Abstract
Ulmus americana (American elm) was an important urban tree in North America prior to the introduction of the Dutch elm disease pathogen in 1930. Subsequently, urban and community forests were devastated by the loss of large canopies. Tree improvement programs produced disease tolerant American and Eurasian elm cultivars and introduced them into the nursery industry. However, consumer acceptance was slow. The National Elm Trial was established to evaluate commercially available taxa of elm across the United States. Established at 16 locations, these plantings monitored survival and growth, as well as disease and insect pressure for 7 to 10 years. ‘Morton’ elm had >90% survival, while 13 cultivars averaged 70% to 90%, and five cultivars ranged from 25% to 69% survival. Trunk diameter growth by location ranged from 0.5 cm/year (Colorado, U.S.) to more than 2.0 cm/year (Iowa, U.S.). By taxa, trunk diameter growth ranged from a low, by ‘JFS Bieberich’ elm (0.7 cm/year), to a high by ‘New Horizon’ elm (1.7 cm/year). Scale insects were minor issues at most trial locations, except Colorado, where scales contributed to the death of several cultivars. Performance ratings (scale 1 to 5) ranged from 2.7 for ‘JFS Bieberich’ elm to 4.5 for ‘New Horizon’ elm. Based on the ratings, the preferred cultivars of American elm were ‘New Harmony’ and ‘Princeton’, and the preferred cultivars of Asian elm were The Morton Arboretum introductions and ‘New Horizon’. These findings will help green-industry professionals determine what elm cultivars will perform the best in different regions.
- Chalkbark Elm
- Japanese Elm
- Lacebark Elm
- Scotch Elm
- Siberian Elm
- Smoothleaf Elm
- Tree Evaluation
- Ulmus carpinifolia
- Ulmus glabra
- Ulmus japonica
- Ulmus parvifolia
- Ulmus propinqua
- Ulmus pumila
- Ulmus wilsoniana
- Urban Forestry
- Wilson Elm
American elm is commonly found in soils that are saturated in spring and autumn; however, it also grows well in deep soils with good drainage. The root system is generally considered shallow but will develop deeper in dry sites with good soil drainage (Harlow et al. 1991). The tall trees with broad, arching branches were a favorite of landscape developers and were once the predominant landscape and street tree across the United States (Gerhold et al. 1993; Plotnik 2000). American elm can tolerate many urban conditions, including soil compaction, flooding, air pollution, and deicing salts (Townsend 2000). Additionally, the species is easy to propagate, grow, and transplant.
In the early 1930s, the introduction of Dutch elm disease (DED) forced the green industry to abandon American elm for more disease-resistant exotic species and encouraged plant breeders to search for or develop disease-resistant elms. The result of many decades of intentional and opportunistic selection has been a rather broad selection of Asian, American, and European species and hybrid elms with various degrees of resistance to DED (Santamour and Bentz 1995; The Morton Arboretum 2015).
Diseases and Insects of Elms
The common diseases of elm (Ulmus) in North America include DED, elm yellows (elm phloem necrosis) and bacterial leaf scorch, bacterial wet wood, and various root, canker, and foliar diseases (Stipes and Campana 1981; Sinclair and Lyon 2005). DED is a vascular wilt disease incited by the exotic fungi Ophiostoma ulmi and O. novo-ulmi. These pathogens move among trees via elm bark beetle vectors and by root grafts (Lanier 1981; Peacock 1981; Webber and Gibbs 1989; Vega and Blackwell 2005). Elm phloem necrosis is widely distributed throughout the eastern portion of the United States (Sinclair and Lyon 2005; Martin 2012). The disease is caused by phytoplasmas (usually ‘Candidatus Phytoplasma ulmi’) that are vectored by leafhoppers and spittlebugs (Karnosky 1982; Riffle and Peterson 1986). Elm phloem necrosis kills phloem tissue in roots and stems, causing death one to two years after symptoms appear (Sinclair and Lyon 2005; Martin 2012). Visual symptoms are similar to DED and the disease may be misidentified by casual inspection. Currently, all North American native elm species are affected, and there are no known preventative or curative treatments (Riffle and Peterson 1986; Martin 2012). Bacterial leaf scorch (Xylella fastidiosa) is predominately found in eastern North America, but it is not known as a limiting factor in the use of elm.
Root decay diseases of elm include Armillaria, Ganoderma, Inonotus, and Laetiporus decays that are occasional issues in the eastern half of North America (Sinclair and Lyon 2005). Cankers are common on Asian varieties in environmentally stressful sites, including Cytospora (Cytospora spp.), black spot Nectria (Nectria nigrescens, anamorph Tubercularia ulmea), and Siberian elm/Botryodiplodia canker (Botryodiplodia hypodermia). Black leaf spot (Stegophora ulmea, formerly Gnomonia ulmea) is the most common leaf spot on American elm cultivars in moist areas of North America. Black leaf spot symptoms include small, black, slightly raised, spots (Sinclair and Lyon 2005).
There are several insect pests that can cause damage to elms (Dirr 2009), and most of the major damaging insects were introduced to North America. Insect pests include various defoliators, leafminers, twig girdlers, bark beetles, borers, and numerous scales and aphids that produce honeydew and resulting sooty mold (Condra et al. 2010; Potter and Redmond 2013). The most common defoliators include Japanese beetle (Popillia japonica), gypsy moth (Lymantria dispar), elm leaf beetle (Xanthogaleruca luteola), and European elm flea weevil (Orchestes alni), all of which can cause unsightly damage to the crown (Condra et al. 2010; Potter and Redmond 2013). Defoliators are generally considered a nuisance, since their populations fluctuate, and even though they can cause enough damage to make elm crowns aesthetically unappealing, they do not normally cause enough damage to kill portions of a tree. However, they can be lethal when combined with other plant stressors. Breeding efforts have been moderately successful incorporating elm leaf beetle feeding resistance into some hybrids (Townsend 2000). European elm scale (Ericoccus spuria) is an introduced and very damaging scale insect that causes branch dieback and can cause great concern from the abundant honeydew production and resultant sooty mold growing on plant and hard surfaces below infested trees.
Breeding and Selection for Resistance to Dutch Elm Disease
Breeding and screening efforts for DED-resistant elms began in the 1930s shortly after the disease was introduced (Townsend and Douglass 2001; Mittempergher and Santini 2004). However, it was the 1970s before efforts at various institutions made significant progress (Townsend 2000). Major elm improvement programs were conducted in Wisconsin, Illinois, Ohio, New York, Washington, D.C., and Maryland, U.S. (Smalley and Guries 1993; Mittempergher and Santini 2004; Townsend et al. 2005). Initial breeding efforts and cultivar releases were based on European and Asian species of elm, because of their inherent resistance to DED (Townsend and Schreiber 1975; Smalley and Guries 1993; Townsend 2000; Townsend and Douglass 2004). Those efforts resulted in the release of several cultivars of elm: ‘Urban’, ‘Dynasty’, ‘Homestead’, ‘Pioneer’, ‘Frontier’, ‘Prospector’, ‘Ohio’, ‘Pathfinder’, and ‘Patriot’ (Townsend 2000).
The development of DED-resistant American elm was slower due to the high mortality rate of seedlings when inoculated with the disease (Smalley and Guries 1993; Mittempergher and Santini 2004). For example, the initial USDA screening of more than 35,000 American elm seedlings for DED tolerance yielded only two trees worthy of further testing (Townsend 2000; Mittempergher and Santini 2004). One of those was subsequently named ‘Delaware’, and was used to successfully incorporate some level of DED tolerance to progeny (Townsend et al. 2005). Townsend and Schreiber (1975) noted good segregation among progeny for disease resistance and growth characteristics. After screening numerous seedlings and survivor/escape trees in areas with high DED pressure, American elms with varying levels of disease resistance began to surface. Trees with potential resistance were propagated, field planted, and inoculated with the causal fungus of DED (Townsend et al. 1995; Townsend and Douglass 2001; Townsend et al. 2005). Symptoms of DED and crown dieback were recorded over time. The extent of crown dieback was compared to known susceptible American elm seedlings and known resistant Asian hybrids.
Two early selections of American elm (‘Valley Forge’ and ‘New Harmony’) are grown commonly at the time of this writing. ‘Princeton’ American elm was selected by Princeton Nurseries in 1922 (prior to the introduction of DED) for its narrow growth habit and rapid growth rate (Townsend et al. 2005). Time has shown it to be DED-tolerant and a popular landscape tree. It should be noted that ‘Valley Forge’, ‘Princeton’, and ‘New Harmony’ selections have shown excellent tolerance to DED over the years but not complete resistance (Townsend et al. 1995; Townsend and Douglass 2001; Townsend et al. 2005). Even in inoculation tests, the data suggests that ‘Valley Forge’ still shows a very low level of susceptibility. ‘American Liberty’ elm (U. americana) was another early release reported to be resistant to DED. However, inoculation studies later determined it sustained crown dieback similar to American elm seedlings (Townsend et al. 1995; Townsend and Douglass 2001; Costello et al. 2004). By the mid-1990s, American elms with reported DED tolerance were undergoing extensive field testing and were then later released (Townsend et al. 1995). Breeding, screening, and selection of American elms with DED tolerance and improved growth habits continues with a goal of increasing the genetic diversity of DED-tolerant elms (Slavicek and Knight 2011).
National Elm Trial
The National Elm Trial was initiated because, in light of the success of DED resistance in American elms, there remains a lack of awareness or lack in confidence among green-industry professionals and the general public in these trees. The National Elm Trial (CSU 2017) was a volunteer education and outreach effort to evaluate the use of commercially available DED-resistant and DED-tolerant elms. The trial was developed by the NCR-193 Agricultural Experiment Station coordinating committee on insects and diseases of woody ornamentals. American, European, Asian, and hybrid elms were included. Trial sites were established across the United States in a variety of growing conditions and USDA hardiness zones where the general public and green-industry professionals could have easy access to the plantings. Individual evaluation plantings were maintained by local research and extension personnel located at land grant universities, colleges, and research stations (Table 1; Table 2; Figure 1). The trial also benefited from generous industry support and knowledge (Table 3).
Location, climate characteristics, and management of U.S. National Elm Trial sites (2005-2015).
Cooperators and locations of U.S. National Elm Trials.
Location of sixteen National Elm Trial sites (2005–2015) across the continental United States.
Taxa and nursery supplier of elms planted in the U.S. National Elm Trial 2005–2015.
The study objectives were to 1) determine the growth and horticultural performance of commercially available DED-resistant and DED-tolerant elm cultivars in various climate regimes in the United States; 2) determine the relative disease, insect, and abiotic stress tolerance of these cultivars; and 3) relate the results of the trial through local, regional, and national reporting to wholesale tree propagators and growers, retail nursery and garden center operators, landscape designers, arborists, and the general public.
MATERIALS AND METHODS
Fourteen to eighteen taxa of elm were planted in 16 locations in the United States (Table 1; Table 2; Figure 1). The cultivars represented a range of hybrids and species of Ulmus and were chosen based on their commercial availability (Table 3). Some cooperators incorporated other cultivars of elm of local interest into their planting, and the performance data of these taxa are available in the state reports on the National Elm Trial website (CSU 2017).
Trees were produced by commercial nurseries on their own roots (rooted cuttings or from tissue culture) or budded onto seedling U. pumila (Siberian elm) rootstocks (Table 3). Bare root trees, 1.5–1.8 m in height, were shipped by nurseries to cooperator sites for planting beginning spring 2005. Due to inventories, availability, and establishment failure, some selections were planted in subsequent years (Table 2).
Planting was done according to local recommendations and site cooperator’s preference. Irrigation and vegetation management was determined by the local cooperator in accordance with need and common local standards (Table 3). All locations used mowing to manage weed and grass between tree rows, and most locations used chemical weed management for a 0.5 m weed free circle around each tree. All trees in all blocks were treated similarly within a site. Data was collected annually by the site coordinator. This data included stem diameter at 1.37 m above the ground (dbh), tree height, crown width in the row, autumn color, occurrence of insects and diseases, and an overall performance/quality rating. The subjective performance/quality rating (1–5 scale; 1 = poor, 2 = fair, 3 = good, 4 = very good, and 5 = excellent) of the cultivars incorporated survival, growth rate, branching patterns, form, and insect and disease damage. Each site had their own evaluator conduct the performance/quality rating at the conclusion of the project. The experimental design was a randomized complete block design with five blocks, each containing one individual of each cultivar.
All data were collected in a similar manner at all locations; summary statistics were prepared using SAS/STAT® software (Version 9.4, SAS Institute Inc., Cary, North Carolina, U.S.). Proc GLM was used to prepare least square means of diameter, height, and crown width growth, as well as insect, disease, and overall quality ratings (1–5), taking into account state as a fixed effect variable. There was a small but significant (P = 0.001) state-by-cultivar interaction that was ignored in an effort to compare cultivar means over all states. When diameter growth was presented by state, cultivar was the fixed effect. Growth rates per year were calculated from the difference between final data collection and initial data collected at planting because not all cooperators collected data annually. Data from West Virginia (WV) was excluded from the analysis because cultivar growth at that location was inconsistent with the remaining sites. Also excluded were data from cultivars planted at only one or two sites [U. × ‘Cathedral’, U. parvifolia ‘Dynasty’ (lacebark elm), U. × ‘Regal’, U. japonica ‘Discovery’ (Japanese elm), U. americana ‘Kuhar II’, and U. rubra (red elm)]. Three cultivars of U. parvifolia (‘BSNUPF’, ‘Emer I’ and ‘Emer II’) were not sufficiently represented (not planted or failed to survive) to provide data in the analysis to produce least square means displayed in Figure 2 through Figure 8. Cultivar performance at each state location is available at the National Elm Trial website (CSU 2017).
RESULTS AND DISCUSSION
Overall Trial Success
Elm cultivars were planted over a wide geographic range (Figure 1). Trial locations had a variety of soils, the average annual precipitation ranged from ≈0.4 m at Colorado (CO) to ≈1.3 at Alabama (AL), and USDA Plant Hardiness Zones (USDA 2012) ranged from Zone 4a in North Dakota (ND) to Zone 9b in California (CA) (Table 1). The American Horticultural Society Heat Zone map (AHS 2017) was also referenced, since summer heat plays a significant role in plant growth, as well as humidity and irrigation requirements. Trial site Heat Zones ranged from Zone 3 in Washington (WA) to Zone 8 in AL, CA, and Kansas (KS). Due to individual site environmental characteristics, cultural practices were at the discretion of the site coordinator. For example, some sites were irrigated for two years to establishment while other were irrigated to prevent excessive summer desiccation. Year-to-year weather variation also influenced cultural practices. In KS, for instance, trees were irrigated bi-weekly during the historic heat wave and droughts of summers 2011 and 2012 to prevent death. However, trees were not irrigated after 2012 and were irrigated only twice per summer prior to 2011. Nearly all sites chose to mow the vegetation within the trial plot, and used herbicides to keep the area surrounding the tree trunk weed free. At most locations, trees were allowed to grow without major structural pruning. Minor pruning was needed to lift the crowns for maintenance and to restore a central leader, if needed. In a typical urban or landscape environment, however, structural pruning will be required on many elms. Co-dominant leaders, narrow branch angles, and excessive shoot growth are negative attributes common among some cultivars of elm that need to be corrected with pruning.
Data were obtained from all sites with the exception of CA, which terminated the study early. Therefore, data from CA were omitted from tables and figures. Results from the first two years at CA can be found in an earlier report (McPherson et al. 2009). The growing conditions at each site are not necessarily representative of growing conditions throughout the region or every region throughout the United States. Therefore, while elm performance at a particular site is useful and informative, it does not necessarily represent how the cultivars will perform in all sites in that region. For example, the extremely slow trunk diameter growth of elms in the CO study does not indicate an overall lack of elm adaptability in other parts of CO. Instead, it indicates elm growth was influenced by the compacted clay soils coupled with irrigation challenges (Figure 2). Additionally, the poor overall survival of trees in WV does not suggest elms should not be planted in that region. Most tree deaths in the WV study occurred from deer browsing prior to the installation of fencing. Additionally, not all states or regions of the United States were represented in the trial. Readers are encouraged to use this information and consult with local experts when contemplating planting elm trees.
Average trunk diameter growth (dbh) per year (cm), with standard-error bars, of all elm cultivars by state, U.S. National Elm Trial 2005–2015. Least square means were adjusted to account for cultivar variation.
Cultivar Survival
Most cultivars grew and survived under the varied growing conditions of the 16 trial sites (Table 4). Only one cultivar, ‘Morton’ elm, had an average survival >90%, with a survival rate of 100% at 13 trial sites. Only Iowa (IA), ND, and WV reported two dead ‘Morton’ elms at each site. These three locations also had the three lowest overall mean survival rates across all cultivars (51%, 56%, and 45%, respectively), suggesting environment, culture, or a combination were unfavorable for the elms. Seven cultivars, including two American elms (‘New Harmony’ and ‘Princeton’) and five hybrid Asian elm selections (‘Morton Stalwart’, ‘Pioneer’, ‘Homestead’, ‘Patriot’, and ‘Morton Glossy’) averaged 80%–89% survival. Six Asian hybrids or species (‘Frontier’, ‘Morton Red Tip’, ‘JFS Bieberich’, ‘Morton Plainsman’, ‘New Horizon’, and ‘Prospector’) had 70%–79% survival. Two American elms (‘Lewis & Clark’ and ‘Valley Forge’) were in the 60%–69% survival group. Both of these cultivars were produced on their own roots from rooted cuttings. Liners of American elms produced from cuttings often have tall vigorous shoots with disproportionally small root systems. These cultivars were no exception at planting (personal observation). This root/shoot imbalance may have resulted in some water deficit stress for these cultivars, thereby lowering their overall survival. The remaining three cultivars in the lowest survival group of 25%–59% were selections of lacebark elm. These cultivars were not planted at many sites, and most were killed by freeze damage. However, the lower survival rate at milder sites could simply be a result of fewer plants or poor performance at the limited number of sites.
Survival (%) of elm cultivars planted in 16 U.S. National Elm Trial locations (2005–2015).
Of the four American elm cultivars, ‘New Harmony’ and ‘Princeton’ had the greatest survival rate at 85.5% and 81.5%, respectively. ‘Lewis & Clark’ and ‘Valley Forge’ had a lower survival rate at 63.6% and 66.7% survival, respectively. The ‘Lewis & Clark’ elm is relatively new (as of this writing) and less known by the nursery and landscape industries. Due to difficulty in sourcing plant material, this cultivar was planted in only 11 of the 16 locations. Limited trial sites and poor performance at three sites may have unfairly influenced the survival rating. The ‘Valley Forge’ elm is very well-known and is considered a fast grower with exceptional DED resistance. Overall survival is not considered a weakness of this cultivar. Again, poor performance at five of the fifteen sites may have biased the data against this tree. Lacebark elm is a staple of the nursery industry in the southern half of the United States. Cold hardiness has been an issue in recommending this tree in northern climates. The poor performance of lacebark elm, particularly in sites with cold winters, was expected. However, 100% survival of ‘Emer I’ elm in Michigan (MI) was not expected. Similarly, 80% survival of ‘BSNUPF’ elm in Vermont (VT) is interesting.
Cultivar Growth
Average trunk diameter growth (dbh) of all cultivars combined per site varied from less than 0.5 cm/year at CO to more than 2.0 cm/year at IA (Figure 2). Trees at CO and WV had such poor growth that those sites would not provide reasonable information on cultivar performance in regards to growth, shape, or form (Figure 2). Trees at CO did not grow well because of extreme soil compaction and related infiltration inhibition during irrigation issues, but the relative growth was similar to other states. Trees at WV were plagued with persistent deer browsing damage, resulting in resprouting from below grafts. Thus, data from WV were not included in any analysis or figure comparing cultivar growth.
Increased dbh and height are good indicators of root growth and overall suitability for specific growing conditions. Overall cultivar dbh performance varied with some cultivars, like the Asian selection ‘JFS Bieberich’ elm growing only 41% (0.7 cm/year dbh) of the growth of New Horizon (1.7 cm/year dbh) (Figure 3). American cultivars performed similarly to Asian hybrids (Figure 3). Height growth by cultivars generally paralleled diameter growth as most of the tallest plants also had the largest diameter growth. Substantial departures were ‘New Harmony’ and ‘Lewis & Clark’ elms, which had relatively small dbh for their height, and ‘Prospector’ and ‘Morton Stalwart’ elms, which had relatively large dbh for their height. The fastest height growth was found with the American elm cultivars ‘New Harmony’ and ‘Princeton’ (0.63 and 0.61 m/year, respectively) (Figure 4). Crown width growth rates were taken at nine of the sixteen locations, and the variation among cultivars ranged from 0.2 to 0.5 m/year (Figure 5). Most Asian and European cultivars had greater crown width growth rates than the American elm cultivars, except for ‘Valley Forge’, which was similar to the widest Asian cultivar, ‘Morton Plainsman’. This result was not surprising considering ‘Princeton’ and ‘New Harmony’ American elms are known for their rather narrow growth habit. Conversely, ‘Valley Forge’ American elm is known for its broad spreading growth as a young tree. ‘Frontier’ and ‘JFS Bieberich’ elms had the least crown width growth per year. They also had the least height growth per year (Figure 4). These data suggest that in this trial, these two cultivars either lacked vigor or were slower-growing trees. This was somewhat surprising since the authors have witnessed individual trees of each cultivar growing vigorously and healthily on several independent occasions. ‘New Harmony’ American elm had a similar crown width growth rate but had the greatest height growth rate, suggesting it has good vigor but a narrow growth form.
Elm cultivar trunk diameter growth (dbh) per year (cm) with standard-error bars, U.S. National Elm Trial 2005–2015. Least square means were adjusted to account for variation across sites (states).
Elm cultivar height growth per year (m) with standard-error bars, U.S. National Elm Trial 2005–2015. Least square means were adjusted to account for variation across sites (states).
Elm cultivar crown width growth per year (m) with standard-error bars, U.S. National Elm Trial 2005–2015. Least square means were adjusted to account for variation across sites (states).
Cultivar Form and Autumn Color
Most cultivars started growing in oval or vase shapes, but some had irregular shapes that became more defined over the 10 years of the study (Table 5). Some elms produced excessive terminal growth within three years of planting. This growth can negatively impact tree architecture and aesthetics. Pruning of elms during their early growth period can encourage a more pronounced vase or oval crowns and prevent excessively long branch growth. Some cultivars commonly have structural defects that need to be corrected with pruning while trees are young. Consulting a certified arborist or other knowledgeable green-industry professional is recommended prior to selecting a specific cultivar of elm.
Growth form, fall color, and preference ranking of elms in the U.S. National Elm Trial at each trial site.
Autumn color is not a significant ornamental feature of most elms. During this trial, there was usually a period in the autumn season in which most elms produced green/yellow fading to yellow or yellow-brown color. ‘Frontier’ elm is known for producing some purple/red autumn color. This is rare among elms, and is one of the most notable attributes of the ‘Frontier’ elm (Table 5).
Cultivar Insect and Disease Issues
Insect defoliators and scales caused the most obvious damage to trees across trial sites. There were also reports of some damage from foliar fungal diseases, stem cankers, and bacterial wet wood (Figure 6; Figure 7). The common foliar and twig insects included the Japanese beetle (Popillia japonica), European elm flea weevil (Orchestes alni), elm leafminer, European elm scale (Gossyparia spuria), and woolly elm aphid (Eriosoma americanum). At the MI trial site, insect defoliation was the major damage to elms, with Japanese beetle causing 60%–75% of the defoliation, and European elm flea weevil the remainder, depending on the cultivar (see MI report, CSU 2017). No cultivar was considered unacceptable at MI based on insect defoliation, except possibly ‘Homestead’, which sustained 40% defoliation. Cooperators in Kentucky (KY) have previously reported on the amount of damage from insect defoliators on the cultivars (Condra et al. 2010; Potter and Redmond 2013). In KY, some cultivars, such as ‘Morton’, had more than 50% defoliation by Japanese beetle, but minor defoliation by other insects, while ‘Homestead’ and ‘Pioneer’ were impacted by both Japanese beetle and European elm flea weevil. The KY site found the cultivars of Asian species U. parvifolia and U. propinqua to be the most resistant to insect defoliators. However, based on the analysis over all National Elm Trial sites, no cultivar was sufficiently damaged by foliar insects to an extent making it unacceptable for landscape planting. ‘Homestead’ elm cultivar was the most severely damaged by defoliators over all trial sites (Figure 6). Adult Japanese beetles feeding on leaves can be quite severe, depending on year, location, and cultivar. The cultivars least damaged by Japanese beetle feeding, based on observations in New Jersey (NJ), MI, and KY, included American elms ‘Valley Forge’, ‘Princeton’, and ‘New Harmony’; and Asian elms ‘JFS-Bieberich’, ‘Prospector’, ‘Emer I’, ‘Emer II’, ‘BSNUPF’, ‘New Horizon’, and ‘Frontier’ (data not presented). Based on the detailed data available at state locations, cultivar selection should carefully consider the state findings in addition to the national averages presented here.
Foliar insect damage rating, with standard-error bars, of elm cultivars, U.S. National Elm Trial 2005–2015 (1 = unacceptable; 5 = excellent). Least square means were adjusted to account for variation across sites (states).
Scale insect rating, with standard error bars, of elm cultivars, U.S. National Elm Trial 2005–2015 (1 = unacceptable; 5 = excellent). Least square means were adjusted to account for variation across sites (states).
Scale insects were a minor issue in most trial locations (avg. rating > 4.2). However, scale populations were higher in CO and KY (avg. rating 3.0). In CO, scale insects, primarily European elm scale, caused complete loss of all ‘Lewis and Clark’ American elm and more than 50% loss of ‘Princeton’ American elm. Across all locations, scale insects were most damaging on ‘New Harmony’, ‘Morton’, ‘New Horizon’, and ‘Morton Glossy’ (Figure 7). In CO, where European elm scale is currently the limiting factor of elms in the urban landscape, ‘New Harmony’, an American elm, and ‘Morton’, ‘Accolade’, ‘Morton Stalwart’, ‘Morton Glossy’, ‘Morton Red Tip’, ‘Morton Plainsman’, ‘Homestead’, and ‘Prospector’ were the most resistant to the insect.
Diseases noted at trial locations included black leaf spot or elm anthracnose (Gnomonia ulmea) at MN, IA, NJ, and NY, as well as bacterial wet wood, DED, and occasional unidentified cankers. DED was noted by the authors on one tree of ‘Valley Forge’ American elm in New York. The disease was not laboratory confirmed, but identified by the experienced eye of the authors. Although ‘Valley Forge’ is highly resistant to DED, the tree had been compromised due to significant storm damage that resulted in heavy bark beetle infestation. Abiotic damages that included freeze damage, drought coupled with freeze damage, and wind breakage were noted at several locations. These abiotic stressors likely accounted for most of the mortality across the trial sites. No elm phloem necrosis were reported at any of the trial sites.
Individual Cultivar Performance Summary
The average performance rating of the 16 well-represented cultivars, across all states, ranged from 2.7 for ‘JFS Bieberich’ elm to 4.6 for ‘Morton’ and ‘New Horizon’ elm (Figure 8). Interestingly, ‘JFS Bieberich’ and ‘New Horizon’ elm had similar survival rates (Table 4) but were on opposite ends of the data for diameter growth (Figure 3), height growth rate (Figure 4), and canopy width (Figure 5). Data collectors may have favored the rapid growth rate of ‘New Horizon’ elm. Fifty-six percent of the cultivars performed well and were rated very good (>4), whereas 31% were rated good (3–4) and 13% were rated fair to poor (<2) (Figure 8). Cultivars ranking with an average rating of 3 or less should be subject to careful scrutiny as a landscape tree, depending on how they performed within the region. It should be noted that each site had their own evaluator. Therefore, some portion of the variability in cultivar performance is likely due to individual evaluator preference or bias, as well as tree cultivar. Based on the ratings, there was not a significant difference among these best performers, including the American elm cultivars ‘New Harmony’ and ‘Princeton’, and the Asian elm cultivars, including all of The Morton Arboretum’s introductions and ‘New Horizon’. Lacebark elm cultivars ‘BSNUPF’, ‘Emer I’, and ‘Emer II’ had poor survival and performance rankings, most likely due to lack of cold hardiness. Based on these observations, lacebark elms would not be recommended in northern locations until more cold-hardy selections are found.
Overall cultivar performance rating, with standard-error bars, of elm cultivars (1 = unacceptable; 5 = excellent), U.S. National Elm Trial 2005–2015. Least square means were adjusted to account for variation across sites (states).
Based on the favorite choices by cooperators, the American elm cultivars ‘Princeton’ and ‘New Harmony’ were the best choices, and of the Asian elm cultivars ‘Morton Glossy’, ‘Morton Stalwart’, ‘Morton’, ‘Morton Red Tip’, and ‘Patriot’ were the most common favorites (Table 5). Interestingly, all selections from The Morton Arboretum have members of the David elm (U. davidiana) complex (U. japonica, U. wilsoniana, and U. propinqua) within their parentage. Based on performance and evaluator preference, this data suggests elm breeding programs should incorporate David elm into their work. It is also worth noting that although decades of efforts to find DED-resistant American elms have been successful, the Asian hybrid elms seem to be favored over the American elms. The authors note that the more restrained growth and the darker green leaves of the Asian hybrids may lead to a more desirable ornamental appearance.
CONCLUSIONS
This study determined, as expected, that growth and horticultural performance of commercially available DED-resistant elm cultivars was different across the various growing conditions and climate regimes of the United States. The variations in soil condition, moisture, temperature, and maintenance precludes any speculation on performance by region of the United States. However, all locations had adequate tree growth to sufficiently evaluate their characteristics and potential landscape uses. Careful selection of elms based on hardiness zones, growth characteristics, and insect resistance/tolerance is critical for future urban tree plantings. Selecting cultivars on overall performance should be tempered with tolerance to insect and disease pressure.
The overall survival data present the best picture of each cultivar’s ability to flourish under various conditions. Ranking of disease and insect damage gives a general overview of how the cultivars performed under pest pressure. Since local disease and insect pressure were different at each location, arborists should consult the National Elm Trial website for local performance and resistance data. Unfortunately, at most locations, the actual cause of mortality was undetermined.
Acknowledgments
We appreciate the technical assistance by Keith Warren and Guy Meacham, J. Frank Schmidt Nurseries; statistical assistance by J. zumBrunnen, Franklin A. Graybill Statistical Laboratory, Colorado State University; and Figure 1 by Darci Paull, Kansas Forest Service. We acknowledge the suggested improvements to the manuscript by anonymous reviewers. Trees were donated by the J. Frank Schmidt and Son Co. nursery and other nurseries, and funding to produce the website and complete the study was provided by the J. Frank Schmidt Family Charitable Foundation and the Colorado State University, Agricultural Experiment Station. We greatly appreciate all the work to plant, maintain, and collect data by volunteer cooperators, student volunteers, local arborists, and university staff around the country.
- © 2017, International Society of Arboriculture. All rights reserved.