereas in the greenhouse host assays many of the cover crops were better hosts for P. penetrans than raspberry. This difference may be attributed to the fact that except for grass mixes 1 and 2, all of the cover crops are annuals and naturally senesced during the summer. However, even in the spring, well before senescence, none of the cover crops were particularly good hosts for P. penetrans in the field, especially when population densities were compared with those on raspberry. Under controlled conditions in the greenhouse, cereal rye, ‘Nora’ oat, and ‘TAM 606’ oat were consistently good hosts (RF > 1), ‘Norwest’ wheat and grass mix 2 were consistently poor hosts (RF < 1), and ‘Rosalyn’ wheat and ‘Trical 103BB’ triticale were maintenance hosts (RF = 1). Differences in observed host suitability of cover crops for P. penetrans in the field and greenhouse assays may be attributed to the harsh environment in the alleyways of raspberry fields. This area of the raspberry field does not receive supplemental irrigation, only rainfall. In northern WA, whereas the winter is very wet, the summer can be dry with average rainfall from June to August of approximately 90 mm. However, the rainfall in the 3 yr during which this study was conducted was far less than the average. The total rainfall from 1 June to 31 August in 2014, 2015, and 2016 was 50.3, 31.24, and 68.33 mm, respectively (WSU AgWeatherNet, 2017a, 2017b, 2017c). In addition to lack of moisture, the alleyway can become very compacted because of repeated movement of equipment to apply pesticides and conduct mechanical harvest. In fact, the 15-cm area closest to the raspberry beds is so compacted and experiences so much tire traffic that resident weeds do not grow there. This was the case in the field where our study was conducted. These factors combined may make alleyways between raised raspberry beds an unsuitable environment for P. penetrans to survive and reproduce and may explain the lack of impact the cover crops had on the raspberry crop.

To our knowledge, this is the first time that many of these cover crop cultivars have been tested as suitable hosts for P. penetrans. Whether suitable or poor hosts, this is important information to have when planting cover crops in soil with a history of P. penetrans. The results of our greenhouse host assays were consistent with the results observed by Thies et al. (1995) in a controlled environment. Several forage grasses were evaluated in the greenhouse for 6 wk, including ‘Starter’ oat, ‘Marshall’ wheat, ‘NK-200’ perennial ryegrass, and generic cereal rye for their host suitability for P. penetrans. Based on the presented data, oat and cereal rye were extremely good hosts with RF values of 3.5 and 5.4, respectively. Wheat and perennial ryegrass were poor hosts with RF values less than 1.0. In two greenhouse studies, Forge et al. (2000) evaluated the host suitability of winter cover crops, including ‘Wheeler’ rye, ‘Saia’ oat, and ‘Amity’ oat for P. penetrans. After growing for 10 wk, ‘Wheeler’ rye and ‘Saia’ oat were maintenance hosts for P. penetrans, with RF values at 0.97 and 0.96, respectively. The RF value for ‘Amity’ oat was slightly less, 0.90, making it a less suitable host. The same study was performed again, but cover crops were left to grow for 20 wk. After 20 wk, ‘Wheeler’ rye and ‘Amity’ oat were very good hosts with RF values of 4.4 and 2.5, respectively. ‘Saia’ oat was still only a maintenance host with an RF value of 1.2. However, ‘Saia’ oat is not an equal comparison to ‘Nora’ and ‘TAM 606’ oats in our study because it is a different species, A. strigosa not A. sativa (Dial, 2014). In addition, ‘Saia’ oat was determined to be a suitable host in a field study (Vrain et al., 1996). In another greenhouse study, the host suitability of ‘Aubade’ perennial ryegrass, ‘Musketeer’ rye, and ‘Ultima’ oats for P. penetrans was evaluated (Bélair et al., 2002). All three cover crops were confirmed as very good hosts with RF values of 8.4, 9.0, and 5.7, respectively. All these greenhouse studies confirm that different cultivars within the same crop species may vary in their host suitability for P. penetrans. The duration of an experiment and the inoculum size are important factors in crop response. In general, however, the results of the previous studies support our findings; oat and cereal rye are good hosts for P. penetrans, wheat can either be a poor host or maintenance host, and perennial ryegrass is a poor host.

Although host suitability is an important factor for growers to consider, the effect a cover crop has on the cash crop yield is their primary concern. In our study, cover crop treatments did not negatively affect yield or fruit quality in either year of our study. In raspberry, alleyway management comparing bare soil, oats, and timothy grass (Phleum pretense L.) did not significantly affect raspberry fruit weight or cane diameter (Sanderson and Cutliffe, 1988), although raspberry cane diameter was numerically lowest in the timothy grass plots. Over the 4 yr of the study, the same yield trends were observed across treatments; yield was highest in the second year and lowest in the fourth year. Throughout the 4 yr, overall yield of raspberry adjacent to timothy grass was significantly lower than bare soil and oats. The yield of raspberry adjacent to oats was the highest throughout the study, but was not significantly higher than the bare soil treatment. An opposing trend was observed in ‘Willamette’ raspberry grown adjacent to alleyways with bare soil or perennial ryegrass over 5 yr (Bowen and Freyman, 1995). Fruit yield, cane height, and cane diameter over 5 yr between bare soil alleyways and alleyways planted to perennial ryegrass were significantly higher in raspberry adjacent to bare soil than raspberry adjacent to perennial ryegrass.

In a 2-yr study of raspberry in two locations, ‘Willamette’ raspberry was again grown adjacent to bare soil or perennial ryegrass alleyways (Freyman, 1989). In location 1, perennial ryegrass was established in a newly established raspberry planting, and reduction in cane diameter in the first year and cane diameter and height over the second year were observed in plants grown in areas with perennial ryegrass compared with bare soil. In the first year of the study at location 1, there were no significant differences in yield. However, in the 2nd year, raspberry adjacent to the bare soil control had significantly higher mean yield than that of perennial ryegrass. In location 2, however, the treatments were established in a 4-yr-old raspberry planting and perennial ryegrass did not reduce cane diameter compared with bare soil. In both years at location 2, raspberry adjacent to the bare soil alleyway had higher yields than raspberry growing adjacent to perennial ryegrass. Vrain et al. (1996) did not report fruit yield or fruit quality, but instead fruiting cane count, biomass, height, and diameter of ‘Willamette’ raspberry grown adjacent to a bare soil control or various cover crops, including ‘Saia’ oats. Plants adjacent to the bare soil control had higher fruiting cane mass, height, and diameter, but not fruiting cane count or total mass per plot than ‘Saia’ oat. Because of the differing ways in which the data were collected and reported in previous studies, it is difficult to make an equivalent comparison to the results in our study. With the exception of Vrain et al. (1996), none of the previous studies comparing alleyway cover crop effects on raspberry yield also included a P. penetrans component. All of these studies took place in British Columbia where P. penetrans is also extremely common in raspberry (McElroy, 1972; Vrain et al., 1996). British Columbia is often considered part of the PNW in terms of climate, soil type, and crop production. There could have been a P. penetrans effect that was not taken into account in many of these studies.

There are also other benefits than just yield which growers may consider when deciding to implement alleyway cover crops. In various production systems, the maintenance of cover crops has been shown to be more economical as compared to repeated tillage (Archer et al., 2007; Sánchez-Girón et al., 2007; Bernstein et al., 2011), especially over the long term. There is also the benefit of decreased soil erosion and increased soil organic matter (Milgroom et al., 2005), which can improve soil quality. In addition, growers may find that mowing the alleyways is simply easier than cultivating from both an equipment and logistics perspective; smaller, cheaper equipment that can maneuver more quickly through the alleyways could reduce cost, time, and labor compared with tillage (Reberg-Horton et al., 2011). Alleyway cover crops can improve soil drainage which may help mitigate standing water in the alleyways, thus allowing growers to enter their fields with machinery earlier in the spring.

The biggest limitation of this study was that it was a 2-yr experiment with a perennial crop that can be in production for at least 5 yr. This study also began during the third year of production. Results may have been different if cover crops had been seeded during the first year of establishment or if the study had been continued for additional years, as in the case of the study conducted by Freyman (1989). However, implementing alleyway cover crops once the raspberry crop has been established may have eliminated the resource competition aspect and may have been responsible for the lack of differences in yield among the treatments. This may be a strategy that growers wish to use—seeding cover crops 1 yr after planting a new raspberry crop.

This study supports the implementation of alleyway cover crops in red raspberry production in the PNW. The potential physical and chemical soil quality benefits in addition to the potential economic and labor advantages outweigh the perceived risks. Our results demonstrate that alleyway cover crops did not affect yield, fruit quality, or P. penetrans population densities in raspberry roots and soil, compared with the current industry-standard practice of repeatedly cultivated bare soil alleyways. The use of alleyway cover crops did not increase alleyway P. penetrans population densities either. Alleyways do not appear to be a hospitable environment for P. penetrans reproduction possibly because of soil compaction and lack of irrigation and fertilization of cover crops. Results varied from season to season within and across the treatments, but the bare cultivated soil control (Till) was consistently not superior to the cover crop treatments. Future studies should examine the effects of cover crops in raspberry alleyways over a longer period of time and also the effects when planted adjacent to other raspberry cultivars of different maturities.