Au stralian Weeds March 1982 25

Chrysolina varians, C. brunsvicensis and Anaitis spp. did not become estab­ REVIEWS (continued) lished, possibly due to predation and unfavourable physical or climatic con­ ditions (Wilson, 1960). C. hyperici , although established, persisted only in very low numbers and was only recently recovered in Victoria. A similar situation exists with Agrilus hyperici , which ap­ parently established at only one site in New South Wales and was rediscovered at this site in 1980 (Delfosse and Cullen, 198Ic). Although Zeuxidiplosis giardi is widely established in Australia, its Biological control of weeds of Mediterranean effects on the weed are normally origin: A progress report insignificant. Thus, of the eight in sects released, E. S. Delfosse and J. M. Cullen only Chrysolina quadrigemina has had significant impact on SI John '5 wort CSIRO Division 01 Entomology, Canberra, ACT (Table I). Thousands of hectares of thi s weed have been eliminated by this species, sometimes in conjunction with pasture improvement. However. there still exist nearly 190 000 hectares in New South Wales alone which are heavily infested (Campbell , 1977). The present Summary When these permits were granted, the distribution of St John 's wort is mainly agents were put into quarantine in Can­ from northern New South Wales to north­ The CSIRO Division of Entomology berra. They were then reared through at eastern Victoria, although it is also pres­ In Canberra is presently conducting least one generation under quarantine ent in South Australia, Western Aus­ research on five terrestrial weeds of conditions, during which time each tralia , Tasmani a and Q ueensland Mediterranean origin: ragwort species was cleared of parasites and most (Parso ns, 1973; Everi st, 1974; Klein­ (Senecio jacobaea), skeleton weed diseases before field release. schmidt and Johnson , 1977; Campbell , (Chondrilla juncea), common heli­ 1977). The terrain in whic h it persists otrope (Heliotropium europaeum), St is generally not conducive to chemical John's wort (Hypericum perforatum) St John's wort or cultural control, and mechanical con­ and Paterson's curse or Salvation Jane trol is not effective. (Echium plantagineum). Biological The hi story of biological control of St The rece ntl y rediscovered population control agents from appropriate Med­ John 's wort was reviewed recently by of Agrilus hyperici is presently under iterranean areas have heen imported Delfosse and Cullen (198Ic) . In Aus­ tralia, the project was begun by CSIR study in Canberra, and releases of this Into Australia, and In some cases re­ in the 1920s, and investigations which leased. Brief accounts of these agents rool-borin g species will be made at the began in England in 1928 were extended fie ld sites if attempts to establish a lab­ and the various programmes are given. to southern France from 1935-40 (Wil­ oratory colony are successful. son, 1960). Several other countries in­ Introduction cluding the United States of America, The current programme Canada, South Africa and New Zealand The CSIRO Division of Entomology has subsequently received insects from Aus­ Chrysolina hyperici and C. quadrige· been engaged in research on biological tralia. Some of these countries also con­ mina have been recollected recentl y by control of terrestrial weeds of Mediter­ ducted their own searches for natural the Montpellier Unit from areas of France ranean origin since the 1920s. Current enemies. which are eco-c1imatically sim ilar to those emphasis is centred on ragwort, skeleton Twelve species of insects were im­ Australian areas where St John's wort weed, St John 's wort, common helio­ ported into Australia between 1929 and is still a problem. Both larval and adult trope and Paterson's curse or Salvation 1953 (Table I): Chrysolina varians, C. Chrysolina spp. fe ed on the foliage , with Jane. brulJsvicensis. C. hyperici. C. quadri­ larvae causing the most severe damage. The selection of weed species as tar­ gem ina (all Coleoptera : Chrysomeli­ Seven field sites have been established gets for biological control was made in dae), Agrilus hyperici (Coleoptera : in New South Wales and Victoria (Del­ consultation with State authorities, and Buprestidae), Lathronympha hypericana fosse and Cullen, 198 Ic), and the first the arthropods and fungi associated with (Lepidoptera: Eucosmidae) , Depres­ releases of C. hyperici occurred in Sep· them examined in detail by the CSIRO saria hypericel/a (Lepidoptera : Oeco­ tember and November 1980. Poor plant Biological Control Unit in Montpellier, phoridae), Anaitis plagiata and A. quality caused by drought conditions France. Initially these examinations in­ efformata (both Lepidoptera : Geome­ during the I 979 and 1980 seasons pre­ cluded literature searches and field sur­ tridae), Actinotia hyperici (Lepidoptera vented further releases, but both species veys of each weed's natural enemies. and : Noctuidae), Aphis chloris (Homoptera are expected to be released at all sites in some cases of their parasites and dis­ : Aphididae), and Zeuxidiplosis giardi by the end of 1981 and their establish­ eases. After apparently effective and host­ (Diptera : Cecidomyiidae) . AJI except menl and success will be monitored. specific agents were found and tested in Actinotia hyperici, Lathronympha hy­ The recently redi scovered population Europe, applications were made to the pericana, Depressana hypericella and ofAgrilus hyperici is prese nty under study Commonwealth Department of Health Aphis chloris were released in the field in Canberra, and releases of this root­ for importation and release permits. between 1930 and 1955. boring species will be made at the field 26 Australian Weeds March 1982

sites if attempts to establish a laboratory tralia (Everist, 1974; Kleinschmidt and 99% mortality. This means that the three colony are successful . Johnson, 1977; Parsons, 1973). populations of D. sealariella which are The Montpellier Unit is currently in­ Chemical, mechanical and cultural presently in quarantine are undergoing vestigating several other species for pos­ methods of control are all possible with a severe genetic bottleneck, the effects sible introduction to Australia for Paterson 's curse but are generally too of which are unpredictable but which biological control of St John's wort. expensive and ephemeral to provide a may adversely affect tbe vigour of this These include the mite Phyllocoples hy­ long-term solution to the problem. species. Should this be the case, the perici (Acari : Eriophyidae), and the populations of D . • eaLarie/La presently moths Arislolelia morphochroma (Lep­ The current programme in quarantine would have to be destroyed idoptera : Gelechiidae) and Aclinolia Four insects have so far been approved and new populations imported should the hyperici (Lepidoptera : Noctuidae) (Del­ by the Department of Health for intro­ programme continue. This would in­ fosseandCullen,198lc). The rmte causes duction and release in Australia and have volve considerable direct costs and pos­ marked dwarfing of the recumbent ro­ subsequently been received in quarantine sibly much more in indirect costs due to sette stems, larvae of Aristolelia mor­ in Canberra (Table I) .. These are the delay in implementing the phochroma cause die-back of shoot tips, Dialecliea scalarielia (Lepidoptera : programme. and larvae of AClinOlia hyper;ci are foli­ Gracilariidae) , Longilarsus aelJeus and So far, the effect of the injunction on age-feeders . It is likely that testing for L. eehii (both Coleoptera : Chrysome­ the other species is less severe. Adults Phyliocoples hyperici will be completed lidae) and Phy loecia coeruleseens of Longitarsus spp. feed on foliage, in 198\ and , if it is sufficiently host­ creating shot-holes, but the greatest dam­ specific and approved for introduction, (Coleoptera : Cerarnbycidae). The larvae of Dialeeliea seaLariella are leaf miners age is created by the larvae which are releases could begin in 1981-82. Aphis root-feeders. Rearing of L. aeneus and chloris has been released in Canada and and three releases were made in 1980 in New South Wales. Follow-up field work L. eelzii is a much slower process, and may be considered again for use here, autumn 1982 would be the ftrst possible whilst new stocks of botb Agrilus hy­ is currently being conducted at all three sites and it is not yet known if this species release time . Spring is the proper time perici and Atzaitis efformata were sent is established. to release the stem-borer Phytoecia from the Montpellier Unit to Canberra A High ·Court interim injunction is coeruleseens , and the injunction has pre­ early in 1981 and have been released in currently in force against this programme vented any releases of this species this the field. which prevents CSIRO from releasing season. Since preliminary trials in rear­ any more insects for the biologicaJ con­ ing P . coeruleseens on artificiaJ diets are trol of this weed. This injunction was encouraging, the major problem of hav­ Paterson's curse or Salvation precipitated by the filing of legal action ing sufficient plants available at the cor­ Jane against CSIRO by two apiarists and two rect stage for use may be partially graziers in July 1980, leading to a hear­ overcome. However, a genetic bottle­ Paterson's curse is a winter-growing an­ ing before the Supreme Court of South neck will be difficult to avoid with nual (or occasionally facultative bien­ Australia. P. eoeruleseens due to the delay caused nial) weed, and biological control was Understandably this injunction has by the injunction, and again, possible suggested over 50 years ago (Tillyard, severely affected the programme of bio­ effects of this forced selection on the 1928). Most activity has taken place over logical control of this weed. For ex­ field suitability of P. eoeruLeseens are the last decade and the history, damage ample, a conservative estimate of the unpredictable. caused by the weed, and controversial number of adult D. scaLarielia which Should the court case be resolved in nature of this programme was discussed would have been released by the end of favour of biological control, there are by Delfosse and Cullen, 198Ib). December 1980 is 50 000. Plants bear­ several other species which have poten­ Paterson's curse occurs in all AustraJian ing larval and pupal D. scalariella had tial as biological control agents for States but is most damaging in the Riv­ to be placed in a cold room at about 6°C Paterson 's curse that may be considered erina areas of New South Wales and in an attempt to store them for an ex­ (Delfosse and Cullen, 1981 b; Wapshere, Victoria and in South and Western Aus- tended period, but this resulted in about 1981).

Table I Natural enemies imported and/or released in Australia for biological control of five terrestrial weeds of Mediterranean origin (some data from Wilson, 1960)

l Plant species Natural enemy Country of Date I Date Effect on origin imported released Status' host plant"

SI John 's won Chrysolina varians (Schall.) (Coleoptera : (Hypericum per/orarum L.) Chrysomelidae) England 1929-33 1930-3 NE C. brunsvicensis (Grav .) (Coleoptera : Chrysomelidae) England 1929-35 1930-4 NE C. hyperici (Forst .) (Coleoptera : Chrysomelidae) England 1930-4 1930-4 E,R L France 1979 1980' SU C. quadrigemirla (Suffr.) (Coleoptera : Chrysomelidae) France 1937-39 1931>-9 E, A S France 1979 1980' SU Agrilus hyperici (Cr.) (Coleoptera : Bupreslidae) France 1931>-40 1939-40 E, R L France 1981 ' 1931-2 Austral ian Weeds March 1982 27

Lathronympha hypericana Hubn . (Homoptera : Eucosmidae) England 193()"'6 NR Depressaria hypericella Hubn . (Lepidoptera: Oecophoridae) England 1932-6 NR Anaitis piagiala L. (Lepidoptera : Geometridae) England 1932-7 1933-8 NE A . efformata Guen . (Lepidoptera : Geometridae) England 1932-7 1933-8 NE France 1938-9 1933-8 NE France 198 1' 198 1- 2 NE Aclinotia hyperici Schiff. (Lepidoptera: Noctuidae) France 1939-40 NR Aphis chloris Koch (Homoptera : Aphididae) England 1932-4 NR Zeuxidiplosis giardi Keiff. (Diptera : Cec idomyiidae) France 1939-40 1953-5 E. S N 1950 California 1953 1953-5 E. S N Paterson '5 curse or Salvation Dia/ec/iea scalariella (Zeller) Jane (Lepidoptera : Gracilariidae) France 1979-80 1980 TE TE (Echium plalllagilleum L.) Portugal 1979-80 1980 TE TE

Lotlgitarsus aelleus Kutsch (Coleoptera : Chrysomelidae) France 1979-80 NYR Portugal 1979-80 NYR L. echii Koch (Coleoptera : Chrysomelidae) France 1979-80 NYR Portugal 1979-80 NYR PhYloec:ia coerulesc:ells (Scopoli) (Coleoptera: Cerambycidae) France 1979-80 NYR common heliotrope Longitarsus albiueus Foudras (Coleoptera (Heliotropium europaeum L.) : Chrysomelidae) France 1979-80 198 1' Greece 1979 1979. 1981' E. P N Crete 1979 NYR ragwoIt Tyria jacobaea L. (Lepidoptera : (Senecio jacobaea L.) Arctiidae) New Zealand 1929-32 193()"'32 NE England 1934-7 1934-7 NE Engl and 1955. 1957 1955-62 NE Italy 1955 1955-6 NE England 1959 1960 NE Switzerland & Austria 1961-2 1962-4 NE Canada 1977 1978" TE TE Switze rl and 1978 1979" TE TE Pegohylemyia seneciella (Meade) (Diptera : Musc idae) England 1934. 1936 NR P . jacobaeae6 (Diplera : Muscidae) New Zealand 1933 NR New Zealand 1959 1959 NE Longifarsus jacobaeae Waterhouse (Coleoptera : Chrysomelidae) France 1977 1979 E. P ske leton weed Puccinia chondrillina Bubak & Syd (Chondrilla jUllcea L.) (Ured inales : Pucciniaceae) (narrow-leaf strain) Italy 197 1 1971 E. A S (strai n TU21) Turkey 1980 1980 E. R TE Cyst;phora schmidt; Rubs. (Diptera : Cecidomyijdae) Greece 1971 197 1 E. A S Aceria chondrillae G. Can . (Acari : Eriophyidae) Greece 1971 197 1 E. A S Bradyrrhoa gilveo/ella Tr. (Lepidoptera : Phycitidae) Greece 1973 1974-6 NE Greece 1977 1978-9 E. P TE Oporopsamma werrheimsleini Rehel (Lepidoptera : Tortricid ae) Iran 1975 . 1977 NR

'NR = not released "N "" neg ligible NYR = nOI yet released. bu t releases expecled or L = light possib le S = significant IE. A = established, abundant TE = too earl y to tell E, P = establishment probable ~More releases to be made E. R = established, rare "Expected E.S = established . spotty occurrence. NE = not established ~ Idenlifi c ati o n un cenllin . Onl y material from New Zealand SU = status uncenain ever exam ined was identified as P. jtlf'lIbal'tl l'. TE = 100 early to tell 28 Auslralian Weeds March 1982

Common heliotrope liotrope, it contains pyrrolizidine alka­ ever, tunnel and feed in crowns of plants Biological control activities for this weed loids and has occasionaly caused and can kill plants directly or in com­ began about 30 years ago, with surveys considerable stock losses, especially in bination with other stresses. They are in Mediterranean Europe and North Tasmania in recent times . well suited to the situation where grazing Africa for potential enemies. The history Ragwort has a long history of attempts by stock or by Tyria jacobaeae causes of the programme has been reviewed by at biological control. The moth Tyria the plant to perennate and produce large Delfosse and Cullen (198 I a). jacobaeae (Lepidoptera: Arctiidae) has crowns. This summer-growing annual weed can been introduced into Australia several Longilarsus jacobaeae has one gen­ cause massive mortality of sheep due to times (Table I) . but each time establish­ eration per year. Following the first re­ ingestion of the alkaloids it contains ment has been un successful due to. mas­ leases some F, adults were recovered at (Culvenor, 1954; Culvenoret al., 1954). sive predation of larvae in the field, both release sites during the 197~80 Common heliotrope is generally distrib­ principally by the scorpion fly Harpo­ summer, but numbers were very small uted west and north of the Great Dividing bittacus nigriceps (Bormemissza, 1966), and establishment mu st still be consid­ Range in New South Wales and Victoria, and because of di sease epidemics. A fur­ ered doubtful. In Europe, the life cycle largely within the zone of 300 to 500 mm ther attempt at establishment of Tyria of L. jacobaeae varies according to the annual winter-dominant rainfall (Moore jacobaeae is currently being made by the area in which it is found, with the fre­ 1956). It is also present in South Aus­ Victorian Department of Crown Lands quent occurrence of aestivation in the tralia, Western Australia and southern and Survey (L. Schmidl, pers. comm .). adult or egg stage. The results of Aus­ Queensland (Everist, 1974). Chemical, T. jacobaeap has been successfull y es­ tralian studies to determine the aestiva­ mechanical and cultural control is fea­ tablished in the United States of Amer­ tion characteristics of the population sible, but too expensive for general use. ica, Canada and New Zealand, with imported from Annonay in France were varying degrees of effectiveness. reported by Cu ll en and Moore (1981 ), The current programme who found that there is no evidence of The flea beetle Longitarsus albineus The current programme adult aestivation and limited expression (Coleoptera : Chrysomelidae) was im­ The nea beetle LongitarsuJ jacobaeae of egg aestivation. The significance of ported and first released in 1979 (Table (Coleoptera : Chrysomelidae) was first the latter and the lack of any develop­ I) (Delfosse and Cullen, 198Ia). Adults released in North America in 1969 fol­ mental delay in the majority of the pop­ of this spec ies feed on the leaves, cre­ lowin g the successful completion of host­ ulati on is still under study in relation to ating shot-holes, while the larvae feed specificity studies. Additional studies Austral ian rag wort . Considerable data on the roots and cause greater damage have now been collected on the popu­ to the plant. Five field sites have been were carried out at the Montpellier Unit lation dynamics of ragwort under dif­ established in New South Wales but, due from 197.3. and it was imported to Aus­ to unfavourable weather, plants were only tralia in late 1977 . Further host­ ferent management regimes and in suitable for release of L. albineus at one specificity st udies of adult feeding, ovi­ different seasons. so that the role of L. jacobaeae of these in 197~80 . The first field re­ position and larval survival were carried can be readily assessed when covery of an adult L. albifleus was made out in quarantine to confirm that it wouid the popu lat ion has built up sufficiently. from this site (near Jugiong) in early not be a threat to the native species February 198 I. Although it is still too Sl)l/ecio lil/eari/olills, S. glomera/Us, S. hispidultts, and S. laulus . The re sults early to claim establishment of L. albi­ Skeleton weed new;. field conditions are excellent and were sati sfactory and the first releases the chances of establishment must be were made in March 1979. Five hundred The biological control of skeleton weed rated as very good . adults were released uncaged at each of is the longest-running of the current pro­ Longitarsus albineus was also re ­ two sites. one in Gippsland in Victoria grammes in Canberra . Accounts have leased near Urana and Corowa in January and the other near Deloraine in Tas­ already been published of several aspects 198 I; no adults have yet been recovered mania . The current programme is a co­ of the programme , for exampl e. the in­ from these sites. We expect that funher operative one with the Victorian itial colonization and spread uf the rust releases of this insect will be made this Department of Crown Lands and Survey Pun.:illia clwlldrillilla (Cullen l'1 al . . year near Mathoura and Trangie. New and the Tasmanian Department of Ag­ 1973), the establi shment of the gall mite South Wales, and near Caniambo, riculture . Stocks of LOligirarstts and gall midge (Cullen, 1974). the popu­ Victoria. jocoho('(l(J were also released to these lation dynamics of the plant (Cullen and Several other insect species are cur­ authorities at the same time. and they Groves. 1977) , the overall success (Cul ­ rently being investigated by the Mont­ have cond ucted further propagation and len. 1978), and the estimated economic pellier Unit (Delfosse and Cullen, 1981a) distribution programmes. benefits (Marsden et af . • 1980). The eco­ and may be imported and released in the [n addition to the two release plots. logical aspects of the weed and its control future. the CSIRO Division of Entomology has were reviewed in Groves and Cullen a control plot at each location. the aim (1981). Ragwort being to analyse the complex interaction The most serious remaining problem Ragwon has the most restricted distri­ between the life cycle of L. jacobaeae with this weed is the increase in density bution of the five species under consid­ and its effects on the density and popu­ and distribution of the remaining two eration, being confined to Tasmania and lari on dynamics of ragwort in typical previously less-common forms of the the high-rainfall areas of Victoria with pasture situ at ions. This should integrate weed. form B (the intermediate-leaf form) only occasional plants in the Adelaide with further st udies of the biology of L . and form C (the broad-leaf form) (Hull hills, the south-western corner of West­ j (l('o/Ja('(I{' and the work on rag wort con­ and Groves. 1973). which are not at­ ern Australia , and south-eastern New trol in relation to pasture management tacked by the strain of P . cholldrillina South Wales (Parsons, 1973). It is pri­ being carri ed o ut by the cooperating imported in 1971 (Burdon. Groves and marily a weed of dairy and beef cattle Departments . Cullen. unpub. data). Recent work has pastures, where it competes with more Adults of L. jllcobol'lIl' feed on the been concerned with the search for fur­ useful species and is difficult to control. fo liage but rarely cause sufficient dam­ ther strains of P . cholle/rillina which are Like Paterson's curse and common he- age to control the weed . The larvae how- virulent against these forms and with two Australian Weeds March 1982 29 other insect species which attack all three was completed . Plans to import more problem areas of St John 's wort are char­ forms of the weed. stock have been disrupted by outside acterized by hiUy terrai n, variable shade factors . (mainly by Eucalyptus spp.) and signi­ The current programme Oporopsamma wertheimsteini attacks fi cant summer rainfall (Delfosse and Extensive surveys have been made in all three forms of the weed. and at least Cullen, 198 Ic). The rainfall occurs at Europe for new strai ns of P. chondrillina under laboratory conditions causes con­ a time which sometimes brings Chry­ (Hasan, 1981 ) and several have been siderable damage when the larvae tunnel solina quadrigemina out of aestivation found which attack the intermediate-leaf around the base of the rosette and top too early, resulting in asynchrony of the form . One of these (strain TU21 from secti on of the root, causing it to rot away. in sect and the weed , so that to be effec­ Manisa in Turkey) was imported to Can­ Under field conditions in Iran there is ti ve the new arthropods imported for berra in 1980, and after two generations onl y one generation per year. with a control of SI John 's wort must be better in quarantine was released in the field pupal aesti vat ion in summer and an egg synchronized with the climate , be able in September 1980 and is currentl y spread diapause in wi nter. These are adaptations to attack the weed in shade, damage the over a small area in New South Wales. to the extreme cont inental climate of the weed so severely that summer recovery Its increase and spread !las been slower areas where O. lVerlheimJteini occurs. is negligible, and achieve control with­ than expected, possibl y because of the and bot h developmental delays can be out pasture improvement. dry season. No strain sufficiently viru­ removed in the laboratory by manipu­ Insects for the control of common le nt against the broad-leaf form has ye t lation of temperature and ph otoperiod. heliotrope, Paterson's curse and rag wort been located in Europe. If eventually released in Australia. the must reduce the weeds to a minor com­ Auempts have been made to establi sh reacti ons of O. ltIertlieim.\·leini to the ponent of the vegetation. This can be the moth Brac/yrrlioa gilveolella (Lepi­ environment will be of considerable ac hieved by a massive reduction in seed doptera: Phycitidae) in the Canberra area interest. production , decrease in growth of indi­ on three occasions. The first two attempts Other insect species attacking skeleton vidual pl ants, or a combination of these. were un successful. almost certainl y due weed are known from Europe and the For annual weeds, reduction in seed pro­ to problems of laboratory-induced selec­ Middle East, but all those examined to duction must be rapid and mu st occur tion encountered during the development date have been rejected because of likely at the beginning of the growth period. of a viable rearing system for this species ineffectiveness, doubts about host-spec­ In general, Ihe stress applied by the bio­ (C ull en, 198 1). Foll owi ng a major ificity or difficulty in establishing ade­ logical conlrol agents must reduce the attempt at establishment in the 1978-79 qu ate specificity. The remaining species competitive ability of these weeds with season using a fre sh colony imported in are currently known only from Russia re spect to ot her, more desirable. past ure 1977 (Table I), recoverie s were made but could be candidates in the future if plants. The complexes of species avail­ in 197'1--80 , at least two generations after necessary. able for each of these weeds seem the last releases. There is every chance capable of achievin g these goals, but that the species is estab li shed at both Discussion unfortunately th e future of the pro­ sites , albeit at low population levels as gramme for Paterson 's curse rests on yet. An attempt is currently being made The current programme covers a wide points of law rather th an biological to establi sh B . gilveolella in a malice site spectrum of weed li fe cycles including principles. and material has been given to th e Vic­ a summer-growing annual (common The requirements for successful con­ torian Department of Crown Lands and heliotrope), a wi nter-growing annual trol of rag wort are complicated by the Survey for further attempts at establish­ which is occasionally a fac ultative bien­ weed 's ability to change from a biennial ment. Larvae of this species tunnel into nial (Paterson's curse), a biennial which to a perennial habit. If it is allowed to and around the top part of the root of is often a facultative perennial (ragwort). remain undisturbed during fl owering but ske leton weed, causing structural dam­ and two perennials (St John 's wort and the production of viable seed or the pro­ age and damaging regenerative and stor­ skeleton weed). It also includes sexuall y duction and survival of seedlin gs is age tissue (Caresl:he and Wapshere. reproducing, outbreeding species (com­ greatly reduced. it shou ld be possible to 1975). There are two to three generations mon heli otrope, Paterson 's curse and gain control, since the plant would retain per year. ragwort) and apomicts (skeleton weed its biennial habit and be dependent on The moth Oporopsamma werlheim­ and St John 's wort); weeds of crops seeds for its persistence. Grazing of the steini (Lepidoptera : Tortricidae) was (skeleton weed and sometim es common pl ant by sheep is comm on however, and imported from Iran in 1975 and 1977 for heliotrope and Paterson's curse), road­ reduction in seeding is rarely high enough ex tensive specificity testing in quaran­ sides (aU five species). and pasture and without overall damage to the plant. tine in Canberra (Table I). This species ran geland (common heliotrope, Pater­ whic h in tum stimulates it to become is almost certainly specific to skeleton son 's curse, St John's wort and ragwort ); perennial . Attack on the mature peren­ weed in the field although early work weeds of low (common heliotrope, St nat ing pl ant is then the onl y possibility. demonstrated the ability of older larvae John 's wort and ragwort ) to extremely and it is hoped that Longitar.ms jaco­ to complete their development on roots hi gh conflicts-of-interest (Paterson's baeae will remove such plants ei ther rap­ of lettuce (Lactuca sativa) under con­ curse); weeds of low (skeleton weed) to idl y (if attack is sufficientl y heavy and fined laboratory conditions (Hasan and high (the remaining species) livestock­ coincides with other stresses on the pl ant) Wapshere , 1977). Host-specificity poisoning potential; and from very old or else more slowly if the plant 's reserves studies on first ins tar larvae were suc­ (St John 's wort and ragwort ) to relatively can be gradually depleted. cessfully completed in Canberra for all new (common heli otrope and Paterson 's A steady depletion of re sources has required test plants (including lettuce) curse) projects. The only common points been the key to cont rol of the narrow­ with no significant damage to any species. between these projects are that they all leaf form of skeleton weed. If the ph ys­ Critical experiments to assess the sig­ deal with very competitive, invasive, iol ogical efficiency of shoot growth aris­ nificance of the earlier observation on herbaceous, terrestrial weeds of Medi­ ing from the perennial root can be reduced mature larvae were commenced, but the terranean origin! sufficientl y. either by natural enemies colony proved difficult to maintain and There are points of interest unique to alone or in combination with competition unfortunately died out before the work each project. Forexample, the remaining from pasture pl ant s (Cullen and Groves. 30 Australian Weeds March 1982

1977), it becomes a drain on the root laneous Publication of the Commonwealth cum perforatum. Proceedings oj the Fifth reserves. Following the introduction of Institute of Biological Control pp. 111-17. IlIIernatiolial Symposium on Biological Puccinia chondrillina, there was contin­ Cullen, J. M . (1978). Evaluating the success Conrrol of Weeds, 1980. CSIRO pp. ued widespread damage to the topgrowth of the programme for the biological control 575-81. ofChondrillajulicea L. Proceedings oJlhe Everist, S. L. (1974). Poisonous Plants of of the narrow-leaf form of skeleton weed Fourth International Symposium on Bio­ Australia. Angus and Robertson. Sydney. but the density of plants did not decrease logical Control oJWeeds. 1976. University 684 pp. significantly until after three to five sea­ of Florida, Gainesville . pp. 11 7-21 . Groves, R. H . and Cullen, J. M. (1981) . sons (Cullen, 1978). It is hoped that sim­ Cullen, J. M. (1981). Considerations in rear· Cholldrilla juncea; the ecological control ilar results can be achieved with the ing Bradyrrhoa gilveolella for the control of a weed. Ecology oj Pests ;11 Australia. intermediate- and broad-leaf forms. of Cho1ldrilla juncea in Australia. Pro· CSIRO. (in press) It should be clear from the preceding c:eedillgs oj the Fijth International Sym­ Hasan, S . (198 1). Present status and pros­ comments that success in any pro­ posium on Biological Comrol oj Weeds. pects of the programme in Europe for the gramme of biological control of weeds 1980. CS IRO pp. 23J...9. microbiological control of Australian will not only be dependent on the amount Cullen, J. M. and Groves, R. H. (1977). The weeds. Proceedings of the Fifth Inter­ of damage a natural enemy can produce population biology of Chondrifla jum:ea national Symposium all Biological Control L. in Australia. Proceedings oj the Eco­ of Weeds, 1980. CSIRO pp. 33J...40. but also on how that damage integrates logical Society oj Australia 10:12 1-34. Hasan , S., and Wapshere, A. J. (1977). Bi· into the particular life system of the weed Cullen , J. M., Kable, P. F. and Catt, M. ology and host specificity of the Chondrilla (and occasionally with political rather ( 1973). Epidemic spread of a rust imported crown moth Oporopsamma wertheimsteini than biological factors). The fie ld eval­ for biological control. Nature 244:462-4. (Rebel) (Lepidoptera : Tortricidae) . Bul· uation programmes for all the current Cullen , J. M. and Moore , A. D. (1981). letin oj Entomological Research 67:6 19-25. projects are designed to a nalyse the Preliminary observations on Lollgitarsus Hull , V. J. and Groves, R. H. (1973). Varia· potential interactions, since only with jacobaeae introduced for the control of tion in ChondrilLa juncea L. in south-east­ such knowledge can predictability and ragwort in Australia. Proceedings oj the ern Australia. Australian Journal oj Botany precision in biological control increase FiJth InternaTional Symposium on Bio­ 21 : 11J...35. and the ground work for its integration logical Control of Weeds . 1980. CSIRO Kleinschmidt, H. E. , and Johnson, R. W . pp. 499-505. (1977) . Weeds ofQ ueensland. Queensland with other management systems be laid . Culvenor, C. J. J. (1954). The alkaloids of Government Printer, Brisbane. 469 pp. References Heliotropium europaeum L. IJ . Isolation Marsden, J. S., Martin , G. E., Parham, D. structures of the third major alkaloid and J., Ridsd ill Smith , T. J. and Johnston , B. Bormemissza, G. F. (1966). An attempt to two minor alkaloids, and isolation of the G. ( 1980) . Returns on Australiml Agri­ control ragwort in Australia with the Cin­ principal N·oxides. Australian Journal oj cultural Research. CSIRO. Melbourne. 107 nabar moth CaWmorpha jac:obaeae (L.) ChemiSTry 7:287-97 . pp. (Arctiidae: Lepidoptera). Australian Jour­ Culvenor. C . J . J., Drummond, L. J. and Moore, C. W. E. (1956). Observations on lIal of Zoology 14:20 1-43. Price, J. R. (1954). The alkaloids of the Autecology of Heliotropium euro­ Campbell , M. H. (1977). Assessing the Area Hefiotropium europaeum L. I. Heliotrine paeum L. in New South Wales and Vic­ and Distribution of Serrated Tussock (Na­ and iasiocarpine. Australiall Journal oj toria. CSIRO Technical Paper No. 7. 13 sella trichotoma). St John 's Wort (Hyper­ Chemistry 7:277- 86. pp. icum perJoratum var. anguslifolium) and Delfosse, E. S. and Cullen. J. M. (198Ia). Parsons, W. T. (1973). Noxious Weeds of Sifton Bush (Cassinia arcuata) in New New activities in biological control of weeds Victoria. [nkata Press, Melbourne. 300 pp. South Wales. New Sowh Wales Dept. oj in Australia. I. Common he liotrope. Tillyard, R. J. (1928). The biological control Agriculture Technical Bulletin No . 18. 23 Heli01ropium europaeum . Proceedings oj of noxious weeds. Transactiolls oj rhe pp . rhe Fifth lmemational Symposium on Bio· International Congress of Entomology. Caresche, L. A. and Wapshere, A. J. (1975). logical Control oj Weeds. 1980 . CSIRO IT haca, 1928 . 4(2):4-9 . Biology and host specificity of the Chon­ pp. 545-61. Wapshere , A. J. (1981). Biological control drilla root moth Bradyrrhoa gilveolella Delfosse, E. S. and Cullen, J. M. (I98 Ib). of Echium planragtneum : Northern Hem­ (Treitschke) (Lepidoptera : Phycitidae). New activities in biological control of weeds isphere studies. Proceedings oj the Fifth Bllllet;" oj Entomological Researl'h in Australi a. II. Echium plantagilleum : lmemational Symposium 011 Biological 65:171-85. Curse or Salvation? Proceedings oj the COllfrol of Weeds , 1980. CS IRO pp . Cull en. J. M. (1974). Seasonal and regional FIJth Illtenwlimwi Sympmiium on Bio­ 599-602. variation in the success of organi sms im ­ logical Comrol of Weed." 1980. CS IRO Wilson, F. (1960). A Review oJthe Biological ported to combat skeleton weed Chondrilla pp . 56J...74. Control of Insects and Weeds in Australia jUll cea L. in Australia. Proceedings oJlhe Delfosse, E. S. and Cu llen, J. M. (198Ic). and Australian New Guinea. Technical Third Illtemational Symposium all Bio­ New activities in biological control of weeds communication No. I , Commonwealth logical Comrol oj WeellJ. 1973. Mi scel- in Australia. III. S1 John 's wort , Hyperi- Institute of Biological Control. pp. 56-9.