What’s New In Biological Control of Weeds? Issue 11, Feb 1999

What’s New In Biological Control of Weeds?

Number 11 February 1999

What’s Inside?

New Attack Launchedon Californian Thistle 1

Hot Gossip 2

Keeping at the CuttingEdge of Weed Issues 3

To Feed or Not to Feed? 4

Hieracium —a Problem Shared 6

Lest We Forget 7

Water, Wattles & Wasps 8

Autumn Activities 9

Trading BiologicalControl Agents 10

Tell Me More... 11

Californian thistle haslanguished in the too-hardbasket for a long time. Althoughit is the most common thistlein New Zealand, it has notbeen easy to find money totackle this perennial nuisance.Three agents have beenreleased in New Zealand to

date, two foliage-feedingbeetles (Lema cyanella, Altica

carduorum) and a gall fly(Urophora cardui), but they arenot sufficiently damaging tomake a difference to thispersistent problem on theirown. More agents are neededto strengthen this attack.

New Attack Launched on Californian Thistle

Manaaki WhenuaLandcare Research

What's New in Biological Control of Weeds? Number 11 February 1999

2

The Scotch thistle gall fly

(Urophora stylata) was releasedfor the first time on two farmsin Auckland’s Rodney Districtjust before Christmas. Thisproject was also made possibleby a cash injection fromAGMARDT and some fundingfrom the Wool Board. Weimported the flies fromAustralia as larvae insideinfested flowerheads and thenew adults emerged inside ourquarantine facility at Lincoln.

Speaking of quarantine andgall flies, the mist flower gall

fly is safely installed insideour quarantine facility atMt Albert, Auckland,undergoing final testing.We also have good newsabout the first agent for mistflower, the white smut

fungus, which was releasedat nine sites in the NorthIsland late last year. JaneFröhlich has been back toeach of the sites and reports

Hot Gossip

Scotch thistle gall fly

that the fungus appearsto be taking hold. Sheobserved symptoms of thedisease at each site, close towhere she had put outinfected plants.

Old man’s beard leaf

miner (Phytomyza vitalbae)would have to get the prizefor the fastest dispersal rateof any of our control agents.In only 2 years the leaf minerhas dispersed from releasesites at Nelson and Blenheimto colonise all old man’sbeard infestations in between(>100 km). Similar rapiddispersal patterns are beingnoticed in the North Islandtoo. Even small, isolatedclumps of old man’s beardhave not escaped attack. Thenumber of mines per squaremetre of leaf material seemsto be increasing rapidly too.

Recently we helped a group offarmers from some of the worstaffected areas, Otago andSouthland, to make a successfulapplication to the Agriculturaland Marketing Research andDevelopment Trust(AGMARDT) for funds topush along this worthy cause.The AGMARDT money willbe spent over the next 3 yearsin two main areas. Firstly, weare joining an international

consortium (with the USAand Canada) to search forpotential new agents inpreviously unexplored partsof the weed’s native range ineastern Europe and westernAsia. Recent evidence suggeststhat these areas, which have upuntil now been inaccessible,may actually be the centre oforigin of Californian thistle,and could therefore yieldmany more rewards than the

parts of Europe previouslysurveyed. Secondly, we plan tolook at how we can maximisethe impact of the three agentsthat have already beenreleased here. For example,we hope to get farmersinvolved in breeding up largenumbers of beetles and flies inspecially designed cages. Boththe Otago and Southlandregional councils will beassisting with this project.

3


Topic Purpose Email Address Command

Crop Weeds to distribute and discuss information cropweed-request@ subscribe

on the management of weeds in farrer.riv.dsu.edu.au cropweed

annual cropping systems

Pasture Weeds to distribute and discuss information majordomo@ subscribe

on the management of weeds in metz.une.edu.au pasture-weeds

pasture systems

Environmental to distribute and discuss information ENVIROWEEDS-request@ subscribe

Weeds on the management of environmental majordomo.nre.vic.gov.au enviroweeds

weeds in natural ecosystems in

Australia

Nursery Weeds to look at how the spread of invasive NURSERY-WEEDS-request subscribe

plants from gardens can be reduced @majordomo.nre.vic.gov.au nursery-weeds

Weeds to assist those involved in education EDUenviroWEEDS-request subscribe

Education and other training on environmental @majordomo.nre.vic.gov.au eduenviroweeds

weeds

Weed Watch to look at how identifying, reporting, WEEDWATCH-request@ subscribe

and recording new or expanding [email protected] weedwatch

weeds in Australia can be improved

Weed to provide progress on annual WEEDBUSTER-WEEK- subscribe

Buster Week national Weedbuster Week [email protected]. weedbuster-week

preparations and activities vic.gov.au

Wild Radish to facilitates the distribution of wild_radish@ None required but

information and discussion on farrer.riv.csu.edu.au please introduce

Raphanus raphanistrum yourself

Email is a wonderful forum,which removes physical barrierssuch as distance and allowsdiscussions on a host of topicsby an ever-increasing audiencefrom diverse backgrounds.Weeds are now making a bigsplash on the email front andthere are currently eight weedlistservers operational inAustralia that will be of interest

to many New Zealanders.They include environmental,pasture, and cropping weeds,which means nearly allecosystems are now covered!

If you are interested inkeeping up with the play, thenyou can subscibe to any of thelistservers. All you need to dois send a message to the

relevant email address with aspecial command in the bodyof your message (see below).Do not put the command inthe subject line, and if youhave an automatic signature atthe end of your message, thenyou need to turn it off. Alllistservers are free of chargeand you can unsubscribe atany time. So subscribe away!

Keeping at the Cutting Edge of Weed Issues



4

To Feed Or Not to Feed?

though testing showed that nativethistles were at risk, a decisionwas made in the 1960s to goahead and release the weevil— biodiversity wasn’t quite sohighly valued at that time! Theecological studies thatprompted these media reportsare extremely useful in thatthey enable us to improve ourunderstanding of the fullconsequences of unleashingbiological control agents onthe environment, and theywill allow researchers to makeeven better decisions in future.

The scientific communitybelieves that safety testsrepresent extreme and unnaturalconditions and usuallyoverestimate rather than under-estimate the real host range ofcontrol agents. Researchers needto carefully consider an agent’sbiology and behaviour whendeciding on the most appropriatekinds of tests to use. Foragents that disperse passivelyno-choice tests (where they aregiven the option of feeding onan alternative host or dying) areconsidered appropriate,because they are continuallyexposed to no-choice situationsin real life. Agents like gorsespider mites and fungi that blowabout on the wind may end upanywhere with little or nochance of being able to escapeonto a preferred host. Theseno-choice tests can often give“false positives” when the agents

the abundance of agents andtheir host plants.

To make sure, researchersrigorously test all proposedagents to assess the risk ofdamage to non-target plants. Itis not possible to test every plantspecies in New Zealand, but aset of internationally acceptedprocedures has been developedto help researchers choose theright sample. These procedureswork, as more than 600 biologicalcontrol introductions have beenmade worldwide without anyunpredicted host changeoccurring. Recently claims havebeen made that attempt to tarnishthis good record. For example,the media have reported thatthe agent we know here as thenodding thistle receptacleweevil (Rhinocyllus conicus) isunexpectedly damaging nativethistles in the USA. Closerinvestigation of this storyreveals that the risks had beenidentified beforehand. Even

Host specificity test set upused for the hieracium gall wasp

The first question on manypeoples’ lips when they hearabout biological control agentsbeing released is “What are theygoing to eat next?” To somepeople biological control soundsfrightening and dangerous. Theyfear massive ecological damage,and often our work is directlycompared with the introductionof rabbits or ferrets. Thiscomparison could not be furtherfrom the truth. Safety issuesare also foremost in the mindsof biological control of weedsresearchers, and only specialistfeeders are considered. Thesespecialists have co-evolvedwith their host plants over along period of time, and havedeveloped adaptations thatallow them to find and feed onthat host plant, and sometimesclose relatives of that plant. Thisspecialisation makes it difficultfor them to change host, and thechance of this happening hasbeen calculated at between onein ten million, and one in one-hundred million (the risk ofnative species unexpectedlybecoming a problem is thesame). Biological control agentswill also never run out of foodbecause they are unable toeradicate their host plants —this is because they can neverfind or utilise every plant. Ifbiological control is successful,plants become increasinglyrare and the agent populationsreduce accordingly, so a newequilibrium forms between

5


Pauline Syrett would like to know if anyone has seen ragwort flea beetles (Longitarsus jacobaeae), cinnabar

moths (Tyria jacobaeae) or magpie moths (Nyctemera annulata) feeding on any plants other than ragwort.

Ragwort flea beetle and cinnabar moth were purposely introduced as biological control agents, but themagpie moth is a native insect that uses ragwort as one of its host plants. The magpie moth isinteresting to us as a model of a potential biological control agent that has a wider host range than asingle plant species, but is probably restricted to a single plant family.

Records and laboratory tests tell us that we should expect to find magpie moths on a wide range of plantsin the daisy (Asteraceae) family, and that cinnabar moths mainly feed on ragwort, but may also befound on other closely related plants. The ragwort flea beetle has a narrower host range, but Paulinewould not be surprised to find it on groundsel (Senecio vulgaris) or marsh ragwort (Senecio aquaticus).

You can help us to improve our knowledge of host specificity testing by filling in the enclosed form. Ifyou have any queries please contact Pauline by phone (03) 325-6701 ext 3724, fax (03) 325-2418, oremail [email protected].

Thanking you in anticipation!

attack plants under artificialexperimental conditions that theyhave never actually been knownto attack in real life. This wasthe case with gorse spider mite(Tetranychus lintearius), whichduring testing laid eggs on dwarfbeans (Phaseolus vulgaris),although it never colonises beanplants under field conditions.

For agents that actively disperse,choice tests (where they aregiven the option of feeding ontheir host and one or morealternative hosts) are consideredmore appropriate and realisticbecause the agents are able tochoose in real life. However,carrying out these tests alonemay still not give you the fullpicture if circumstances canforce the agent into no-choice

situations. For example, thissummer we were surprised todiscover broom seed beetles(Bruchidius villosus) on treelucerne. The beetles do notattack tree lucerne seeds in theirhome range because the plantdoes not grow there. Since treelucerne does grow in Australasiaand it is closely related to broom,researchers on both sides of theTasman included this plant inchoice tests before the beetleswere released down under —the beetles chose broom everytime. We have discovered thatin New Zealand because, thebeetles become reproductivelyactive much earlier in the springthan in their homeland, they donot always have a choice. Theyare able to begin egg-layingweeks before broom begins to

flower when only tree lucernepods are available to them. Weplan to investigate this situationfurther and we will keep youposted as we find out more.

If we are to make the bestpossible decisions whenevaluating risk, then we needto continue to develop moresophisticated tests that moreaccurately reflect real-lifesituations, and to improve ourinterpretation of the results weobtain. For this reason PaulineSyrett is undertaking a study tocompare the predicted targetsof introduced biological controlagents with their actual targetsin the field after release. Youcan help too by providing uswith some really valuableinformation (see box below).

Can You Help?


6

Hieracium — A Problem Shared

Last year Pauline Syrett spentseveral months working in theUnited States and she wasable to catch up with Americancolleagues working on similarbiological control projects. Oneproject that we have in commonis hieracium. Pauline was

shown extensive infestationsof field hawkweed (Hieracium

caespitosum) by two farmers whorun the Hawkweed ActionGroup — a fund-raising groupthat is modelled on our ownHieracium Control Trust. Fieldhawkweed is not regarded asthe worst of the four weedyhawkweeds we have inNew Zealand. “At present it isonly a serious problem inMarlborough, but that may bebecause it is a more recentarrival,” warned Pauline. Weneed to keep a close eye onfield hawkweed as theAmericans are collectingimpressive data that shows ithas massive a growth rate andreproductive potential. Whilefield hawkweed is not ourprimary target for control, three

Agent Species of hawkweed attacked Method of attack Status

Hieracium Mouse-ear* Attacks the leaves, Widely establishedrust slowing growth

Plume moth Mouse-ear Caterpillars graze rosette Mass-rearing underwaycrowns & leaves to enable widespread

releases to begin

Gall wasp Mouse-ear Larvae gall the stolons 2 field releases madein February 1999.Mass-rearing underway

Gall midge Mouse-ear, field* Larvae gall the plant Undergoing final testing& king-devil*

Root Mouse-ear, field Larvae feed on roots Undergoing final testinghover fly & king-devil

Crown Mouse-ear, field, Larvae feed within leaf Undergoing final testinghover fly king-devil & orange* axils and rosettes crowns

* Mouse-ear hawkweed = H.pilosella, field hawkweed = H. caespitosum, king-devil hawkweed = H. praealtum,orange hawkweed = H. aurantiacum

of the six agents (the two hoverflies and the gall midge) that theHieracium Control Trust hopeto release in New Zealand willattack it (see below). Americanresearchers plan to look foradditional agents that willspecialise on field hawkweed.

Mouse-ear hawkweed Field hawkweed

7


The town of Eureka inHumboldt County,Northern California, wasso grateful to the greaterSt John’s wort beetle(Chrysolina quadrigemina) thatit erected a monument in itshonour in the 1960s. St John’swort was once a huge problemin this area, but it is nowunder control, thanks to thebeetle. Perhaps we shouldbe doing the same sort ofthing here?

The curse of St John’s worthas been forgotten by mostNew Zealanders because theyhave never experienced it.Earlier this century it was one

Lest We Forget

of the four worst weeds,up there with gorse, ragwort,and blackberry. The plantcontains a nasty chemical

called hypericin, whichcauses aphotosensitisation

disease in sheep andcattle that have

eaten it.Affectedstock also

becomehypersensitive towater resulting inviolentconvulsionsand musclespasms. Inresponse to theserious impacts

of this plant, thelesser St John’s

wort beetle (C. hyperici)was first introduced here in1943, followed by the greaterSt John’s wort beetle in 1965.

The impact of the defoliatingbeetles and improved landmanagement techniques havecaused St John’s wortpopulations to shrink intoobscurity. These days you aremost likely to see the plantalong roadsides where it ismost obvious during flowering— look out for yellow,ragwort-like flowers, and youmay also see the shiny metallicbeetles feeding on them.

In the 1990s the tables havewell and truly turned on StJohn’s wort. The plant is backin favour as a herbal remedyfor treating depression,sleeping disorders, seasonalaffective disorder, and as anantioxidant and antiviralagent. Some councils haveeven granted exemptions thatallow people to grow thisformer menace undercontrolled conditions.

Pauline Syrett with a memorial dedicated to the St John's wort beetle


8

South African researchershave been investigating thenatural enemies of wattles inAustralia. Unfortunately someof these potential biologicalcontrol agents are rare in theirhomeland, possibly becausethey, in turn, have their ownnatural enemies andcompetitors. Fortunately NewZealand has been able to cometo the rescue. We have a smallblack wasp (Bruchophagus

acaciae) that feeds insidewattle seeds. Since we don’thave any native wattles here,the wasp must have originallycome from Australia. It wasfirst “discovered” at Lincolnin 1910, before it was knownin Australia. The wasp isextremely common, and candestroy up to 90% of theannual seed crop of silverwattle (Racosperma dealbatum)around Lincoln. No wonderthat the South African PlantProtection Research Institute

Water is one of South Africa’smost precious commodities.Invasive pines and Australianwattles (now calledRacosperma in New Zealand,but previously known asAcacia) are seriouslythreatening that resourceby sucking rivers dry,displacing native vegetation,increasing the risk of fire,and interfering withagriculture. The SouthAfrican government hasstarted a “Working for Water”programme, which aims toreduce the exotic tree cover inkey areas. It is a huge,expensive undertaking. Theplants are physically clearedbut, with so much seed in theground, this is only atemporary solution . A morepermanent solution,incorporating biologicalcontrol, is required andNew Zealand is playing asmall part in trying to help.

Water, Wattles & Wasps

(PPRI) are keen to get theirhands on some!

Dr Richard Hill (Richard Hill& Associates) has beenworking for PPRI to assesswhether this wasp is a suitablecontrol agent, collecting seedfrom over 40 species of wattlesto see which ones are attacked.Preliminary results suggestthat the wasp only attacks asmall number of closelyrelated Australian wattles,and fortunately this is thegroup that causes mostproblems in South Africanriverbeds. Richard has alsobeen looking at the biology ofthe wasp, its seasonal activitypattern, fecundity, distribution,and impact on seed production.He has already sent a shipmentof the wasps to South Africaso that PPRI can do furthertesting in containment.

There are other naturalenemies commonly seenattacking wattles in NewZealand. Look out for a weevil(Storeus sp.) that also feeds onthe seeds, and a fungus(Uromycladium spp.) thatforms galls on stems andpods. The galls are initiallygreen and later turn brownwhen spores are formed.Without these natural enemiesit is likely that we would havemore problems with wattleshere in New Zealand too.Autumn is a good time for

Collecting seed-feeding wasps from silver wattle using sleeve cages

9


harvesting and redistributingragwort flea beetles, noddingthistle crown weevils(Trichosirocalus horridus),nodding thistle gall flies(Urophora solstitialis), gorsepod moths (Cydia succedana),old man’s beard leaf miner,and old man’s beard fungus(Phoma clematidina).

For the first four agentslisted refer to the appropriatepages in “The BiologicalControl of Weeds Book” fordetailed instructions on howto go about this. See also“Tell Me More...” (later inthis newsletter). Two wordsof caution when shiftingagents around. Avoid sealingup ragwort flea beetleswith large quantities ofragwort in non-breathablecontainers. It is possible thatthe plant releases toxic gasesthat can prove fatal to theinsects under these conditions.Where possible collect on acool day (or during the coolerparts of the day), put theinsects in paper bags (notplastic bags), and provideonly just enough plantmaterial to tide the insectsover while they are in transit.Also be careful to sort throughany material that you collectwith your garden-leaf vacuumso that you don’t shift anypests, like the clover rootweevil (Sitona lepidus), at thesame time.

Old man’s beard leaf minersare dispersing extremelyquickly throughout thecountry, but if you do comeacross any uninfested patchesbefore they do, then you caneasily help to fill any gaps intheir distribution. Simplycollect as much mined leafmaterial as you can from aheavily infested site and leavethis on the ground at the newsite where it won’t get blownaway. You may notice smallbrown pupal cases stuck to the

sites. Collect blackened,infected leaves and washthe spores off by swillingthem around in a bucket ofwater. Transfer the resultingliquid into a sprayer (it isbetter to keep one especiallyfor this purpose to avoidcontamination withherbicides etc.) and applyit to some old man’s beardfoliage. It is preferable tosoak one area thoroughlythan to apply the mixturetoo thinly. Because the

Autumn Activities

undersides of leaves. Newadults will quickly emergefrom these and colonise thenew sites. Any larvae stillmining the leaves willcomplete development on thecut material, and provideadditional new adults forestablishment.

Likewise it is easy to shift oldman’s beard fungus to new

Old man's beard leaf miner pupae

fungus needs moisture to beeffective, where possiblechoose shady, damp sites. If itrains heavily soon afterapplication, then the sporesmay get washed off and youmay need to repeat theprocedure.


10

Trading Biological Control Agents

The following register listsagents that are now soabundant in some regionsthat the councils areprepared to share (or swapthem) with other councils.If you also have agents thatyou would like to trade,please contact Lynley Hayesso you can be includedin the next newsletter.

Agent Available from

Alligator weed beetle Peter Joynt, Northland Regional Council, Ph 09 439 6662

Cinnabar moth Dave Galloway, Auckland Regional Council, Ph 09 426 7643Lindsay Scott, Manawatu-Wanganui Regional Council, Ph 06 374 6700Kevin Worsley, Wellington Regional Council, Ph 06 378 2484

Gorse pod moth Rob McCaw, Canterbury Regional Council, Ph 03 365 3828

Gorse spider mite Dave Galloway, Auckland Regional Council, Ph 09 426 7643Bob Morgan, Manawatu-Wanganui Regional Council, Ph 06 376 7758Ken Massey, Northland Regional Council, Ph 09 438 4639Kevin Worsley, Wellington Regional Council, Ph 06 378 2484

Nodding thistle Phil Crotty, Canterbury Regional Council, Ph 03 688 9069crown weevil

Nodding thistle Phil Crotty, Canterbury Regional Council, Ph 03 688 9069gall fly

Ragwort flea beetle Laurence Smith, Canterbury Regional Council, Ph 03 314 8014Lindsay Scott, Manawatu-Wanganui Regional Council, Ph 06 374 6700

11


Tell Me More ...

This is a new section wherewe will deal with some of thequestions that we are mostcommonly asked or topicalissues that have arisen.Please feel free to submit anyquestions you may have toLynley Hayes (address below).

Question: Leaf miners are now

becoming established in a

residential area where we require

people to control old man’s beard.

What is the best advice that we

can give these people so that the

weed is controlled without

wasting the leaf miners?

Answer: Tell them to leave thecut vines on the ground for atleast a fortnight. This way anyflies that complete developmentand emerge will still survive andbe able to seek out other old

man’s beard infestations nearby.Alternatively the cut vinescould be taken straightaway toother sites where control is notmandatory or feasible.

Question: What’s the optimum

time of year for collecting ragwort

flea beetles for redistribution?

Answer: Studies in theManawatu suggest that egglaying occurs over a muchlonger period than waspreviously thought. Even aslate as May there was noindication of a decline in thenumber of eggs obtained fromfield-collected samples. Thismeans that even if beetles arecollected late in autumn theyare still likely to lay largenumbers of eggs. However, we

suggest that any time betweenmid-February and mid-Marchwould be best.If you collect too early, someof the beetles that you releasemay disperse or die beforethey begin to lay eggs. If yourelease the beetles too latethey will have fewer eggs tolay before they succumb toadverse weather.

Question: How far will

ragwort flea beetles have

dispersed after 3–4 years?

Answer: Dispersal seem to behighly variable. We’ve hadreports of ragwort flea beetlesbeing recovered severalkilometres away from arelease site in the first season,but these might have been

Ragwort flea beetle


12

This newsletter is produced three times a year by:Manaaki Whenua – Landcare Research NZ LtdPO Box 69LincolnNew ZealandPh +64 3 3256 700Fax +64 3 3252 418

Editor: Lynley HayesLayout: Kirsty CullenCartoons: Anne Rose

ISSN 1173-762X

Manaaki Whenua. Manaaki Tangata:Care for the land. Care for the people.

This information may be copied and distributed to others without limitations, provided LandcareResearch and the source of the information is acknowledged. Under no circumstances may a chargebe made for this information without the expressed permission of Landcare Research.

from an unrecorded,unofficial release nearby.More commonly we findthat a release of 300 beetlesin one corner of a largepaddock will have coveredthe whole paddock (severalhundred metres), but notmuch further, by the nextyear. Conversely, at one of ourexperimental sites that washeavily infested with ragwort,the beetles took 3–4 years tospread over the wholepaddock. When a ragwortinfestation is severe, beetlesmay spread slowly because

they have no need to travel farto find good plants on whichto lay eggs. The beetles aregood at finding ragwortplants, so when ragwortinfestations are sparse, withwidely scattered plants, fleabeetle populations maydisperse rapidly. Some of ourcolleagues in Oregon put apotted ragwort plant on theroof of a tall building in themiddle of a city, manykilometres from the nearestnaturally growing ragwortplant, and ragwort flea beetlesquickly discovered it.

Question: Do ragwort flea

beetles prefer rosettes over

flowering plants?

Answer: Yes. They lay moreeggs per weight of ragwortfoliage on rosettes than onflowering plants. This makessense, as flowering plantsusually die and larvae thatcannot complete theirdevelopment before the rootsand crown senesce musteither transfer to a nearbyplant or perish.

Documents

What’s New In Biological Control of Weeds? Issue 11, Feb 1999