Abstract
We evaluated several horticultural cultivars and species of Monarda, a genus native to North America with a center of diversity in the Southeast and advertised as beneficial to wildlife, to assess landscape performance with respect to vegetative habit, flower production, and disease tolerance in Georgia Piedmont and montane habitats. We established two experimental sites: the State Botanical Garden of Georgia in Athens (USDA Zone 8b) and the Georgia Mountain Education and Research Center in Blairsville (USDA Zone 7b). We then tracked plant performance over 2 years after establishment. Our study included 10 samples of Monarda, representing five cultivars and four species. Estimated height and width at flowering showed M. bradburiana, M. Sugar Buzz® Grape Gumball, and the M. punctata ecotypes were smaller than other tested taxa. M. fistulosa had the most flowers at the Blairsville site and equal flowering with M. punctata at the Athens location, but most taxa flowered for 2 to 3 months with ∼100 flowering stems per stand. All samples were susceptible to powdery mildew, but M. bradburiana displayed the highest level of tolerance. Otherwise, cultivars tended to be more tolerant to powdery mildew than species. The observed variations in horticultural characteristics and performance highlight the high value of this genus for Georgia landscapes.
Monarda, colloquially called beebalm, is known for its showy floral display suited to managed borders, roadsides, and right of ways and is naturally found in woodlands, meadows, and floodplains (Weakley 2022). The flowers, seeds, and shoots of this North American forb provide forage and habitat for many invertebrate and vertebrate species of wildlife, including butterflies, bees, wasps, and birds (Prather et al. 2002). The high concentration of thymol in the leaves of Monarda spp. has led to its human use for food and as a flavoring agent in the food industry (Escobar et al. 2020). The genus has also been long used in traditional American medicine to treat digestive disorders, parasitic worms, cough, and stings. More recently it has been extensively studied for its antimicrobial properties (Chevallier and Nanba 2000; Escobar et al. 2020; Savickiene et al. 2002; Zhilyakova et al. 2009).
Monarda has many cultivars established in the trade and several species native to Georgia (Coombs 2016; Hawke 1998; Weakley 2022). It has been the subject of extensive trials at the Chicago Botanic Garden (Chicago, IL, USA) and Mt. Cuba Center (Hockessin, DE, USA), but in-depth studies from southeastern locales are scarce. In the global horticulture trade, the best known species include M. citriodora (lemon bergamot), M. didyma (sweet bergamot), M. fistulosa (wild bergamot), and M. punctata (dotted monarda). These taxa, M. bradburiana and M. ×hybrida Hort. (hybrids of M. didyma and M. fistulosa) have been extensively studied for their essential oil composition and ethnobotanical uses (Collicutt and Davidson 1999; Dudchenko et al. 2020; Mattarelli et al. 2017; Tabanca et al. 2013). We chose to include M. bradburiana in our trial and not M. citrodora due to the US Department of Agriculture (USDA) zones of the sites selected for study, species distribution maps and the potential for M. bradburiana to show high tolerance to powdery mildew (Weakley 2022).
The popularity of Monarda as a medicinal plant and as a nectar source for wildlife makes the perennial a great candidate for landscape use and garden interest. Monarda taxa are often used in seed mixes to provide summer forage to wildlife (Gray et al. 2007; Otto et al. 2017; Quinlan et al. 2021; Rubio et al. 2022; Wolf et al. 2022). Robust evaluations from the Chicago Botanic Garden and Mt. Cuba Center on species and cultivars of native plants in the Midwest and Northeast provide growers and gardeners with regionally specific information on morphometrics, phenology, and disease tolerance (Coombs 2016; Hawke 1998). The Mt. Cuba trials further identify wildlife observed on Monarda flowers within the trial garden (Coombs 2016). Comparatively, the climate in the Southeast is typified by longer growing seasons and milder winters. Additionally, plants grown in the Southeast are subjected to different and often significant disease pressures. Yet systematic data on horticultural performance of Monarda in southeastern locales is lacking.
The diversity of floral and inflorescence morphologies and pollination systems within the genus make it an excellent candidate for studying plant–pollinator interactions among wild and cultivated forms (Prather et al. 2002). Of the 10 taxa included in our study, eight are distinct in the coloration of floral parts. For example, within M. punctata, the Georgia and New Jersey (GA and NJ) ecotypes vary in bract color; M. punctata GA has pink bracts whereas M. punctata NJ displays white (Table 1). The role of hybridization in the evolution of Monarda (Prather et al. 2002) and the availability of cultivars in the trade make the genus an appropriate candidate to study the relationship between floral traits and wildlife. Alterations of the floral characteristics (e.g., corolla color and size) may have implications for nectaring insects (Kalaman et al. 2022). Along with assessing establishment and resiliency in the landscape, it is also important to provide empirical evidence on wildlife value. In the first part of the study, we focus on the plants’ horticultural performance. We assess floral morphology and nectar production in the second part of the study.
Monarda species and cultivars trialed in Blairsville and Athens, GA, USA.
Monarda spp. are susceptible to a variety of pathogens including powdery mildew [Podosphaera pannosa (syn. Sphaerotheca pannosa Wallr. Ex. Fr.), Golovinomyces biocellatus and G. monardae, and Erysiphe cichoracearum], aster yellows phytoplasma, stem rot (Sclerotium rolfsii), downy mildew (Peronospora monardae), and rust (Collicutt and Davidson 1999; Davidson 2002; Han et al. 2011; Holcomb 1994; Hwang et al. 1997; Perry 1998; Salgado-Salazar et al. 2020; Xu and Zhan 2022). We focused on powdery mildew because the genus is highly susceptible and the disease is a significant problem for perennial growers (Perry 1998). Powdery mildew is the common name for many obligate, polycyclic fungi in the order Erysiphales. The fungi cover plant surfaces in epiphytic, sometime endophytic, white growth, especially in conditions with high humidity and temperature fluctuations (Heffer et al. 2006; Perry 1998). The susceptibility of Monarda to powdery mildew has made tolerance to the disease an objective of many breeding and trial programs within the genus (Coombs 2016; Hawke 1998; Perry 1998).
We evaluated select Monarda at two sites in the Georgia mountains and Piedmont regions to assess horticultural performance in the landscape, flower phenology, and powdery mildew tolerance of 10 Monarda popular in the trade. We hypothesized that there would be differences in vegetative characteristics, reproductive behavior, and disease tolerance among the taxa.
Materials and Methods
Plant material sources
Species, cultivars, and greenhouse culture.
Five cultivars, the two species with cultivars, and two distinct species without known cultivars were chosen to demonstrate the diversity of habit, phenology, and powdery mildew tolerance of Monarda available in the trade (Table 1). Plants were received as plugs from New Moon Nursery (Woodstown, NJ, USA) and North Creek Nursery (Landenberg, PA, USA) in 2019, then grown into 0.95-L (quart-sized) containers in a climate-controlled greenhouse at 21 to 24 °C with 20% shade and natural photoperiod at the University of Georgia UGArden in Athens, GA, USA. The growing media was a mix of peat, perlite, and composted pine bark amended with 12.30 kg·m−3 N–P–K fertilizer (Plant-tone Organic 5–3–3, 0.4% ammoniacal nitrogen, 1.6% other water-soluble nitrogen, 3.0% water insoluble nitrogen, 3.0% P2O5, and 3.0% K2O; Epsoma, Millville, NJ, USA). Plants were fertilized with 100 ppm N liquid feed (Jack’s Acid 20–20–20 General Purpose, 3.83% ammoniacal nitrogen, 6.07% nitrate nitrogen, 10.10% urea nitrogen, 20% P2O5, and 20% K2O; J.R. Peters, Inc. Allentown, PA, USA) monthly until transplanting into the field. Containerized plants in the greenhouse were hand watered as needed over Summer 2019.
Site location, soil conditions, bed preparation, and climate conditions.
The first site was planted 14 Oct 2019 in raised beds at the University of Georgia Mountain Research and Education Center in Blairsville, GA, USA (34.838800, –83.927941, USDA zone 7b). Plants were allowed to establish in Fall and Winter 2019–20 for data collection starting Spring 2020. This experiment ran from April to October in both 2020 and 2021. Raised beds were made up of mushroom compost, composted pine bark, and 5 cm of pine bark nuggets for mulch. Before amendments, soil tested very high in phosphorus and calcium, and high in potassium, magnesium, zinc, and manganese. The raised bed pH was 6.8. No lime was added to the beds. An additional 0.45 kg of 15–9–12 (Osmocote Smart-Release Plant Food, 8.4% ammoniacal nitrogen, 6.6% nitrate nitrogen, 9% P2O5, 12% K2O; The Scotts Company, Marysville, OH, USA) was equally distributed among plants and top-dressed early spring each season.
The second site was the State Botanical Garden of Georgia’s Mimsie Lanier Center for Native Plant Studies (MLCNPS) in Athens, GA, USA (33.902371, –83.391072, USDA zone 8b). The site was planted on 1 May 2020 for plant establishment and data collection in Spring 2021. Data were collected from April to October in both 2021 and 2022. Beds comprised native soil (clay ultisol), gravel, and 5 cm of hardwood mulch. The bed area was divided into two subplots for the soil tests. Before amendments, the north side of the bed tested medium in calcium, zinc, and manganese. Phosphorus, potassium, and magnesium tested low. The pH of the north side was 5.7. The south side of the bed tested very high in calcium and medium in phosphorus, potassium, magnesium, zinc, and manganese. The pH of the south side was 6.1. No lime was added to the plot. An additional 0.45 kg of 15–9–12 (Osmocote Smart-Release Plant Food, 8.4% ammoniacal nitrogen, 6.6% nitrate nitrogen, 9% P2O5, 12% K2O) was equally distributed among plants and top-dressed early spring each season.
Plant data
Vegetative parameters.
The height and width of plants were measured at peak flower. Shoot height was measured to the tallest node with foliage. Plant width was measured at the widest point of mature vegetation. Due to the symmetrical nature of the canopy, we deemed one width measurement as a sufficient estimation.
Reproductive parameters.
Open inflorescences were counted during peak bloom at each site in 2021 and at the MLCNPS in 2022 (methods adapted from Ruane et al. 2014). Floral density was defined as the counted number of flowering stems with open inflorescences. Phenology (total length of bloom) for each site and year was calculated at the date of first flower opening and last flower senescing per treatment. Because peak bloom varied greatly between repetitions depending on plant health, we recorded this parameter when ∼50% of set floral buds had opened or showed color. The end of peak was determined when ∼50% of flowers had senesced.
Plant health.
Health (defined as disease progression) was evaluated as the percentage of plant leaf area infected with powdery mildew, a foliar disease known to be problematic for the genus (Davidson 2007). Health was evaluated eight times during the April to October season in 2020, 2021, and 2022 for the respective sites. Percent of infected foliaged was assessed on a 1- to 5-point scale in 2020, 2021, and 2022, where 1 = no infected foliage, 2 = up to 25% infected foliage, 3 = 26% to 50% infected foliage, 4 = 51% to 75% infected foliage, and 5 = 76% to 100% infected (per methods in Long et al. 2010). The area under the disease progression curve (AUDPC) was calculated as the sum of trapezes (area) under the disease progression curve, which considered disease rating over time (de Mendiburu 2021). The score allows multiple observations of disease over a season to be compared among taxa using a single value.
Experimental design and data analyses
At both sites, plots were planted with the 10 Monarda species and cultivars (treatments). Each combination of species/cultivar of a species was considered a treatment. Each plot had seven rows, and each row contained the 10 treatments randomized in a unique order for a total of 70 plants per site. Plants were spaced in 1.5-m centers. Each site and year were analyzed independently. Treatment was considered the main effect and error consisted of each plant. Data were subjected to analysis of variance (ANOVA) followed by post hoc means separation by Tukey’s honestly significant difference test (HSD) using R Core Team 2018 (R Core Team 2018), with statistical significance determined at alpha ≤ 0.05. Model assumptions were checked visually with the residuals. Graphs and figures were generated using ‘tidyverse’ and ‘ggplot2’ (Wickham et al. 2019 and Wickham 2016). Individual parameters, transformations, and models are addressed next.
Vegetative parameters.
Morphometric data (height and width) were subjected to ANOVA with the Monarda samples as the main effect and followed by post hoc mean separation by Tukey’s HSD. Standard errors were calculated with ‘plotrix’ (Lemon et al. 2021).
Reproductive parameters.
Floral density, as count data, was subjected to a generalized linear model fitted to a negative binomial distribution with the following packages: ‘stats’, ‘MASS’, ‘pscl’, ‘sandwich’, ‘lmtest’, ‘car’, and ‘emmeans’ (Fox and Weisberg 2019; Jackman 2020; Lenth et al. 2022; R Core Team 2018; Venables and Ripley 2002; Zeileis et al. 2020; Zeileis and Hothorn 2002; Zeileis 2004, 2006).
Plant health.
The AUDPC scores were calculated with ‘agricolae’ (de Mendiburu 2021). Standard errors were calculated with ‘plotrix’ (Lemon et al. 2021).
Results and Discussion
Based on our two-site, 2-year study, we have sufficient data to support the hypotheses that morphology and phenology vary among Monarda. We provide detailed data on plant heights and widths, floral density, flowering phenology, and powdery mildew tolerance for Monarda in Georgia’s Piedmont and Montane regions.
Height and width.
Height and width significantly varied among Monarda taxa (Table 2) in both sites and both years. In the first season after establishment at Blairsville (2020) M. bradburiana and M. Sugar Buzz® Grape Gumball were shorter than other samples. This trend continued in 2021 in Blairsville and Athens, although some overlap occurred between M. punctata GA in Blairsville and M. punctata NJ in Athens. These four tended to be shorter than other, excluding some overlap between the M. punctata ecotypes and M. fistulosa, M. didyma, and M. didyma ‘Jacob Cline’ during the first year after establishment. Monarda × ‘Judith’s Fancy Fuchsia’ and M. ‘Raspberry Wine’ ranked tallest in the first year after establishment in Blairsville. Otherwise, M. fistulosa was the tallest (Table 3).
Analysis of variance results for effect of 10 Monarda taxa on height and width during peak bloom in Blairsville, GA, USA, 2020 and 2021 and Athens, GA, USA 2021 and 2022. Each site and year were subjected to independent analysis.
Mean (± SE) height (cm) and width (cm) of 10 Monarda taxa per year and site grown in Georgia. Measurements taken at peak bloom for seven replicates per taxa at each site (70 plants per site). Means within a column followed by different letters are significantly different (P ≤ 0.05). Each site and year were subjected to independent analysis and comparisons are valid within height and width columns.
Our results show two distinct groups concerning height. Height at flowering ranged from just under .5 m to just over 1.5 m depending on site and sample (Table 3). The shorter group included M. bradburiana, M. Sugar Buzz® Grape Gumball, and the M. punctata ecotypes. M. fistulosa ‘Claire Grace’ was more compact than the straight species, except in Blairsville 2020. The range of heights within Monarda provides a basis for selecting plants for different landscape uses. M. bradburiana and M. Sugar Buzz® Grape Gumball are well suited for the front border, while M. punctata is more appropriately placed in the front or middle of a bed. The remaining taxa are tall enough to be visible from the back in a border planting or the center in an island planting.
Plant width at flowering was also distinct among taxa. Trends in width loosely paralleled trends in height. M. bradburiana, M. Sugar Buzz® Grape Gumball, and M. punctata ecotypes had smaller spreads. The widest taxa were M. × ‘Judith’s Fancy Fuchsia’, M. ‘Raspberry Wine’, and M. fistulosa (Table 3). Shorter taxa tended to spread less. We observed a more compact habit in M. didyma cultivars than the species. The M. fistulosa cultivar Claire Grace was also bred to be more compact. M. Sugar Buzz® Grape Gumball, which was bred as a dwarf cultivar of M. didyma, along with M. bradburiana could be spaced on half-meter centers due to their shorter stature and tendency to mound. M. punctata ecotypes can be spaced on 1-m centers. All other taxa will spread well over 1 m and should be on 1.5-m spacing.
Spread of Monarda throughout the garden is best managed by removing rhizomes from the center of the clump in the spring and summer to increase air flow through the stand (Thompson 2007). M. punctata behaves uniquely relative to the other trialed taxa. It spreads via stolons, unlike the other taxa that have rhizomes. Growth of this species during the second season was less robust than the first, yet M. punctata will readily reseed each fall.
Floral density.
In Blairsville, M. punctata GA had the fewest flowering stems, and M. fistulosa had the most. Other taxa fell in between, with all but M. Sugar Buzz® Grape Gumball setting more than 100 flowering stems on average (Table 4). On average in Athens, M. bradburiana set the least number of flowering stems and M. fistulosa ‘Claire Grace’ set the most. M. didyma taxa, M. bradburiana, and M. × ‘Judith’s Fancy Fuchsia’ had under 100 flowering stems during peak bloom in Athens in 2021. M. didyma taxa M. bradburiana, and M. Sugar Buzz® Grape Gumball had fewer than 100 flowering stems during peak bloom in Athens in 2022 (Table 4).
Mean floral density (± SE) of 10 Monarda taxa per year and site grown in Blairsville and Athens, GA, USA. Floral density was defined as the number of flowering stems during peak bloom (70 plants per site). Means within a column followed by different letters are significantly different (P ≤ 0.05). Each site and year were subjected to independent analysis and comparisons are valid within columns by year.
Floral density differed among taxa, with less pronounced differences observed in Blairsville than in Athens (Table 4). Flowering periods varied between site and season (Fig. 1). When the number of flowers and flowering period are both considered, it becomes difficult to determine which taxa have the most “flower power.” Rather it depends on the gardener’s needs and the context of the planting site.
Monarda phenology in Blairsville, GA, USA 2020 and 2021, and Athens, GA, USA 2021 and 2022.
Citation: HortScience 59, 6; 10.21273/HORTSCI17793-24
Regarding flower power, M. fistulosa stands out in Blairsville, whereas M. fistulosa and M. punctata ecotypes stand out in Athens. M. didyma taxa tended to set fewer flowering stems, but each inflorescence is quite large (Table 4). M. × ‘Judith’s Fancy Fuchsia’ and M. ‘Raspberry Wine’ provided an even split, each with relatively high flower set and long flowering period.
Phenology.
In Blairsville in 2020, observations of start of bloom were not carried out for M. didyma, M. ‘Raspberry Wine’, M. Sugar Buzz® Grape Gumball, nor M. fistulosa taxa due to the COVID-19 lockdown in June. Regardless of site or year, M. bradburiana was the first Monarda taxon to bloom, spanning the months of late April and May, and ending in June. M. punctata ecotypes bloomed from late June, July, or August and end in September to October, depending on the plant and the site. All other taxa bloomed throughout the summer, starting around June and ending between July and October. We observed M. didyma and M. didyma ‘Jacob Cline’ senesced in July and rebloomed in September. Peak bloom lasted between 7 and 21 d for most taxa. Notably, M. Sugar Buzz® Grape Gumball did not peak in Athens 2021 (Fig. 1). Lack of establishment resulted in no more than 10 flowers per repetition, and we did not consider this a “peak bloom.”
Phenology is important for producers, consumers, and wildlife, with longest bloom being most desirable; although we did not analyze it statistically, our observations allowed for a good estimate. In 2021, we noted that the growing season in the more southern site (Athens) was longer than in the more northern site (Blairsville). Flowering generally started later and ended earlier in the latter (Fig. 1). There were several notable groupings, with M. bradburiana in spring, M. didyma and M. fistulosa taxa in midsummer, and M. punctata categories in late summer or early fall. M. punctata ecotypes bloom later than all other taxa, starting anywhere from late June to July or August and ending in September or October depending on the plant age and USDA zone.
By using Monarda taxa that bloom early, mid, or late summer, careful selection by gardeners can provide continuous and robust floral displays for human enjoyment and wildlife consumption. With minimal care and fertilization, most taxa flower for 2 to 3 months, producing ∼100 flowering stems, with peak bloom lasting 1 to 3 weeks (Table 4 and Fig. 1). We observed that M. didyma and M. didyma ‘Jacob Cline’ peak flower in June and again in late summer or early fall. The repeated flowering observed in the previously stated taxa was not observed among other M. didyma cultivars. M. fistulosa and M. fistulosa ‘Claire Grace’ had more flowering stems, but the length of bloom was longer for M. didyma samples. Whereas M. bradburiana and M. Sugar Buzz® ‘Grape Gumball’ could be considered low bloomers relative to other evaluated samples, M. bradburiana set more than 100 flowers in Blairsville and M. Sugar Buzz® ‘Grape Gumball’ can set 100 flowers in the first season after establishment (Table 4).
Health.
The AUDPC was significantly different between Monarda taxa (Blairsville 2020, P < 2e-16; 2021, P = 2.15e-05; Athens 2021, P = 2.12e-12; 2022, P = 0.000647). The disease progression of powdery mildew throughout the growing season is shown in Fig. 2. Differences in disease tolerance among samples were more pronounced in the first year of data collection than in the second. Although all showed some level of infection, M. bradburiana was the least affected in each site and year. M. didyma and M. fistulosa had high scores (i.e., exhibited more symptoms) in each site and year. Cultivars showed some tolerance relative to parent species (Table 5). M. punctata ecotypes were excluded from disease analysis due to their short life cycle; however, it also exhibited signs of infection in the second year. M. Sugar Buzz® ‘Grape Gumball’ and M. didyma were excluded from AUDPC scoring because at least three repetitions of each died over Winter 2021–22.
Powdery mildew progression among Monarda taxa in Blairsville, GA, USA (2020 and 2021) and in Athens, GA, USA (2021 and 2022).
Citation: HortScience 59, 6; 10.21273/HORTSCI17793-24
Calculated area under the disease progression curve [calculated as the sum of trapezes (area) under the disease progression curve, which considered disease rating over time (de Mendiburu 2021)] averages and standard errors for powdery mildew. Values are means of eight observation dates for each site and season. Each site and year were subjected to independent analysis; statistical comparisons are valid within columns.
The progression of powdery mildew for each site and season and calculated AUPDC scores rating tolerance among Monarda taxa are presented in Table 5 and Fig. 2. The short lifecycle of M. punctata excluded the ecotypes from AUDPC scoring and analysis. We observed more differences in tolerance in the first season than in the second season at both sites. Cultivars were more tolerant of powdery mildew than the parent species M. didyma and M. fistulosa, especially in the first season. We also saw good tolerance in M. bradburiana. It is conceivable that cultural methods such as spring and summer divisions or removal of diseased stems and overwintering material would reduce the powdery mildew within stands, but we do not know how this would impact wildlife. Specifically, Monarda stems are hollow and could provide nesting material for cavity-nesting wasps and bees (O’Neill and O’Neill 2010). Removing senesced foliage from the ground may be sufficient to minimize disease spread even if stem stubble is left to encourage insect nesting.
In terms of establishment, M. Sugar Buzz® Grape Gumball was notably different from other tested samples, and especially other M. didyma cultivars. Members of the Sugar Buzz® series from Walters Gardens (n.d.) are bred as compact hybrids significantly shorter than other cultivars in our trial. The plants did not persist past one season in Athens (data not shown). The performance in the first season after establishment was impressive at both sites, so we recommend its use in annual displays; its longevity in more southern climates remains doubtful and would warrant further study.
Compared with trials conducted at the Chicago Botanic Garden and Mt. Cuba Center, the Georgia seasons are longer (Coombs 2016; Hawke 1998). Season length could explain the differences among results between the three studies. Not all taxa in our study were trialed by the Chicago Botanic Garden and Mt. Cuba Center and vice versa, so we can only compare taxa that were included in our trial and at least one other study. Plant height at flowering, width after 2 years, and peak flower coverage were similar between the three studies. Phenology and tolerance to powdery mildew varied. Floral initiation started earlier and ended later in Georgia compared with Chicago Botanic Garden and Mt. Cuba Center trials. Notably M. didyma and M. didyma ‘Jacob Cline’ had recurrent flowering in all sites and years of our study. All studies showed that M. fistulosa had the least tolerance to powdery mildew, whereas M. bradburiana showed excellent tolerance. We note that the powdery mildew on M. fistulosa was not detrimental to its floral display. Chicago Botanic Garden and Mt. Cuba Center drew more distinctions in powdery mildew tolerance among samples than we observed in our study (Coombs 2016; Hawke 1998).
Conclusions
These studies were designed to inform the ornamental industry, landowners, park managers, civic entities, and consumers on the performance of select Monarda species and cultivars. Morphology, phenology, and tolerance to powdery mildew varied among samples for each site and year. Samples showed two height groups (i.e., less than or greater than 1 m), and three flowering seasons (early, middle, and late summer). We found that M. punctata is short-lived and as an individual plant did not last more than two seasons. Yet it tends to produce viable seeds that can spread throughout the garden or a naturalized right of way. This behavior is typical of short-lived perennials, and it differs from the other taxa we evaluated. Most samples set around 100 flowering stems at peak bloom, except for M. bradburiana, M. didyma, and M. didyma ‘Jacob Cline’ and M. punctata GA in Blairsville. Cultivars exhibited some tolerance to powdery mildew relative to straight species, except for M. bradburiana, which has no known cultivar, and M. punctata, which has a shorter life span. We conclude most tested Monarda are suitable candidates for adding summer color to gardens located in the piedmont and montane regions of Georgia. M. punctata ecotypes provide an appealing texture, M. bradburiana exhibited high tolerance to powdery mildew, and dwarf varieties such as M. Sugar Buzz® Grape Gumball would be well suited for annual beds. This information will be of value to floriculture producers, landscape architects, and consumers in marketing their plants and designing their landscape uses. Our data can be used to determine not only which plants to include in the design, but also the amount of space taxa may require in a planting—which for landscape professionals is important in deciding number of plants and inputs—and potentially could lead to reduced installation costs, lower replacement rate, and savings.
References Cited
Chevallier T, Nanba S. 2000. The encyclopedia of medicinal plants. Seibundo Shinkosya Publishing Co., Ltd., Tokyo, Japan.
Collicutt LM, Davidson CG. 1999. ‘Petite Delight’ Monarda. HortScience. 34(1):149–150. https://doi.org/10.21273/HORTSCI.34.1.149.
Coombs G. 2016. Research report: Monarda for the Mid-Atlantic region. Mt. Cuba Center, Hockessin, DE, USA.
Davidson CG. 2002. ‘Petite Wonder’ Monarda. HortScience. 37(1):235–236.
Davidson CG. 2007. Chapter 28: Monarda, Bee-Balm: Monarda didym, p 757–779. Flower Breeding and Genetics Issues, Challenges and Opportunities for the 21st Century. Springer Netherlands, Dordrecht, The Netherlands.
Dudchenko VV, Svydenko LV, Markovska OY, Sydiakina OV. 2020. Morphobiological and biochemical characteristics of Monarda L. varieties under conditions of the southern steppe of Ukraine. J Ecol. Engin. 21(8):99–107.
Escobar A, Perez M, Romanelli G, Blustein G. 2020. Thymol bioactivity: a review focusing on practical applications. Arabian J Chem. 13:9243–9269. https://doi.org/10.1016/j.arabjc.2020.11.009
Fox J, Weisberg S. 2019. An R companion to applied regression (3rd ed). Sage Publications, Thousand Oaks, CA, USA. https://socialsciences.mcmaster.ca/jfox/Books/Companion/. [accessed 1 Jun 2020].
Gray SS, Simpson T, Baccus J, Manning R, Schwertner TW. 2007. Seasonal diet and foraging preference of greater kudu Tragelaphus strepisiceros in the Llano Uplift of Texas. Wildl Biol. 13(1):75–83. https://doi.org/10.2981/0909-6396.
Han KS, Park MJ, Cho SE, Shin HD. 2011. First report of powdery mildew caused by Golovinomyces biocellatus on Monarda didyma in Korea. Plant Dis. 95(12):1590. https://doi.org/10.1094/PDIS-07-11-0569.
Hawke RG. 1998. Plant evaluation notes: Monarda and powdery mildew resistance. Chicago Botanic Garden, 12. https://www.chicagobotanic.org/downloads/planteval_notes/no12_monarda.pdf.
Heffer V, Powelson ML, Johnson KB, Shishkoff N. 2006. Identification of powdery mildew fungi. The Plant Health Instructor. https://doi.org/10.1094/PHI-I-2006-0706-01.
Holcomb GE. 1994. First report of stem rot of Monarda didyma caused by Sclerotium rolfsii. Plant Dis. 78(2): 208–208.
Hwang SF, Chang KF, Howard RJ. 1997. First report of a yellows disease of monarda (Monarda fistulosa L.) in Canada caused by a phytoplasma. Z Pflanzenkr Pflanzenschutz. 104(2):173–181.
Jackman S. 2020. pscl: Classes and Methods for R Developed in the Political Science Computational Laboratory. United States Studies Centre, University of Sydney, Sydney, New South Wales, Australia. R package version 1.5.5. https://github.com/atahk/pscl/. [accessed 1 Aug 2021].
Kalaman H, Wilson SB, Mallinger RE, Knox G, Kim T, Begcy K, van Santen E. 2022. Evaluation of Native and Nonnative Ornamentals as Pollinator Plants in Florida: II. Floral Resource Value. HortScience. 57(1):137–143. https://doi.org/10.21273/HORTSCI16124-21.
Lemon J, Bolker B, Oom S, Klein E, Rwolingson B, Wickham H, Tyagi A, Eterradossi O, Grothendieck G, Toews M, Kane J, Turner R, Turner R, Witthoft C, Stander J, Petzoldt T, Durrsma R, Biancotto E, Levy O, Dutang C, Solymos P, Engelmann R, Hecker M, Steinbeck F, Borchers H, Singmann H, Toal T, Ogle D, Baral D, Groemping U, Venables B, The Cran Team, Murdoch D. 2021. Plotrix: various plotting functions. R (≥ 3.5.0) Version 3.8-2. https://CRAN.R-project.org/package=plotrix. [accessed 1 Sep 2021].
Lenth RV, Buerkner P, Herve M. 2022. Emmeans: estimated marginal means, aka Least-Squares Means. R (≥ 3.5.0) Version 1.7.5.
Long MC, Krebs SL, Hokanson SC. 2010. Field and growth chamber evaluation of powdery mildew disease on deciduous azaleas. HortScience. 45(5):784–789. https://doi.org/10.21273/HORTSCI.45.5.784.
Mattarelli P, Epifano F, Minardi P. 2017. Chemical composition and antimicrobial activity of essential oils from aerial parts of Monarda didyma and Monarda fistulosa cultivated in Italy. TEOP. 20(1):76–86.
de Mendiburu F. 2021. Agricolae: Statistical Procedures for Agricultural Research. R (≥ 2.10) Version 1.3-5. https://CRAN.R-project.org/package=agricolae.
O’Neill KM, O’Neill JF. 2010. Cavity-nesting wasps and bees of central New York State: The Montezuma Wetlands complex. Northeastern Naturalist. 17(3):455–472.
Otto CRV, O’Dell S, Bryant RB, Euliss NH, Bush RM, Smart MD. 2017. Using publicly available data to quantify plant–pollinator interactions and evaluate conservation seeding mixes in the northern Great Plains. Environ Entomol. 46(3):565–578. https://doi.org/10.1093/ee/nvx070.
Perry LP. 1998. Herbaceous perennials production: A guide from propagation to marketing (NRAES-93). Northeast Regional Agricultural Engineering Service, Cooperative Extension, New York, NY, USA.
Prather LA, Monfils AK, Posto AL, Williams RA, Botany SS, Mar NJ. 2002. Monophyly and phylogeny of Monarda (Lamiaceae): Evidence from the Internal Transcribed Spacer (ITS) region of nuclear ribosomal DNA. Syst Biol. 27:127–137.
Quinlan GM, Milbrath MO, Otto CRV, Isaacs R. 2021. Farmland in U.S. conservations reserve program has unique floral composition that promotes bee summer foraging. Basic Appl Ecol. 56:358–368. https://doi.org/10.1093/ee/nvx070.
R Core Team. 2018. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.
Ruane LG, Rotzin AT, Congleton PH. 2014. Floral display size, conspecific density and florivory affect fruit set in in natural populations of Phlox hirsuta, an endangered species. Ann Bot. 113(5):887–893.
Rubio A, Wright K, Longing S. 2022. Bee and flowering plant communities in a riparian corridor of the lower Rio Grande River (Texas, USA). Environ Entomol. 51(1):229–239. https://doi.org/10.1093/ee/nvab108.
Salgado-Salazar C, LeBlanc N, Wallace EC, Daughtrey ML, Crouch JA. 2020. Peronospora monardae, Hyaloperonospora daughtreyae and H. iberidis: New species associated with downy mildew diseases affecting ornamental plants in the United States. Eur J Plant Pathol. 157:311–326.
Savickiene N, Dagilyte A, Barsteigiene Z. 2002. Identification of flavonoids in the flowers and leaves of Monarda didyma L. Medicina (B Aires). 38:1119–1122.
Tabanca N, Bernier UR, Ali A, Wang M, Demirci B, Blythe EK, Khan SI, Baser KHC, Khan IA. 2013. Bioassay-guided investigation of two Monarda essential oils as repellents of yellow fever mosquito Aedes aegypti. J Agr Food Chem. 61:8573–8580.
Thompson P. 2007. Summer division of perennials. Plantsman (Lond, Engl). 6(2):89–91.
Venables WN, Ripley BD. 2002. Modern applied statistics with S (4th ed). Springer, New York, NY, USA.
Walters Gardens, Inc. n.d. Monarda ‘Grape Gumball’ PP27498: Walters Gardens, Inc. Monarda ‘Grape Gumball’ PP27498. https://www.waltersgardens.com/variety.php?ID=MONGG.
Weakley AS, Southeastern Flora Team. 2022. Lamiaceae: Monarda Linnaeus 1753 (Bergamot, Beebalm), p 1485–1488. In: Flora of the southeastern United States. University of North Carolina at Chapel Hill Herbarium and Botanical Garden.
Wickham H. 2016. ggplot2: Elegant graphics for data analysis. Springer-Verlag, New York, NY, USA.
Wickham H, Averick M, Bryan J. 2019. Welcome to the tidyverse. J Open Source Softw. 4(43):1686. https://doi.org/10.21105/joss.01686.
Wolf AT, Watson JC, Hyde TJ, Carpenter SG, Jean RP. 2022. Floral resources used by the endangered rusty patched bumble bee (Bombus affinis) in the Midwestern United States. Nat Areas J. 42(2):301–312. https://doi.org/10.3375/22-2.
Xu YY, Zhan JT. 2022. First report of powdery mildew caused by Golovinomyces monardae on Scarlet Beebalm (Monarda didyma) in China. APS Publications: Plant Disease. 106(5):1525.
Zeileis A. 2004. Econometric computing with HC and HAC covariance matrix estimators. J Stat Softw. 11(10):1–17. https://doi.org/10.18637/jss.v011.i10.
Zeileis A. 2006. Object-oriented computation of sandwich estimators. J Stat Softw. 16(9):1–16. https://doi.org/10.18637/jss.v016.i09.
Zeileis A, Köll S, Graham N. 2020. Various versatile variances: An object-oriented implementation of clustered covariances in R. J Stat Softw. 95(1):1–36. https://doi.org/10.18637/jss.v095.i01.
Zeileis A, Hothorn T. 2002. Diagnostic checking in regression relationships. R News. 2(3):7–10. https://CRAN.R-project.org/doc/Rnews/. [accessed 1 Sep 2020].
Zhilyakova ET, Novikov OO, Naumenko EN, Krichkovskaya LV, Kiseleva TS, Timoshenko EY, Novikova MY, Litvinov SA. 2009. Study of Monarda fistulosa essential oil as a prospective antiseborrheic agent. Bull Exp Biol Med. 148:612–614.
