Biological Control of Purple Loosestrife in North America
From: Exotic Pests of Eastern Forests, Conference Proceedings - April 8-10, 1997, Nashville, TN, Edited by: Kerry O. Britton, USDA Forest Service & TN Exotic Pest Plant Council
In recent years, interest in a biological method to control problem plants
in natural areas in the United States has grown (US Congress 1993). All
federal agencies must comply with standards to reduce the use and dependence
on chemical control of weeds. But, biological methodologies are not readily
available, nor have they been well-endorsed or financially supported. Despite
an excellent safety record (Harris 1988; Crawley 1989), skepticism concerning
the safety and effectiveness of exotic insect introductions for weed control
remains high among the general public, administrators, and even scientists.
The successful control of Hypericum perforatum (Huffaker & Kennett
1959) and others that followed, have demonstated that long-lasting, cost-effective,
environmentally sound and effective control programs, can be implemented.
But, despite an increase in the number of programs initiated, the ability
to select and to establish control agents has not progressed to a point
where the rate of success has improved (Crawley 1989). Basic questions about
the kind of herbivore species to introduce, impact of single and multiple
species herbivory, and release strategies, remain unanswered. The control
program targeting purple loosestrife (Lythrum salicaria L.), a Eurasian
wetland perennial responsible for the degradation of many prime wetlands
throughout temperate regions of North America (Thompson et al. 1987; Malecki
et al.1993), is intended to emphasize the need for research investigations
during pre- and post-release phases of the program.
The Control Agents
Detailed investigations in Europe began in 1986 with surveys for potential
control agents and investigations about their life-history, distribution,
impact, and host-specificity (Blossey 1993; Blossey et al. 1994a, b; Blossey
and Schroeder 1995; Blossey 1995b). Biological attributes of herbivores
(host specificity, fecundity, impact, etc.), have served as guidelines for
selection of control agents (Harris 1973; Goeden 1983); however, such characteristics
are often difficult to observe in the field. Therefore, species proposed
for introduction were selected based on information about (a) impact on
the target weed in the field, (b) host specificity, (c) distribution, and
(d) feeding niche on L. salicaria.
Six species were selected as the most promising control agents for further
investigations. These were a root-mining weevil, Hylobius transversovittatus,
attacking the main storage tissue of purple loosestrife; two leaf-beetles,
Galerucella calmariensis and G. pursilla capable of completely
defoliating individual plants and entire L. salicaria populations; a flower
feeding weevil Nanophyes marmoratus; a seed feeding weevil N.
brevis; and a gall midge, Bayeriola salicariae, attacking leaf
and flower buds.
Demonstrated host specificity is of overriding importance before any
control organism can be released. During the screening program for purple
loosestrife, we conducted various tests and compared results of different
methods (Blossey et al. 1994a, b; Blossey and Schroeder 1995). After initial
host specificity screening results became available, a questionnaire concerning
the potential impact of this program was sent to the Departments of Agriculture
and Natural Resources in 32 states (Blossey et al. 1994a). The questionnaire
asked for the occurrence, special concerns (rare or endangered) and ecological
importance of Lythrum alatum and Decodon verticillatus, two
plant species where some feeding by potential control agents had occurred.
The questionnaire asked whether respondents would favor a release of biological
control agents over a potential negative impact on D. verticillatus
and alatum. While the majority fav releases, responses ranged from
extreme opposition to enthusiastic support (Blossey et al. 1994a). Often
a split occurred between the two agencies in a state, with the most common
concern being lack of sufficient information to appropriately evaluate danger
to native plants. This, and the second most commonly expressed concern,
that the introduction of another exotic species might create another problem
similar to purple loosestrife, illustrate the necessity to assess and publish
the impact on target and non-target host plants after insects have been
released. Without scientific evaluation, the safety of biological control
will remain subject to doubt and if public concerns are not taken seriously,
suffer further restrictions. Conflict resolution will always be a part of
biological control, and only sound scientific analysis can offer guidance
to necessary decisions. For example, based on the available information,
one of the agents under consideration, B. salicariae, because of
a wider host range, was not proposed for introduction (Blossey and Schroeder
Based on the available knowledge at the time of introduction of the first
control agents in 1992, the following predictions emerged (Malecki et al.1993):
(1) all species will become established throughout the current range of
L. salicaria in North America; (2) the root feeder H. transversovittatus
and the two leaf-feeders G. calmariensis and G. pusilla, will
be most important in reducing large populations. The flower and seed feeders
will stabilize smaller populations, further reducing seed output in such
a way that not every disturbance will lead to a new outbreak of L. salicaria;
(3) combinations of agents will have greater control effect than any species
alone; (4) control of L. salicaria will be achieved more rapidly
in mixed plant communities with competition for space and nutrients; and
(5) purple loosestrife abundance will be reduced to 10% of its current level
over 90% of its range.
North America – 1992 to Present
Despite a long history of using insects for weed control and a considerable
improvement in procedures, only about 60% of released agents become established
(Crawley 1989). The influence of factors such as agent taxonomy, climatic
pre-adaptations, number of individuals released, numbers and timing of releases,
predators, and weather conditions in determining the fate of releases, lack
scientific evaluation and are largely observational (Crawley 1989; Lawton
1990). In the control program against L. salicaria, agents were collected
from climatically different source populations and releases occurred across
North America. Experiments were started to determine the best release procedure.
Agents became established across the entire continent regardless of the
source populations, the number of agent releases, time of release, stage
released, or whether caged releases or openfield releases were conducted
(Hight et al. 1995).
Harris (1981) proposed that biocontrol agents be considered stress factors;
the aim being to increase stress load until the balance is tipped towards
the disadvantage of the target weed population. Myers (1985) argued that
frequently good control has been achieved by a single agent replacing another
less successful one. Introducing several control agents could potentially
result in the suppression of a formerly successful species by a competitively
superior species (Ehler and Hall 1982). Crawley (1989), however, could not
find any evidence that multiple species introductions have ever led to the
replacement of effective agents by economically less successful ones. On
the contrary, agent combinations were recently reported to be more destructive
to plants than a single species alone (Fowler and Griffin 1995). Masters
et al (1993) found that spatially separated herbivores interact via their
common host plant. Root-feeders showed a reduced performance if their host
plant was simultaneously attacked by an above-ground herbivore. Above-ground
herbivores showed improved performance on plant individuals simultaneously
attacked by a root-feeder. Whether these interactions have any influence
on the success of weed biocontrol in systems where above- and below-ground
herbivores were released needs further study. We are currently conducting
these experiments for the L. salicaria-Galerucella-Hylobius
system. This is a good example of how an on-going biological control program
can benefit from simultaneously conducted basic research, and vice versa.
Mass rearing is often an integral part of a biological control program
since control agents are generally in short supply. A major concern has
been potential negative side effects of laboratory mass rearing (e.g., adaptations
to rearing conditions) and reduced quality of the produced insects (Hopper
et al. 1993). We have experimented with various field and laboratory mass
rearing techniques (Blossey and Hunt 1997), and found a reduced fecudity
and increased mortality associated with increasing duration of artificial
rearing conditions. We now prefer to mass produce all species outdoors for
one generation and allow subsequent overwintering. Between 1994 and 1996,
over 400,000 leaf-beetles were shipped to 26 different states and Canada
to collaborators in a wide range of organizations (universities, State Departments
of Agriculture and Natural Resources, National Wildlife Refuges, Bureau
of Reclamation, Tennessee Valley Authority, and Animal Plant Health Inspection
Service); many have started their own mass rearing program. We believe that
we need to be concerned about the quality of insects released, not the quantity,
and recommend outdoor mass rearings. Releasing fewer, but fitter, individuals
might be a much more successful approach and quality control should accompany
every mass rearing program.
Increased attention is given to follow-up studies to monitor target plant
and control agent populations. The lack of published evaluations might (hopefully)
reflect the lag time between releases and documented successes since the
biocontrol community has long agreed on the necessity of these studies (Schroeder
1983; Sheppard 1992). The future of biological weed control is intimately
linked to the demonstrated safety and efficacy of our programs. For example,
releases of control agents against L. salicaria in the state of Wisconsin
were only allowed once the Department of Natural Resources agreed on a monitoring
plan for insect and plant populations.
An important consideration is the many different ways to monitor insect
or plant populations. Our goal has been to develop standardized monitoring
guidelines sophisticated enough to allow valuable scientific evaluation,
but at the same time, simple enough to allow participation by wildlife managers
or their staff with little guidance. Preliminary versions of a monitoring
guide have been tested in 1995 and 1996, and a final version will be distributed
by the end of 1997.
The Magic Formula For Success?
A number of factors have contributed to the rapid growth of a coordinated
biocontrol effort for purple loosestrife in the United States. L. salicaria,
based on its rapid spread, projected range, and severity of impact, was
identified among the most harmful non-indigenous species in the United States
(US Congress 1993). This designation created interest for improvements in
management approaches, including biological control, across the entire continent.
From its inception, the biological control program against L. salicaria
has been a multi-agency effort. The overseas exploration by the International
Institute of Biological Control was conducted in association with the USDA
Agricultural Research Service (ARS), and the US Fish and Wildlife Service.
The initial success of the interagency effort led to the formation of
a scientific advisory group (Purple Loosestrife Working Group, [PLWG]),
with representation from several US federal and state agencies, universities,
and Canada. Since 1986, this working group provided continual guidance on
all aspects of our biological control program.
One of the major accomplishments has been to keep federal and state agencies
actively involved, informed through internal annual reports, and through
participation in decision-making processes. This broad-based involvement
has facilitated maintenance of secure funding since 1985. Particularly important
was the ability to pool resources from a variety of sponsors, thus, in the
absence of major grants, the cooperation across political and agency boundaries
has been extremely benefcial. Once the first insects became available in
1992, they were distributed to 7 states and to Canadian cooperators. Workshops
held in Colorado and Minnesota in spring 1993 introduced interested agencies
to life-history of control agents, mass rearing techniques, follow-up studies,
and monitoring techniques. In addition to regular meetings of the PLWG,
we now conduct annual planning meetings for the future of the control program.
Purple loosestrife is not an agricultural weed. People actively involved
in the control program are often resource managers, essentially a new audience
for biological weed control. Their willingness to participate in basic research
has enabled us to implement a scientific approach to the entire program
with the intention to improve biological control as a science. The leadership
provided by Cornell and the willingness to share research results has created
a unique cooperative environment that allowed the program to move forward
at a fast pace. Last, but not least, early results indicate that the selected
control agents are going to be effective.
Blossey, B. 1993. Herbivory below ground and biological weed control:
life history of a root-boring weevil on purple loosestrife. Oecologia 94:380-387.
Blossey, B. 1995a. A comparison of various approaches for evaluating
potential biological control agents using insects on Lythrum salicaria.
Biol. Contr. 5:113-122.
Blossey, B. 1995b. Coexistence of two competitors in the same fundamental
niche. Distrbution, Adult Phenology, and Oviposition. Oikos 74:225-234.
Blossey, B., and Hunt, T. 1997. Mass rearing methods for Galerucella
calmariensis and G. Pusilla (Coleoptera: Chrysomelidae), biological control
agents of Lythrum salicaria (unpublished manuscript).
Blossey, B., and Ehlers, R.U. 1991. Entomopathogenic nematodes (Heterorhabditis
spp. and Steinernema anomali) as potential antagonists of the biological
weed control agent Hylobius transversovittatus Goeze (Coleoptera: Curculionidae).
J. Invert. Pathol. 58:453-454.
Blossey, B., and Schroeder, D. 1995. Host specificity of three potential
biological weed control agents attacking flowers and seeds of Lythrum salicaria.
Biol. Contr. 5:47-53.
Blossey, B., Schroeder, D., Hight, S.D., and Malecki, R.A. 1994a. Host
specificity and environmental impact of the weevil Hylobius transversovittatus,
a biological control agent of purple loosestrife (Lythrum salicaria). Weed
Blossey, B., Schroeder, D., Hight, S.D., and Malecki, R.A. 1994b. Host
specificity and environmental impact of two leaf beetles (Galerucella calmariensis
and G. pusilla) for the biological control of purple loosestrife (Lythrum
salicaria). Weed Science 42:134-140.
Crawley, M.J. 1986. The population biology of invaders. Philosphical
Transactions of the Royal Society. London B. 314:711-731.
Crawley, M.J. (1989). The successes and failures of weed biocontrol using
insects. Biocontrol News and Information 19:213-223.
Ehler, L. E., and Hall, R.W. 1982. Evidence for competitive exclusion
of introduced natural enemies in biological control. Environmental Entomology
Fowler, S.V., and Griffin, D. 1995. The effect of multi-species herbivory
on shoot growth in gorse, Ulex europaeus. In: Proceedings of the VIII International
Symposium on the Biological Control of Weeds, pp. 579-584 E.S. Delfosse
and R.R. Scott (eds). February 2-7, 1992, Canterbury, New Zealand. DSIR/CSIRO,
Goeden, R.D. 1983. Critique and revision of Harris scoring system for
selection of insect agents in biological control of weeds. Protection Ecology
Harris, P. 1973. The selection of effective agents for the biological
control of weeds. Canadian Entomologist 105:1495-1503.
Harris, P. 1981. Stress as a strategy in the biological control of weeds.
Pp. 333-340 In: Biological Control in Crop Production. G.C. Papavizas (ed.).
Allanhead, Osman and Co., Totowa, New Jersey.
Hight, S.D., Blossey, B., Laing, J., and DeClerck-Floate, R. 1995. Establishment
of insect biological control agents from Europe against Lythrum salicaria
in North America. Environmental Entomology 24:967-977.
Hopper, K.R., Roush, R.T., and Powell, W. 1993. Management of genetics
of biological control introductions. Annual Review of Entomology 38:27-51.
Huffaker, C.B., and Kennett, C.E. 1959. A ten-year study of vegetational
changes associated with biological control of Klamath weed. Journal of Range
Lawton, J.H. 1990. Biological control of plants: a review of generalizations,
rules, and principles using insects as agent. Pp. 3-17 In: Alternatives
to the Chemical Control of Weeds. C. Bassett, L.J. Whitehouse, and J.A.
Zabkiewicz (eds.). Proceedings of an International Conference, Rotorua,
New Zealand, July 1989. FRI Bulletin 155, Ministry of Forestry.
Malecki, R.A., Blossey, B., Hight, S.D., Schroeder, D., Kok, L.T., and
Coulson, J.R. 1993. Biological control of purple loosestrife. Bioscience
Masters, G.J., Brown, V.K., and Gange, A.C. 1993. Plant mediated interactions
between above- and below-ground insect herbivores. Oikos 66:148-151.
Myers, J.H. 1985. How many insects are necessary for successful biocontrol
of weeds? In: Proceedings of the VI International Symposium on Biological
Control of Weeds. Pp. 77-82. E.S. Delfosse (ed.). August 19-25, 1984, Vancourver,
Canada. Agriculture Canada, Ottawa.
Schroeder, D. 1983. Biological control of weeds. In: Recent Advances
in Weed Research. Pp. 41-78
W.E. Fletcher (ed.). Commonwealth Agricultural Bureau, Farnham Royal,
Sheppard, A.W. 1992. Predicting biological weed control. Tree 7:290-291.
Thompson, D.Q., Stuckey, R.L., and Thompson, E.B. 1987. Spread, impact,
and control of purple loosestrife (Lythrum salicaria) in North American
wetlands. U.S. Fish and Wildlife Service. Fish and Wildlife Research 2.
US Congress, Office of Technology Assessment. 1993. Harmful non-indigenous
species in the United States. OTA-F-565, Washington, D.C., US Printing Office.
Wapshere, A.J. 1985. Effectivensss of biological control agents for weeds:
Present Quandries. Agriculture, Ecosystems, and Environment 13:261.