Landcare Research - Manaaki Whenua

Landcare-Research -Manaaki Whenua

Solving Pampas Puzzles

Pampas flowerhead infected with black floral smut.

Pampas flowerhead infected with black floral smut.

The work to determine if there are suitable biocontrol agents for pampas continues to encounter some intriguing challenges, but the picture is gradually becoming a little clearer.

Pampas can be found throughout New Zealand but is mainly problematic in the northern regions. There are two main weedy species, Cortaderia selloana and C. jubata, both originating from South America. Molecular studies have also detected at least one other South American Cortaderia entity, which is much less common. This could become a weed in the future, so ideally a biocontrol programme would find agents to control it too. Two plant-hoppers (Saccharosydne subandina and Lacertinalla australis), which attack the foliage, have been identified as the most promising insect agents for pampas. While similar in appearance, the two species can be distinguished with the aid of a hand lens or microscope. Both have proven difficult to work with, as they are very fragile and easily damaged by handling and shipping. However, perseverance and hand-carrying of carefully packed shipments eventually paid off when a colony of S. subandina was recently successfully established in containment in Auckland.

The host range of both plant-hoppers was described in the literature as Cortaderia, rye (Secale cereale) and garlic (Allium sativum). “However, this struck us as odd since these species are not closely related, and when we tested the pampas plant-hoppers on rye and garlic early on we quickly determined they weren’t hosts,” said Quentin Paynter. This made the team working on the project suspicious that cryptic species might be involved, whereby the plant-hoppers look identical and differences between them can only be determined by examining their DNA. This means that what is reported to be a species with a wide host range can turn out to be a complex of entities with narrower host ranges.

“We then studied the DNA of the pampas plant-hoppers we had collected from a range of sites in Chile, which showed at least three cryptic species within S. subandina and two within L. australis,” said Quentin. This opened up the possibility that pampas plant-hoppers might have different host ranges within Cortaderia. It might also explain why it was such a struggle to successfully establish colonies of some of the imported material. They were possibly being offered a sub-optimal host, which they could feed on if offered no alternative but that would not allow them to produce healthy offspring, leading to colony failure after a couple of months.

The team are studying plant-hoppers from Chile first, because it is currently too difficult to get permission to export them from other South American countries. Cortaderia jubata is not present in Chile, and New Zealand material has come from Peru or Ecuador. Despite searching widely throughout South America, only a handful of C. selloana plants have been found that match the New Zealand material, all deliberately planted as ornamentals in Chile. As a result, most of the plant-hoppers have been collected from Cortaderia species that don’t occur in New Zealand. The plant-hopper now well established in containment is – not surprisingly – S. subandina, collected off C. selloana, which is known to be a good match. Lacertinella australis has only been found at higher altitudes in Chile on Cortaderia species probably not present in New Zealand, and it has failed to thrive in containment on material available in New Zealand. This would seem to rule out this plant-hopper as a contender unless further surveys in Chile can find more, preferably wild-growing, C. selloana populations with L. australis on them.

Once molecular studies confirmed that we were dealing with cryptic species, individual females were reared in isolated lines to create populations of known identity. Once the female had produced offspring, a molecular test was used to determine the entity she belonged to, and then like progeny could be combined with like. “Working with cryptic species is tricky, time consuming and expensive!” said Quentin. At this point, since only one entity was doing well, the other lines were culled to avoid any possible contamination, and the team were finally able to begin proper host-range testing.

What happened next was quite unexpected. The plant-hoppers showed no signs of attacking native toetoe (Austroderia spp.), which was hugely promising, apart from one A. fulvida plant. The other five replicates of this species were unharmed. The attacked plant died soon after and was possibly unusually susceptible to plant-hopper attack because it was already compromised in some way. The team checked and ruled out that the result could be explained by misidentification, so more testing is required to explore this unusual result, to see, for example, whether toetoe plants are less well defended against herbivores when young. Testing has shown that this plant-hopper entity does not do that well on C. jubata, meaning that further studies of the plant-hoppers found on this host in Ecuador might need to be carried out in the future.

Since other similar species of plant-hoppers are associated with species of phytoplasma (specialised bacteria that cause plant disease), the phytoplasma status of the pampas plant-hoppers has been explored using molecular tools. The first shipment ever received from Chile returned a positive result: it was a 99% match for a bunch of phytoplasmas that relate to Candidatus Phytoplasma australiense, the organism responsible for disease in cabbage trees and flax, previously thought to occur only in New Zealand and Australia. Other shipments of the plant-hoppers have tested negative, including the current populations being studied in containment. However, further work is needed to explore the risk that pampas plant-hoppers could vector phytoplasma already in New Zealand. In New Zealand, Candidatus Phytoplasma australiense is spread by a native plant-hopper (Zeoliarus oppositus), which occasionally frequents pampas, so it may have spread the disease to pampas already, where it may remain uncommon without a suitable vector. “We are running some trials where we cage the native plant-hopper onto pampas plants to allow phytoplasma infection to occur. Next we put the pampas plant-hoppers onto these plants for a time and then move them to clean plants to see if they can successfully transmit the disease,” said Quentin. The results should be in later this year. There are still many unknowns, but if the pampas plant-hoppers are highly host-specific and can vector phytoplasma already present in pampas in New Zealand, this could prove to be a highly damaging combination.

Meanwhile, work to discover the secrets of a black smut fungus (Ustilago quitensis) has also been continuing. This fungus attacks the flowerheads and could potentially reduce the ability of pampas plants to produce seed, but the infection process is poorly understood. Chantal Probst has been working with fungal material, also from Chile, trying a range of inoculation techniques. This work is complicated by the length of time that can elapse between infection and the emergence of infected floral plumes (months or even years later), and the logistics of growing on and then housing large flowering-age plants in a containment facility, let alone getting them to flower on demand.

This has created the need to find alternative study methods. On artificial media the smut becomes a yeast, which needs to mate with another yeast form, with different mating type genes, to be able to infect plants. Attempts to do this were unsuccessful. Next mini pampas plants were grown in test tubes and pots and attempts made to infect them in various ways. “Recently I made a key breakthrough using this approach by injecting a teliospore suspension into the plants,” said Chantal. “I was able to successfully detect the smut in the tissue of these plants using molecular techniques a few months later.” The next step is to bring plants starting to produce flowers into containment and inject the teliospores in the forming inflorescence to see if the smut develops on the flowerheads. Once a reliable inoculation method has been developed, host range testing can begin.

There is still much to learn about the black smut, phytoplasma and plant-hoppers, and work to be done, but the team hasn’t yet ruled out the possibility of biocontrol for this tricky target in the future.

This project was originally funded by a grant from the Sustainable Farming Fund and more recently by the National Biocontrol Collective. Manaaki Whenua – Landcare Research is using some of its discretionary funding provided by the Ministry for Innovation, Business and Employment to support the black floral smut and phytoplasma studies. We acknowledge the assistance of our Chilean collaborator, Dr Hernan Norambuena, in providing shipments of organisms for study.

Contacts: Quentin Paynter ( and Chantal Probst (

Subscribe to Weed Biocontrol alerts

via email