News from the ‘Rearing’ End
Rearing insects in containment.
The failure of weed biocontrol agents to establish in the field can be costly in terms of resources, but also because of ongoing harm to the environment and productive landscapes.
Whether or not we can easily rear biocontrol agents can have a significant bearing on agent establishment and, therefore, on the outcome of weed biocontrol projects. Quentin Paynter has recently led a study to analyse which key factors influence our ability to successfully rear biocontrol agents.
In his analysis Quent compiled a list of agents that had been reared both in and out of containment in New Zealand. He used published records and the Landcare Research release database to classify the resulting 82 species into three categories: species that could be reared easily for several generations; species that could be reared through at least one generation, but only in low numbers, and species that could not be reared at all. Quent then tested whether rearing success was influenced by feeding guild, taxon and novelty (whether the agent had been reared successfully overseas). Quent also explored whether mandatory disease testing, introduced in 1984, had had an impact on the establishment of the 43 species introduced after this time.
Of the 82 agents considered, 13 species could not be reared in containment. Six of these 13 species did not mate and as a result did not produce fertile eggs. Four species didn’t reach maturity due to poor host plant quality and the remaining three species failed to emerge from hibernation. Thirteen species were unable to be reared in large numbers, but the reasons for low breeding success were not so easy to identify.
“Generally speaking, we have fewer problems rearing agents that have already been used overseas in other biocontrol programmes as more is known about their lifecycles and preferences. However, if an agent is novel, we have to develop the rearing techniques ourselves, which can require some degree of trial and error,” Quent added.
“The feeding strategy of the agents (feeding guild) can also influence our ability to rear them. For example, root and rosette feeders are more difficult to rear than defoliators,” explained Quent. “Beetles seem to be generally easier to rear than moths, and sawflies are more difficult again as they are very particular about the conditions they need for hibernation. “Agents such as stem borers, that require mature stems, can be logistically difficult because of the size or maturity of the plants required," he commented.
The containment phase, where all new agents must be housed in a secure facility under artificial conditions for several months, can be especially challenging. This phase is essential to ensure that the correct species has been imported, and that no unwanted hitchhikers have come along for the ride. Often the biggest obstacle is successfully growing the weeds themselves under artificial conditions, along with the added challenge that most of the problem weeds in New Zealand originate from the Northern Hemisphere. Therefore, any insects newly imported into containment from that part of the world will be out of phase with Southern Hemisphere conditions, which may necessitate manipulating their hosts to flower or produce fruit or new growth out of season in containment, which can be tricky. “To get around this problem we can do the host range testing overseas, but at some point insects progressing towards release must be rephased by manipulating light and temperature regimes, and this is easier for some species than others,” said Quent.
Some plants sulk when grown in pots, and none do as well under artificial light regimes as in natural sunlight. If host plants are not of sufficient quality the insects will also do poorly and may not survive. Outbreaks of pests, like aphids, can also be a problem since they cannot be controlled in ways that might also harm the biocontrol agents,” he added. Other frustrations include agents failing to emerge from winter hibernation since the normal triggers are absent or confused inside containment, and staggered emergence times (a strategy thought to have evolved to avoid parasitism). Sometimes males and females emerge at different times with little overlap, posing an additional complication.
“A recent example, where we had major rearing challenges was with the white admiral butterfly (Limenitis glorifica) for Japanese honeysuckle (Lonicera japonica). These butterflies need a lot of room to do elaborate courtship displays, which is not possible in containment, and the end result was that we could not get the butterflies to mate. Attempting a hand-pairing technique developed for other related Limenitis species, and releasing the white admiral into a spacious butterfly house were also unsuccessful,” said Quent. The only remaining option was to release new adults in the field where they could breed naturally. Despite the fact that only a small number could be released, for a variety of logistical reasons, success has been achieved (see article this issue).
The downside of not being able to mass rear agents and having to rely on natural breeding under field conditions instead, is that it can take longer before widespread releases can begin. Climate-controlled rearing rooms allow colonies to be kept at the optimal temperature and not at the mercy of the weather, predators or parasitoids. It is also obviously easier to keep tabs on them indoors and collect them up when it’s time for releases to be made.
“It is pretty well recognised that agent establishment is closely linked to the number of individuals that are released and the number of release sites,” said Quent. “This was supported by our analysis, which showed a positive correlation between release size and establishment success. With the sole exception of one unsuccessful Californian thistle agent (Hadroplontus litura) back in 1976, all of the establishment failures associated with small release size have occurred since 1984, when disease-testing became mandatory,” explained Quent. Prior to 1984, if rearing proved difficult large direct field releases from the country of origin were made instead. For example, around 150 000 gorse seed weevils (Exapion ulicis) were directly released. However, this type of activity is no-longer possible. Traditionally, disease screening has required sacrificing a percentage of the population to be released and getting an insect pathologist to examine them in minute detail under a microscope. Such time-consuming and painstaking work is not practical for large samples, and so only relatively small numbers (<400) have been direct field released since 1984. More recently, non-destructive molecular methods have become feasible, such as screening for evidence of disease in the faeces (frass). However, such tests need to be developed on a species by species basis and are not yet widely available.
In summary, Quent’s analysis demonstrated that difficulties have been experienced with rearing about one-third of all biocontrol agents in New Zealand and c. 16% could not be reared at all. The best predictors of rearing success were novelty (i.e. whether or not an agent had already been reared successfully in another country) and the feeding guild the agent belonged to. Since rearing difficulties have contributed to the establishment failure of five out of 34 agents released between 1985 and 2014 (i.e.about one in seven agents approved for release) this suggests that the cost-effectiveness of weed biocontrol programmes could potentially be enhanced by:
- prioritising (but not discounting) candidate agents based on feeding guilds that are likely to be straightforward to rear and/or doing small direct field releases and then redistributing the agents once establishment has been achieved.
- developing better information pathways between research institutes to improve the techniques used to rear agents in containment and identify the main causes of rearing failure.
- developing improved direct-release methods for agents that are difficult to rear, such as using non-lethal molecular analysis allowing a larger number of individuals to be released.
- releasing agents in more climatically favorable sites first to establish source populations that can be used to support releases in more climatically extreme environments.
This project is funded by the Ministry of Business, Innovation and Employment as part of Landcare Research’s Beating Weeds programme.