Evaluation location: Oregon, United States
Abstract. A lack of genetic diversity in cultivated black raspberry (Rubus occidentalis L.) germplasm has been widely recognized as a major factor limiting progress towards breeding improved cultivars. Despite this, little effort has been made since the early twentieth century to systematically collect and evaluate wild black raspberry for germplasm improvement. In recent years, there has been renewed interest in black raspberry breeding to replace existing cultivars that lack durability and disease resistance. We planted seedlings from 109 wild black raspberry populations, representing 24 US states and two Canadian provinces, in the field in replicated trial plots in Corvallis, Oregon (USA), to evaluate their performance. These populations showed wide variation in morphology, architecture, fruiting season, vigor, and apparent field tolerance to Verticillium wilt. For nearly every trait examined, wild black raspberry germplasm exhibited a range of variation beyond existing cultivars, and showed great potential for use in future breeding. While most populations were uniform phenotypically, segregation for fruit gloss and possible tolerance to Verticillium wilt was noted in a few, indicating the possibility of simple inheritance of these traits. A few populations with unusual morphology, such as spinelessness or flower abnormalities, were identified, as were populations that flowered on first year canes and produced fall fruit. Populations from the southern edge of the range appear to be particularly well adapted to western Oregon, with vigorous upright growth, strong canes, and some with a low incidence of Verticillium wilt. This germplasm will be of great value to breeders interested in developing improved black raspberry cultivars.
Plant materials and field maintenance
Collection of plant materials has been previously described (Dossett and Finn 2010; Dossett et al. 2012b). Briefly, during the summer of 2006, friends and colleagues living in eastern North America, within the native distribution of R. occidentalis, (USDA NRCS 2015) were solicited to send seed or fruit from wild plants in their area. Additional seed was obtained in 2007 through a similar request and from collecting trips across the southern and western edges of the native range (Hall et al. 2009; Hummer et al. 2008a, b). Through these efforts, seeds were obtained from more than 150 locations across the range, including 27 states and two Canadian provinces. Additional seed was obtained from all R. occidentalis seed accessions heldat the National Clonal Germplasm Repository (NCGR) in Corvallis, OR. Seeds were treated to promote germination as described by Dossett and Finn (2010).
Sufficient seed germination was obtained from 109 wild populations for establishment of replicated field plantings. Due to differences in timing of seed acquisition, seedlings of 78 populations were planted in the field in September 2007, while 31 additional populations were planted in the field in May 2008 (Table 1). On each planting date, seedlings were planted in a randomized complete block design with four replications of at least three and no more than four plants, for a total of 12–16 plants from each popula-tion, depending on the total number of seedlings available. Four-plant plots of the cultivars Jewel, Mac Black, and Munger were also included in each replicate block for comparison. Extra seedlings and seedlings from seed lots that failed to produce at least 12 plants were planted as border rows of the field. Because of the difference in planting dates, popula-tions planted in the field in September 2007 and May 2008 were laid out separately with independent randomization and replication, but were directly adjacent to each other in the field.
Plants were planted 0.91 m apart in rows spaced 2.74 m apart at the USDA-ARS North Farm (Corval-lis, OR) and were trained to a three-wire trellis system with a lower wire at 0.50 m and two parallel wires hung 0.15 m apart at 0.91 m. Primocanes were trained between the parallel wires and primocane tips were pruned approximately 5–10 cm above the wires to induce branching in early June, just before fruit ripening. In the fall, primocane branches were pruned near the tips to help prevent them from rooting in the row. Floricanes from the previous fruiting season were removed during each dormant season. In late winter, new floricanes were pruned so that floricane branches were approximately 30 cm in length. Plants were fertilized, irrigated, and chemical weed controls applied per standard practices for commercially grown black raspberries in Oregon.
Collection of data and fruit samples
Evaluations of phenology and plant performance were performed in a similar manner as Dossett et al. (2008). In 2009 and 2010, dates of first bloom (first fully open flower) and fruit ripening (first fully colored fruit) were recorded for each plant. In addition, each plant was rated on a 1–9 scale for primocane vigor (1 = very poor vigor, 9 = extremely vigorous) in the spring of both years. Symptoms of Verticillium wilt (Verticillium dahliae Kleb.) were recorded in early fall 2008, after the first season of establishment, and again in the fall of 2009. Severity of symptoms was scored on a 0–6 scale (0 = no infection, 6 = all primocanes showing extensive discoloration and stunting). In the spring of 2009 and 2010, plants were scored on a 0–5 scale for the amount of cane death over the winter (0 = no floricane death or injury to cane tips, 5 = 80–100 % of floricanes dead). Samples of 25 randomly picked fruit were collected from each plant for weighing and evaluation. Fruit were scored for gloss on a 1–5 scale (1 = most pubescent, 5 = most glossy). In a few cases, 25 ripe fruit were not available from a given plant, so as many ripe fruit as could be collected were picked and weighed. Fruit were picked when they were fully colored and separated readily from the receptacle but before they were overripe. In 2009, plants were also rated on a 1–5 scale for primocane stiffness (1 = least stiff, 5 = most stiff) and lateral branching angle (1 = *75 –90 ,5= B15 ). Other morphological features were recorded in the field as they were noted. All traits were recorded in 2009 and 2010 for the seedlings planted in 2007. Due to the immaturity of plants established in 2008, observations of fruiting phenology, fruit characteristics, and winter floricane injury were recorded for the seedlings during 2010 only. A single seedling from each of ORUS 3815, ORUS 3827 and ORUS 4108 had morphological traits (e.g. canes with dense spines, poorly set reddish/purplish fruit, differences in leaflet shape) indicating they were probably the result of natural hybridization with wild red raspberry (R. strigosus Michx.). Data from these individuals were excluded from analysis. A few off-types of ‘Munger’ were noted, which ripened fruit approximately 7–10 days later than expected for this cultivar and which had a slightly different flower cluster shape (Dossett et al. 2012a). Data from these individuals were also excluded from analysis.
Statistics
Following visual inspection of residual plots and Levene’s test to confirm assumptions of normality and homoscedasticity, the GLM and CORR procedures in random mix from these seed lots
SAS (version 9.1; SAS Institute, Cary, NC) were used for analysis of variance for all traits as well as correlations (Pearson) between ratings of Verticillium wilt symptoms, winter cane injury, vigor, branching angle, and cane stiffness.