Objectives

This module is designed to introduce the concept of biological control, in all its forms, with an emphasis on its practice in Australia. At the end of this topic you will:

• understand the aims of biological control;
• be familiar with the ecological basis behind biological control;
• appreciate the advantages and disadvantages of biological control over other types of weed management;
• be able to explain the four main types of biological control (broad-spectrum, conservative, classical and inundative) with an emphasis on classical biological control; and
• describe the steps involved in a classical biological control project and the practices that should be employed to increase the likelihood that a project is successful

Module Outline

• Introduction
• Principles to biological control
• Biological control in Australia
• Myths about biological control
• Types of biological control
• Classical biological control
• Examples of classical biological control
• Inundative or augmentative biological control
• Examples of inundative biological control
• Conclusion
• References and further reading
• Press release

Go to Activity 13-1

---

Introduction

Definition

Biological control can be defined as: ‘the utilisation of organisms for the regulation of host plant densities’.

This is a basic and all inclusive definition of biological control. The term ‘organisms’ may refer to insects, pathogens , other animals and even other species of plants.

The aim of biological control

Biological control aims to reduce the density of a target weed to a level below that which causes economic or environmental damage and to maintain the weed density at this lower level (Figure 13.1).

The acceptable weed density may vary depending on the species of weed and the situation in which it occurs.

Biological control is achieved through the introduction of one or several of the natural enemies of the target weed.

Biological control is used to maintain an ecological balance between a weed and its natural enemies in its introduced range. Unlike some other control methods, complete eradication of the weed species is NOT the objective of biological control.

Figure 13.1 The effect of the application of a successful biological control agent on the density of a target weed over the course of time.

Principles to biological control

The ecological basis of biological control

Plants evolve in communities along with other plants, herbivores and diseases which limit their population size.

Thus, an ecological balance is developed in natural communities between plants and their competitors, predators and diseases. Some of these herbivores and diseases may be highly host-specific (i.e. they attack only a single species).

Plants are often moved to new areas as a result of human activities.

Plants may be introduced deliberately for ornamental or for agricultural purposes (e.g. as garden plants, crops or pasture species). They may also be introduced accidentally (e.g. as contaminants in grain and other produce, in clothing, or on transport vehicles).

However, the organisms that limit plant population size are usually not present in these new areas.

The introduction of a plant species into a new location disrupts the ecological balance at that location because the organisms that limit its population in its natural environment are generally not present at the new location and usually not introduced along with it.

This may result in the introduced plant out-competing native or other desirable species and cause the introduced plant to be classified as a ‘weed’.

A species is usually considered to be a serious ‘weed’ when it begins to cause noticeable environmental or economic damage.

Importation and release of an introduced plant’s natural enemies may decrease its competitive ability and allow the native or desirable vegetation to recover.

By introducing the natural herbivores and diseases of an introduced plant into the new location, the natural balance in that environment may be restored.

Advantages of biological control

Biological control is usually self-sustaining, does not require intervention, and provides long term control.

Unlike herbicides, which may need to be applied each year, biological control agents are usually self-sustaining once they have become established at a site and will generally provide control indefinitely.

Biological control is useful for weeds that cannot otherwise be managed (e.g. environmental weeds).

For example, some weed species infest large areas that have little economic value and control by other methods (e.g. using herbicides) is not cost effective. Control using other methods may also damage environmentally sensitive non-target species, whereas biological control agents are usually very host-specific.

Biological control can provide high benefits for the overall costs provided.

Even though it may be expensive to set up a biological control program, there is little additional cost to provide control over a very large area. The agents usually disperse naturally to cover the entire weed infestation and therefore the amount of control obtained for the initial cost can be very substantial.

Biological control generally only affects the target weed.

Because biological control agents are thoroughly host-tested and are usually specific to the host species, normally only the target weed is affected. Where non-target effects have occurred, they have usually been foreseen and the risk has been considered acceptable (i.e. the benefit greatly outweighs the risk).

Biological control is environmentally sound as it reduces pesticide use, environmental contamination and health risks to people.

Effective biological control often replaces chemical control practices using herbicides, or reduces the need for such practices.

Herbicides usually have negative environmental impacts and can also have residual effects on the environment. As biological control has no such impact or residual effects, it is considered to be much more environmentally sound.

The use of herbicides can also be a health risk to the primary producers that use them. They are also a potential health risk to other people and animals that may come into contact with these chemicals.

Disadvantages of biological control

Not all weeds are suitable targets for biological control.

Some weed species are unsuitable for biological control because they are closely related to important crop or native plant species. In this situation potential control agents may be rejected as they could also attack these important non-target plant species. For example, fireweed (Senecio madagascariensis) is very closely related to the Australian native plant known as variable groundsel (Senecio lautus) and it has been difficult to find a suitable agent that does not attack both species. Other species may not have any natural enemies that are sufficiently host-specific or damaging to be potential candidates for biological control.

The initial cost of a biological control project may be quite high.

The initial cost of setting up a biological control program can be quite expensive due to the requirement for foreign exploration, quarantine, etc. A typical program from conception to completion could cost more than AUS$ 1 million.

It can take a long time for a biological control project to start to have an impact, and therefore such projects are not suitable for situations requiring fast-acting or short-term weed management.

It may take several years to identify, test, import, rear and release a suitable biological control agent, and sometimes several agents may be required. Also, once a species has become established it can take several years to have a significant impact. For example, the leaf-feeding beetle Zygogramma bicolorata was released in 1980 to control parthenium weed (Parthenium hysterophorus), but significant damage was not reported until the early 1990’s.

Generally biological control projects cannot be sold, so they do not attract industry support, and usually require significant government support.

Apart from mycoherbicides and similar products, biological control agents cannot be sold and so they do not attract companies interested in profits. Therefore significant government funding is usually required to undertake a biological control program.

Released biological control agents may not necessarily contribute to the management of the weed.

The overall success rate of biological control is thought to be less than 30% world-wide. Agents may be unsuccessful for several reasons (e.g. they are difficult to rear, they will not become established in the new location, they are not sufficiently damaging to the target species, etc.). Hence, a great deal of time and money can be spent for very little benefit.

Biological control in Australia

Australia is one of the world leaders in weed biological control.

Only the USA has undertaken more biological control projects and introduced more biological control agents than Australia. Australia also has one of the highest success rates in biological control projects.

Over 60 weed species have already been targeted in Australia.

In almost half of these cases some degree of control has been achieved, and for 19 weed species a substantial degree of control has been achieved. These figures undoubtedly underestimate the effectiveness of biological control programs in

Australia, as many projects are still ongoing and control may be achieved by the introduction of new agents, or by the increased effectiveness of previously released agents.

The majority of Australian projects have targeted weeds of pastures, rangelands and natural environments.

Until recently little effort has been made to target weeds of intensely managed crops using biological control. This is because regular cultivation and rotation of crops was perceived to be detrimental to agent survival and impact. However, mycoherbicides are now being considered in some instances and classical biological control in a cropping situation has been achieved. For example, the control of some forms of skeleton weed (Chrondrilla juncea) with a rust fungus.

These Australian projects have involved a broad range of plant life forms, however relatively few grass weeds have been targeted.

Herbs and forbs have been most commonly targeted, reflecting their relative abundance among problem weeds. Various programs have also targeted succulents, shrubs, vines and trees. Programs against succulents have been particularly effective, for example the program against prickly pears (Opuntia spp.). The grasses are a large and very similar family of plant species and little work has been conducted on them due to the fear that sufficiently host-specific agents would be difficult to find. However, research into controlling grass weed with pathogens, which tend to be more stringently host-specific, is becoming more common.

Myths about biological control

Biological control can be dangerous.

It is a fact that the cane toad was introduced into Australia in the 1930’s to control the sugar cane beetle. It failed to have any significant impact on the sugar cane beetle and has itself become a very serious pest.

However, the cane toad was introduced prior to the introduction of host-range testing of agents and against the advice of biological control researchers. Today agents are subject to stringent testing before they can be imported and released. The results of this testing must pass the scrutiny of a panel of scientists.

Biological control is a ‘magic bullet’.

The dramatically effective control of prickly pear (Opuntia stricta) over very large areas by Cactoblastis cactorum in the 1930s is often used to portray biological control as a magic bullet.

However, in reality, it took many years of background research, and the release of several agents, for one of these to be an almost ‘overnight’ success. On the other hand, the biological control of lantana (Lantana camara) was first investigated in the early 1900s and is still yet to be achieved to this day, despite numerous projects in Australia and overseas and the release of more than twenty agents into Australia for its control.

The control of a weed is often only achieved by a combination of several complementary biological control agents. Complementary agents that attack different parts of the plant during different stages of its life history may be required to provide adequate control (e.g. a seed-feeder, a root-borer, a stem-borer and a defoliator).

Types of biological control

Introduction

There are four recognised types of biological control:

• broad-spectrum
• conservative
• classical, and
• inundative or augmentative.

The first two types, broad-spectrum and conservative, are either rarely used or often not strictly considered a true form of biological control, and hence they are only covered briefly here. The last two types, classical and inundative/augmentative, are by far the most common and well known forms of biological control and are described in greater detail.

Broad-spectrum biological control

The broad-spectrum approach involves the use of herbivores to manage weeds through vegetation removal. Basically, polyphagous herbivores (i.e. those that feed on a very wide range of species) are introduced to an area and they remove most of the vegetation, including the weeds. This type of biological control is only useful in heavily infested areas (i.e. areas dominated by weeds) and is only a short-term management option.

It may be applied to both aquatic and terrestrial habitats, for example using grass carp to control underwater plants or using goats to control thistles and woody weeds in pastures. In Australia, broad-spectrum biological control is more often considered a form of grazing management rather than a type of biological control.

Conservative biological control

Conservative biological control involves the management and protection of existing natural enemies to maintain or enhance their impacts on weeds. The idea of this form of biological control is to maintain or enhance the beneficial impact of natural enemies.

One example of this would be the use of the natural enemies of the native water plant Hydrilla verticillata (viz. the hydrilla leaf-mining fly; Hydrellia balciunasi and the hydrilla stem-boring weevil; Bagous hydrilla), employed in the control of the closely related, but introduced invasive weed elodea (Elodea canadensis).

Another example would be the re-distribution of native scale insects to areas where Chinese scrub or Sifton bush (Cassinia arcuata), a native plant, has become invasive in pasture situations in eastern Australia.

This approach of biological control is best suited to invasive native plants and has not been widely used in Australia as most weeds are introduced. However this is beginning to change, with more and more native plants becoming invasive outside their native ranges. Hence this form of biological control may become more important in the future.

Classical biological control

Classical biological control involves the use of introduced natural enemies against invasive alien plants. It is the oldest, most common, and most effective type of biological control. Once the agents are established in the environment, no further intervention is required. This form of biological control is discussed in detail in a following section.

Inundative biological control

Inundative biological control involves the selection and release of a biological agent against a weed, as an inundative ‘biological herbicide’. It is a newer form of biological control, less common than the classical approach and has been attempted in a few locations around the world. This form of biological control is discussed in detail in a following section.

Table 13.1

Classical biological control	Inundative biological control
the regulation of populations of a target alien weed by the introduction of natural enemies from the weed's native range.	the regulation of populations of a target weed by the repeated introduction of natural enemies (often fungi) of the weed.
based on the ecological principles that herbivores can limit the populations of a plant and that within the native range of the plant there will have evolved a specialised guild of herbivores, some of which are highly host-specific.	often applied over the whole weed population in a similar fashion to the application of a herbicide
aims to restore the ecological balance between a plant and its herbivores in the introduced range and to reduce the weed populations to lower, less damaging level	aims to introduce large doses of a natural weed parasite and to reduce the weed populations to lower, less dangerous level.

Table 13.2 The traditional view on biological control strategies appropriate for use in the various natural and agricultural production systems

Form of biocontrol	System
	Natural	Pasture	Horticultural	Crop
Classical	•	•
Inundative			•	•

Back to top

Classical biological control

Introduction

A schematic representation of the desired course of events in a classical biological control project is shown in Figure 13.2.

Figure 13.2 A schematic representation of the desired course of events in a classical biological control project.

This Figure gives a stage by stage outline of the desired course of events in managing a weed incursion with a biological control program. The stages are as follows:

Introduction of weed and the development of a weed problem

Initially a weed species will become introduced to a new location, whether by accident or deliberate actions. The time taken for a weed to develop into a serious threat may vary depending on the ecology of the species and its life-history, and it may be also dependant on certain climatic events.

The damage threshold level is reached

A species usually becomes a problem once its population reaches a damage threshold level. This is the point where its population is great enough to cause some form of damage (e.g. economic or environmental). In a cropping or pasture system this is usually an economic threshold, where the species begins to have a detrimental effect on the productivity of that system. In a natural system this is usually a threshold above which the species causes environmental damage (e.g. it may reduce the species diversity at a location or endanger a species to the point of extinction). Some species may only be damaging during some seasons (particularly annual species) while others may have an impact which varies little between seasons (particularly perennial shrubs and trees).

The example illustrated in Figure 7.2 is most representative of an annual weed species.

An equilibrium position before the release of the biological control agent is reached

If left uncontrolled, or inadequately controlled, a weed’s impact will eventually reach an equilibrium level where it is causing some form of economic or environmental injury, and this may vary from season to season.

The introduction of the biological control agents

When a classical biological control program is implemented, and an agent is released, there is generally not an immediate impact on the weed. A significant amount of time is usually required for the agent to build up its population before it becomes damaging to the target weed. The agent’s population will normally reach a climax, where it becomes extremely abundant, as the population of the host species is very high at this stage. The agent will then usually have a dramatic effect on the target weed species and significantly reduce its population. Once the population of the target species becomes significantly reduced this will then lead to a significant reduction in the agent’s population (in some cases this agent over-population may lead to it briefly attacking non-target species as it searches for an alternative food source).

An equilibrium position after the release of the biological control agent

Eventually a new equilibrium position will be reached, hopefully one where the agent is limiting the population of the target species to a point where it no longer causes economic or environmental injury. The agent will not cause the eradication of the species, but it may make eradication possible using other control methods. From this point onwards the population of each species is usually closely linked to that of the other, and there may be variations in the population of both species due to seasonal factors.

The steps in a classical biological control project

There are a number of possible steps involved in a classical biological control project. The list below includes some that are always relevant and others that may not always be required. Each step will be discussed individually in the following notes:

• project initiation;
• project approval;
• exploration in the weed’s native range (foreign exploration);
• ecological studies – introduced range;
• ecological studies – native range;
• evaluation and selection of potential agents;
• host-specificity studies;
• obtain approval to release agents;
• importation of agents;
• rearing and release of agents;
• distribution of agents; and
• evaluation and monitoring.

Project initiation

Prior to the commencement of a biological control project, background research first needs to be conducted. Any relevant literature concerning the target species and its natural enemies should be reviewed. Background research is important as it may reduce the chances of spending large amounts of time and money duplicating research that has already been conducted by other researchers in the past. It may also help determine if the species is a suitable candidate for a biological control program. In addition, it will also familiarize the people involved in the project with the target species and its natural enemies.

The existing knowledge discovered should then be compiled and documented. Compiling and documenting this information may be invaluable, especially for any additional staff that are added to the project at a later stage, or if there is a change in personnel for some reason. It may also be of use to future biological control projects on this species or to projects attempted in other countries.

Project approval

The background research discovered can then be used to obtain approval from the relevant authorities for the commencement of a biological control project on a particular weed species.

In Australia there is a ‘target species list for biological control’, which lists those weed species that have been approved as candidates for biological control. Prior to 1985 additions to this list were approved by the Australian Weeds Committee. Approvals are now governed by the Natural Resource Management Standing Committee under the Biocontrol Act.

If the species that is to be the target of a project is not already on the approved target list, the background research can be useful in putting forward a case to the relevant authorities for its addition to the list.

The next step is to obtain the required funds for conducting the project. The information discovered can also be used to help put forward an excellent case to prospective funding bodies for support for the work. It may provide convincing evidence about the necessity for the project and its chances of success.

The following three steps (i.e. Foreign Exploration, Ecological Surveys in the Introduced Range and Ecological Surveys in the Native Range) should be conducted simultaneously if at all possible. This will reduce the time and money required to conduct the project, and the information gathered from each of these steps will hopefully be mutually beneficial to the other steps.

Foreign exploration

Foreign exploration is one of the most vital and expensive stages of a biological control program, and good planning is required to make the search for potential agents as effective as possible.

The native range of the weed might be the first area explored for potential biological control agents. However, before this can occur the native range needs to first be accurately determined. This is important because some potential agents may only be present over a portion of the target species’ native range and they may be missed if only part of the native range is investigated. In particular, the parts of the weed’s native range that are most eco-climatically similar to the areas infested should be determined and searched.

The native range of the weed needs to be searched by suitably qualified field based staff (e.g. entomologists and pathologists) for the natural enemies of the target weed.

This work may be assisted by pre-existing information gathered from the literature prior to the commencement of field studies. The exploration should focus on the regions that are most eco-climatically similar to the infested areas. Introducing agents from such locations will increase the chance that they will establish in the new environment, because they are already well-adapted to those eco-climatic conditions.

Ecological surveys

The introduced (or exotic) range of the weed should then be surveyed to obtain a thorough understanding of the biology and ecology of the weed in its new habitat. This is useful in deciding whether biological control is an appropriate strategy and what types of species might be best suited as control agents. For example, the life-history of the species (i.e. whether it is annual, biennial or perennial) and its means of dispersal can be critical to determining a biological control strategy. Other important ecological information may include: how the species reproduces; how important and long-lived its seed bank is; how it is affected by climatic conditions, and how competition affects its growth. This research is best conducted in the introduced range for two main reasons: it is usually less expensive than foreign research; and there may be different forms or biotypes of the weed in its natural range that are different to those that are required to be controlled in the introduced range.

During the ecological survey in the introduced range it is a good idea to record what fauna and other organisms are attacking the weed, and determine the origin of these organisms. A list of the species that are already attacking the weed in its introduced range should also be determined, as some of its natural enemies may already be present and some native species may have become adapted to attacking the weed. This will prevent wasting time and money studying potential control agents in foreign countries that are already present in the introduced range.

If at all possible the native range of the weed should be surveyed so information is gained on its ecological behaviour in its natural environment. Studying the weed in its natural environment and measuring the impact of its natural enemies on its population dynamics may also provide valuable information as to which species or which types of agents may be the most suitable candidates for controlling the weed. Any variation in the weed should be studied, and research into potential agents should focus on those that attack the same form of the weed that is present in its introduced range.

At the same time it would be beneficial to study its natural enemies. Briefly studying these potential control agents and determining their native range may be useful in determining which species will be well-adapted to the eco-climatic conditions present in the weed’s introduced range.

It is also important to determine the host range of these natural enemies. Some study of the host range of these natural enemies (i.e. at least whether they are generalists or relatively specific to the target) may help to refine a list of potential candidates (i.e. eliminate some less-specific species and highlight those that have the most potential as candidates).

Evaluation and selection of potential agents

The information gathered can then be used to determine which agents are the most suitable candidates for introduction.

The information that has been gathered from the Foreign Exploration Studies and the Ecological Studies (in both the weed’s foreign and introduced range) can be compiled and used to evaluate which species will have the highest chance of success as biological control agents. This is important as it is normally only possible to expend the limited financial resources of a project on testing and introducing one or at most a few species. Therefore, it is essential to accurately determine which agents will be most suitable and to focus on these species in order to provide the greatest chance of success for the project. Sometimes the agents are ranked in order of suitability and then the species with the highest rankings may be tested in order, until the funding for a project runs out.

These potential agents must be sufficiently damaging to have an impact on the target weed.

There is no point introducing an agent if it does not have a significant impact on the target weed. This should have the highest priority when determining the potential suitability of candidates. When working with pathogens in particular, it is important to determine which strain is most virulent against the target weed in order to increase its chances of success as an agent (the most virulent strains of a pathogen against a target weed are also often the most host-specific as well).

These potential biological control agents must be able to exist in the weed’s introduced range.

While some agents may be able to adapt to Australian conditions over time, it is better to focus efforts on agents that are already well-adapted to the climatic conditions in the weed’s introduced range.

These potential biological control agents must exhibit some degree of host-specificity.

If the potential agent is unlikely to be given approval to be imported and introduced into Australia because it may attack non-target species, there is little point spending resources on it.

There are also other factors that should be taken into account when deciding on the suitability of a potential agent.

They are only applicable in a small number of cases:

• The candidate species must be capable of reproducing rapidly enough to build up enough numbers to damage the target weed (this is particularly important when the weed is an annual or ephemeral species).
• It must be possible to develop suitable rearing methods for the candidate species, otherwise it will not be possible to release enough of the agent to achieve successful establishment in the field.

Host-specificity studies

The host-specificity of the agent should be determined by preparing an appropriate list of test plants.

This list will consist of economically important plants and native species that are closely related to the target weed. It usually contains several plants within the same genus, followed by some representatives from closely related genera, then other species within the same family, and then occasionally others that are increasingly less related to the weed. Sometimes quite unrelated plants may also be included if they have chemical or morphological similarities with the target weed. When preparing such a list, consultation with a plant taxonomist familiar with the taxonomic group in which the target species belongs is usually necessary.

The host-specificity should be submitted for approval from the relevant authorities and making any modifications to this list that are deemed necessary.

The relevant authorities that review this list may require additional species to be added to it. It is necessary for these additional species to be included in the host-testing procedure if approval to import the agent is to be sought from these authorities at a later date.

The next step is the conducting of the rigorous host-testing trials.

The trials that are required to be conducted vary depending on the type of agent to be host-tested (i.e. insects and pathogens have different host-testing requirements). Insects usually undergo cage tests, where the agent is confined in a cage with the test plant (i.e. no-choice tests) or with the test plant and the target weed (i.e. choice tests).

The feeding, laying of eggs (i.e. oviposition) and survival of the agent is observed in these tests. Pathogens are treated quite differently, with the test plants being exposed to the pathogen and the reaction of the test plant (and the level of infection) being observed microscopically as well as macroscopically.

Approval for introduction of agent

Once an agent is deemed to be suitable for release, approval for its introduction must then be obtained by a two step process.

Firstly, reports are submitted to the relevant authorities.

These reports outline the benefits of control of the target weed, and the results of the host-testing procedures, and are circulated to 21 reviewers from a variety of organizations. Approval is only given once all reviewers accept them. They must contain detailed information about the identity, origin and biology of the agent and any potential interactions it may have with other beneficial organisms. It is important to include information about the damage caused by the target weed and the benefits of control as well as the results of the host-testing procedures. A control agent may still be introduced even if it does attack beneficial non-target species, as long as it can be demonstrated that the likely benefits of its introduction significantly outweigh any negative impacts and that any risks are acceptable.

Secondly, additional experiments are conducted, and other issues needing to be resolved are resolved.

The reviewers of such proposals may accept or reject them as they are, or they may require additional experiments to be conducted. The results of these further studies will then need to be reviewed again before final approval can be given for the release of the agent.

Importation of agent

If the host-specificity studies are to be conducted in Australia, then approval to import the agent and conduct these studies in a quarantine facility will need to be obtained prior to their being conducted.

The importation and release of agents is regulated by The Quarantine Act (which is administered by the Australian Quarantine and Inspection Service – AQIS) and the Wildlife Protection Act (which is administered by Environment Australia). Permits for release are required from both bodies.

Once an agent has been approved it can be imported for release, but only certified clean material should be obtained for importation. This reduces the risk of importing any other parasites or pathogens along with the agent.

Once imported the agent must be initially kept in quarantine to ensure the elimination of any parasites or pathogens before it is released. Even though certified material has to be imported, there is still the requirement to rear the agent through at least one generation in quarantine to ensure that it is not carrying any parasites or diseases. This will not only prevent a potential new threat to non-target organisms, but ensure that the effectiveness of the agent is not reduced by these organisms.

Rearing and release

The most appropriate procedure for the mass rearing of the agent needs to be determined and then carried out. A quick, inexpensive method for mass rearing the agent may be necessary if adequate agent material for successful releases is to be made.

The agent can then be released at the most suitable sites in the field, ensuring that it has the best chance of survival and establishment. The nature of material that is released will depend on the type of agent involved and its life cycle. With insects this most commonly involves the release of adults (directly or in cages) or larvae/pupae contained within potted plants. Pathogens are normally released by transporting infected plants into the field or by spraying an inoculum of fungal spores onto plants in the field.

Distribution of agents

Once established, the agent then needs to be distributed over the introduced range of the target weed. Some agents will naturally disperse rapidly and widely, while others may require active distribution because they have a slow dispersal rate.

For example, the stem-galling moth Epiblema strenuana (which attacks parthenium weed and annual ragweed) dispersed naturally over very large areas of Queensland and New South Wales in just a couple of years. However, the stem-boring weevil Listronotus setosipennis (which also attacks parthenium weed) was restricted to a very small area of central Queensland for many years.

Agent redistribution program may require, or be enhanced by, collaboration with other institutions, government authorities and community groups. For example, the redistribution of the leaf-feeding beetle Zygogramma bicolorata (which attacks Parthenium weed) was greatly enhanced by a community group, known as the Parthenium Action Group (PAG). This program included establishing and maintaining ‘colonies’ (i.e. nursery sites) of the agent on its target weed for later distribution.

Evaluation and monitoring

After the agent has been released and distributed, field evaluation studies will then be conducted to determine if and/or where it has become successfully established in the field.

It is important to determine if the agent is successfully established, as this will confirm whether further rearing and releases of the agent are required or not.

Monitoring is then undertaken to determine its rate of spread and new distribution within the target weed’s introduced range. Such monitoring will help to determine whether active redistribution of the agent is required, and where this intervention is most necessary.

Some estimation should then be made of the effect that the agent is having on the target weed. This is important for a variety of reasons. Firstly, it can provide positive evidence as to whether the agent has been successful – partially or totally – or not, or whether some other factor has been involved. This will determine if further research and the introduction of other agents is required or not. Secondly, this information can be used to document the benefits of the program (especially if only partial success is achieved), which can be used to justify the continuation of the program or other biological control programs.

At the same time, some effort should be made to determine if the agent is having any unforeseen effect on non-target species.

Achieving successful classical biological control

The choice of agents and the methods used to rear, release and distribute them are all critical to the success or failure of a classical biological control project. With the experience that has been accumulated over the years, and with more rigorous methodical planning of biological control programs these days, there is a greater likelihood of choosing a suitable agent. Experience has also led to better processes for rearing, releasing and distributing of agents.

Equally important, but less often recognized, is the need to monitor and evaluate agent survival and impact so that the degree of success can be quantified, and the reasons for failure understood.

As mentioned previously, this is important to determine if further work is required. As the benefits of biological control are often not immediately obvious and are usually spread over a vast area, it is important to document them in order to give an accurate representation of the true value of such projects (e.g. using economic analyses, cost/benefit ratios, etc.).

Understanding the reasons for failure will prevent similar results occurring in the future, and hopefully result in an increase in the success rate and effectiveness of biological control programs.

Reading ‘Classical biological control’ (pages 161-180) in Australian Weed Management Systems. ‘Community involvement in the distribution and evaluation of biological control agents: Landcare and similar groups in Australia’ by DT Briese and DA McLaren 1999, Biocontrol News and Information, vol. 18, pp. 39-49.

Examples of classical biological control

Alligator weed (Alternanthera philoxeroides)

Alligator weed is a semi-aquatic weed native to South America. It can grow along the edges of water bodies, as mats of vegetation on the water surface, or as a terrestrial plant in damp environments. It is a serious weed in other parts of the world, but it is not yet very wide-spread in Australia. However, due to its potential threat it has been deemed to be on the 20 Weeds of National Significance (WoNS).

The alligator weed flea-beetle (Agasicles hygrophila) was introduced into Australia from South America. This species has a short life cycle of only 30 days and both adults and larvae feed on alligator weed.

This agent has successfully controlled alligator weed in some parts of Australia and the USA. The insect is very effective against the aquatic form of the weed, especially

in temperate regions, however it is not as effective against the terrestrial form. Therefore another agent may need to be released to fill in this gap in the control of this species.

Other examples of classical biological control in Australia

There are a number of examples of successful projects including those on skeleton weed (Chondrilla juncea), rubber vine (Cryptostegia grandiflora), salvinia (Salvinia molesta), water hyacinth (Eichhornia crassipes), annual ragweed (Ambrosia artemisiifolia) , Paterson’s curse (Echium plantagineum) and ragwort (Senecio jacobaea).

In addition, there have been a few examples of unsuccessful projects, including those on creeping lantana (Lantana montevidensis), mistflower (Ageratina riparia), spiny emex (Emex australis) and common heliotrope (Heliotropium europaeum).

Software Weed Biocontrol Select the ‘Weed Biocontrol’ CD and run it on your computer. Go through the tutorial until you understand the operation of the CD. You can browse the information, photos and videos that explain the process of biological control of weeds. Pay particular attention to the steps involved in the process. This software product is available from the UQ Gatton Library.

Back to top

Inundative or augmentative biological control

Introduction

Inundative biological control involves the mass production and release of natural enemies in high doses at regular intervals to reduce populations of the targeted weed.

Pathogenic micro-organisms (i.e. those that cause plant disease) are most commonly used in inundative biological control (e.g. fungi, bacteria, viruses etc.). Inundative biological control is best applied at times when the plants are most susceptible to damage.

These agents are normally applied to the weeds in a similar manner to conventional chemical herbicides, and so they are often termed ‘bioherbicides’.

Unlike classical biological control programs, this method of biological control is most often developed by industry and sold commercially (i.e. just like herbicides). Due to the characteristics of ‘bioherbicides’, they are most commonly employed in cropping situations.

Advantages of inundative biological control

As large quantities of the agent are applied to a given area it does not require a lengthy time period for the agent to reproduce and disperse before it has an impact.

As cropping systems usually involve short-lived seasonal crops, they require relatively short-term intervention and control. This is the main reason that bioherbicides are most commonly employed in such situations.

The agent does not have to be as host-specific. Generalist agents (i.e. those that have a wide host range) can be used in this situation because they are being applied in vast quantities to the target species (note – as long as they do not also attack the non-target crop species). Therefore, indigenous or naturalised pathogens are more commonly used in inundative biological control because such agents have the advantage of requiring no testing, importation or approval prior to their use.

Disadvantages of inundative biological control

Inundative biological control only provides short-term control, and because it is not self-sustaining it may need to be reapplied at a later date. For example, in cropping situations bioherbicides usually need to be reapplied each season (unless the organisms can survive for long periods in the soil), and therefore there is an ongoing cost involved.

The formulation of the agent may have a limited shelf life (i.e. it may lose its effectiveness within a short period of time). This is a disadvantage because it means that any bioherbicide ‘product’ that is not used in the first season is wasted, and additional supplies will need to be purchased or obtained for future seasons.

Procedures: mass production

If an organism is to be used as a bioherbicide a cost-effective technique first needs to be developed for its mass production, particularly if it is to be sold commercially. The two most common methods of mass production of bioherbicide agents are submerged culture fermentation and solid substrate fermentation.

The submerged culture fermentation technique is the most common and best method for the mass production of bioherbicide agents.

The solid substrate fermentation technique is less desirable, but is sometimes the only one available to produce the bioherbicide.

Procedures: formulation and application

A formulation and method of application of the ‘bioherbicide’ must be developed which maintains the activity of the agent and results in the optimum infection of the target weed. The formulation is the mixture of the bioherbicide agent and the carriers, spreaders or other materials that are required to improve the storage, mixing and/or application of the product. Most formulations are either liquid-based or granular. Agents can also be formulated as dusts, but these are not yet preferred because they are vulnerable to movement away from the target plants by wind.

Procedures: submerged culture fermentation

Submerged liquid culture fermentation involves the use of a fermentor or a ‘bioreactor’. Here the agent is grown in a liquid culture medium containing nutrients for its growth. The harvested products of this fermentation process may either be fungal mycelium or spores.

Fungal mycelium can be homogenised or separated from the culture media and it may also be treated to induce sporulation (i.e. spore production). More commonly spores are the harvested product, and they are usually separated from the mycelium by filtration and centrifugation.

Growth of the agent can be optimised by adjusting the balance between nutritional elements, temperature and aeration in the culture vessel. It is important to get the growing conditions right for optimal growth and spore production of the agent. For example, the carbon concentration and the ratio of carbon to nitrogen can strongly influence the amount and type of spores produced in submerged culture fermentation.

Procedures: solid substrate fermentation

Solid substrate fermentation involves the growth of bioherbicide agents on plant materials. Large-scale solid substrate fermentation systems are generally not readily available or economically feasible for commercial industry. They are more common in developing countries, where labour costs are cheaper and the raw materials are more freely available.

Cereal products such as straw, wheat or oat grains, and corn meal are examples of commonly used substrates. Because of the nature of these substrates, one drawback of this method is that the propagules of the agent may be difficult to separate from the substrate.

This method is not applicable on a very large scale, but may be necessary for some pathogenic agents. This is because some pathogens will only produce propagules on solid culture media.

Procedures: liquid-based formulations

Liquid-based formulations are applied as sprays. To achieve this they may be formulated as liquids that fall into one of two main sub-types, wettable powders and solutions.

A wettable powder is a powder to which water is added. Such formulations are sold or distributed as a powder to which water is added prior to its application. Wettable powders are often preferred as they are usually easier to store than solutions and often have a longer shelf life.

The alternative is a solution with the agent held in suspension. These formulations are normally distributed in concentrated form and water is also added to the solution in order to dilute it prior to application.

Solutions are the most appropriate means of applying foliar herbicides. Foliar bioherbicides are those that are applied to, and infect, the leaves of the target weed. Liquid-based solutions have a tendency to remain at least partially on the leaf surface, and may also have the added advantage of increasing leaf wetness or extending the dew period.

Procedures: granular formulations

Granular formulations are applied by spreading and are the most suitable formulations for soil-applied bioherbicides. Granules tend to penetrate the foliage layer of the crop or pasture to which they are applied, and the vast majority usually end up on or near the soil surface.

Granular formulations usually require activation by some form of precipitation. Unlike foliar-applied liquid-based formulations, granular formulations are not necessarily immediately activated once they are applied. They require soil moisture or rain for the agent to be released from the granules and become active.

Back to top

Examples of inundative biological control

Control of northern jointvetch (Aeschynomenevirginica)

Northern jointvetch is a common weed of rice and soybean in North America. The spores of the fungal pathogen Colletotrichum gloesporioides f. sp. aeschynomene have been sold commercially in North America as a ‘bioherbicide’ to control northern jointvetch.

The product called ‘Collego’ is marketed in the form of a wettable powder and has a shelf life of about 2 years. With Collego, the active ingredient is dry spores (i.e. viable

conidia) that are commercially produced by submerged culture fermentation. This product has exhibited a success rate of 90% weed kill in the field.

Control of strangler vine (Morrenia odorata)

Strangler vine is a weed of citrus orchards in Florida, USA. Mycelial fragments of the fungal pathogen Phytophthora palmivora have been sold commercially in North America as a ‘bioherbicide’ for the control of this weed.

The product called ‘DeVine’, is marketed as a wet formulation, but has an expiry date of 6 weeks and must be kept in refrigerated storage. Mass production of chlamydospores by fermentation was possible, but a long shelf life could not be achieved. Therefore it has to be marketed as mycelial fragments in suspension as a wet formulation. The mycelial fragments consist of hyphae (i.e. the wet body of the

fungus) and chlamydospores (i.e. unicellular spores produced by asexual reproduction). This product has exhibited a success rate greater than 90% weed kill in the field.

Control of American blackberry (Prunus serotina)

American blackberry is a weed of pine forests in Europe. Mycelial fragments of the fungal pathogen Chondrostereum purpureum have been sold commercially in North America as a ‘bioherbicide’ for its control.

The product called ‘Biochon’, is marketed as an agar suspension and requires a wound to be made to the target weed before application. This weedy tree is usually cut mechanically and the cut surfaces are painted or sprayed with the product.

Control of starfruit (Damasonium minus) in Australia

Starfruit is a native plant that is a problem weed in Australian rice fields. This species is a significant problem in recent times because it has developed resistance to the only group of herbicides available for effective control against all rice weeds.

Rhynchosporium alismatis is an endemic fungus that naturally causes disease in the weed. A mycoherbicide is currently being developed using this pathogen. This pathogen also has the advantage of potentially controlling other closely related introduced weeds that are also present in rice crops (i.e. Alisma lanceolatum).

Reading ‘Inundative biological control of weeds - the bioherbicide tactic’ (pages 193-206) in Australian Weed Management Systems.

Conclusion

Classical Biological Control has proven to be an effective and relatively safe means of managing widespread weed species, particularly those in natural ecosystems and places that are difficult to access.

Classical Biological Control has been shown to provide a very high benefit to cost ratio on average, so it is not only effective but also cost-effective. It is also particularly useful, and often the only management option available, in difficult to reach areas such as swamps, water bodies, ravines, mountainsides, etc.

It is quite possible that the role of bioherbicides may continue to become more important in the future, particularly as organic farm products become more popular and effective alternatives to chemical herbicides are more greatly sought after. Chemical herbicides cannot be used in organic crops – but inundative biological control may be used in such situations. Weed species are also becoming increasingly resistant to chemical herbicides reducing their effectiveness.

Video/DVD DVD2: Biological Control Watch ‘Assault on the Sepik’ on the DVD ‘Weeds in the way’.

Go to Activity 13-2

Back to top

References and further reading

Briese, DT & McLaren, DA 1997, ‘Community involvement in the distribution and evaluation of biological control agents’, Biocontrol News and Information, vol. 18, pp. 39-49.

Crawley, MJ 1989, ‘The successes and failures of weed biocontrol using insects’, Biocontrol News and Information, vol. 10, pp. 214-223.

Harris, P 1997, ‘Monitoring and impact of weed biological control agents’, in DA Andow, DW Ragsdale and RE Nyvall, (eds) Ecological interactions and biological control, Westview Press, Boulder, CO, pp. 215- 223.

Jupp, PW, Briese, DT & Groves, RH, (eds) 1997, ‘St. John’s wort: Integrated control and management workshop proceedings’, Plant Protection Quarterly, vol. 12, pp. 51-108.

Wapshere, AJ, Delfosse, ES & Cullen, JM 1989, ‘Recent developments in the biological control of weeds’, Crop Protection, vol. 8, pp. 27-250.

Woodburn, TL, Briese, DT & Corey, S (eds) 1996, ‘Thistle management workshop proceedings’, Plant Protection Quarterly, vol. 11, pp. 231-292.

Press release

15 August 2006

Back to top

---

Self assessment Do you know: the theory behind classical biological control the differences between classical and inundative biological control the key steps needed to complete a biological control project and appreciate the potential benefits and constraints of biological control?