JOurn.31 ofr in.nerr.3du.3a- Re--search
..Oluii 5, is.ue 2 - rio.-e l-i er 2i:,:,3
Characterizing Potential Invasiveness of Fourteen Buddleja Cultivars in
Plant growth, visual quality, flowering, and seed production were determined for 14 Buddleja species.
Buddleja davidii x B. lindleyana and B. x weyeriana 'Honeycomb' had the greatest growth index and shoot dry
weight of all cultivars. Each of the 14 cultivars evaluated produced seed. The shape and number of seed pods
per infructescence varied with cultivar. The influence of light and temperature on germination were determined for
6 Buddleja davidii cultivars. Regardless of temperature or cultivar, light was required for germination. With or
without light, less than 2% germination occurred at 33 OC for each cultivar. At 15 or 24 OC in light, germination
was greatest for 'Nanho Purple' followed by 'Dartmoor'. Germination of 'Black Knight' and 'White Profusion'
was consistently lower than that of other cultivars, regardless of temperature.
The State of Florida is the second largest producer of ornamental plants in the U.S. with an estimated 9.908 billion
in total industry sales during 2000 (Hodges and Haydu, 2002). While most intentionally introduced species remain
in their cultivated settings, some escape cultivation and invade natural areas. Many of the plants that are
invasive today were originally introduced as ornamentals and highly prized by both producers and consumers for
their rapid growth, durability, and ease of propagation. Of the 235 woody plant species that have escaped
cultivation and naturalized in North America, 85% were introduced primarily for the landscape trade (Reichard
and Hamilton, 1997). Of the 124 plant species listed as invasive by the Florida Exotic Pest Plant Council
(FLEPPC), 67% were introduced as ornamentals (www.fleppc.org). Today, approximately 1.9 million acres of
Florida's remaining natural areas have been invaded by exotic plant species and more than 240 million dollars
have been spent in Florida to control invasive, exotic plant species since 1980 (www.fleppc.org).
The Tampa Bay Wholesale Growers, Florida Nurserymen and Growers Association, and the American Nursery
and Landscape Association Board of Directors are among the first in the nation to adopt voluntary codes of
conduct for the nursery industry with regards to invasive plants (www.mobot.org). The codes of conduct
involve adoption of risk assessment methods that consider plant characteristics and prior observations or
experience with the plant elsewhere in the world before it is released and distributed. Research on invasive plants
is critical to provide scientific evidence of whether a plant is currently invasive or has great potential to
become invasive. This information can then be relayed to the growers, who in turn can make an informed decision
as to whether or not a particular plant should be produced commercially.
It has been shown that seed production and viability vary significantly across cultivars of wildtype plants
(Krumfolz and Wilson, 2002). Clearly, it is not appropriate to deem an economically important species with
high ornamental value as invasive without considering all cultivars. Buddleja davidii is a widely cultivated,
extremely popular, flowering shrub with attractive foliage and a range of flower colors, some with
exceptional fragrance. A member of Loganiaceae, Buddleja is native to China and valued for summer flowering
of large inflorescences that vary in length and profusely attract butterflies. Numerous cultivars have replaced
the species. Dirr (1998) lists 70 B. davidii taxa that can be grouped by white, lanvender/blue, pink/rose/
mauve, purple/violet/magenta, and yellow flowers. Unfortunately, some of the same properties that lend it as a
great landscape plant (i.e., long flowering period, adaptability to a range of environmental conditions, ease
of propagation, and high vigor) also predispose it as a potentially invasive woody shrub. The species B. davidii
has escaped cultivation in 19 states and Puerto Rico (www.plants.usda.gov) and is considered a weed in
Hawaii (Staples et al., 2000), New Zealand (Kay and Smale, 1990), Australia (Csurhes and Edwards, 1998) and
Great Britain (Crawley, 1987). In the US, it is not listed as a noxious weed, however, it is considered by many
as weedy, as it is commonly found along roadsides, streamsides and other disturbed areas of its growing range.
In Florida, there are no vouchered specimens of its escape into natural areas, although it is commonly utilized
in landscapes of the north and central part of the state. Buddleja lindleyana, a separate species with darker
green leaves and longer flower panicles, has escaped cultivation in Florida (www.plantatlas.usf.edu) as well as
8 other states (www.plants.usda.gov).
Because of its invasive characteristics, extensive use in the landscape, numerous cultivars, and history as an
invasive plant in other countries, Buddleja is a good candidate to evaluate for seed production and seed
germination in Florida. Specific objectives of the full study were to (1) assess the plant growth, visual quality,
and seed production and viability of 14 Buddleja cultivars planted in west (Milton) and central/south Florida
(Fort Pierce). Partial results of the Fort Pierce study will be presented here.
MATERIALS AND METHODS
Fourteen Buddleja cultivars were selected for this study based on availability, popularity, and performance in
the landscape (Table 1). Each cultivar was clonally propagated (West Florida Research and Education
Center, WFREC), transferred to 3.8 L pots, and fertilized with a slow release fertilizer prior to planting.
The experiment was conducted at Fort Pierce (USDA Zone 9B) where 9 uniform plants of each cultivar were
planted (4 June 2003) 8' on center on raised beds covered with white plastic mulch. Plants were watered by
seep irrigation as needed. Visual quality assessments (plant color, form and flower) and seed production data
(date of first visible seed pods) were taken monthly by three individuals for each cultivar in each block.
Assessment of color and form was performed on a scale from 1 to 5 where 1=very poor quality, not
acceptable, severe leaf necrosis or yellowing, 2=poor quality, not acceptable, large areas of necrosis or
yellowing, poor form, 3=fair quality, marginally acceptable, somewhat desirable form and color, 4=good quality,
very acceptable, nice color without yellowing, good form, marketable, and 5=excellent quality, very
marketable. Flowering was assessed by recording the percent of plant canopy covered with flowers on a scale from
1 to 5 where 1=no flowers present, 2=25% of canopy with flowers, 3=50% of canopy with flowers, 4=75% of
canopy with flowers, and 5=100% of canopy with flowers.
Plants were harvested after 25 weeks. After final visual quality measurements were taken, growth indices
were calculated as an average of the measured height and two widths ((height + north-south width + east-
west width)/3). Panicle infructescences were removed from shoots prior to severing plants at the crown. Seed
pods were separated from the flowers, weighed and allowed to dehisce at room temperature. Separately, shoots
and flowers (no pods) of each plant were oven dried at 70 OC for 1 week and weighed. Vegetative re-growth from
the severed base of each plant was assessed and recorded 4 months after plants were harvested.
Preliminary seed germination studies were conducted on seed collected from 6 cultivars of B. davidii
('White Profusion', 'Pink Delight', 'Dartmoor', 'Nanho Alba', 'Black Knight', and 'Nanho Purple') grown in the
Milton Botanical Gardens under open pollination. Seed was collected from plants at maturity and allowed to dehisce
at room temperature (22 to 25 OC) in paper bags for 1-2 weeks. Individual treatments for all incubator
germination experiments consisted of five replications of 25 seeds placed on two sheets of Reeve angel filter
paper (Whatman Inc., Clifton, NJ) in 9 cm petri-dishes. The filter paper was moistened with 10 mL nanopure
water containing 0.15% liquid Captan to prevent fungal growth. All dishes were sealed with Parafilm to
prevent desiccation. Germination studies were conducted in temperature and light controlled incubators
(Precision Model 818) equipped with fluorescent lamps. Seed were placed in light or darkness at 15, 24, 33
and 30/20 -C. A 12 hour photoperiod was maintained in each chamber with a photosynthetic photon flux of 22-
30 pmol.m-2.s-1. Dishes assigned to dark treatments were wrapped in two layers of heavy-duty aluminum foil
and remained unopened until the final day of the experiment. Germination of seed exposed to light was
monitored daily for 28 days. Seed germination in incubators was concurrent with seed germination in
the greenhouse. Twenty-five seeds of each species were placed in 8.9 cm pots filled with soilless germination
mix (Fafard, Inc., Apopka, FL). Two treatments were implemented whereby seeds are placed either on the
soil surface or 1 cm below the medium. Diurnal greenhouse temperatures were recorded with a Universal
K thermograph (Ryan ï¿½ Instruments, Redmond, WA). A seed was considered germinated when a cotyledon
was visible above the medium.
RESULTS AND DISCUSSION
Plant growth, visual quality, flowering and re-growth varied significantly with each cultivar (Table 2).
Buddleja davidii x lindleyana and B. x weyeriana 'Honeycomb' had the greatest growth index and shoot dry weight
of all cultivars. At 8 weeks, 'Dartmoor' had the highest flower rating, with almost 75% of the shoot canopy
having open flower panicles or buds. Each of the 14 cultivars evaluated produced seed. The shape and number
of seed pods per infructescence varied with cultivar (data not shown). For example, on average, a single plant
of 'White Profusion' (2.5' tall) produced 13.6 g of seed, each pod containing approximately 52 seeds; whereas,
a single plant of 'Nanho Blue' (1.9' tall) produced 5.01 g of seed, each pod containing approximately 82 seeds.
Large differences in the amount of viable seed have been reported for other B. davidii cultivars. As part of a
variety trial including 50 Buddleja taxa at Longwood Gardens (Kennett Square, PA), Anisko and Im (2001)
reported that only two ('Orchid Beauty' and 'Summer Rose') out of the 35 B. davidii cultivars tested produced
less than 0.1 gm of seed per infructescence. Visual quality ratings (foliage color and plant form) were highest
for 'Dartmoor', 'Honeycomb' and B. davidii x lindleyana, indicating that even at the end of the season, plants were
of good quality and marketable. All cultivars had some vegetative re-growth from the severed crown of the
original plants, with the exception of 'Nanho Purple'. Highest re-growth (56%) occurred for 'Moonlight'.
Regardless of temperature or cultivar, light was required for germination (data not presented). This suggests that
in natural conditions, seeds may not germinate if shaded by liter or a leaf canopy or following burial in soil. With
or without light, less than 2% germination occurred at 33 OC for each cultivar (data not presented). At 15 or 24 ï¿½C
in light, germination was greatest for 'Nanho Purple' followed by 'Dartmoor' (Figure 1). Germination of 'Black
Knight' and 'White Profusion' was consistently lower than that of other cultivars, regardless of temperature.
The results from these studies suggest that invasive properties of Buddleja may differ greatly among cultivars.
While long flowering periods, prolific seed production, lack of seed dormancy, and aggressive growth don't
constitute a plant as invasive, it does merit caution when utilizing this species in the Florida landscape and it
does support the need for suitable sterile alternatives.
Authors gratefully acknowledge the Florida Department of Environmental Protection for funding the initial
germination research on B. davidii and B. lindleyana and the University of Florida - IFAS Invasive Plant
Working Group for funding the cultivar evaluation study.
Anisko, T. and U. Im. 2001. Beware of butterfly bush. American Nurseryman July, 46-49.
Crawley, M.J. 1987. What makes a community invasible? In A.J Gray, M.J. Crawley, and P.J. Edwards
(ed.) Colonisation, Succession and Stability. Blackwell Scientific Publications, London.
Csurches, S. and R. Edwards. 1998. Potential Environmental Weeds in Australia: Candidate species for
preventative control. Biodiversity Group, Canberra ACT.
Dirr. M.A. 1998. Manual of Woody Landscape Plants. Stipes Publishing, L.L.C., Champaign, IL.
Hodges, A.W. and J.J. Haydu. 2002. Economic Impacts of the Floriculture Environmental Horticulture Industry,
2000. Economic Information Report El 02-3, University of Florida, Gainesville, FL.
Kay, M. and M.C. Smale. 1990. The potential for biological control of Buddleja davidii Franchet in New Zealand.
Forest Research Institute, Rotorus, New Zealand. Pp. 29-33.
Krumfolz, L.A. and S.B. Wilson. 2002. Varying Growth and Sexual Reproduction across Cultivars of
Ruell//ia brittoniana. Proc. Southern Nursery Assoc. 47:99-103.
Reichard S.H. and C.W. Hamilton. 1997. Predicting invasions of woody plants introduced into North
America. Conserv. Biol. 11:193-203.
Staples, G.W., D. Herbst, and C.T. Imada. 2000. Survey of invasive or potentially invasive cultivated plants
in Hawai'i. Bishop Museum Press, Honolulu.
Back to the Journal of Undergraduate Research
College of Liberal Arts and Sciences I University Scholars Program I University of Florida I| I l I UNIVERSITY of
ï¿½ University of Florida, Gainesville, FL 32611; (352) 846-2032.