plant material.

Bare root liners of seven pennisetums (Tift 5, Tift 6, Tift 10, Tift 11, Tift 13, Tift 15, and Tift 26) were received on 22 June 2010 and 3 Mar. 2011, respectively, for the 2 years of the study. The selections were the result of a breeding program that used germplasm with cytoplasmic male sterility, short day (<10.5 h) flowering selection, and complex pedigree that resulted in highly sterile plants. These selections were chosen for the study because they exhibited superior performance in previous evaluations. There was also a range of habits, from nearly prostrate to upright, and foliage colors. Liners were potted the day they were received into 0.7-gal containers filled with bark-based soilless substrate composed of 7 douglas fir (Pseudotsuga menziesii) bark : 2 peat : 1 pumice (by volume). Potted containers were top-dressed with 16 g/pot of 18N–2.6P–9.9K controlled-release fertilizer (Apex; Simplot Professional Products, Lathrop, CA). Containers were maintained in a glasshouse with day/night set temperatures of 20/15 °C. In 2010, plants were potted (for the container evaluation) or planted (for field evaluation) directly after being moved from the glasshouse as described below. In 2011, plants were moved on 4 April to an outdoor growing facility in Corvallis, OR, under microirrigation using spray stakes (Netafim USA, Fresno, CA) with a flow rate of 0.2 L·min⁻¹ to maintain a 0.4 leaching fraction (140% of container capacity applied at each irrigation event). On 15 July 2010 and 19 May 2011 three replicates of the seven selections were either potted for container evaluation at the Corvallis, OR, location or moved to the Aurora, OR, location for field evaluation. At this time the plants had a mean growth index of 14.4 cm in 2010 and 36.0 cm in 2011.

Growing conditions.

Plants were evaluated under container and field conditions. Container performance was evaluated in Corvallis, OR [U.S. Department of Agriculture (USDA) zone 8b (USDA, 2012)]. Plants were potted into 6.5-gal (#10; Nursery Supplies, McMinnville, OR) containers filled with the bark-based soilless substrate described above and top-dressed with 140 g/pot of 18N–2.6P–9.9K controlled-release fertilizer (Apex). Plants were microirrigated using spray stakes (Netafim USA) as previously described. In 2010, there was a second container site in Corvallis, OR, under overhead irrigation that included three replications of the seven selections as well as the napiergrass cultivar Princess.

Field performance was assessed at the North Willamette Research and Extension Center in Aurora, OR [USDA zone 8b (USDA, 2012)]. The native soil was 53B-Latourell loam (Fine-loamy, mixed, superactive, mesic Ultic Haploxeralfs) with ≈15% clay, pH 5.0 to 6.5, 10 meq/100 g soil cation exchange capacity, 0.5% to 2.0% organic matter, and 16% available water holding capacity (USDA, 1985). The soil was well drained and positioned on terraces with 0% to 3% slope. Mean soil temperature was 12.2 to 13.3 °C measured using an onsite certified weather station (Aurora, Oregon AgriMet Weather Station). Plants were placed on 3-m centers in 1 × 1-m plots that were maintained weed free by hand weeding, mulched with 2-inch douglas fir bark, and overhead irrigated as needed.

Measurements and ratings.

Final size measurements and ratings were collected 22 Oct. 2010 and 27 Oct. 2011. Size index was calculated as (height × width 1 × width 2)^1/3; width 1 and width 2 were measured perpendicular to each other with one being at the widest point. Growth form rating was performed by having five individuals rate each plant on a 1 to 5 scale, where 1 = very poor growth form and 5 = excellent growth form. Individuals were allowed to use personal preference with regard to growth form; however, general guidelines were provided to base the rating on uniformity of growth and overall aesthetic impression. Individuals often regarded more upright selections with more tillers and finer texture as superior. Anecdotal information was collected on texture and habit of selections. Similarly, color rating was performed by five individuals rating each plant on color, where 1 = green and 5 = deep purple or red. The five ratings for growth form and color were averaged for each plant, such that for statistical analysis there was a single value for each plant. In 2011, when final measurements were made, the predominant leaf color of each selection was assigned to a color patch of the RHS Colour Chart (Royal Horticultural Society, 2007). Plants remained in situ overwinter without protection to assess survival. Leaves were not removed and therefore, may have provided minimal protection to the crown. To assess survival of the 2010 and 2011, we collected plantings survival ratings 1 May 2011 and 2012. Presence or absence of leaf spot (Helminthosporium sp.) was recorded at each location, but severity was not assessed. We also made anecdotal observations on leaf width, plant texture, and plant habit.

Experimental design and data analysis.

The experimental design at both locations was completely randomized with three replications for each of the seven hybrids. Data for each location were collected and analyzed separately. Size index, growth form rating, and color rating were subjected to analysis of variance using GLM procedures within SAS (version 9.2; SAS Institute, Cary, NC). When appropriate, interactions were analyzed using simple effects. When significant differences were observed means of main effects or simple effects were separated using Fisher’s least significant difference at a 0.05 level of probability.

Container performance.

Significant main effects and interactions for SI, growth form rating, and color rating for the container study were observed (Table 1). The significant year by selection interaction for SI was primarily the result of inconsistent performance of Tift 6 and Tift 10 from 2010 to 2011. Inconsistent rankings for Tift 10 and Tift 15 between years also made the combined analysis of growth form and color rating inappropriate. Therefore, data were analyzed and presented separately by year (Table 2).

Table 1.

Analysis of variance table for size index, color rating, and rating of growth and form of hybrid pennisetum selections (S) produced over 2 years (Y) in containers or in a field in Corvallis, OR, and Aurora, OR, respectively.

View Table

Table 2.

Comparison of seven selections of pennisetum hybrids grown in 6.5-gal (24.61-L) containers in Corvallis, OR, for size index, growth form rating, and color rating.

View Table

Significant differences were observed among selections for color rating in both 2010 and 2011, as well as growth form rating in 2010 (Table 2). In 2010, Tift 6, Tift 11, Tift 13, and Tift 15 received the highest mean growth form ratings. There were no differences among selections for growth form rating in 2011, but all selections had improved ratings from 2010 except Tift 15, which had a slight reduction in growth form rating, from 3.2 to 2.9. Significant differences were observed for color rating among selections in 2010 and 2011 (Table 2). Tift 5, Tift 10, and Tift 26 were among the four highest rated selections for color in both years; these selections received consistently high color ratings that corresponded to patch 187A or 187B in the grayed-purple group of the RHS Colour Chart.

Field performance.

In the field study, the main effect of year was significant for SI and color rating, but the interactions between year and selection were not significant (Table 1). As these interactions were insignificant, we pooled SI and color rating across years (Table 3). A significant year by selection interaction for growth form was observed.

Table 3.

Comparison of seven selections of pennisetum hybrids field grown in Aurora, OR, for size index, growth form rating, and color rating pooled across years (2010 + 2011) or by year when statistically appropriate.

View Table

Differences existed among selections for SI with Tift 5, Tift 10, Tift 11, and Tift 26 being the largest (Table 3). In 2010, Tift 6, Tift 11, and Tift 13 received high growth form ratings, but in 2011, Tift 5, Tift 10, Tift 11, Tift 15, and Tift 26 were highly rated and not statistically different from each other. Tift 10 received the highest color rating in the 2 years combined, followed by Tift 5, Tift 11, Tift 15, and Tift 26, which were not different from each other (Table 3). These selections with high color ratings all corresponded either to patch 185B, 187A, or 187B in the grayed-purple group of the RHS Colour Chart.

No leaf spot was observed at the Corvallis container site using spray stakes or the overhead irrigated field site in Aurora. At the Corvallis site that received overhead irrigation in 2010, leaf spot was observed on all three plants of ‘Princess’ and at least one plant of all selections being evaluated except Tift 11 (data not presented). Even though we did not conduct ratings to evaluate severity, it was clear that the seven selections being evaluated were more resistant to leaf spot than ‘Princess’. Similarly, Hanna et al. (2010) observed increased resistance over ‘Princess’ in ‘Tift 17’ and ‘Tift 23’, which share a similar complex, three-species pedigree to the selections included in the current study.

Anecdotally, we observed variation in leaf width and overall plant texture (data not shown). Selections such as Tift 5, Tift 10, Tift 15, and Tift 26 appeared to have wider leaves and coarser texture, similar to the napiergrass cultivars Prince and Princess (Hanna and Ruter, 2005), while Tift 6, Tift 11, and Tift 13 appeared to have thinner leaves and finer texture more like ‘Tift 23’ (Hanna et al., 2010).

In 2011, plants were planted 56 d earlier than in 2010. This likely accounted for the difference between years observed in SI in the field study. Mean SI for all selections was 101.2 cm in 2010 and 151.0 cm in 2011, which was a 67% increase. There was a slight increase in SI in 2011 over 2010 in the container study but less than in the field site. In 2010, the mean SI was 103.6 cm and in 2011, it was 111.8 cm. It is possible that increased temperatures from growing in black plastic containers on black ground cloth accounted for the lack of marked difference in SI between years in the containerized plants.

There were differences in color between field and container evaluations. Selections were not consistently ranked between locations or between years. However, Tift 5, Tift 10, and Tift 26 were among the highest rated selections for color in both years of the container study and in the field study. In addition, Tift 15 was among the four highest rated selections in 2010 of the container study and in the field study. Selections also exhibited different habits in containers compared with the field. All selections were slightly more upright in containers. This was particularly noticeable in Tift 5, Tift 11, and Tift 26, which were lower-growing and spreading in the field evaluation but upright-spreading in containers.

Ornamental Pennisetum selections with similar parentage are reported to be hardy in USDA hardiness zone 8 (Hanna and Ruter, 2005; Hanna et al., 2010), but none of the selections survived winter at any of the evaluation sites during either year. In Corvallis, plants were exposed to 37 d with temperatures below 0 °C and a low of −7 °C in 2010 to 2011, and 40 d with temperatures below 0 °C and a low of −5 °C in 2011 to 2012. In Aurora, plants were exposed to 34 d with temperatures below 0 °C and a low of −8 °C in 2010 to 2011, and 44 d with temperatures below 0 °C and a low of −5 °C in 2011 to 2012. It is unclear why these selections were killed at temperatures that they have survived in previous field evaluations in Tifton, GA, during exposure to similar low temperatures. It is possible that overall cooler temperatures at the current locations or increased rainfall during winter led to mortality. Our findings indicate that in the Willamette Valley of Oregon, these selections will be annuals or potentially tender perennials if protected. However, their vigor, relative leaf spot resistance, and attractive foliage make several of them viable options for landscapes of the Pacific northwestern United States.

Units

View Table