Biological Control - CSIRO Research Publications Repository

1

NOTICE: This is the author’s version of a work that was accepted for publication in

CSIRO Publishing title: Biological Control of Weeds in Australia, eds Mic Julian, Rachel

McFadyen, Jim Cullen. Any changes resulting from the publishing process, such as peer

review, editing, corrections or structural formatting may not be reflected in this

document. Changes may have been made to this work since it was submitted for

publication. Link to Publisher’s version: <http://www.publish.csiro.au/pid/6509.htm>

2

Word count: 5,209

Heliotropium europaeum L. – common heliotrope

Andy W. Sheppard, Louise Morin and Jim Cullen

CSIRO Ecosystem Services, GPO Box 1700, Canberra, ACT 2601, Australia

ABSTRACT

Common heliotrope is a summer growing annual weed of crops and pasture, considered

native to Mediterranean Europe, Turkey and western North Africa, that is present in all states

and territories of Australia except Tasmania. It can be an important component of annual

pastures in southern Australia, where it often dominates fallow, disturbed or ploughed land

from December to April. When eaten by livestock it causes cumulative liver damage and

causes the highest fatality rate from primary pyrrolizidine alkaloid poisoning of any plant.

The biological control program targeting common heliotrope started in the 1950’s, with a

peak of activities from the early 1970’s to the early 1990’s, and consisted of natural enemy

surveys of more than 15 annual Heliotropium spp. in the western Mediterranean, eastern

Africa and Iran (the centre of origin of the genus). Nearly 40 arthropods and pathogens were

identified from common heliotrope out of the 132 species found across Heliotropium spp.

Host-specificity testing in Europe was undertaken for three arthropod candidate agents; the

root and leaf feeding flea beetle, Longitarsus albineus, the root weevil, Pachycerus segnis

and the flower cyme feeding moth Ethmia distigmatella, and two plant pathogens; the rust

Uromyces heliotropii and the leaf-blotch fungus, Cercospora heliotropiicola. Thousands of

flea beetles were released in Australia from 1979 until the late 1980’s following the

development of effective laboratory mass-rearing techniques, but the insects never persisted

more than a few years at a site and this failure was never adequately explained. The rust

fungus was released between 1991 and 1993, widely established and is still occasionally

found, but has had no impact on the abundance of the weed. None of the other candidate

agents were released. The inability of released agents to reduce weed populations was put

down to the difficulty of achieving biological control of an ephemeral summer annual

agricultural weed with highly variable between-year dynamics.

3

Key words: Boraginaceae, Longitarsus albineus, Pachycerus cordiger, Ethmia

distigmatella, Uromyces heliotropii, Cercospora heliotropiicola, Cercospora taurica.

INTRODUCTION

Heliotropium europaeum L. (Boraginaceae), common heliotrope, is a summer annual herb

considered native to the Mediterranean region and south-west Asia. It is a weed in North

America (Parsons and Cuthbertson 2001), but its impacts are far worse in Australia. Its

distribution in Australia is given in Figure 1. It is most abundant north and west of the Great

Dividing Range in the south east of the continent, where it occurs in a broad arc from

southern Queensland, through NSW and northern Victoria to SA. It is also abundant in the

lower south west of Western Australia (Parsons and Cuthbertson 2001). This corresponds to a

zone of between 300-500mm annual rainfall (Moore 1956). In southern Australia it

germinates from spring to early summer on bare soil with soil moisture once soil

temperatures reach 24°C and grows from November to May, with abundance in any given

year driven by the seasonal distribution of summer rain and the degree of litter accumulation,

which provide favourably moist conditions for plant growth (Moore 1956, Hunt et al., 2008).

It is a very ephemeral summer annual that by its germination requirements rarely occurs in

the same place twice in consecutive years in the same season. It often dominates fallow,

disturbed or ploughed land and can be an important component of annual pastures. When

eaten by stock it causes cumulative liver damage and leads to the highest fatality rate from

primary pyrrolizidine alkaloid poisoning of any plant. Common heliotrope is six times more

toxic than Paterson’s curse, which also contains pyrrolizidine alkaloids (O'Dowd and Edgar

1989). Historically pyrrolizidine alkaloidisis accounts for more than 63% of sheep losses in

some years (Edgar et al. 1941), although the situation more recently is unclear. Outbreaks of

common heliotrope do not occur every year, but in years of high abundance (e.g., 1991-93)

farmers generally face economic losses due to the impact of the weed and costs of

implementing largely ineffective chemical or cultural control (Parsons and Cuthbertson

2001). It has been estimated that on average in the 1980s, common heliotrope cost the meat

and wool industry at least $46 million pa ($17.5 m from livestock mortality, $14 m from

decreased productivity, and $25 m in control costs) (Cullen and Delfosse 1990). In a

4

cropping context it is considered to use up valuable soil moisture over summer prior to

autumn/winter crop sowing (Delfosse and Cullen 1980).

Common heliotrope is a hairy, ephemeral summer-annual herb usually 10-30 cm high with a

well-developed tap root system and a somewhat offensive odour. It reproduces only by seed.

The grey-green erect or semi-prostrate shoot has alternate elliptical to ovate leaves. Flowers

are small, white with a yellow throat, in two rows on the flowering coiled cymes, that

straighten as seeds ripen. The fruit has four nutlets with brown to black small subglobular

seeds (Parsons and Cuthbertson 2001) (Figure 2). Plants produce flowers and seeds

concurrently with vegetative growth, soon after germination, thereby creating a persistent

seedbank of up to 3,000,000 seeds m-2. They senesce in autumn if they are not previously

killed by drought or rain-induced leaf-blight outbreaks over the summer.

In a comparative study of common heliotrope in both its native and exotic ranges, Sheppard

et al. (1996) found maximum plant densities and basic demography (e.g., seed production

versus density) to be similar in both regions although the average infestation size and density

were greater and their occurrences more frequent in Australia. The higher summer rainfall

and more extensive rotational agriculture in Australia appeared to be the dominant causes of

the plant’s abundance status. These conditions however, could not fully explain why the

seedbanks in Australia were ten times larger than in the native range.

Control options for common heliotrope are well documented in Parsons and Cuthbertson

(2001) and include cultivation, herbicides and crash grazing by sheep, but each of these

options has significant economic, practical and animal health constraints. The persistent

seedbank is the other major hindrance in rotational cropping systems. Classical biological

control was identified very early, in the 1950’s. When major activity in this biological

control program commenced there was no regulatory requirement to either nominate a weed

as a biological control target, or consider risks that potential arthropod agents might pose to

native Australian plants, though this was already an issue for plant pathogens.

BIOLOGICAL CONTROL HISTORY

Australia is the only country to have attempted a biological control program against common

heliotrope. The program started with survey work by Frank Wilson from about 1950 in

5

Europe and North Africa, but was deemed "not especially promising" (Delfosse and Cullen

1980) and was shelved until 1971 when exploration was renewed in Europe, the Middle East

and North Africa. The early surveys and the work from 1971 to the late 1980s was funded by

CSIRO and then jointly by CSIRO and the Meat Research Corporation from 1988 when the

program received encouragement from producers, politicians and state agencies with across

jurisdictional support through the Australian Weeds Committee. Adding biological control to

existing management practices was considered the quickest way towards a sustainable control

strategy. While the early focus of the program was almost entirely on arthropod agents, one

of which was eventually released, the success of the skeleton weed rust in Australia prompted

consideration of potential pathogen agents in the 1980’s (Hasan 1985; 1992).

PLANT TAXONOMY

Common heliotrope is a member of a complex of closely-related Heliotropium species in the

eastern part of its native range in Eurasia, which has caused confusion on whether certain

natural enemies collected during early surveys had indeed been found on the target. Only

common heliotrope, H. europaeum, is considered to be present in Australia, but there are

several Australian native Heliotropium spp.

EXPLORATION

Frank Wilson's initial exploration included southern France, eastern Spain, coastal Algeria

and Tunisia. He then conducted surveys in Kenya, Eritrea and the Sudan. Eventually he

concluded that the prospects for biological control of common heliotrope were low (Delfosse

and Cullen 1980). There was no further work until 1971 when increasing pressure in

Australia to find some solution to the problem, plus the presence of a long-term base in

Europe encouraged renewed exploration. This base allowed more extensive exploration to the

east including the Middle East known to have a high diversity of species in the genus.

Surveys were subsequently conducted in Iran in 1971 and 1974 and in Tunisia and Algeria in

1974 mainly by John Huber. Iran was chosen as a primary site because it has a large number

of annual Heliotropium spp., some of which are similar to H. europaeum, and because of the

climatic similarity to areas of Australia where H. europaeum is a problem. These surveys

yielded a greater diversity of insects on Heliotropium spp. in northern Iran than in

6

Mediterranean Europe or North Africa, though the most promising agent, the flea beetle

Longitarsus albineus (Foudras) (Coleoptera: Chrysomelidae) was present everywhere.

Surveys continued through the 1970s in Mediterranean Europe and Turkey with the

promising root feeding weevil Pachycerus segnis Germar (=Pachycerus cordiger Germar

,=Pachycerus scabrosus Brullé) (Coleoptera: Curculionidae) being found in Greece and

Turkey as well as in Iran.

A number of later surveys were made to Eastern Europe from CSIRO’s European Laboratory

in the 1980’s and 1990’s, either to collect live material or to try to understand host species

usage by agents under consideration.

CANDIDATES

Natural enemies

The surveys of Heliotropium spp. from 1950 until 1980 in the native range found 132 species

of natural enemy 37, of which were considered to have been found on common heliotrope

(list in Delfosse and Cullen 1980). Table 1 lists those species considered to have biological

control potential. Diversity of natural enemies on common heliotrope was higher towards the

eastern end of its distribution and closer to the presumed evolutionary centre of origin of the

genus. Five species were considered to be potentially specific to at least the genus from the

literature at the time (Delfosse and Cullen 1980). Longitarsus albineus (flea beetle),

Utethesia pulchella (L.) (Lepidoptera: Arctiideae), Cercospora heliotropiicola A.K. Kar &

M. Mandal (Moniliales: Dematiaceae) and the weevil P. segnis are found throughout the

native range of common heliotrope, Ethmia distigmatella Erschoff (Lepidoptera: Ethmiidae)

is only found in the eastern part of the range. Delfosse (1985) discussed strategies for release

stating that the original proposed order of release had been the flea beetle followed by the

weevil then E. distigmatella and finally the rust fungus (U. heliotropii ), but that this

changed to the rust followed by the moth and then the weevil and finally the flea beetle based

on the status of the host specificity testing and availability. As it turned out only the flea

beetle and the rust were ever released. Four species were found on common heliotrope in

Australia prior to the release of exotic agents; a native flea beetle Longitarsus sp. (nr L.

victoriensis Blackburn) (C Reid pers. comm. 1993), a native arctiid moth, Utethesia

pulchelloides (L.), a native gelechiid moth Stomopteryx isosceliscantha (Lower) (Delfosse

7

and Cullen 1980; CSIRO Entomology 1994) and a probably introduced leaf-blotch fungus

Cercospora taurica Tanzschel (J Walker pers. comm. 1991; CSIRO 1994).

Natural enemies tested but not released.

Curculionid weevil; Pachycerus segnis. This large univoltine weevil completes its lifecycle

on species in the Boraginaceae with larvae developing in an earthen cell on the roots of

common heliotrope. The taxonomy has been confusing (Huber and Vayssières 1990) being

considered as an eastern European species P. cordiger with a sister western European species

P. scabrosus by two weevil systematists or as one pan European species later accepted as P.

segnis by others. Huber and Vayssières (1990) found that adults from Eastern Europe

(Turkey and Greece) were larger than museum specimens of either species from Western

Europe and so tested weevil populations collected from common heliotrope in Greece and

Turkey against seven European Boraginaceae and 31 economically important crop species

and. They also found no Pachycerus species commonly on common heliotrope in western

European. They concluded that the tested material was specific to Boraginaceae in the

laboratory and to Heliotropium spp. in the field, and that therefore should be safe for

introduction to Australia. Later concerns that no Australian species in the Boraginaceae had

been included in the original tests led to further testing in Europe. Testing of eastern

European populations over four years in the laboratory in France and in the field in Greece

showed that no Australian species of Heliotropium tested (H. asperrimum R.Br., H.

bacciferum Forsk and H. crispatum F. Muell. Ex Benth.) was suitable for larval

development, but other Australian Boraginaceae; Cynoglossum australe R.Br. and Myosotis

australis R.Br. supported adult feeding, oviposition and larval development (MJ Cullen

unpublished). Whether or not the phenologies of these Australian species might reduce their

exposure to use in the field in Australia by P. segnis was hard to assume and meant the level

of risk to them was difficult to assess and work on this insect was suspended. A further study

discovered a morphologically identical Pachycerus species to that from Eastern Europe

restricted to H. europaeum populations in Western Europe, adding host range support that

there was one pan-European species (Brun et al. 1993), which has since been recognised as

P. segnis (Alonso-Zarazaga 2010). Given these results and the concerns over specificity, no

further application was made to release this species in Australia.

Ethmiid moth; Ethmia distigmatella. This flower-cyme feeding moth recorded from

common heliotrope in Crete, Turkey and south west Asia was only once successfully reared

8

in quarantine in Australia on common heliotrope in small numbers despite several attempts.

An initial host-specificity test carried out using freshly emerged adults found no oviposition

on common heliotrope or the Australian native Heliotropium ovalifolium Forsk. tested.

Using the Flora of Turkey (Davis 1984), a survey of the complex of closely-related species

including H. europaeum, and the insects on this complex in western Turkey in 1991 failed to

confirm whether or not the moth had indeed ever been recorded from common heliotrope in

the native range. Based on this it was considered that common heliotrope is not the preferred

host of this moth (CSIRO Entomology 1994).

Leaf-blotch fungus; Cercospora heliotropiicola. Following rainfall, this plant pathogen can

produce severe disease epidemics on common heliotrope in Europe that can prematurely kill

populations prior to much seed production (Brun et al. 1995; Hasan et al. 1995). However in

the 1980’s another Cercospora species, C. taurica, was observed causing epidemics on

common heliotrope in Australia (J Walker pers. comm. 1991). This species was also known

from eastern Europe, Russia and western Asia (Chupp 1953). The taxonomic status of the

two Cercospora spp. attacking common heliotrope was explored morphologically and using

isozyme electrophoresis (Brun et al. 1996). Results showed consistent differences between

isolates classed as either C. heliotropiicola (from France) or C. taurica (from Turkey and

Australia) used in the study.

Before host-specificity testing of C. heliotropiicola could be justified it was important to see

whether C. heliotropiicola (ex. France) was more pathogenic than C. taurica, which already

occurred in Australia. A comparative demographic study of common heliotrope in Australia

and Europe demonstrated that Cercospora epidemics were both more common and more

damaging in Europe than in Australia, potentially contributing to the observed differences in

the seed bank sizes between the native and exotic range (Brun et al. 1995; Sheppard et al.

1996). Brun et al. (1996) also showed that C. heliotropiicola (ex. France) was more

pathogenic on detached leaves of common heliotrope than C. taurica (ex. Australia) and that

all five Mediterranean Heliotropium species tested were susceptible to both species.

C. heliotropiicola was therefore imported from France into quarantine in Australia in 1996 to

undertake detailed comparative studies with C. taurica. In experiments using whole plants,

both Cercospora species produced a similar number of leaf lesions on plants exposed to the

same temperatures (14, 22, 30°C) and wetness periods (8, 16, 24, 48 h) during the initial

9

infection phase (Morin and Trueman 1997). Under optimum conditions, the pathogens had a

similar impact on plant growth and reduced plant dry weight by 63%. In the first 24 hours,

C. taurica produced more than five times the number of conidia produced by C.

heliotropiicola on infected necrotic tissue placed in high humidity conditions. For both

species, production of conidia increased as the period under high humidity increased,

although C. taurica always produced the highest number of conidia. These results suggested

that under the same (controlled) conditions, C. heliotropiicola did not appear to have greater

value than C. taurica for biological control of common heliotrope and was consequently not

tested further nor released in Australia.

Natural enemies released.

Chrysomelid flea beetle; Longitarsus albineus . This multi-voltine flea beetle feeds on the

exteriors of fine roots as larvae, pupating in the soil and feeding on foliage as adults. In the

field in the native range adult feeding was recorded on four annual Heliotropium species and

typical feeding damage was observed on another 12 species in the genus. A method for

observing and rearing the root-feeding larvae was developed in the late 1970’s and the beetle

was found to have three to four generations a year under field conditions (Huber 1979, 1981).

A parasitic nematode (13% of collections) and a hymenopteran parasitoid were recorded from

this insect in the native range (Huber 1981). Host-specificity testing was carried out against

41 non-target plants using no-choice adult oviposition and larval starvation tests. The test

plant list included only two Australian native species; a Eucalyptus sp. and an Acacia sp.

Adult feeding was observed on 20 of the test plant species, but severity of feeding was less

than a tenth of that seen on common heliotrope. Oviposition was observed across a wide

range of test plants, although the number of eggs laid was a third or less than that found on

common heliotrope. Larval development was recorded on the Heliotropium peruvianum L.

and Myosotis alpestris F.W.Schmidt, but adults emerged only from the former species. At

the time assessing risks to native Australian species was not a requirement for arthropod

agents and therefore Huber (1981) considered that this agent was specific enough for release,

despite the test results suggesting that some Australian native species in the Boraginaceae

were likely to be within its host range.

Common heliotrope seedlings were killed by only a few larvae in tests so significant impact

in the field was predicted. Huber (1981) also concluded that L. albineus was physiologically

ill suited to breeding on non-summer Heliotropium or other Boraginaceae, but that its broad

10

climatic adaptability, its host finding ability and its rapid rate of increase meant that it should

have no difficulty establishing in Australia and that survival of overwintering adults would be

the main determinant of its effectiveness under Australian conditions.

The flea beetle was imported into Australia and first released in 1979. Beetles were

recovered the following season at most sites, but establishment beyond this first season was

not confirmed. A mass-rearing program was successfully developed in Canberra that

produced over 31,000 beetles, which were then released at seven sites across four states

(NSW, SA, Vic and WA) from 1987 to 1989. Measures, such as irrigation, were taken to

ensure common heliotrope presence in subsequent summers. Successful recoveries of flea

beetles were made at four of these sites in lower numbers in the next year and individuals

were recovered at some sites until 1992, but not after. As no flea beetles persisted at any site

longer than three years and it was clear from the search effort that no spread had occurred

monitoring was terminated in 1994. As no further sightings have been made it has been

assumed that this agent failed to establish in Australia. Field experiments in the native range

suggested that unless the post-release densities of flea beetles were very high compared to

those observed in the native range, this agent would have little impact on common heliotrope

abundance, given the already highly variable between year dynamics of this weed (CSIRO

Entomology 1994).

Rust fungus; Uromyces heliotropii. This macrocyclic and autoecious rust was originally

discovered on common heliotrope in Turkey, but was subsequently observed in France, and is

generally found, though only sporadically, across its native range. Field experiments in the

native range suggested that this rust could significantly impact on and kill common heliotrope

seedlings and could survive winter conditions and reinfect plants the following season (Hasan

and Ayres 1990; Hasan and Aracil 1991). Host-specificity tests were carried out on 96 non-

target plant species including eight species of Heliotropium native to Australia and 16 other

native species in the Boraginaceae (Hasan et al. 1992). The Australian native Heliotropium

crispatum F. Muell. Ex Benth. and two exotic Heliotropium spp. present in Australia were

susceptible to the rust, albeit at a much lower level than common heliotrope (Hasan et al.

1992). Any potential threat posed by the rust to H. crispatum was discounted as this species

is both temporally (winter growing) and geographically (northern Australia) separated from

common heliotrope (Hasan and Delfosse 1995).

11

An isolate from Turkey was first released in 1991 at a site in NSW, followed by 19 further

releases at 13 sites across four states (NSW, SA, Vic and WA) over three years, using

sprayed spore suspensions in water, covered by plastic sheeting in the first 24 hours, at each

release point. Local spread of the rust up to 500 m from the release plot was observed at 11

sites within the season of release (Sheppard et al. 1993) and overwintering teliospores were

seen at many sites. Six sites were monitored in the subsequent season and natural re-

infection of common heliotrope by the rust was observed at three of them, but only on one or

a few plants per site and only at sites where good spread had been observed in the initial year.

Infection in the second year never spread nor multiplied the way it did in the initial year and

the rust could only be found at one of the monitored sites as spot infections two years after

the releases. While climate appeared limiting in WA (too dry), the main cause of poor

establishment of this agent is probably the inability of the overwintering teliospores to

synchronise germination with the reappearance of the target in spring (S Hasan pers. com.

1992). The annual life cycle and ephemeral nature of common heliotrope may also limit the

abundance and therefore the potential impact of the rust, which, following winter without its

host, must undergo a sexual cycle on re-infection each spring. While this agent has had no

apparent impact on the target in Australia, its persistence continued to be confirmed through

occasional sightings on common heliotrope at sites across south-eastern Australia unrelated

to those where original releases were made.

DISCUSSION

As one of the longest running of Australia’s biological control programs, with very extensive

native range research activities across Europe, North Africa and the Middle East, the program

against common heliotrope is most notable by its failure to deliver any kind of effective

control of the target. Frank Wilson, the earliest scientist to work on it, did not rate the

prospects of biological control very high. His 1950 report has since been lost so we can not

be sure of his reasoning, but it is likely he came to this conclusion after his initial efforts did

not find many apparently specific candidate agents in the 1950’s (listed in Delfosse and

Cullen1980), and it was only pressure from the agricultural sector for CSIRO to do

something, plus the opportunity to explore closer to the centre of origin of the species, that

stimulated more work. Also the limited chance of success of biological control against

summer annual cropping weeds is due to a number of factors, including their ephemeral

nature (Chaboudez and Sheppard 1995) was probably not as well appreciated at the time.

12

Annual plants are also likely to support a smaller community of natural enemies than longer

lived species because of this.

The only insect to be released, L. albineus, was as Huber (1981) rightly considered “a good

looking agent on anyone’s books”. That it was so hard to establish, despite a highly

successful mass-rearing program, is a significant enigma, unless Australian soil conditions or

summer temperatures or over-wintering conditions were just too unsuitable in some way for

successful beetle survival. As it was, there was every likelihood this agent would have had to

reach very high numbers very quickly within a growing season to have any impact on

common heliotrope populations and would have also attacked many native Heliotropium spp.

and possibly other Boraginaceae had it become widespread (CSIRO Entomology 1994).

This program also fell at a time when Australia was enthusiastic about plant pathogens as

potential agents following the successful use of a rust fungus for the biological control of

skeleton weed in Australia, another cropping weed. Expectations that the two promising

pathogens found on common heliotrope, including a rust fungus, could achieve similar levels

of success would have been quite high. With hindsight, could the failure of the common

heliotrope rust to impact on the weed have been predicted from its rarity in the native range?

Perhaps, but this program was the first to show that rust fungi do have an Achilles heal when

their host is so ephemeral. Most of the teliospores, the stage of the life-cycle of rusts that

enables survival over winter without a host, could have germinated with moisture in the

spring when environmental conditions were suitable and produced infective basidiospores,

but before temperatures were high enough for germination of the host plant. That is their may

be a problem of phenological synchrony. In comparison, the Cercospora spp. found on

common heliotrope cause a much more damaging disease, often resulting in severe epidemics

in both the native range and Australia (Brun et al. 1995; Sheppard et al. 1996). The impact

of these pathogens however, is driven by summer rainfall, which provides optimal conditions

for infection and associated rain splash necessary for local spread. Keeping the infected litter

in the paddocks may be a way of encouraging such epidemics, provided the period until new

host foliage is present at the sites is not too long.

This common heliotrope program provided a reality check on over optimism that successful

biological control of cropping weeds was possible. It also provided key lessons on the need

to focus on the target weed species rather than the target genus when there are con-generic

13

species native to Australia in order to not waste time investigating ineffective candidate

agents. Also since the early 1980’s agents to be released in Australia have had to be more

specific anyway to address the legislative requirements of the Wildlife and Countryside Act

of 1981.

ACKNOWLEDGEMENTS

We gratefully acknowledge the contribution of Laurent Brun, Ernest Delfosse, Siraj Hasan,

John Huber, Bobby Lewis, Carey Smith, Tony Wapshere, Frank Wilson, Tim Woodburn and

many technical staff to this biological control program.

REFERENCES Alonso-Zarazaga, MA (2010), Fauna Europaea: Pachycerus segnis (Germar 1824). Fauna

Europaea version 2.2, http://www.faunaeur.org

Brun LA, Sheppard AW and Carrara A (1993) Host range of Pachycerus cordiger Germar

(=P. scabrosus Brulle) (Col.: Curculionidae). Entomophaga 38, 537-539

Brun LA, Sheppard AW, Hasan S, Jourdan M, and Delmotte E (1995) Development of

Cercospora blight epidemics and effect on the summer annual weed Heliotropium

europaeum L. in the field. Annals of Applied Biology 127, 137-150.

Brun LA, Sheppard AW and Lopes V (1996) Variation in pathology of isolates of

Cercospora spp. attacking Heliotropium spp. Annals of Applied Biology 128, 565-570

Chaboudez P and Sheppard AW (1995) Are particular weeds more amenable to biological

control? - a reanalysis of mode of reproduction and life history. In: Proceedings of the VIIIth

International Symposium on Biological Control of Weeds. 2-7 February 1992, Canterbury,

New Zealand. (Eds ES Delfosse and RR Scott) pp. 95-102. DSIR/CSIRO, Melbourne,

Australia.

Chupp C (1953) A monograph of the fungus genus Cercospora. Ithaca, New York. pp. 667.

Comment [jul007 1]: Use en dashes rather than hyphens in the refs.

14

CSIRO Entomology (1994) Project CS.104 Biological control of common heliotrope,

Heliotropium europaeum. Contracted Research Report for Meat Research Corporation.

Cullen JM and Delfosse ES (1990) Progress and prospects in biological control of weeds. In:

Proceedings of the Ninth Australian Weeds Conference. 6-10 August, 1990, Adelaide. (Ed.

JW Heap) pp. 452-470. Crop Science Society of South Australia, Adelaide, Australia..

Davis P (1984) Flora of Turkey and the East Aegean Islands. Edinburgh University Press.

Delfosse ES (1985) Re-evaluation of the biological control program for Heliotropium

europaeum in Australia. In: Proceedings of the VI International Symposium on Biological

Control of Weeds. 19-25 August 1984, Vancouver, Canada. (Ed. ES Delfosse) pp 735-742.

Agriculture Canada, Ottawa, Canada.

Delfosse ES and Cullen JM (1980) New activities in biological control of weeds in Australia.

I. Common heliotrope, Heliotropium europaeum. In: Proceedings of the V International

Symposium on Biological Control of Weeds. 22-27 July 1980, Brisbane, Australia. (Ed. ES

Delfosse) pp. 545-561. CSIRO, Melbourne, Australia.

Edgar G, Hindmarsh WL, Keast JC and Rose AL (1941) The distribution and incidence of

enzootic or toxaemic jaundice in south-eastern Australia with special reference to New South

Wales. Australian Veterinary Journal 17, 120-130.

Hasan S (1985) Prospects for biological control of Heliotropium europaeum by fungal

pathogens. In: Proceedings of the VI International Symposium on Biological Control of

Weeds. 19-25 August 1984, Vancouver, Canada. (Ed. ES Delfosse) pp. 617-623. Agriculture

Canada, Ottawa, Canada.

Hasan S (1992) An update of the biological control of common heliotrope in Australia using

plant pathogens. In: Proceedings of the First International Weed Control Congress, Vol. 2.

17-21 February 1992, Melbourne, Australia. (Ed. RG Richardson) pp. 212-214. Weed

Science Society of Victoria Inc., Melbourne, Australia.

15

Hasan S and Aracil E (1991) Biology and effectiveness of Uromyces heliotropii Sred., a

potential biological control agent of Heliotropium europaeum L. New Phytologist 118, 559-

563

Hasan S and Ayres PG (1990) Tansley Review No. 23. The control of weeds through fungi :

principles and prospects. New Phytologist 115, 201-222.

Hasan S and Delfosse ES (1995) Susceptibility of the Australian native, Heliotropium

crispatum, to the rust fungus Uromyces heliotropii introduced to control common heliotrope,

Heliotropium europaeum. Biocontrol Science and Technology 5, 165-174

Hasan S, Delfosse ES, Aracil E and Lewis RC (1992) Host-specificity of Uromyces

heliotropii, a fungal agent for the biological control of common heliotrope (Heliotropium

europaeum) in Australia. Annals of Applied Biology 121, 697-705.

Hasan S, Jourdan M, Brun L and Sheppard AW (1995) Axenic culture and Influence of

wetness period and inoculum concentration on infection and development of cercospora

blight of Heliotropium europaeum. European Journal of Plant Pathology 101, 491-496

Huber JT (1979) A method for observing and rearing the root-feeding larvae of Longitarsus

albineus (Foudras) (Col. Chrysomelidae). Bulletin de la Société Entomologique Suisse 52,

431-433.

Huber JT (1981) Observations on the heliotrope flea beetle, Longitarsus albineus (Col.:

Chrysomelidae) with tests on its host specificity. Entomophaga 26, 265-73.

Huber JT and Vayssieres JF (1990) Life cycle and host specificity of the heliotrope weevil,

Pachycerus cordiger (= madidus Auct.) [Col.: Curculionidae]. Entomophaga 35, 475-484.

Hunt JR, Cousens RD and Knights SE (2008) The Biology of Australian Weeds 51.

Heliotropium europaeum L. Plant Protection Quarterly 23, 146-152

16

Moore CWE (1956) Observations on the autoecology of Heliotropium europaeum L. in New

South Wales and Victoria. CSIRO Division of Plant Industry Technical Paper No 7. CSIRO,

Melbourne.

Morin L and Trueman HE (1997) Pathogenicity and inoculum potential of two Cercospora

species on the weed Heliotropium europaeum. In: Programme and Summaries: 11th Biennial

Conference of the Australasian Plant Pathology Society. p. 49.

O’Dowd DJ and Edgar JA (1989) Seasonal dynamics in the pyrrolizidine alkaloids of

Heliotropium europaeum. Australian Journal of Ecology 14, 95-105.

Parsons WT and Cuthbertson EG (2001) Noxious Weeds of Australia. Second Edition.

CSIRO Publishing, Collingwood, Victoria.

Sheppard AW, Lewis RC and Delfosse ES (1993) The establishment of Uromyces heliotropii

Sred., a biological control agent of Heliotropium europaeum L. In: Proceedings of the 10th

Australian & 14th Asian -Pacific Weed Conference.6-10 September 1993, Brisbane,

Australia. p. 89-93. Queensland Weed Society, Brisbane, Australia.

Sheppard AW, Brun LA, Lewis RC. (1996) A demographic comparison of common

heliotrope, Heliotropium europaeum L.: southern Australia and southern France. Proceedings

of the 11th Australian Weeds Conference. 30 September-3 October 1996, Melbourne,

Australia. (Ed. RCH Shepherd) pp. 286-290. Weed Science Society of Victoria Inc.,

Melbourne, Australia.

Table 1. Species considered of possible biological control potential discovered during surveys of H. europaeum.

Species Common name S. France,

E. Spain,

Algeria, Tunisia

1950

Iran 1971 1974

Tunisia, Algeria

1974

Med. Europe, Turkey. 1970s

Coleoptera

Comment [jul007 2]: More details?

Comment [van215 3]: These aren't really "common names" as such

Comment [van215 4]: Perhaps replace with "feeding damage/type" column

Comment [van215 5]: Not clear what dates refer to and what their significant is. Perhaps just have a column with "distribution" and a text field summarising it. (possibly with survey dates in brackets if that is important)

Comment [van215 6]: An additional column summarising state of play would be useful. E.g. Released 1979. Status unconfirmed since last citing in 1992; Tested extensively but not sufficiently host-specirfic;

17

L. albineus Flea beetle X X X Released 1979. Status unconfirmed

since last citing in 1992;

X

P. cordiger Root weevil X Tested extensively

but not sufficiently

host-specirfic;

X

Neoliturus opacipennis (Leth)

Cicadellid X X* ?

Austroagallia sinuata (M.R.)

Cicadellid X X* ?

Atomoscelis signaticorus Reut.

Mirid X ?

Polymerus sp. Mirid X ?

E. distigmatella Cyme moth X Failed to establish

culture, non-preferred host

X

Utethesia pulchella

Defoliating moth X ?Already in Oz X

Uromyces heliotropii

Rust fungus X Released 1991, widely

eatsblished but epidemics

never observed

C. heliotropiicola

Leaf blotch fungus Abundant in Oz etc

X

* Recorded as Neoliturus sp. and Austroagallia sp.

Comment [van215 7]: Spell out

Comment [van215 8]: Not consistent with text?

18

Figure 1. Distribution of common heliotrope, Heliotropium europaeum, in Australia.

(Reproduced from Australia’s Virtual Herbarium with permission of the Council of Heads of

Australasian Herbaria Inc; map generated on 1 May 2010).

19

A

B

Figure 2. Common heliotrope Heliotropium europaeum A. The plant. B. An infestation near

Gundagai, NSW.

20

A

B

Figure 3. A. The flea beetle Longitarsus albineus B. The rust fungus, Uromyces heliotropii.