application for licence to release a non-native species ... · 3 see rebeca work package 5 –...

Application for licence to release a non-native species for biological control purposes in England, under Sections 14 and 16 of the Wildlife and

Countryside Act 19811,2

This application form and associated guidance was developed by the EU-funded ‘REBECA’ project

(Regulation of Biological Control Agents) (http://www.rebeca-net.de/) to support the use of macrobial biological control agents.

This application form and associated guidance may be used for any applications to release a non-native invertebrate biological control agent into the environment in England under Sections 14 and

16 of the Wildlife and Countryside Act 1981. The competent authority in England for these applications is The Department for Environment, Food and Rural Affairs (Defra). Defra has

nominated the Central Science Laboratory (CSL) to deliver the service and all applications should be submitted to CSL.

Details for submission of an application are provided below. It is not obligatory to use this form for

applications providing all information specified in Annex 2 of the Act is provided. Using this form

This form should be used for the submission of an application to Defra for a permit to license the release of a non-native invertebrate natural enemy used for the biological control of invertebrate and plant pests (Invertebrate Biological Control Agent or IBCA). Organisms include invertebrates as well as entomopathogenic nematodes3, but not micro-organisms. Guidance on the completion of this form is provided in the accompanying Guidance Document1. This form is valid for an application relating to a single biological control organism. An organism is characterized as any identifiable and recognisable taxon of the IBCA, either a species, or recognized sub-species, population, strain or biotype. After CSL has received your application (administrative forms and documentation/dossier), you will receive an acknowlegdement of receipt within 5 working days. The application will then be checked for completeness and subjected to a risk assessment in relation to the purpose of your application (e.g. for research under quarantine conditions, or a commercial release). The risk analysis will be conducted by an established group of experts. CSL will conduct a risk analysis in the light of the information provided, or any other sources they have available. CSL may need to contact you to clarify parts of the application or to seek further information. At all times and in all communication, including that with external experts, your application will be regarded as confidential. After the risk assessment has been completed, CSL will make a decision as to whether to grant a permit. CSL aims to issue a decision within 12 weeks from the receipt of application and will keep you informed of the progress of your application. The licence to permit an import and/or release will be valid for a fixed period of time, assigned by Defra, after which a renewal may be sought, or a request may be made to place the organism on the EPPO Positive List. In the case of mixed products, an application should be made for each separate component. Information required to complete this form

This application form and related information requirements for the release of non-indigenous IBCAs contains 5 parts (numbered 1-5) and is structured in a step-wise way: depending on the origin of the organism and the purpose of the application, the sequence of assessments and level of information required is related to the perceived level of risk. An application for any specified organism should include the following information:

1 Guidance on the completion of this application form is provided in a separate document 2 This application form and guidance was developed by the EU-funded ‚REBECA’ project (work package on macrobial

biological control agents). Key authors: A.J.M. Loomans, F. Bigler, G. Sterk and J.S. Bale. For all correspondence contact [email protected].

3 See REBECA work package 5 – Recommendations for regulation requirements for entomopathogenic nematodes

WCA Application form V1 April 2008 of 31

http://www.rebeca-net.de/

mailto:[email protected]

Part 1. Application information A. Information on the applicant B. Purpose of the application and use

Part 2. Information on the invertebrate biological control agent

A. Taxonomy and origin B. Product information

Part 3. Information requirements for the intentional release of a non-indigenous IBCA:

A. Biology and ecology B. Assessment of risks and benefits

a. Establishment, b. Host specificity c. Dispersal d. Direct and indirect effects

Part 4. Submission of forms and signature

A. Submission details B. Agreement: safeguards and signature

Part 5. Appendices – any additional relevant information Submitting an application and further information:

All applications and queries should be directed to: Sarah Hugo Central Science Laboratory Sand Hutton York YO41 1LZ, UK Tel: + 44 (0) 1904 462223 Fax: + 44 (0) 1904 462250 Email: [email protected] or [email protected] Further information is available at http://www.csl.gov.uk/nonnativebiocontrol




http://www.csl.gov.uk/nonnativebiocontrol

Application for licence to release a non-native species for biological control purposes in England, under Sections 14 and 16 of the Wildlife and Countryside Act 1981 Part I. Application Information A Information on the Applicant

Name of organization

Cornwall County Council on behalf of the Japanese knotweed Natural Control Project Board (awaiting conformation and legal advice)

Name of applicant* TBA

Affiliation of applicant

Address

Postal code

City

Phone

Fax

E-mail

1.1 Who will apply for the permit? *only a legally authorized person is allowed to apply. Include confirmation of the person’s authorization with the application.

Chamber of Commerce # N/A

Name of contact person Dr. Richard Shaw

Affiliation of contact person

Principal Investigator at CABI E-UK

Visiting Address Bakeham Lane, Surrey

Postal code TW20 9TY

City Egham

Phone Tel: + 44 (0) 1491 829025

Fax + 44 (0) 1491 833508

1.2. Who is the contact person? Contact person, research manager and/or quarantine officer.

E-mail [email protected]


B Purpose of Application and Use

Application type Renewal First Application

Renewal (application number and expiry date)

Positive List organism Yes No Relation with previous/ other applications None

Application or registration elsewhere in Europe No

1.3. Information on application

Licence period requested May 2009 – August 2012

Import Research (Mass) rearing

Release Trials Commercial*

Type of biocontrol programme

Classical Weed Biocontrol using an Insect Biological Control Agent (IBCA)

1.4. Purpose of use *To include full scale release of a classical biocontrol agent

Area of release Full, classical, intentional field release into the UK’s natural environment

Address N/A

Postal Code N/A

Location

Agent already held under quarantine licence PHL 199B/5746 (11/2007; amended 04/2008) in quarantine facility at CABI Europe-UK

Facility N/A

Contingency plan N/A

Standard Operating Procedures

N/A

Quality control management

N/A

1.5. Facilities and procedures Describe how the risks, and the extent or probability of escape into the wild will be managed

Accreditation N/A

1.6. Information on target organism(s) Give a description of the biology and ecology of the target pest(s), including weeds Target host taxon

Order: Polygonales; Family: Polygonaceae; Genus: Fallopia; See below for species names of host knotweeds and closely related congeners/species varieties that can be considered targets under the Japanese knotweed Natural Control Project (CABI ref VM10036 formerly VM03021)


Names of target pests

Main target species : Japanese knotweed - Fallopia japonica var. japonica (Houtt.) syn. Polygonum cuspidatum syn. Reynoutria japonica Invasive alien species in the UK, restricted under the Wildlife and Countryside Act (1981). Bohemian knotweed – Fallopia x bohemica (Chrtek & Chrtková) JP Bailey. Hybrid of F. japonica & F. sachalinensis, widespread non-native pest in the UK and appears to be capable of spreading faster than F. japonica (Mandák et al., 2004) Giant knotweed – Fallopia sachalinensis (F.Schmidt ex Maxim.) Ronse Decraene (This non-native weed has limited distribution but can be problematic where it is found).

Original area of distribution of the pests

F. japonica is native to Japan, Sakhalin Island, the Kurile Islands, Korea, SW China, Taiwan, and Vietnam (Ohwi 1984, Jäger 1995) but the main target for this application originates from and is a widespread plant in Central and Southern Japan F. x bohemica is a hybrid that has formed independently in the invasive range after fertilisation of the female F. japonica var japonica with pollen from Giant knotweed, Fallopia sachalinensis which has its origins in the Northern part of Japan/Sakhalin Island. F. x bohemica is more restricted than Japanese knotweed but its distribution is not well reorded owing to difficulty of identification. It occupies a similar habitat to the main target Japanese knotweed


Biology of pests

F. japonica and the hybrid F x bohemica (Bailey et al., 1996), are widespread invasive weeds throughout the UK from Cornwall to Scotland and the Isles (Beerling et. al., 1994; Bailey & Conolly, 2000) and are likely to expand their range and prevalence further north under global warming (Beerling, 1993). These Japanese knotweeds are found in riparian habitats and derelict land as well as roadsides, lay-bys and railway embankments. All these invasive knotweeds have serious consequences on biodiversity (Maertz et al., 2005, Gerber et al., 2008, Topp et al., 2008). F. japonica is a herbaceous perennial plant (hemicryptophyte), and in its native range it is considered functionally dioecious (Maruta, 1981). In the introduced range F. japonica reproduces vegetatively by means of its extensive rhizome system which acts as a carbohydrate store in the winter months and which is often fragmented during disturbance. Cut stem tissue can also develop and initiate shoots. The plants grow very quickly in spring (main growing period in May to June) and reach a height up to 3 m (Beerling et al., 1994) which is generally taller than in its native range in Japan (Holzner & Numata 1982) where it is recorded as being 0.3-1.5 m tall (Makino 1997). In the introduced range, most, if not all F. japonica plants are female and thought to be clonal since they are derived from a very small number of initial introductions (Hollingsworth & Bailey 2000). Although pollen of F. japonica is mostly lacking, the plant often produces seeds but these are usually of hybrid origin and have not been found to be viable in the field

Target crops None – this is a proposed release against a weed that is not associated with crops


Part 2. Information about the Invertebrate Biological Control Agent A Taxonomy and Origin

Class Insecta

Order Hemiptera

Family Psyllidae

Genus Aphalara

Species Itadori

Sub-species The “Kyushu strain” collected from Mt. Aso region of Kyushu

Common names Itadori madarakijirami (Japanese for “Japanese knotweed speckled psyllid”)

Alternative names English translation : Japanese knotweed psyllid

2.1. Identity For what species/organism is the application made? Indicate which species is involved (a single species per application) and full scientific name and taxonomy

Associated organisms

A single eulophid parasitoid has been reared from Aphalara itadori in Japan and this appears to be a Tamarixia species. The psyllid culture, from which any release would be made, has been cleaned of this organism and any other arthropod and fungal parasites through line rearing and has remained free from pests for the past 3 years so it should not act as a host or vector of any associated pests from its native range. There are no records of economic pests that either specialise on the psyllid genus or exist on the host plant, that could be vectored in its current area of distribution.

Authority

A. itadori was first described in the genus Psylla by Shinji, O (1938). It was later transferred to Aphalara by Miyatake (1964). The original description is provided in Appendix 1 as is the more useful redescription by Burckhardt & Lauterer (1997).

Methodology

Morphological and taxonomical (see above) and confirmation from Dr D. Hollis Dept of Entomology, the Natural History Museum, London (collection reference number 24205)

2.1.1. ID-Confirmation Indicate means, methods of ID-confirmation and vouchers.

Voucher deposits Dept of Entomology, the Natural History Museum, London (collection reference number 24205)


2.2. Characterization of IBCA Specify life-stages, strains or taxonomic constraints

Diagnostic descriptions

A. itadori is somewhat unusual in the genus, and gives its name to a species-group containing only itself and one other species, A. taiwanensis. For more diagnostic information see Appendix 1. The damage caused by the nymphal stage of the psyllid can result in the death of potted knotweed hosts but it is most easy to spot from the prolific wax production by the feeding nymphs. Adults can be spotted on the leaves and stems of the plant but damage with adult feeding damage only evident by sporadic honeydew production.


Specific characteristics

A. itadori can be distinguished morphologically from other members of the same genus. There appears to be a difference in host preference between populations of A. itadori collected from F. sachalinensis in Hokkaido in the far North of Japan and this strain collected in Kyushu studied for potential release in the UK (Rob Bouchier & Fritzi Grevstad, personal communication, Jan 2008). Thus, strain or sub-species distinctions between psyllid populations may be possible in the future. A. itadori is found from sea level to altitudes of 2,150 m.a.s.l., so it can be assumed to be very tolerant of climatic extremes. Initial lab studies show that systemic pesticides are highly effective against psyllid populations with various stages but contact insecticides are less effective (see Appendix 6). Eggs are relatively resilient and are often laid under the papery sheaths surrounding the nodes of the host plants giving them some degree of protection to topical pesticide application but it is assumed that systemic insecticides will still be effective since the egg period should be less than the period off insecticide protection provided by commercial systemic pesticides. At high numbers, nymphs conglomerate and produce a protective and highly hydrophobic waxy flocculence that could provide protection against contact pesticides. The life-cycle is summarised in Appendix 2 and consists of egg, 5 nymph stages and followed by adult, all mobile stages of which feed on the host plant. The nymphs are sedentary but early instars move to the papery sheaths surrounding the nodes of the knotweed stem. The final moult to adult normally takes place on the underside of an older leaf.

Taxonomic characteristics

For full description of taxonomic characteristics see Appendix 1 and details of life stages, see Appendix 2

2.3. Origin and Origin Non-indigenous to Great Britiain


Field collected

From F. japonica var. japonica populations on and around the Mt. Aso region of Kyushu, Kumamoto Prefecture ; 850-1,500m a.s.l. Latitude: 32o 53.920’ N Longitude: 131o 03.161’ E

Laboratory culture

The psyllid culture was established from adults collected in Southern Japan in mid 2004 primarily on and around Mount Aso in Kumamoto prefecture, Japan. Subsequently, all laboratory culturing was carried out in controlled environment rooms (13 hours daylight and 11 hours darkness at a temperature of 22oC ± 1.5 oC with 50-85% humidity) under quarantine conditions in Silwood Park, Ascot, Berkshire and from 2008 in CABI’s quarantine facilities in Bakeham lane, Egham, Surrey under Plant Health licence PHL 199B/5746 (11/2007; amended 04/2008)

Producer/Supplier N/A

Original area and distribution

The distribution of A. itadori is recorded as Japan, Korea, Russia, Sakhalin, Kurile Islands (Kwon 1983, Gegechkori & Loginova 1990). Although the psyllid considered by this PRA is from the Mt. Aso region of Kyushu – The “Kyushu strain”.

Distribution IBCA What is the immeditate source of the organism. Include details of the origin and distribution of the IBCA (species or lower taxon)

Areas introduced before The pest has only ever been introduced into licensed quarantine facilities in the UK and Canada.


B Product Information

Product/Trade name N/A

Producer/Supplier N/A

Method of supply Single species

Life stages Adults of both sexes

Label information N/A

Storage N/A

2.4. Product Information

Method of use

Specialist sap sucking insect. Intentional release as an IBCA against Japanese knotweed. It is normal for establishment to be achieved with a single release of a large number of biocontrol agents. If establishment fails then repeated releases are anticipated.

Co-formulants None 2.5. Product Composition

Contaminants None


Part 3. Information requirements for intentional release of a non-indigenous IBCA

A Biology and Ecology

Life cycle – generations/ year

In its native Japan, Aphalara itadori adults along with eggs, were observed from late April to mid August. It is a multivoltine species and according to the climatic assessment presented in Appendix 4, at least one generation per year is possible across most of the UK.

3.1. Information on Biology and Ecology Give a description of the biology and ecology of the IBCA

Developmental biology

Aphalara is a multivoltine species and reproduces very rapidly in ideal conditions with fertile adult females producing an average of 637 ± 121.96 (±1SE, n = 11) eggs each over a production period of 37.5 days ± 5.85 days (mean ±1SE, n = 11) and up to 75 days (at 23 oC) The developmental period taken from Table 2, Appendix 2) is as follows (±SE, n=21) Egg – 9.2 days ± 0.1 N1 – 4.8 ± 0.2 N2 – 3.3 ± 0.2 N3 – 3.9 ± 0.3 N4 – 4.5 ± 0.1 N5 – 7.5 ± 0.3 Total development 32.9 ± 0.8


Mechanisms of survival

During the 4-year survey period (2004-2007) in Japan, A. itadori was never found on any other plant species, other than the Japanese knotweeds encountered. However, the over-wintering habit of the psyllid is not well known and its shelter plants, in the absence of the primary knotweed host, which dies back at the first frost, are presumed to be evergreen trees. Baba & Miyatake (1982) and Miyatake, (1973; 2001) recorded over-wintering A. itadori on Japanese pine Pinus densiflora [Pinaceae] and Japanese cedar Cryptomeria japonica [Taxodiaceae] in Japan. Miyatake (1973) states "it is unknown whether adults suck their shelter plants such as Pinus spp. and Cryptomeria spp" (translated from Japanese). Currently there is no evidence to suggest that over-wintering of the psyllid is reliant on a particular host species, or that it causes any noticeable damage to other species in the absence of its normal food plant, Fallopia japonica. Hodkinson, (1974) indicated in a review of general Psyllid biology that it is not known whether over-wintering adults feed on shelter plants, though a consideration of their moisture requirements would suggest they do. Ossiannilsson, (1992) stated that “most of our [Scandinavian] psyllids hibernate in the adult stage, a few on their host-plants (Cacopsylla pyricola-the pear psyllid) but most species on "shelter plants", usually conifers, or in crevices in bark or other protected sites”. Ossiannilsson goes on to say “whether psyllids hibernating on conifers do actually feed on them, as was supported by Reuter (1909), is still unknown”. Dolling (1991) comments that the four Aphalara spp. currently found in Britain feed on various species of dock, bistort and knotgrass, but adults are often encountered sheltering on evergreens from late summer through to Spring.


Mechanisms of dispersal

Only the adult stage is truly mobile and adults are not strong flyers, relying more on jumping and wind currents for migration to local plants. Kristoffersen & Anderbrant (2005) indicated that the carrot psyllid, Trioza apicalis could cover distances of up to one kilometre and that there was a trend to spread in the direction of the prevailing wind. However, the same authors later found the wind direction seemed to be irrelevant but that most psyllids were collected within 250m of the crop (Kristoffersen & Anderbrant, 2007). NB - Intentional transportation or redistribution by humans may play a role in distribution, if the biocontrol agent is approved for release and perceived as a solution to invasive knotweed populations by the general public.

Climatic conditions

Though Japan’s eco-climatic conditions are not directly comparable with those of the UK, its volcanic nature provides higher altitude sites which are comparable to the UK even in Kyushu Island (Appendix 4). Populations of A. itadori can be found on Kyushu Island from sea level to the top of Mount Aso (over 1,500m. above sea level) so it can be assumed to be very tolerant of climatic extremes. The culture under consideration was collected from and around Mt. Aso, at altitude. On-going studies of its thermal tolerance and degree-day requirements show that no development took place at 10ºC and very little at 12ºC. As a rule of thumb, species with a 10ºC minimum threshold have been considered to be marginal for the UK (Baker & Bailey, 1979; Baker, 2002). A detailed assessment of the thermal tolerance of A. itadori is presented in Appendix 4. This study suggests that it would establish successfully across most of the UK and would only be excluded from parts of Scotland.


Habitat range

A. itadori has been found at sea level in Kagoshima and on the upper slopes of Mt. Aso in Kyushu Island so it is very adaptable and tolerant of climatic extremes. It’s habitat is primarily determined by oviposition behaviour which is determined by the presence of its host, Japanese knotweed since this host or F. sachalinesis and their hybrid F x bohemica appear to be required for the completion of the lifecycle and therefore sustenance of a population. However, over winter the preferred habitat is likelky to switch to evergreen trees as has been suggested by numeros authors (see “mechanisms for survival)


Host range

During the 4 year survey period (2004-2007) in Japan, A. itadori was never found on any host plant other than knotweeds. During the extensive experimental host range studies reported in Appendix 2 this very narrow host range was confirmed. It can be seen that adult mortality data on various hosts show that survival on anything other than knotweed hosts is severely compromised and as such these plant species can be considered unsuitable hosts and incapable of sustaining viable psyllid populations. However, the choice of true host is determined by the oviposition behaviour of the adult psyllids since nymph stages are sedentary. In multiple-choice tests the location of over 146,885 eggs were recorded and 98.5% of these were laid on plants in the invasive knotweed group, Fallopia japonica, F. sachalinensis and F x bohemica. Thus 1.5% were laid on non-target species or varieties with other ornamental knotweeds, F. conollyana and F. japonica var compacta along with the invasive Russian vine F. baldshuanica accounting for another 1,422 eggs (0.97%). Importantly, only 5 UK native species, namely Fallopia dumetorum, F. convolvulus, Fagopyrum dibotrys, Rumex hydrolaphum and Oxyria digyna received eggs (a total of 150 eggs =0.1%). Not one of the eggs laid in multiple choice tests was able to develop successfully to adult. However, forced nymph transfer studies under elevated humidity conditions did reveal that the non-native Meuhlenbeckia complexa could support development to viable adult and. These studies also revealed that a few of the native non-target species could support limited feeding and development. However, since oviposition rates on these species in the presence or absence of the target knotweed were tiny this is highly unlikely to happen in the field.



Natural enemies

In the south of Japan, the species is rarely found in high numbers which could be due to parasitism, though only one unidentifiable eulophid parasitoid has been reared from a late nymph (possibly a Tamarixia species (A. Polaszec, pers. com.) The possibility of UK natural enemies is dealt with in the report by A. Polaszec (Natural History Museum) which is presented in Appendix 1. It would seem that the likelihood of significant attack by UK native parasitoids on A. itadori is low. According to Noyes (2007), no known parasitoids of Aphalara itadori are listed, and out of a total of 90 records of Psyllidae parasitoids, only two are associated with Aphalara species. Both apparently concern North American species, and both are likely to be Psyllaephagus species (Encytidae), although one is recorded as “Encyrtus aphalarae”. Of the 90 records of chalcidoid parasitoids of Psyllidae, more than 60 concern Psyllaephagus spp. On the face of it, there is therefore some probability that an indigenous British Pyllaephagus species, of which there are currently two, could use Aphalara itadori as a potential host. However, of these two species, one (P. lusitanicus Mercet) is known only from the scale family Asterolecaniidae, and the other (P. pilosus Noyes) is known only from Ctenarytaina eucalypti – a psyllid living on eucalypts. It is therefore very unlikely that Aphalara itadori will be attacked by any known British chalcid parasitoid. Japanese knotweed plants possess extra-floral nectaries (EFN) and such structures have been shown to attract predators and parasitoids (Bouchery et al., 1975; Pemberton & JangHoon, 1996). This is also the case with Japanese knotweed in Japan (Kawano et al., 1999) though no evidence of this association has been seen in the PRA area and attempts to induce EFN activity through psyllid feeding were variable. In Japan, high populations of the psyllid have been observed despite the presence of ants in the local area so it can be concluded that they should not prevent establishment. Limited studies using generalist predators have shown that most generalist can include the psyllid in their diet in a no-choice situation but in choice studies there was an apparent preference for the normal (aphid) host.

B Assessment of Risks and Benefits Human health N/A

Animal health N/A

3.2. Safety and Health Effects Potential hazards of IBCA, product or any co-formulants, and measures taken to limit operator exposure

Measures of prevention N/A

History of previous releases or introductions1 None

3.3. Information on Environmental Risk Assessment (ERA) All fields should normally be completed (but see exemptions listed below), but may be weighted differently in the evaluation of risks

Outcome of previous risk assessments1

A revised PRA has been submitted to CSL in parallel to this application

3.3.1. Potential for establishment2

Physical constraints

In the absence of co-evolved natural enemies and given the high fecundity of gravid females, if the host plant is present in good condition then populations should establish easily. Indeed, small sustainable populations have been routinely found on surveys in Japan. A. itadori has been found at sea level in Kagoshima and on the upper slopes of Mt. Aso in Kyushu Island so it is very adaptable and tolerant of climatic extremes. Establishment is most likely because the psyllid has been shown to be capable of breeding in the climatic conditions available in the UK (Appendix 4) and there is no shortage of the host plant in the area of intended release (UK). Based on climatic matching studies, the geographical areas where the temperature threshold of the psyllid for development would be inadequate have been identified as having no or limited knotweed populations (see Appendix 4). The psyllid is also likely to be released again if the first attempts fail though a review of the reason for failure would be necessary prior to repeat attempts. Its tolerance to pollution, which may be higher in parts of the UK than much of its more mountainous range in Japan, is unknown, though it thrives on the slopes of volcanic fumaroles where hydrogen sulphide gas escapes are a frequent occurrence. The culture has been found to be vulnerable to low relative humidities and dry airflow, a combination which is rarely if ever found in the UK.


Resource constraints

The target knotweed hosts are widespread across the area of introduction so lack of resources is not considered to be a significant constraint in establishment. Establishment would be unlikely however, if release takes place outside the growing season of the host plant or too late in the year for a generation to be completed . It is not known whether the psyllid actually requires, i.e. feeds upon, any other species of woody plant over winter or whether they are really just providing a structural shelter. If A. itadori has host-specific requirements over winter then establishment would be compromised.

Survival data and methods used

The absolute thermal minimum for this organism has not yet been determined but adults have been kept alive in the absence of their normal knotweed host at 5oC for up to 8 weeks in a cooled and illuminated incubator. In the field in Japan it survives cold winters when much of the habitat is under snow.

Evidence of establishment

Experimental studies presented in Appendix 4 estimated the day-degrees requirement of the psyllid to be 462.5 from egg to adult, where day-degrees are the total amount of heat required, above the development threshold, for an organism to develop from one stage to another stage of its life cycle. The possible distribution of the psyllid throughout the UK, based on temperature was then considered using this data. In the south east of the UK, the psyllid could establish two generations, whereas in the south west and central England the psyllid would establish one-possibly two generations. Scotland and higher altitude areas of Wales, would establish one generation of the psyllid whilst some areas of the Lake District and the Scottish highlands, -showing the lowest day-degrees- are areas free from or which have a low abundance of Japanese knotweed and establishment could be precluded


Wild hosts known

No other wild hosts are known in the introduced range. A. itadori is a biological control agent that has been selected based on its high level of host specificity and its ability to establish and spread.

3.3.2. Host range assessment3

Organisms tested

See Appendix 5 for full test plant list and the justification for their selection. The proposed UK test plant list contains seventy-three species including representatives from 19 families, consisting of 33 plants native to the UK, 15 introduced species, 3 native to Europe, 13 ornamentals and 10 economically important species. It should be noted that the psyllid has been tested on 87 species since data from test plants of significance to North America have been included in this application. There was no mechanims for agreeing this test plant list with the authorities in advance of the submission of the application, unlike the standard procedure in all other countries active in classical biological control.


Procedures used for host range testing

(see Appendix 2 for details of methods for host range testing and the results). Field observations in Japan were used to provide useful information on the host range of all organisms associated with Japanese knotweed in its native range including A. itadori. Experimental studies were carried out under quarantine conditions in ventilated Perspex cages into which three knotweed plants and 3 replicates of two other non-target plant species were randomly allocated positions in a 3x3 grid. Thirty adults of A. itadori, less than 1 week old, were released into each cage and allowed to lay eggs for 7 days, after which the number and location of eggs was recorded. Six plants of each of the test species were exposed to oviposition, normally in two cages of 3 replicates per cage. Any cage which produced fewer than 100 eggs in total was discarded. The whole experiment was carried out over 27 months. Those plant species which received eggs in the multiple-choice oviposition test above were maintained, so any feeding and development by hatched nymphs could be recorded through observation of wax production by the nymphs and eventual adult emergence. Target-absent multiple choice tests were also carried out on closely related species and compared to multiple choice tests, as well as nymph development studies on those plant species that received eggs in the above-mentioned multiple-choice studies, using hand transferred 1st instar nymphs. Adult survival on a number of non target hosts was also measured in lieu of adult feeding tests as a correlate of performance



Target and non-target host plants

The host range testing carried out (see Appendix 2) indicates that A. itadori feeding and development from egg to adult is restricted to the Japanese knotweeds. Fallopia japonica is the natural host plant for this psyllid and the following species can be included as hosts following experimental host range testing as detailed in Appendix 2. Japanese knotweed – Fallopia japonica var. japonica (widespread pest) Bohemian knotweed – Fallopia x bohemica (hybrid of F. japonica & F. sachalinensis, widespread pest) Compact knotweed – Fallopia japonica var. compacta (ornamental- not commonly grown) Giant knotweed – Fallopia sachalinensis (limited pest and not preferred host) Conolly’s knotweed – Fallopia conollyana (hybrid of F. japonica & F. baldshuanica) Evergreen trees may provide the overwintering host/shelter for adult psyllids in the absence of knotweed between the first frost and its emergence in Spring. Some egg laying did occur on 11 non-target species (Fallopia baldschuanica, Fallopia dumetorum, Fallopia convolvulus, Oxyria digyna, Rheum palmatum, Rheum hybridum-Glaskins, Muehlenbeckia complexa, Persicaria polystachyum, Fagopyrum esculentum,Fagopyrum dibotrys, Rumex hydrolapathum (see Appendix 5 for profiles of the plants). However, none of these eggs were able to develop through to adult. Nymph transfer studies were also carried out on these 11 species and these revealed that after two weeks nymphs were still alive on 8 species, albeit at very reduced numbers and four test plants F. conollyana, M. complexa, F. dumetorum and R. hydrolapathum still supported nymphs at day 28. These were the only species able to support development beyond 3rd instar and only M. complexa could produce adults (4 out of 60). However, these species received a tiny fraction of the number of eggs laid on Japanese knotweed in choice tests and the very low survival rate even after nymph transfer suggest that these would never be preferred or even likely host plants in the field. Adult survival is severely compromised when they are held on non-knotweed hosts (Appendix 2) which suggests that adult feeding range is very narrow too.

3.3.3 Dispersal4 Ability to disperse

The psyllid is capable of reproducing rapidly and repeatedly in the intended release area. However, it is only the adult stage that is truly mobile and adults are not strong flyers, relying more on jumping and flying along with air currents. Populations of another psyllid biocontrol agent, Boreioglycaspis melaleucae intentionally released in Florida, USA to control the Australian paperbark tree, Melaleuca quinquenervia, dispersed 2.2-10.0 km/year, with the slower rates in dense, continuous Melaleuca stands and faster rates in fragmented stands (Center et al., 2006). Any transportation by humans will almost certainly be intentional not least because the growing of its host plant in the wild is prohibited by law in the UK. If the biocontrol agent is perceived as a solution to invasive knotweed populations by the general public, it is likely that demand for the psyllid will be high and its intentional redistribution will take place repeatedly. In summary the psyllid would be able to disperse in large numbers from the release sites but will require its knotweed host plants to be present to sustain a population.


3.3.4. Direct and/or indirect non-target effects5

Summary of available information and conclusions on risks

Based on the data in Appendix 2 it would appear that the psyllid Aphalara itadori would pose no direct threat to any non-target test plants were it to establish and spread in the UK following an intentional release. There is a slim possibility that, in the period soon after release, when psyllid numbers could be expected to be very high, adults may be found on the most closely-related non-target species if their ordinary host has been eradicated. If it were to occur, this spill-over effect would be temporary since the adults inflict little damage (as supported by similar studies by Purcell et al.,1997and Wineriter at al., 2003) and any eggs have been shown to be incapable of developing to adult, and therefore sustaining a population, on all species except M. complexa. Host selection by the ovipositing female is very restricted and any eggs laid on non-target hosts have not been shown to be capable of developing to adult (only under forced nymph transfer). These findings were supported by adult survival studies and observations on other Polygonaceae in Japan. It is notoriously difficult to predict any secondary, tertiary etc. effects on the rest of the food chain (Pearson and Callaway, 2005) such as apparent competition (Bonsall and Hassell, 1997). The likely lack of specialist, native natural enemies reported in Appendix 1, should mean that these risks are minimised but some indirect effect due to a new organism in a habitat can never be ruled out. Generalist predators have been shown to be capable of feeding on all stages of the psyllid in the laboratory but they may even be be tended by native ant species as has been observed with numerous sap-suckers. The most likely effect of any unanticipated predation would be to limit the effectiveness of the biological control agent (see Ireson et al., 2003; Pratt et al., 2003, McFadyen & Spaffod Jacob 2004). However, if any secondary level effects do occur they would almost certainly be local to areas where knotweed is present (and already disruptive) and short lived since the agent is believed to be effective and therefore its host will become a limited resource


1 Where application for renewal of a previously issued licence is sought, applicants must state whether release under the previous licence led to any new information becoming known, and whether any new information has been published about the organism since the previous application. If any new information affects the previous risk assessment this must be discussed in section 3.3 and a revised risk assessment provided. 2 When outdoor establishment of the IBCA is very unlikely and predicted to die out rapidly (as indicated by the data provided), the subsequent fields need not be completed, and no further risk assessments are necessary; 3 When outdoor establishment of the IBCA is necessary or likely to occur, host range information is essential for the risk assessment; 4 Dispersal test results are not required for glasshouse releases, but should be provided when IBCAs are released into open fields or structures that do not prevent escape (e.g. polytunnels) and long term establishment is very unlikely; 5 A summary of known direct and indirect non-target effects should always be given, irrespective of whether host range and/or dispersal have been assessed.

Method(s) to determine efficacy See efficacy summary Appendix 3 3.4. Efficacy and

benefits of the IBCA Assessment of efficacy, economic and environmental benefits

Results of efficacy trials

The sap-sucking activities of the psyllid are capable of killing potted plants under high loads. Nymphal stages cause the most significant damage and manipulative studies in the UK quarantine laboratory revealed that it can reduce the rate of growth and significantly increase leaf count under relatively low nymph loads (White, 2007). In the field in Japan, Aphalara itadori has been observed causing significant damage to large stemmed plants, stunting growth, limiting leaf expansion and reducing flowering (see Appendix 3). Studies on the use of the psyllid in combination with current control measures have bene funded by the Environment Agency and appear to show a positive relationship between psyllid presence and the efficacy of current control measures though these studies are not complete.


Economic benefits

Culliney (2005) reviewed the economics of weed biocontrol using ex ante estimated benefit/cost ratios from 32 projects for which adequate data existed. The ratios varied considerably around a mean of over 200: 1 (range = 2.3: 1 to 4,000: 1) but were all positive. It is important to realise that these ratios are likely to be underestimates due to the difficulty of valuing benefits to nature and ecosystem services (Goulder and Kennedy, 1997). Psyllids have been used in biological control of weeds before, namely Boreioglycaspis melaleucae against M. quinquenervia in Florida USA and Heteropsylla spinulosa against Mimosa invisa (now called Mimosa diplotricha) in Papua New Guinea (Kuniata Korowi, 2004; Center et al., 2006). Both of these were successful with the latter leading to reductions in the cost of chemical control. Another psyllid Arytainilla spartiophila released against scotch broom Cytisus scoparius has had a patchy performance. If successful, the release of A. itadori will provide positive economic returns through cost savings on knotweed management, an improvement in the built environment where knotweed infestation is often an indicator of lack of investment/dereliction. Even if the psyllid was only capable of limiting the spread of Japanese knotweed this would be a significant cost saving. If the target host plant ceases to be a serious riparian invader then any contracted work directly associated with the plant’s ability to exacerbate flood risk, i.e. flood recovery would be reduced but this would be offset by reduced insurance claims and premiums.


Environmental benefits

A reduction in the range and dominance of Japanese knotweed is anticipated as a result of the activities of the psyllid. This should take place slowly and there would be a reduced impact of Japanese knotweed on the environment and associated biodiversity, whilst native species will be allowed to re-establish, It would seem logical that a reduction in knotweed cover and its replacement with native plant species would be of benefit to any vertebrates that had colonised knotweed stands but this has not been investigated. There is also the risk that the niche will be filled by other non-native invasive species but the slow nature of classical biological control should reduce that risk. We have only found negative impacts of knotweed on other species in the scientific literature (e.g. Maertz et al., 2005; Gerber et al., 2008; Topp, et al., 2008). A reduction in the use of chemical herbicides is also anticipated


Part 4. Submission of forms and signature

A Submission Details Information requirements

Literature reference copies

Identification of applicant

Chamber of Commerce

4.1. Appendices Check for completeness of application

Authorization payment No payment required in England

Name Sarah Hugo

Organisation Central Science Laboratory

Address Sand Hutton

City York, North Yorkshire, UK

4.2. Where to submit the application

Postal code YO41 L1Z

B Agreement

4.3. General safeguards The applicant or authorized user undertaking the release proceeds under the conditions of the authorisation for release, taking into account of the following requirements: • All appropriate safety procedures should be put in place. • Any relevant information on adverse effects, which might relate to the released IBCA, should be

reported immediately to the National Competent Authority (The Department for Environment Food and Rural Affairs, Defra).

• Information on sites and dates of supply or release of the IBCA should be made available to Defra, if requested.

• Information requirements have been supplied according to the most recent knowledge. • The conditions made by Defra will be respected.

Date

Applicant’s name

4.4. Signature* *completed by a legally authorized person Signature

Part 5. Appendices Any additional, relevant information or data should be provided in appendices.


References

Baba, K. and Miyatake, Y., (1982). Survey of adult psyliids in winter in Niigata Prefecture. Transactions of the Essa Entomological Society 54, 55-62.

Bailey, J.P., Child, L.E. and Conolly, A.P. (1996). A survey of the distribution of Fallopia x bohemica (Chrtek and Chrtková) J. Bailey (Polygonaceae) in the British Isles, Watsonia 21,187-198.

Bailey, J.P., Conolly, A.P., (2000). Prize-winners to pariahs - a history of Japanese knotweed s.l. (Polygonaceae) in the British Isles. Watsonia 23, 93-110.

Baker, C. R. B., & Bailey, A.G. (1979). Assessing the threat to British crops from alien diseases and pests. In D. L. Ebbels and J. E. King (eds), Plant health, 43-54, Blackwell Scientific Publications, Oxford, United Kingdom.

Baker, R.H.A. 2002. Predicting the limits to the potential distribution of alien crop pests. In Hallman, G.J. & Schwalbe, C.P. (Eds). Invasive Arthropods in Agriculture. Problems and Solution, 207-241. Science Publishers Inc. Enfield USA.

Beerling, D.J. (1993) The Impact of Temperature on the Northern Distribution Limits of the Introduced Species Fallopia japonica and Impatiens glandulifera in North-West Europe, Journal of Biogeography 20, 45-53.

Beerling, D.J., Bailey, J.P., Conolly, A.P., (1994). Fallopia japonica (Houtt.); Ronse

Bonsall, M.B. and Hassell, M.P. (1997) Apparent competition structures ecological assemblages, Nature (London) 388 (6640), 371-373.

Bouchery, Y., Rabasse, J.M. and Lafont, J.P. (1975) Ecological role of the extra-floral nectaries of broad bean in ant-aphid relationships, Sciences Agronomiques Rennes, 139-142.

Burckhardt, D., Lauterer, P., (1997). Systematics and biology of the Aphalara exilis (Weber & Mohr) species assemblage (Hemiptera: Psyllidae). Entomological Scandinavica 28, 271-305.

Center, T.D., Pratt, P.D., Tipping, P.W., Rayamajhi, M.B., Van, T.K., Wineriter, S.A., Dray, F.A.,Jr. and Purcell, M. (2006) Field colonization, population growth, and dispersal of Boreioglycaspis melaleucae Moore, a biological control agent of the invasive tree Melaleuca quinquenervia (Cav.); Blake, Biological Control 39 (3), 363-374.

Culliney, T.W. (2005) Benefits of classical biological control for managing invasive plants, Critical Reviews in Plant Sciences 24 (2): pp.131-150.Dolling W.R., (1991). The Hemiptera. Oxford University Press, NY, UK.

Gegechkori, A.M. & Loginova, M.M. (1990) Psillidy SSSR. Gosudarstv. Tbilisi. 1-164.

Gerber, E, Krebs, C. Murrell, C., Moretti, M., Rocklin, R. and Schaffner, U. (2008). Exotic invasive knotweeds (Fallopia spp.) negatively affect native plant and invertebrate assemblages in European riparian habitats. Biological Conservation 141, 646-654.

Goulder, L.H. and Kennedy, D. (1997) Valuing ecosystem services: philosophical bases and empirical methods, pp.23-47 in: G.C. Daily (ed.) Nature's Services: Societal Dependence on Natural Ecosystems. Island Press, Corvelo, CA. USA.Hodkinson, I.D., (1974). The biology of the Psylloidea (Homoptera): a review. Bulletin of Entomological Research, 64, 325-339.


Hollingsworth, M.L., Bailey, J.P., (2000). Evidence for massive clonal growth in the invasive weed Fallopia japonica (Japanese Knotweed). Botanical Journal of the Linnean Society 133 (4), 463-472.

Holzner, W., Numata, M., (1982). Biology and Ecology of Weeds. W. Junk Publishers, The Hague.

Ireson, J.E., Gourlay, A.H., Kwongc,R.M., Holloway, R.H. & Chatterton, W.S. (2003) Host specificity, release, and establishment of the gorse spider mite, Tetranychus lintearius Dufour (Acarina: Tetranychidae), for the biological control of gorse, Ulex europaeus L. (Fabaceae), in Australia Biological Control 26, (2), 117-127

Kawano, S., Azuma, H., Ito, M. and Suzuki, K. (1999) Extrafloral nectaries and chemical signals of Fallopia japonica and Fallopia sachalinensis (Polygonaceae), and their roles as defence systems against insect herbivory, Plant Species Biology 14 (2), 167-178.

Kristoffersen, L. and Anderbrant, O. (2005) Winter host ecology of the carrot psyllid (Trioza apicalis); Proceedings of the IOBC/WPRS Working Group on Integrated Protection in Field Vegetable Crops, Deinze, Belgium, 12-16 October 2003, Bulletin OILB/SROP 28 (4): 129-132.

Kristoffersen, L. and Anderbrant, O. (2007) Carrot psyllid (Trioza apicalis) winter habitats - insights in shelter plant preference and migratory capacity, Journal of Applied Entomology 131 (3), 174-178.

Kwon, Y.J. (1983) Psylloidea of Korea (Homoptera: Sternorrhyncha) . Insecta Korea 2: 1-181.McFadyen, R.E.C. (1998) Biological control of weeds, Annual Review of Entomology 43, 369-393.

Maerz, J.C., Blossey, B. and Nuzzo, V. (2005) Green Frogs Show Reduced Foraging Success in Habitats Invaded by Japanese knotweed. Biodiversity and Conservation 14, 2901-2912.

Makino, T. and revision team, (1997). Revised Makinos Illustrated Flora In Colour. Hokuryukan, Tokyo.

Mandák, B., Pyšek, P. and Bímová, K. (2004) History of the invasion and distribution of Reynoutria taxa in the Czech Republic: a hybrid spreading faster than its parents, Preslia 76, 15-64.

Maruta, E. (1981) Seedling establishment of Polygonum cuspidatum on Mt. Fuji, Japanese Journal of Ecology 26, 101-105.

McFadyen, R. and Spafford Jacob, H. (2004) Insects for the biocontrol of weeds: predicting parasitism levels in the new country, pp.135-140 in: J.M. Cullen, D.T. Briese, D.J. Kriticos, W.M. Lonsdale, L. Morin, and J.K. Scott (eds.) Proceedings of the XIth International Symposium on the Biological Control of Weeds. 27 April – 2 May, 2003. Canberra, Australia.

Miyatake, Y., (1964). Psyllidae in the collection of the Osaka Museum of Natural History, with description of a new species (Hemiptera: Homoptera). Bulletin of the Osaka Museum of Natural History 17, 19-32.

Miyatake, Y., (1973). Psyllids and their life Nature Study 19 (1), 5-11. (In Japanese)

Miyatake, Y. (2001). Psyllids in the southern Osaka (4) Minami-Osaka no Konchu 3 (4), 6-10. (In Japanese)

Noyes, J. (2007) Universal Chalcidoidea Database (www. http://internt.nhm.ac.uk/jdsml/research-curation/projects/chalcidoids/ )


Ossiannilsson, F., (1992). The Psylloidea (Homoptera) of Fennoscandia and Denmark. Fauna Entomologica Scandinavica 26, 1-346.

Pearson, D.E. and Callaway, R.M. (2003) Indirect effects of host-specific biological control agents, Trends in Ecology & Evolution 18 (9), 456-461.

Pemberton, R.W. and JangHoon, L. (1996) The influence of extra-floral nectaries on parasitism of an insect herbivore, American Journal of Botany 83 (9), 1187-1194.

Pratt, P.D., Coombs, E.M.& Croft, B.A. (2003) Predation by phytoseiid mites on Tetranychus lintearius (Acari: Tetranychidae), an established weed biological control agent of gorse (Ulex europaeus) Biological Control, 26, 40-47

Purcell, M.F., Balciunas, J.K., Jones, P. (1997) Biology and host range of Boreioglycaspis melaleucae (Hemiptera: Psyllidae), a potential biological control agent for Melaleuca quinquenervia (Myrtaceae). Environmental.Entomology 26, 366–372.

Reuter, O.M., (1909). Charakteristik und Entwicklungsgeschichte der Hemipterenfauna der palaearktischen Coniferen. Acta Soc. Scient. fenn. 36: pp. 1-129.

Shinji, O.. (1938). Five new species of Psylla from Japan. Kontyû 12, 146-151.

Topp, W., Kappes, H. and Rogers, F. (2008) Response of ground-dwelling beetle (Coleoptera) assemblages to giant knotweed (Reynoutria spp.) invasion. Biological Invasions. 10, 381-390.

White, S. D. (2007) The efficacy of Aphalara itadori as a biological control agent of Japanese knotweed (Fallopia japonica) . Unpublished Masters Thesis, Imperial College, University of London.


The Natural History Museum Cromwell Road London SW7 5BD United Kingdom +44 (0)20 7942 5000 www.nhm.ac.uk

Appendix 1

Aphalara itadori (Shinji) (Insecta: Psyllidae)

Taxonomy & Natural Enemies

Andrew Polaszek, Dept of Entomology, the Natural History Museum, London

Taxonomy A. itadori was described originally in 1938 in the genus Psylla. It was transferred to Aphalara by Miyatake (1964). The original description, including a single figure, is reproduced below in its entirety, as is the later and more useful redescription by Burckhardt & Lauterer (1997). A. itadori is somewhat unusual in the genus, and gives its name to a species-group containing only itself and one other species, A. taiwanensis. Shinji, O.1938; Five new species of Psylla from North-Eastern Japan. p. 149-150. 12. Psylla itadori Shinji n. sp. (Itadori-madarakijirami) Adult: - Body dirty yellow, with usually brownish tint. Frontal tubercles conical, large, slightly infuscated. Head broader than long, divided in middle by a black longitudinal suture, infuscated a little and black at the margins. Eyes large, black, arising from the sides of the head. Ocelli reddish. Proboscis reaching beyond the front leg, apical article black. Prothorax broader than the head but not so broad as the head including the eyes. Meso- and meta-thoracic parts infuscated with 5 grayish longitudinal stripes on the dorsum of the mesothorax. Legs with tibiae and their spurs and claws black, the remaining portion being concolourous with the body. Hind and often the middle tarsi with about 12 large blunt and black spurs and the first hind tarsi with about 4 of them. Wings subhyaline, broad and rounded at the apex. Fore wings with radius almost perpendicular to subcosta and radial sector which is nearly one and half times as long as the radius, almost straight. Medial stalk arising at about the middle part of the radial stalk, basal portion usually obscure, almost as long as radial sector, both M1 and M2 short and subequal. Cubital stalk arising at about ¼ part of medial stalk which arises in turn at about the middle part of the cubital stalk, C1 much curved up, about 3 times as long as C2 which is almost straight. The following places are maculated or infuscated: a wide area on both sides of C1; along C1 and continuing to cover the apical half of the cell enclosed by it, a transverse region extending from the infuscated part of the cell and traversing it to the costal margin, and area extending along the radial sector and M1 and M2 as well. Legs short with femora infuscated, the remaining parts being concolorous with the body, hind tibia with about 10 black and blunt spurs and the first tarsus with about 4. Abdomen yellowish in colour with infuscated transverse band on each of the segments II~VIII. Genital valves with some long hairs, the upper being much longer than the lower. Measurements in mm: Length of body 1.70 Length of fore wing 2.00 Width of abdomen 0.90 Length of antenna 0.65 Length of hind tarsus 0.35

Host plant: Polygonum reynoldi Type locality: Morioka, Iwate Prefecture, Japan. Date of collection: Sept. 20, 1937. Notes: This species attacks the underside of leaves as well as shoots and flowers. When attacked leaves curls in toward the underside and enclose the insects within. In the case of flowers they become stunted in growth and congregate all in a body. The insects are very common during the months of Jury (sic) to October.

Fore wing of Aphalara itadori (from Shinji, 1938)

Redescription of Aphalara itadori (Shinji) by Burckhardt & Lauterer, 1997 (Entomologica Scandinavica 28: 271-305)

Aphalara itadori (Shinji, 1938) (Figs 26, 46, 62) Psylla itadori Shinji, 1938: 149. Syntypes, Japan: Iwate Prefecture, Morioka, 20.ix.1937, Polygonum reynoldii {depository?), not examined. Aphalara itadori; Miyatake 1964. Aphalara nebulosa (Zetterstedt); Matsumura 1916 (misidentification). Aphalara kunashirensis Klimaszewski, 1983: 8. Holotype ♂, Russia: Primorskiy Kray, Amursk region, Sev. Kunashir, okr. volc. Tjatja, 2-4.vii.1976 (V. Zherihin) (ZMUM?), not examined. Syn. n. Description. - Adults. Head. Anteorbital tubercles and macroscopic setae on vertex absent. Anterior tubercles on vertex small. Outer anterior vertex margin straight or concavely rounded. Tubercle between antennal insertion and eye small, flattened. Clypeus short, not or hardly visible from above, conical, apically subacute. Forewings. Membrane semi transparent to weakly coriaceous; with light or dark brown pattern consisting of well-defined spots and patches. Surface spinules forming regular cellular pattern. Genitalia. Parameres in profile lamellar, with small antero-basal projection; slender in the middle, flattened apically; anterior subapical process removed from apex, small, claw-like. Apical dilatation of distal segment of aedeagus small relative to shaft; with large dorso-apical membranous sack; apico-ventral process small, pointed, hooklike, strongly sclerotised distally and convex proximally. Female proctiger subacute apically; subgenital plate subacute apically; valvulae dorsalis and ventralis curved. Measurements and ratios (1♂, 1♀). HW 0.61-0.65; AL 0.61-0.73; ALHW 1-1.12; WL 2.12-2.32; WLHW 3.48-3.57; WLW 1.97-1.98; cell m1 ratio 1.07-1.23; cell cu1 ratio 1.86-2.13;

TLHW 0.74; MP 0.17; MPHW 0.28; PL 0.22; AE 0.19; FP 0.69; FPHW 1.06; FPCP 3.83; FPFS 1.50. Larvae and eggs unknown. Host plants. - Polygonum cuspidatum, P. reynoldii, and P. sabulosum (=P. sachalinense). Distribution. - Japan; Korea; Russia: Sakhalin, Kurile Islands (Kwon 1983, Gegechkori & Loginova 1990). Material. - Japan: Kyushu, Honshu (MHNG). Russia: Kurile Islands (ZISP). Remarks. - Type material of Psylla itadori could not be traced and seems to be lost (Y. Miyatake pers. comm. to DB, 24.viii.1996). Type material of A. kunashirensis was unavailable to us; 1♀ labelled as paratype which we could examine (ZMUM) belongs to Cacopsylla. No other type I specimen could be traced in the ZMUM (S. Kurbatov pers. comm.) nor in the SIUK, and we think that the type material is lost. A series of A. itadori from the Kurile Islands (ZISP) fits the description of A. kunashirensis which is therefore synonymised with the former.

(Figures modified from Burckhardt & Lauterer, 1979). Natural enemies of Aphalara itadori According to R. Shaw (personal communication) a single eulophid parasitoid has been recorded attacking the species in Japan (see photographs below). From the photographs, this appears to be a Tamarixia species (considered by some a synonym of Tetrastichus), although Tamarixia species are usually associated with Triozidae. The six British Tamarixia species are all associated with Triozidae, but could conceivably move over to Aphalara.

Chrysonotomyia species (Eulophidae) have been reared from Psyllidae twice, but never from Aphalara spp. Noyes’ (2007) “Universal Chalcidoidea Database” (www. http://internt.nhm.ac.uk/jdsml/research-curation/projects/chalcidoids) lists no known parasitoids of Aphalara itadori, and out of a total of 90 records of Psyllidae parasitoids, only two are associated with Aphalara species. Both concern North American species (California), and both are Psyllaephagus species (Encyrtidae), although one is recorded as “Encyrtus aphalarae”. At least one of these records is attributable originally to Jensen (1957), who also records a cecidomyiid predator of Aphalara maculipennis from Germany. Of the 90 records of chalcidoid parasitoids of Psyllidae in the Noyes database, more than 60 concern Psyllaephagus spp. On the face of it, there is therefore some probability that an indigenous British Psyllaephagus species, of which there are currently two, could use Aphalara itadori as a potential host. However, of these two species, one (P. lusitanicus Mercet) is known only from the scale family Asterolecaniidae, and the other (P. pilosus Noyes) is known only from Ctenarytaina eucalypti – a psyllid living on eucalypts. Furthermore, none of the four indigenous British Aphalara species is known to be attacked by any parasitoid species. According to Dr Ian Hodkinson (psyllid specialist) there is very little known about the parasitoids of Aphalara species. Having worked extensively on populations of Craspedolepta species (the next genus) he has never found high levels of parasitism. It seems, therefore, very unlikely that Aphalara itadori will be attacked to any major extent by any known British chalcid parasitoid.

Unparasitised Aphalara nymph Parasitised Aphalara nymph

Parasitoid pupa with host remains removed Adult eulophid parasitoid (?Tamarixia sp.) (photos by Naoki Takahashi)

References Burckhardt, D. & Lauterer, P. (1997) Systematics and biology of the Aphalara exilis (Weber & Mohr) species assemblage (Hemiptera: Psyllidae). Entomological Scandinavica 28: 271-305. Gegechkori, A.M. & Loginova, M.M. (1990) Psillidy SSSR. Gosudarstv. Tbilisi. 1-164. Klimaszewski, S.M. (1983) New data about the psyllids (Homoptera: Psylloidea) of Far Eastern USSR. Acta biologica Katowice 13: 7-21 Jensen, D.D. (1957) The parasites of the Psyllidae. Hilgardia 27:71-99. Kwon, Y.J. (1983) Psylloidea of Korea (Homoptera: Sternorrhyncha). Insecta Korea 2: 1-181. Matsumura, S. (1916) Applied Entomology 1: 1-713. Miyatake, Y. (1964) Psyllidae in the collection of the Osaka Museum of Natural History, with description of a new species (Hemiptera: Homoptera). Bulletin of the Osaka Museum of Natural History 17: 19-32. Shinji, O. (1938) Five new species of Psylla from Japan. Kontyû 12: 146-151

Appendix 2 Revised draft of a paper submitted to Biological Control (Elsevier) entitled: “The life history and host range of the Japanese knotweed psyllid, Aphalara itadori Shinji: Potentially the first classical biological weed control agent for the European Union”

The life history and host range of the Japanese knotweed psyllid, Aphalara

itadori Shinji: Potentially the first classical biological weed control agent for

the European Union

1

2

3

4

5

6

7

8

9

10

11

12

Richard H. Shaw1, Sarah Bryner2 and Rob Tanner1

1 – CABI Europe-UK Centre. Bakeham Lane, Egham, Surrey, TW20 9TY UKa

2 – Department of Agricultural and Food Science, Swiss Federal Institute of Technology

(ETH), CH-8092 Zurich, Switzerland

a [email protected], Tel 00 44 (0)1491 829 129, Fax – 00 44 (0)1491 829 123

1


Abstract 13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

Japanese knotweed is a serious invasive weed in the UK, North America and parts of

mainland Europe. Current control measures are both difficult to implement, unreliable and

expensive. In 2003, a classical biological control programme was initiated, one that could

lead to the first ever authorised release of a classical biocontrol agent against a weed in the

European Union. Literature studies and surveys in the native range in Japan revealed over

180 species of arthropod natural enemies but only a psyllid, Aphalara itadori, has reached the

point of official assessment for release. A. itadori passes from egg to adult through five

nymph stages in just under 33 days at 23oC and the timing and physical appearance of these

stages is presented. Multiple-choice oviposition studies using 87 species/varieties of test

plants showed that only 1.52% of 146,885 eggs were laid outside what we call the invasive

knotweed group. None of these eggs laid on non-targets plants were able to develop to adult,

however, subsequent nymph transfer experiments revealed limited development on some

closely related members of the Polygonaceae and the development to adult in 7% of cases on

Meuhlenbeckia complexa. Adult survival, used as a surrogate for feeding tests, further

revealed that survival on non-targets was severely compromised. These results coupled with

an assessment of likely overwintering habits show A. itadori to have a narrow physiological

and behavioural host range making it an ideal candidate for the first official release of a

classical biological control agent against an alien invasive weed in the European Union.

Key words

Japanese knotweed; Fallopia japonica; Reynoutria japonica; Polygonum cuspidatum;

Aphalara itadori; psyllid; host range; oviposition; life-cycle

2

1. Introduction 36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58 59

60

61

Japanese knotweed, Fallopia japonica (Houtt.) Ronse Decraene var japonica [Polygonaceae]

is arguably the most troublesome invasive alien plant in Europe and North America (Lowe et

al., 2000; Weber, 2003) and has serious consequences for biodiversity (Maertz et al., 2005;

Gerber et al., 2008; Topp et al., 2008). The history of alien species of Polygonum and

Reynoutria in the UK has been well reviewed (Conolly, 1977, Bailey and Conolly, 2000,

Bailey, 2005) and is typical of many Victorian introductions which are now invasive

following a long lag-phase. After its arrival in Europe as a prized ornamental in the first part

of the 19th Century (Synge, 1956), the plant has spread exponentially in Britain (Figure 1)

where it has increased its range to include most regions of the country (Preston et al., 2002).

The plant requires both a heat sum ≥2,505 degree days and an absolute minimum temperature

≥-30.2oC and its range is expected to spread northwards under climate change scenarios

(Beerling, 1993).

Japanese knotweed, hereafter referred to as Fallopia japonica or F. japonica, is a vigorous,

herbaceous perennial with annual, glabrous, tubular stems that ascend from a large bulbous

rhizome crown which acts as a carbohydrate store in the winter months. These stems are light

green in colour, often with reddish flecks, and branched, reaching up to 3 metres in height

(Beerling et al., 1994) terminating in creamy-white flower panicles. In the exotic range plants

are generally taller than their Japanese counterparts (Holzner and Numata, 1982; Makino and

revision team, 1997), the latter being closer to 2 m (Author’s obs.). For a more detailed

account on the plant’s morphology see Lousely and Kent (1981) and Beerling et al. (1994).

Remarkably, the expansion in much of the exotic range of F. japonica has been achieved

without the advantage of sexual reproduction, which normally contributes to long-distance

dispersal (Levin, 2000) since it is believed that only a male-sterile clone was introduced

3

(Bailey, 1994). In the UK, F. japonica has spread rapidly by vegetative means as a result of

disturbances such as flood events which aid transportation of rhizome fragments throughout

river catchments. Anthropogenic disturbance, in particular the illegal dumping of waste and

top soil contaminated with rhizome fragments, has also aided the distribution of this species

throughout the UK. Control is extremely difficult since the plant benefits from an extensive

rhizome system from which it can reappear after, apparently successful, chemical control and,

therefore, official knotweed management guidance has been issued (Environment Agency,

2007). After a thorough meta-analysis, Kabat et al. (2006) were unable to conclude long-term

efficacy for any of the traditional control measures they assessed. The chemicals permitted

for use in its preferred riparian habitats are restricted by law in Europe and mechanical control

can often be counter-productive due to the unwitting redistribution of viable material in soil.

Consequently, control costs are high, with estimated costs for a UK-wide control programme,

were it to be attempted, being in excess of £1.5 billion (Defra, 2003).

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

This prohibitive cost means that the only economically-viable control option is classical

biological control. Fallopia japonica is a highly appropriate target for classical biological

control (Holden et al., 1992), not only because it has arrived without any of its specialist

Japanese natural enemies, but also because it is believed to be clonal (Hollingsworth and

Bailey, 2000) and thus unable to show variation in resistance or tolerance to arthropod or

microbial infection. Plants that reproduce asexually are expected to be more susceptible to

biological control (Burdon and Marshall, 1981; Crawley, 1990) if a suitable agent can be

found.

Fallopia japonica is not the only invasive knotweed in the UK since giant knotweed, Fallopia

sachalinensis (F.Schmidt ex Maxim.) Ronse Decraene, can also be found in similar habitats

4

though not causing the same level of concern, at least in Europe. Unfortunately, F.

sachalinensis can form the hybrid Fallopia x bohemica (Chrtek and Chrtková) J.P. Bailey

when crossed with F. japonica var japonica and this hybrid is also proving highly invasive

(Mandák et al., 2004). Though Fallopia x bohemica is able to produce fertile seeds, there is

almost no evidence of germination in the wild in the UK in contrast to North America where,

after first being noted in 2003 (Zika and Jacobsen, 2003), this hybrid appears to be producing

genetically diverse populations as is the case for Central Europe (Mandák et al., 2004).

These three plant taxa are not desirable and hereafter will be collectively referred to as the

invasive or target knotweeds. In its native range the distinction between species of target

knotweeds is not as clear but there is a general trend from an exclusive occurrence of giant

knotweed on the northern island of Hokkaido, a wide range of knotweed varieties in the

central Japanese island of Honshu, to a plant very close to F. japonica var japonica in the

Nagasaki region of Kyushu island, southern Japan.

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

Considering the scale of classical biocontrol efforts around the globe (Cameron et al., 1989;

Cock, 1985; Rao et al., 1971) and the world-leading biological pest management strategies

implemented in European covered crops (Eilenberg et al., 2000; Minks et al., 1998), it is

surprising how little effort has been targeted at exotic weeds in Europe. No full classical

weed-biocontrol programme has been carried out in any EU Member State (though see Baker

et al. (1972) for a possible exception), and this is not due to the lack of potential targets but

for many diverse reasons (Shaw, 2003; Sheppard et al., 2006).

With respect to the biological control effort against F. japonica, literature reviews and

continued survey efforts revealed that the plant, in its native range, has 186 species of

associated phytophagous arthropods. The ensuing four years of research funded by a

5

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

consortium of UK and North American sponsors, demonstrated that one arthropod agent had

the highest potential as a classical biocontrol agent, and this paper reports the scientific

studies that revealed the life-cycle and host range of that insect, Aphalara itadori Shinji.

Aphalara itadori was described originally in 1938 in the genus Psylla (Shinji, 1938) from

where it was transferred to Aphalara by Miyatake (1964). For a recent re-description see

Burckhardt and Lauterer (1997) who point out that A. itadori is somewhat unusual in the

genus, and gives its name to a species-group containing only itself and one other species, A.

taiwanensis.

2. Materials and methods

2.1 Surveys

A thorough review of the printed and electronic literature (CAB Abstracts, Google Scholar,

ISI Web of Science and various Japanese search engines) in both the English and Japanese

language using all the common synonyms of F. japonica was carried out. This literature was

combined with 7 natural enemy surveys of Japanese knotweed at more than 140 sites on all

four main islands of Japan to reveal a total of 186 species of associated phytophagous

arthropods. The presence, behaviour and impact of Aphalara itadori, along with other natural

enemies of interest, were recorded at each site and any parasites that emerged from collected

individuals were retained for identification.

2.2 Culturing

All laboratory culturing and experimental studies were carried out initially under quarantine

conditions in Berkshire UK in controlled environment rooms (13 hours daylight and 11 hours

6

dark at a temperature of 22oC ± 1.5 oC with 50-80% humidity) and subsequently in Surrey

under the same conditions though humidity varied between 65% and 85%.

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

The psyllid culture was established from adults collected in Kumamoto prefecture, Kyushu

Island, Southern Japan in mid 2004. On a monthly cycle, around 50 adults were exposed to

F. japonica plants grown from rhizome material collected from various sites in southern

England. All plants were grown in 13 cm diameter pots in an equal mix of John Innes no. 3

soil and a multipurpose peat-free compost and pests were kept under control using

commercially–available biocontrol agents. The psyllids and F. japonica plants were

maintained in 0.4 m x 0.4 m x 0.5 m ventilated Perspex cages under the Japanese summer

light regime. After 7-10 days all adults were removed and the nymphs were allowed to

develop on the plants. As the cohort neared adulthood, most leaf material was removed and a

small F. japonica plant was provided so that new adults would move onto the fresh material,

making them easier to harvest for establishing another rearing cage.

2.3 Aphalara itadori life-cycle and fecundity

In order to provide details on the psyllid’s life-cycle in the laboratory, newly-emerged A.

itadori adults were collected into a glass vial and mating pairs were removed and released

onto small (7-11 cm tall) F. japonica plants in 13 cm diameter pots under tailor-made,

ventilated Perspex cloches. The pre-oviposition period was recorded, as was the time to egg

hatch. Forty-five new nymphs were transferred individually to same sized F. japonica plants

using an artist’s paintbrush and survivorship and moulting was recorded during daily

observations. The study continued until all adult had emerged.

7

In order to estimate fecundity, newly-emerged adult mating pairs were exposed to small F.

japonica plants (as above) and each pair was transferred weekly onto a new plant. Each week

the number of eggs laid was recorded until egg production had ceased or the female had died.

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

2.4 Selection of test plants

A test plant list was complied using the principles of the centrifugal phylogenetic method of

Wapshere (1974) with additional modifications, as suggested by Briese (2005) and Briese and

Walker (2008), to include recent advances in phylogenetics (for Polygonaceae and

Caryophyllales see: Kim and Donoghue, 2008; Lamb Frye and Kron, 2003). Native plant

species were confirmed against Preston et al., (2002) and ornamental species were selected

with the aid of the Plant Finder (Royal Horticultural Society, 2003). Plant species were

selected initially based on their relatedness to the target species including all species from the

sub family Polygonoideae, genus Fallopia native and introduced to the UK. Species selection

then moved from the genus to the Tribe Polygoneae to include all 8 species native to the UK

as well as 5 introduced species. Subsequently representative species were selected from the

remaining tribes of the sub-family, namely from the Persicarieae, Coccolobeae and Rumiceae,

if such species are present in the UK either as natives or ornamentals. Species selection then

moved out of the sub-family Polygonoideae into the sub-family Eriogonoideae which is

poorly represented in the UK, hence the inclusion of just one species, Erigonium umbellatum

(Torr.). Additional species were selected from all major families of the Caryophyllales. Also

selected were plant species which have a similar morphology and biochemical composition to

the target species. The selection of Malus domestica (Borkha) and Cytisus scoparius (L.) was

influenced by the fact that some of the classical biocontrol agents under consideration at the

time have congeneric species recorded or known as arthropod pests of these plant species in

the UK. Finally, 10 economically important safeguard species were included in the test plant

8

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

list to satisfy donors and the public. The test plant list is included alongside the results of the

oviposition trails in Table 2.

In total, the test plant list contained ninety species and varieties including representatives

from 19 families. The list consisted of 37 plants native to the UK, 23 species introduced to the

UK, 3 species native to Europe, 13 ornamental and 10 economically important UK species.

In the absence of a regulatory system tailored to accommodate modern classical biological

control, it was impossible to have this proposed all-inclusive test-plant list agreed by the

relevant authorities in advance to the specificity testing of the psyllid. Thus inevitably some

irrelevant test plant species were included to satisfy anticipated requests by future reviewers

of any application and the general public.

2.5 Host range testing

2.5.1 Adult multiple-choice oviposition

Multiple-choice oviposition studies were carried out in ventilated Perspex cages as detailed

above, into which three F. japonica plants and three replicates of two other test plant species

were randomly allocated positions in a 3 x 3 grid. Thirty adults, less than one week old, were

released into each cage and allowed to lay eggs for a period of 7 days, after which time the

number and location of each egg (leaf adaxial, leaf abaxial and petiole sheath) was recorded.

In total six plants of each of the test species were exposed to oviposition, normally in two runs

of three replicates per cage, though replication was increased if any oviposition was observed.

Any run which produced fewer than 100 eggs on an individual F. japonica plant was

discarded. The whole experiment was carried out over 27 months.

2.5.2 Target-absent oviposition

9

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

Target-absent multiple-choice tests were undertaken using only the most closely-related

species present in the UK, namely F. baldshuanica (Regel) Holub, F. convolvulus (L.) Á.

Löve and F. dumetorum (L.) Holub. These plant species were exposed to oviposition by 30

adults, in the absence of F. japonica. The number and fate of any eggs laid was recorded.

The results were compared with those obtained in multiple-choice studies with F. japonica

present, using a generalised linear model with a Poisson error structure (R version 2.3.1). In

cases where the data were over dispersed an empirical scale parameter was estimated.

2.5.3 Nymph development and survival

Those plant species which received eggs in the multiple-choice oviposition test were

maintained and any nymphal feeding and development to adult was recorded.

More detailed nymph development studies were made on any non-target plant species which

received eggs in the oviposition trials (11 species) as well as Polygonum arenastrum Jord. ex

Boreau which was chosen due to its wide and common distribution in the UK. These studies

involved the hand transfer of 10 first instar (N1) nymphs to each of 6 test plants of the

respective species alongside 3 replicates of F. japonica as controls. The plants were

maintained in Perspex cages as before but care was taken to ensure that the plants did not

touch each other or the side of the cage. The development of nymphs was recorded as well as

the number alive on each plant and whether feeding was evident through the production of

exuded wax. These detailed studies were carried out in CABI’s Egham quarantine facility

where humidity was elevated and airflow reduced because under normal conditions first instar

nymph survival was found to be compromised on the knotweed control. Since daily

recording was not always possible over the 5 month period of the study, some survival counts

were estimated based on the pervious and following count. Where survival counts were

higher in the days following a low record it was assumed that the nymph(s) had been missed

10

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

and the higher figure was accepted as the true count. Nymph survival on different test plant

species was compared on day 3,7,14 and 28 after transfer and the number of adults that

emerged from each cohort was recorded.

2.5.4 Adult survival

Given the small size and sporadic waste production of the adults, as well as the challenge of

distinguishing feeding from resting positions, it was decided, after initial trials, that survival

should be used as a correlate of performance. Thus, six newly-emerged adults were

constrained on each of 10 potted test plants of various species under a cloche. The species

chosen were the only three Fallopia spp. found in the UK, namely the two natives F.

dumetorum, F. concolvulus and the ornamental Russian vine F. baldshuanica along with the

F. japonica target. In addition buckwheat, Fagopyrum esculentum Moench was included

since this is an important related crop species in North America where the psyllid may also be

considered for release. A section of plastic ornamental rose was used as a control and a

similar section was provided as a further control with a honey water feeder. The number of

adults alive was recorded at regular intervals and in all cases except for F. japonica plants the

experiment was terminated once all adults were dead.

3. Results

3.1 Surveys

During the surveys in Japan, observations on the many sympatric Polygonum, Persicaria and

Rumex species consistently failed to reveal any evidence of A. itadori presence or previous

activity on these non-target species. The psyllid was only ever collected from species

identifiable as “close to” what we recognise as F. japonica, F. sachalinensis and F. japonica

11

var compacta and F x bohemica, since even with expert help the vast range of Japanese

knotweeds in Japan are hard to distinguish.

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

3.2 Life-cycle and fecundity

Field observations revealed Aphalara itadori to be a wide-ranging species which was

recorded from southern Kyushu to as far north as mid-Hokkaido. It appears to be tolerant of a

wide temperature range and was found from sea level up to 2,150 m.a.s.l. Adults were

observed from late April, along with eggs, to mid August. On Kyushu, the psyllid was rarely

found to be abundant. However, at one site near Nagano in Gunma Prefecture on the more

northerly Honshu Island, a high population of the insect was observed inflicting significant

damage to F. japonica in the field. Although only one individual of an unidentified eulophid

parasitoid (possibly a Tamarixia species (A. Polaszec, pers. com.)) has been reared from a late

nymph, it could not be concluded that parasitism limited populations in the field.

In the life-cycle studies, 42 nymphs (93.3%) survived to adult giving a mean developmental

period, from egg to adult, of 32.2 days ± 0.5 (±1SE, range 28-42 days, n = 42). It was only

possible to record each stage by observing the moulted exuviae in 21 cases, and the

summarised data are presented in Table 1. The morphological characteristics of each stage are

presented diagrammatically in Figure 2.

Of the 24 pairs that were set up in the fecundity studies, 13 pairs failed to produce any eggs,

and were excluded from the analysis. The remaining 11 pairs produced a total of 7,010 eggs,

giving a mean egg production per female of 637 ± 121.96 (±1SE, n = 11). The mean period

of egg production was 37.5 days ± 5.85 days (±1SE, n = 11).

12

3.3 Host range 283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

3.3.1 Adult multiple-choice oviposition

During the multiple-choice oviposition experiments, the location of a total of 146,855 eggs

was recorded. As many as 3,708 eggs were laid on a single caged F. japonica plant in the

laboratory though the mean count was 433.7 ±23.3 (± 1SE, n=324). The vast majority of eggs

were laid on the upper surface of the leaves (mean = 304.5 eggs per plant ±19.9 (±1SE,

n=324)) with the remainder evenly distributed between the lower surface of the leaf (mean =

64.6 eggs per plant ±5.7 (±1SE, n=324)) and under the sheath’s surrounding the nodes at the

base of the petioles (mean = 54.6 eggs per plant ±4.3 (±1SE, n=324)). Each experimental run

was found to produce a mean of 1,298 eggs per cage ± 95 (± 1SE, n = 75) and only two test

runs produced less than 250 eggs in any cage. The summarised results are presented in Table

2.

Out of the 87 test-plant species and varieties tested (including North American species), only

18 species received eggs. This small group included the invasive knotweeds as well as F.

japonica var. compacta (Hook. f.) and the hybrid F. conollyana, both of which are

uncommon horticultural curiosities in the UK. In addition, F. convolvulus (UK and USA

origins), F. dumetorum, F. baldshuanica, Fagopyrum dibotrys (D. Don.) Hara, Fagopyrum

esculentum, Muehlenbeckia complexa (A.Cunn.) Meissn., Oxyria digyna (L.) Hill, Persicaria

polystachyum (Wall. ex Meisn.) H. Gross, Rheum palmatum (L.) and R. x Glaskin (L.), and

Rumex hydrolapathum Hudson also received eggs. The majority of these species received

less than 10 eggs per plant and the highest mean egg count per species were recorded for

Rheum palmatum, with 17.3 ±7 eggs/plant (±1SE, n=12), buckwheat with 16.9 ±

6.7eggs/plant (±1SE, n=14) and wire plant, M. complexa, which received 14.7 ±5 eggs/plant

13

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

(±1SE, n=7). These rates compare with 433.7 eggs per plant ± 28.3 (± 1SE, n = 324) for the

target F. japonica.

3.3.2 Target-absent oviposition

The oviposition behaviour in the presence and absence of F. japonica is recorded in Table 3

and shows no significant difference in the number of eggs laid on the same plant species

between tests during the 7 day exposure. It is notable that during the target-absent oviposition

multiple-choice tests, 80% of the adults used in the test were dead after a week in the absence

of F. japonica. This compares with a mean mortality of 37% recorded in the main oviposition

study with knotweed present.

3.3.3 Nymph development and survival

In the original multiple-choice oviposition study, which was subsequently maintained as a

development study, only a subset of the genus Fallopia, i.e. F. japonica and its hybrids, were

found to be suitable hosts for successful development of A. itadori. The mean development

time, assuming all eggs were laid on day 1 of the 7 day exposure, was found to be

significantly longer on F. sachalinensis plants (47.13 ± 2.19 days, n = 16) than on the F.

japonica plants used in the previous development period study (33.17 ± 0.48 days, n = 42) (t

test p<0.001, rising to p = 0.0056 if the eggs on F. sachalinensis are assumed to have been

laid on day 7). This was not the case for F. conollyana (t test p = 0.091).

In the development study which involved nymph transfer onto non-target plants and F.

japonica controls, more thorough data on actual development were generated and these are

summarised in Table 4. It can be seen that on test plant species other than the very closely

related F x bohemica survival rates of the psyllid were severely compromised. Nonetheless,

14

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

two weeks after transfer nymphs were still developing on 8 non-target species, albeit at very

reduced numbers and four test species, namely F. conollyana, M. complexa, F. dumetorum

and R. hydrolapathm, still supported nymphs at day 28. Only 5 plants were able to support

development beyond third instar, namely Japanese knotweed and its hybrid F. conollyana as

well as F. dumetorum, F. baldshuanica and M. complexa. In contrast to the results of the

multiple choice/development study where none of the 103 eggs laid on M. complexa were

able to develop to adult, this nymph transfer experiment revealed that A. itadori can indeed

develop on this species although only 4 out of 60 individuals achieved this.

3.3.4 Adult survival

The adult survival data on the most closely-related hosts (all the representatives of the same

genus present in the UK) is presented graphically in Figure 3 and shows that survival on

species other than the F. japonica host is severely compromised. Only F. dumetorum was

capable of supporting adults beyond 9 days, the time after which all adults were dead on the

artificial plant with honey water feeder. At 12 days, 16 adults (26.7%) were still alive on F.

dumetorum but this is considerably lower than the 58 adults (96.7%) still alive on the F.

japonica plants after the same period. However, it does suggest that some feeding has taken

place on F. dumetorum.

4. Discussion

Despite the considerable number of phytophagous species recorded on F. japonica in Japan,

the thorough host range testing process carried out on some of the most promising agents

collected from Fallopia japonica in Japan, has left only one potential classical biocontrol

insect, namely the psyllid Aphalara itadori. Based on the fact that A. itadori has only ever

been recorded from the target plant, and its varieties and hybrids, in the field in Japan

15

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

(Burckhardt and Lauterer, 1997; Shinji, 1938; Author’s obs.), and that it could be highly

damaging to its host, it was apparent that this psyllid holds considerable potential as a

classical agent. The more in-depth host range studies reported in this paper confirm that A.

itadori is a highly specialist natural enemy of F. japonica with an extremely narrow

fundamental and realised host range, and suggest that the psyllid would not pose an

unacceptable threat to non-target plants in the UK.

Given that the nymphs of this species are sedentary, the key to host specificity can be

assumed to lie with adult oviposition preference. If the adult does not adhere an egg to the

host plant then it is highly unlikely that first instar nymphs would ever be able to locate

another suitable host before their death given their size and locomotary ability. In the caged

laboratory studies established to test ovipositional preference, egg-laying behaviour on non-

target plants in the absence of the preferred F. japonica host was not significantly different

from tests when the knotweed was present. In all cases, a very limited number of eggs were

laid on non-targets in multiple-choice oviposition experiments, with the target present or

absent, indicating very high host selectivity by gravid females.

Overall 1.52% of 146,855 eggs recorded in these multiple-choice tests were laid on plant

species outside the invasive knotweed group, though this decreases to 0.63% if the two

uncommon horticultural varieties, F. connolyana and F. japonica var compacta are ignored.

Of the remaining 928 eggs, not one was able to develop fully to adult on any of these plant

species. This finding was largely supported by the nymph transfer development studies which

showed that development on non-target species is severely restricted. However, F.

conollyana, F. dumetorum and F. baldshuanica were able to support development to late

stages and M. complexa was found to be capable of supporting development of the psyllid

16

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

through to adult. The reasons for this apparent change in performance on a non-target is

likely to be due to the increased humidity in the experimental cage relative to that used in the

multiple-choice/development study allowing development that would not be likely to occur in

the field. Furthermore, in using first instar nymphs that are likely to have fed on F. japonica

albeit briefly, we may have inadvertently bypassed crucial stages in the process of nymph

survival i.e egg mortality/hatch on a non-host and the need for its preferred food source in the

first moments of life. Nonetheless, this result is less surprising when one considers that the

M. axillaris has recently been placed in the Polygonum clade based on molecular

characterisation (Lamb Frye and Kron, 2003).

Fallopia conollyana is not listed in the Royal Horticultural Society’s Plantfinder website. On

the other hand F. baldshuanica and M. complexa, both vigorous climbers that can be

problematic, are listed. The former, sold as Russian vine is often seen escaping from gardens

and smothering vegetation and is also recognised as invasive in parts of the USA under its

synonym Polygonum aubertii L. Henry. Muehlenbeckia complexa can also smother adjacent

vegetation and is known to form huge “humps and carpets” on the low cliffs of the Scilly

Isles. One species that would certainly be of concern were it to be threatened through a

release of the psyllid is F. dumetorum which is a protected species in the UK. However,

copse bindweed rarely co-occurs with the target knotweeds and our studies have shown that if

F. japonica is present, less than 2% of the total eggs laid in the multiple-choice oviposition

experiments are laid on this plant, and A. itadori cannot sustain a population on the plant nor

is damage likely to be perceptible.

Adult survival, used as a surrogate for feeding trials, on non-target plant species, further

indicates that in no-choice situations adult survival is severely compromised, even on the

17

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

most closely-related plants. This suggests that, in the field, adults would prefer to feed on

their normal knotweed host and be unlikely to feed on anything else. Our field and laboratory

observations indicate that adults do not inflict any perceptible damage in the absence of

nymphs which is supported by the work reported by Wineriter et al. (2003).

Since F. japonica dies off after the first frost, the question arises as to where the adult psyllids

over-winter. Ossiannilsson (1992) reported that “most of our [Scandinavian] psyllids

hibernate in the adult stage, a few on their host-plants but most species on "shelter plants",

usually conifers, or in crevices in bark or other protected sites”. Ossiannilsson goes on to say

that “Whether psyllids hibernating on conifers do actually feed on them, as was supported by

Reuter (1908), is still unknown." As far as known all Aphalara species overwinter on

conifers (Burkhardt, pers. comm.). Aphalara itadori adults are reported to disperse from their

host to shelter plants for over-wintering in autumn and Pinus densiflora Zieb. and Zucc. as

well as and Cryptomeria japonica D. Don, have been mentioned as shelter plants (Baba and

Miyatake, 1982; Miyatake, 1973; Miyatake, 2001). Again it is not clear whether the psyllids

actually feed on these shelter plants (Miyatake, 1973), however, there is no report that

overwintering psyllids of any species have ever damaged their shelter plants.

Consideration of other psyllids should provide some insight to the likely habit. Cacopsylla

pyricola (Förster), which feeds on pear, is able to over-winter as adults concealed in crevices

in the bark of the host tree from where they emerge in early March, in Scotland (Lal, 1934).

The blackberry psyllid, Phylloplecta tripunctata (Fitch), is known to spend the winter

hibernating on the needles of spruce, pine and cedar trees but do not seem to injure them

(Peterson, 1923). Some species, such as the Hackberry psyllids, Pachypsylla celtidivesicula

Riley and P. celtidismamma (Fletcher), actually become pests in houses whilst seeking over-

18

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

wintering sites, which clearly indicates that they do not feed during the winter period. The

exact behaviour of the adult of A. itadori throughout the winter is being studied by a team at

the University of Kyushu, who hope to determine in the near future which host plant(s), if

any, is used.

Psyllids are recognised vectors of viruses and phytoplasmas (e.g. Carraro et al., 2004). In

order to consider the risk of disease transmission associated with a release of A. itadori in the

UK, assuming feeding on pine species does occur, it is necessary to identify any potential

transmissible diseases. Since there are no reports of known plant pathogens attacking F.

japonica in the UK and Europe as yet, the risk to forestry in its introduced range posed by the

release of the psyllid can be considered minute. If any spill-over feeding of A. itadori should

take place adults are unlikely to create a new pressure from vectored diseases since other

oligophagous or polyphagous sucking insects such as Aphalara polygoni Förster already feed

on, and between, members of the Polygonaceae. The population of A. itadori that would be

considered for release will originate from a line-reared culture that has been taken through

more than 30 generations with no disease symptoms ever appearing in the insect or any

exposed test plant. Hence the risk of the introduction of a Japanese disease must be

considered as negliable.

In summary Aphalara itadori shows high fidelity to its normal host Fallopia japonica in the

field in Japan and in the laboratory host range tests we have subjected it to. Mated females

are highly unlikely to lay eggs on any species outside the target knotweed group. Even if they

do lay eggs on non-target plants, these will only be in areas close to invasive knotweeds, in

low numbers and only a very small proportion of these would be capable of hatching, feeding

and developing to any significant extent. Our studies further show that, only in the unlikely

19

event of eggs being laid on M. complexa could adult production be expected to be successful

and even then numbers will probably be so low as to preclude a population from being

maintained.

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

This programme has pioneered the first weed biocontrol agent host range risk assessment for

Europe. It is hoped that the inclusion of phylogenetically distant related economic and native

species considered unnecessary by Briese (2003) will not set a precedent for the rest of

Europe, whose use of classical biological control is expected to increase. This approach was

taken because of a lack of procedure for agreeing the list with the responsible authority and

we recommend that this situation is resolved as soon as possible to avoid wasting resources.

The data presented in this paper will be combined with thermal tolerance and performance

data and will be used to support a Pest Risk Analysis dossier to provide the UK authorities

with the necessary information to consider the release of this psyllid as a biological agent for

the permanent suppression of Fallopia japonica and its hybrids. We believe that this psyllid

is adequately specific to be suitable for release in the UK and further anticipate it being

damaging in the field, based on preliminary efficacy studies which are on-going. If

successfully released, A. itadori should prove that biological control is a worthy addition to

the integrated management of one of our worst weeds and should pave the way for many

more introductions of classical biological control agents in Europe. This would allow its

Member States perhaps to catch up with the rest of the world in utilising all the tools available

for invasive alien plant management.

Acknowledgements

20

481

482

483

484

485

486

487

488

489

490

This work would not have been possible were it not for the consortium of funders comprising

the Welsh Assembly Government, Department of Environment Food and Rural Affairs,

Environment Agency, South West Regional Development Agency, Network Rail and British

Waterways, all coordinated by Cornwall County Council. Some activities reported here were

also funded in part by the United States Department of Agriculture Forest Service.

Collections were facilitated by Professor Masami Takagi and Daisuke Kurose and further

field observations and literature reviews were made by Naoki Takahashi, all of Kyushu

University. Experimental support was provided by Ghislaine Cortat, Djamila Djeddour, Lynn

Hill and Sacha White. Thanks are also due to Sadie Rhodes and to anonymous reviewers who

provided valuable comment on an earlier version of this manuscript.

21

References 491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

Baba, K. and Miyatake, Y., 1982. Survey of adult psyllids in winter in Niigata Prefecture.

Transactions of the Essa Entomological Society 54, 55-62.

Bailey, J., 2005. The history of Japanese knotweed. Ecos - British Association of Nature

Conservationists 26 (3), 55-62.

Bailey, J.P., Conolly, A.P., 2000. Prize-winners to pariahs - a history of Japanese knotweed

s.l. (Polygonaceae) in the British Isles. Watsonia 23, 93-110.

Baker, C.R.B., Blackman, R.L., Claridge, M.F., 1972. Studies on Haltica carduorum Guerin

(Coleoptera:Chrysomelidae) on alien beetle released in Britain as a contribution to the

biological control of creeping thistle, Cirsium arvense (L.) Scop. Journal of Applied Ecology

9 (3), 819-830.

Beerling, D.J., 1994. The impact of temperature on the northern distribution limits of the

introduced species Fallopia japonica and Impatiens glandulifera in north-west Europe.

Journal of Biogeography 20, 45-53.

Beerling, D.J., Bailey, J.P., Conolly, A.P., 1994. Fallopia japonica (Houtt.); Ronse Decraene

(Reynoutria japonica Houtt,; Polygonum cuspidatum Sieb. and Zucc.). Journal of Ecology 82,

959-979.

Briese, D.T., 2003. The centrifugal phylogenetic method used to select plants for host-

specificity testing of weed biological control agents: can and should it be modernised?;

Improving the selection, testing and evaluation of weed biological control agents. CRC

Technical Series No.7, pp. 23-33.

22

BSBI, 2000. Botanical Society of the British Isles Atlas, 2000 online

(

512

http://www.bsbimaps.org.uk/atlas, accessed October 2007) 513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

Burckhardt, D., Lauterer, P., 1997. Systematics and biology of the Aphalara exilis (Weber

and Mohr) species assemblage (Hemiptera: Psyllidae). Entomological Scandinavica 28, 271-

305.

Burdon, J.J., Marshall, D.R., 1981. Biological control and the reproductive mode of weeds.

Journal of Applied Ecology 18 (2), 649-658.

Cameron, P.J., Hill, R.L., Bain, J., Thomas, W.P., 1989. A Review of Biological Control of

Invertebrate Pests and Weeds in New Zealand 1874 to 1987. CAB International, Wallingford.

Carraro, L., Ferrini, F., Labonne, G., Ermacora, P., Loi, N., 2004. Seasonal infectivity of

Cacopsylla pruni, vector of European stone fruit yellows phytoplasma. Annals of Applied

Biology 144 (2), 191-195.

Cock, M.J.W., 1985. A review of biological control of pests in the Commonwealth Caribbean

and Bermuda up to 1982. Commonwealth Institute of Biological Control Technical

Communication, Slough.

Conolly, A.P., 1977. The distribution and history in the British Isles of some alien species of

Polygonum and Reynoutria. Watsonia 11, 291-311.

Crawley, M.J., 1990. Plant life-history and the success of weed biological control projects. In:

Delfosse, E.S. (Ed), Proceedings of the VII International Symposium on Biological Control of

Weeds. CSIRO Publications, Victoria, Australia, pp. 17-26.

Cronquist, A., 1981. An integrated system of classification of flowering plants. Columbia

University Press, New York.

23

http://www.bsbimaps.org.uk/atlas

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

Defra, 2003. Review of non-native species policy - report of the working group. PB8072, pp.

1-90.

Eilenberg, J., Enkegaard, A., Vestergaard, S., Jensen, B., 2000. Biocontrol of pests on plant

crops in Denmark: present status and future potential; International symposium on biological

control agents in crop and animal protection, Swansea, UK, 24-28 August, 1999. Biocontrol

Science and Technology 10, 703-716.

Environment Agency, 2007. Managing Japanese knotweed, 2007 (February 27).

Holden, A.N.G., Fowler, S.V., Schroeder, D., 1992. Invasive Weeds of amenity land in the

UK: Biological control – the neglected alternative. Aspects of Applied Biology 29, 325-332.

Hollingsworth, M.L., Bailey, J.P., 2000. Evidence for massive clonal growth in the invasive

weed Fallopia japonica (Japanese Knotweed). Botanical Journal of the Linnean Society 133

(4), 463-472.

Holzner, W., Numata, M., 1982. Biology and Ecology of Weeds. W. Junk Publishers, The

Hague.

Julien, M.H., Griffiths, M.H., 1998. Biological Control of Weeds; A World Catalogue of

Agents and their Target Weeds (Fourth edition). CABI Publishing, Wallingford.

Kabat, T.J., Stewart, G.B. and Pullin, A.S., 2006. Are Japanese knotweed (Fallopia japonica)

control and eradication interventions effective? Systematic Review No. 21 Centre for

Evidence-Based Conservation, Birmingham, UK.

Lal, K.B., 1934. The biology of Scottish Psyllidae, Transactions of the Royal Entomological

Society of London 82 (2), 363-385.

24

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

Lamb Frye, A.S. and Kron, K.A. 2003. rbcLPhylogeny and character evolution in the

Polygonaceae. Systematic Botany 28 (2), 326-332.

Levin, D.A., 2000. The Origin, Expansion and Demise of Plant Species. Oxford University

Press, New York.

Lousely, J.E., Kent, D.H., 1981. Docks and Knotweeds of the British Isles. BSBI, London.

Mabberley, D.J., 1997. The Plant Book : A Portable Dictionary of the Vascular Plants.

Cambridge University Press, Cambridge.

Makino, T. and revision team, 1997. Revised Makino's Illustrated Flora In Colour.

Hokuryukan, Tokyo.

Mapmate. Mapmate, Checklist of recorded species page, (Online), .

Minks, A.K., Blommers, L.H.M., Ramakers, P.M.J., Theunissen, J., 1998. Fifty years of

biological and integrated control in Western Europe: accomplishments and future prospects;

Proceedings, 50th international symposium on crop protection, Gent, 5 May 1998. Part I.

Mededelingen - Faculteit Landbouwkundige En Toegepaste Biologische Wetenschappen,

Universiteit Gent 63 (2), 165-181.

Miyatake, Y., 1973. Psyllids and their life. Nature Study 19 (1), 5-11.

Miyatake, Y., 1964. Psyllidae in the collection of the Osaka Museum of Natural History, with

description of a new species (Hemiptera: Homoptera). Bulletin of the Osaka Museum of

Natural History 17, 19-32.

25

Miyatake, Y., 2001. A new species of the genus Aphalara from the cool temperate zone of

Honshu and Shikoku, Japan (Homoptera: Psylloidea: Aphalaridae). Special Publication of the

Japan Coleopterological Society, 123-127.

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

Ossiannilsson, F., 1992. The Psylloidea (Homoptera) of Fennoscandia and Denmark. Fauna

Entomologica Scandinavica 26. E.J. Brill, Leiden, The Netherlands.

Peschken, D., Friesen, H.A., Tonks, N.V., Banham, F.L., 1970. Releases of Altica carduorum

(Chrysomelidae : Coleoptera) against the weed Canada thistle (Cirsium arvense) in Canada.

Canadian Entomologist 102 (3), 264-271.

Peterson , A. (1923) The Blackberry Psyllid, Trioza tripunctata, Fitch, New Jersey

Agricultural Experiment Station Bulletin 378, 32.

Pridham, A.M.S., Schwartzbeck, R.A., Cozart, E.R., 1966. Control of emigrant Asian

perennials. Boikemia 11, 6-8.

Rao, V.P., Ghani, M.A., Sankaran, T., Mathur, K.C., 1971. A review of the biological control

of insects and other pests in south-east Asia and the Pacific region. A Review of the

Biological Control of Insects and Other Pests in South-East Asia and the Pacific Region, pp.

xiii-149.

Reuter, O.M., 1908, Charakteristik und Entwicklungsgeschichte der Hemipterenfauna der

Palaearktischen Coniferen. Druckerei der Finnishen Litteratur-Gesselschaft. Helsinki,

Sweden.

Royal Horticultural Society, 2003. RHS Plant Finder. Dorling Kindersley, UK.

Shaw, R.H., 2003. Biological control of invasive weeds in the UK: opportunities and

challenges. In: Child, L., Brock, J.H., Brundu, G. (Eds.), 6th International Conference on the

26

596

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

Ecology and Management of Alien Plant Invasions (EMAPi), 12 to 15 September 2001,

Loughborough, UK. Plant Invasions: Ecological Threats and Management Solutions, pp. 337-

354.

Sheppard, A.W., Shaw, R.H., Sforza, R., 2006. Top 20 environmental weeds for classical

biological control in Europe: a review of opportunities, regulations and other barriers to

adoption. Weed Research 46 (1), 1-25.

Shinji, O.. 1938. Five new species of Psylla from Japan. Kontyû 12, 146-151.

Stace, C.A., Ellis, R.G., Kent, D.H., McCosh, D.J., 2003. Vice-County Census Catalogue of

the Vascular Plants of Great Britain, the Isle of Man and the Channel Islands. Botanical

Society of the British Isles, London.

Synge, P.M., 1956. The Royal Horticultural Society Dictionary of Gardening. Royal

Horticultural Society, Oxford.

Wapshere, A., 1974. A strategy for evaluating the safety of organisms for biological weed

control. Annals of Applied Biology 77, 201-211.

Welsh Development Agency, 1998. The eradication of Japanese knotweed: Model tender

document. Welsh Development Agency, Cardiff.

27

Table 1 - Mean developmental periods for all stages of A. itadori in days (± 1SE, n = 21) 612

613

Egg 1st

instar

2nd

instar

3rd

instar

4th

instar

5th

instar

Complete

life cycle

Mean±1SE 9.2 ± 0.1 4.8 ± 0.2 3.3 ± 0.2 3.9 ± 0.3 4.5 ± 0.1 7.1 ± 0.3 32.9 ± 0.8

Range 9 - 10 4 - 6 2 - 5 3 - 8 4 - 6 5 - 11 28 - 42

614

28

Table 2 – Results of multiple-choice oviposition studies carried out with adult A. itadori

using various test plants. Development success was scored according to the relative number of

adults that emerged (Good = >50%, Moderate = 10-50%, Poor = <10%). Test plant status in

the UK is as follows: N= Native; I= Introduced; SH= can occur in same habitat as the target

F. japonica; NR= National Rare; Pr= Protected; A= American; O= ornamental; E=

Economically Important Species

615

616

617

618

619

620

621 Test plant

Test plant Status

Reps Eggs laid total

Mean number of egg count

(±1SE)

Development success of A.

itadori Polygonaceae

Sub family Polygonoideae Tribe Polygoneae Fallopia japonica(Houtt.) I 324 140,517 433.7 (± 28.3) Good Fallopia japonica USA (Ithaca) A 3 1,523 507.7 (±173.1 ) Good Fallopia x bohemica UK (Chrtek & Chrtkov) I 12 2,033 169.4 (± 41.5) Good Fallopia conollyana(Bailey) Hybrid 15 866 57.7 (± 17.7) Poor Fallopia japonica var compacta(Hook) I 12 441 36.8 (± 9.9) Poor Fallopia sachalinensis(F.Schmidt ex Maxim) I 18 547 30.4 (± 7) Moderate Fallopia dumetorum(L.) N, NR, Pr 12 93 7.8 (± 2.9) - Fallopia convolvulus (L.) ex. USA A 11 78 7.1 (± 3.1) - Fallopia convolvulus(L.) N, SH 14 11 0.8 (± 0.5) - Fallopia baldschuanica(Regel) I 15 100 6.7 (± 2) - Polygonum arenastrum(L.) N, SH 6 0 0 - Polygonum aviculare(L.) N, SH 6 0 0 - Polygonum boreale (Lange) N, NR,Pr 0 - - - Polygonum maritimum(L.) N, NR, Pr 9 0 0 - Polygonum oxyspermum(L.) N 6 0 0 - Polygonum rurivagum(Jord. Ex Boreau) N 5 0 0 - Tribe Persicarieae - Polygonum pensylvanicum(L.) A 6 0 0 - Persicaria sagittatum (L.) (USA) A 9 0 0 - Persicaria virginianum(L.) (USA) A 3 0 0 - Persicaria amphibia(L.) N, SH 6 0 0 -

Persicaria affins(L.) I,O 6 0 0 - Persicaria bistorta(L.) N, SH 9 0 0 - Persicaria campanulata(Hook.f.) I 12 0 0 - Persicaria capitata(Buch.-Ham. Ex D.Dom) I, O 9 0 0 - Persicaria hydropiper (L.) (USA) A 6 0 0 - Persicaria hydropiper (L.) N, SH 6 0 0 - Persicaria hydropiperoides (Michx.) (USA) A 10 0 0 - Persicaria lapathifolia(L.) N, SH 6 0 0 -

29

Persicaria maculosa(Gray) N, SH 6 0 0 - Persicaria minor (Hudson) Opiz N, NR, Pr 0 - - - Persicaria mitis(Schrank) Assenov N, SH, Pr 0 - - - Persicaria mollis(Schrank) I, O 6 0 0 - Persicaria polystachyum(Wall. Ex Meish) I, SH 6 15 2.5 (±1.8) - Persicaria tinctoria(Aiton) I, O 9 0 0 - Persicaria vivipara(L.) N 6 0 0 - Fagopyrum esculentum Moench I, SH 14 237 16.9 (±6.7) - Fagopyrum dibotrys(D. Don) I 9 1 0.1 (±0.1) - Tribe Coccolobeae

-

Muehlenbeckia complexa(A.Cunn.) I 7 103 14.7 (±5) - Tribe Rumiceae - Rheum palmatum(L.) I, SH, O 12 208 17.3 (±7) - Rheum hybridum Glaskins(L.) I, SH, E 9 8 0.9 (±0.9) - Rumex acetosa(L.) N, SH 6 0 0 - Rumex acetosella(L.) N, SH 6 0 0 - Rumex alpinus (L.) (USA) A 6 0 0 - Rumex aquaticus(L.) N, SH 6 0 0 - Rumex conglomeratus(Murray) N, SH 12 0 0 - Rumex crispus(L.) N, SH 6 0 0 - Rumex longifolius(DC.) N 6 0 0 - Rumex maritimus(L.) N 6 0 0 - Rumex obtusifolius(L.) N, SH 9 0 0 - Rumex orbiculatus(Gray) (USA) A 15 0 0 - Rumex palustris(L.) N, SH 6 0 0 - Rumex pulcher(L.) N, SH 6 0 0 - Rumex rupestris(Le Gall) N, NR,Pr 6 0 0 - Rumex sanguineusL.) N,SH 6 0 0 - Rumex scutatus(Jacq.) NE,O 6 0 0 - Rumex hydrolaphum(Hudson) N,SH 6 2 0.3 (±0.2) - Oxyria digyna(L.) N 12 43 3.6 (±3.1) -

sub family Eriogonoideae - Erigonium umbelatum(Torr.) I,O 6 0 0 -

Other Caryophyllales -

Family Chenopodiaceae - Beta vulgaris(L.) E 6 0 0 - Chenopodium album(L.) N,SH 6 0 0 -

Family Caryopyllaceae - Cerastium fontanum(Baumg) N,SH 6 0 0 - Cerastium glomeratum(Thurill) N,SH 15 0 0 -

Family Nyctaginaceae - Bougainvillea spp. O 9 0 0 -

Family Aizoaceae - Delosperma cooperi(Hook) O 6 0 0 -

Family Plumbaginaceae - Limonium bellidifolium(Gouan) N, NR 5 0 0 -

30

Limonium binervosum(AGG) N,NR 6 0 0 -

Family Frankeniaceae - Frankenia laevis(L.) I,O 6 0 0 -

Family Portulacaceae - Lewisia columbiana(Howell ex Gray) O 9 0 0 -

Family Phytolaccaceae - Phytolacca americana(L.) O 6 0 0 -

Family Cactaceae - Notocactus magnificus(F. Ritter) O 3 0 0 - Echinocereus subinermis(SD ex Scheer) O 6 0 0 - Carpobrotus edulis(L.) I,O 6 0 0 -

Family Amaranthaceae - Celosia argentea(Hook) I,O 6 0 0 -

Family Tamaricaceae - Tamarix gallica (cut branchlet) I,O 11 0 0 -

Morphologically similar - Calystegia sepium(L.) N,SH 9 0 0 - Houttuynia cordata(Thunb.) I,O 7 0 0 -

Biochemically similar - Rubus fruticosus(Sens.) N,SH 6 0 0 - Vitis vinifera(L.) I,O 6 0 0 -

Agent specific species - Cytisus scoparius(L.) N,SH 3 0 0 - Malus domestica (Borkh) E 9 0 0 -

Economic safeguard species - Hordeum vulgare(L.) E 6 0 0 - Lycopersicon esculentum(L.) E 6 0 0 - Phaseolus vulgaris (L.) E 9 0 0 - Solanum melongena(L.) E 6 0 0 - Solanum tuberosum(L.) E 6 0 0 - Triticum aestivum(L.) E 6 0 0 - Vicia faba(L.) E 9 0 0 - Zea mays(L.) E 6 0 0 - Rosa spp.(L.) E 6 0 0 - Brassica napus(L.) E 6 0 0 -

622

31

Table 3 - The mean number of eggs received per test plant in the presence and absence of the

Fallopia japonica in caged multiple choice experiments (± 1SE) and the significance of any

difference as determined by a generalised linear model using Poisson error structure. P values

corrected for overdispersion. (dispersion parameters for F. dumetorum, F. convolvulus and F.

baldshuanica were 12.609, 14.956 and 7.55 respectively).

623

624

625

626

627

628

Test Plant F.japonica present F. japonica absent Chi-square

value

p value

F. dumetorum 7.75 ± 2.895 (n=12) 5.167 =/- 3.188 (n=6) 0.323 0.578

F. convolvulus 3.65 ± 1.775 ( n=20) 6.833 ± 2.664 (n=6) 0.643 0.431

F. baldshuanica 7.75 ± 2.336 (n=12) 3.833 ± 1.887 (n=6) 1.379 0.257

629

32

Table 4 – Percentage survival over time and development of nymphs of A. itadori hand

transferred onto various host plants. (* indicates values that have been estimated based on

counts on previous and following days)

630

631

632

633 Test plant Day 3 Day 7 Day 14 Day 28 Ultimate stage

Fallopia japonica 73.1 70.7 68.6 66.2 Adult (>66%) Fallopia conollyana 61.7 61.7 53.3 48.3 Adult (48.3%) Meuhlenbeckia complexa 40.0 26.7 11.7 8.3 Adult (6.7%) Fallopia dumetorum 33.3 28.3 13.3 3.3 N5 (10%) Rumex hydrolapathum 38.9* 25.0 13.3 3.3 N3 (8.3%) Rheum palmatum 27.5* 10.0 1.7 0 N3 (1.7%) Polygonum arenastrum 11.7* 3.3 1.7 0 N3 (1.7%) Fallopia baldschuanica 51.7 22.5* 0.8* 0 N4 (1.7%) Oxyria digyna 16.7 6.7 0 0 N3 (1.7%) Fallopia convolvulus 40.0 2.2* 0 0 N1 (41.7%) Fagopyrum esculentum 8.3 1.7 0 0 N3 (15%) Persicaria polystachyum 5* 0.4* 0 0 N2 (8.3%) Rheum Glaskin's 23.3 0 0 0 N2 (8.3%) Fagopyrum dibotrys 5.6* 0 0 0 N2 (5%)

33

Figure 1 - Plot of log number of UK hectads containing F. japonica over time (Data from

Botanical Survey of the British Isles).

634

635

636 637

638

639

640

641

Figure 2 - The individual stages of the life cycle of A. itadori (clockwise from top right –

egg, N1, N2, N3, N4 N5 and adult)

Figure 3 - Survivorship curves showing the number of adult A. itadori alive over time on

different hosts (see legend)

34

642

1

1.5

2

2.5

3

3.5

4

1900 1920 1940 1960 1980 2000

Year

Log

num

ber o

f hec

tads

643

35

644

36

0

10

20

30

40

50

60

0 1 2 3 4 5 6 7 8 9 10 11 12

Day

No.

Aliv

e

F. japonicaF. bladshuanicaF. dumetorumF. convolvulusF. esculentumPlastic plantPlastic plant (H2O)

37

Appendix 3 Efficacy of the psyllid In Nagano Prefecture in Japan, Aphalara itadori nymphs have been observed causing significant damage to large plants of Japanese knotweed, stunting growth, limiting leaf expansion and reducing flowering (Author’s observations) and indeed the sap-sucking activities of the psyllid nymphs are capable of killing potted plants of Japanese knotweed under high loads. However, the populations from wthe culture was established were much smaller. It should be borne in mind that thespecialist natural enemies present in Japan have been removed by the line rearingprocess and as such the psyllid should bmore likely to reach higher populations than those observed in the field without thirestriction. We anticipate and hope that the psyllid will provide significant cothe UK, both through reduction in JK vigour and where populations do occur inumbers, death of plants. The psyllid is not expected to eliminate the knotweed entirely since the pest/host system should be self-regulating and should settle at equilibrium after initially greater population fluctuations. It would be expected toreduce the overall vigour of knotweed as well as its competitive advantage over other plants. This should allow other species, including natives, to grow where they werepreviously hindered. There would also be an associated reduction in current coneffort and costs, not least through reduced use of chemicals.

hich

e

s ntrol in n high

trol

ome impact studies have been carried out in the laboratory and are summarised here:

es of knotweed rhizome (0.52g – 5.32g) were planted in 9cm pots and the

, R

ence of feeding nymphs was found to have a significant negative impact on

Fig 1 – Psyllid impact in Japan

SMethods Small piecresultant plants exposed to 10 replicates each of first instar nymph loads of 0, 10 and 40 (control, low, maximum) under the same conditions as detailed in the culturing section above. The nymphs were transferred a fine paintbrush and placed near the base of the petioles of the upper leaves and the plants were covered with a Perspex cloche. At this time the plant height and number of leaves was recorded. Similar measurements were made weekly for three weeks and comparisons made of the increase in height and increase in number of leaves under each treatment (Anovaversion 2.3.1) Results The presplant performance as measured by height and shown in Figure 1. The mean increase in growth relative to initial height were 1.39, 0.99 and 0.55 for control, 10 (Low) and 40 (maximum) nymphs respectively (Anova F = 4.306, p = 0.024).

Control Low Maximum

0.5

1.0

1.5

2.0

2.5

3.0

Nymph load

Pro

porti

onal

Incr

ease

in h

eigh

t

Figure 1 - Boxplot of proportional increase in knotweed height after 3 week exposure to varied nymph loads of A. itadori. There is no significant difference between control and low (p = 0.122) but maximum nymph load causes a significant reduction in height change (p = 0.015) Nymph feeding was also found to have a significant effect on leaf production at both the 10 and 40 nymph level as displayed in Figure 2 (Anova F = 7.69, p = 0.002).

Control Low Maximum

510

1520

2530

35

Nymph load

Incr

ease

in le

aves

Figure 2 - Boxplot of the increase in leaf count under differing A. itadori nymph loads It would appear that the impact of the psyllid will be felt even at low densities. and that with increased leaf area induced, integration with chemicals could lead to greater control. A reduction in the range and dominance of Japanese knotweed is anticipated as a result of the activities of the psyllid. Laboratory studies, funded by the Environment Agency, are on-going to investigate integration of the psyllid with current control measures. Preliminary results indicate

that psyllid feeding can reduce the amount of the most commonly-used chemicals required to have an impact on the plant. This is also proving true for cutting and digging (using different sized rhizome as a surrogate for the after effects of digging). In short, the presence of the psyllid can improve the results from traditional control efforts replicated in the quarantine lab.

Appendix 4 – Climatic comparisons One of the key factors in determining if a biological control agent will establish in an area outside of its native range is temperature (Boivin, et. al., 2006). If, in the area of introduction the temperature extremes are significantly higher or lower than the area of origin, the agent may have difficultly establishing and have a higher risk of death. The climate of Japan is strongly influenced by the Asian continent, the Pacific Ocean and the associated air masses. It is the collision of the Siberian air mass into the moist Pacific air mass that causes high snow fall in the west of the country. Three main ocean currents influence the climate, the Kuro-shio current, a warm current flowing north washing the southern and eastern side of Japan, the Tsushima current, another warm current flowing west of Kyushu and the Oya-shio current, a cold water current flowing south past the east side of Hokkaido. Overall Japan and the UK have temperate climates with four clearly defined seasons. There is considerable variation in temperature within each country with Japan showing the greater extremes. Japan extends over 25о of latitude, 3,200km from northeast to south west. The southern-most part of Japan, including the islands of Yaku-shima and Tokara-retto, have a sub-tropical climate, where the average winter temperature rarely drops below 13o. The yearly average temperature range is 7

oC to 17

оC and average precipitation per year ranges from 1000-

2500mm. Figure 1 shows the average monthly temperature of 5 regions in Japan at an altitude of 200 metres above sea level.

-5

0

5

10

15

20

25

30

Jan

Feb

Mar

Apr

il

May

June

July

Aug

Sept Oct

Nov

Dec

Month

Tem

pert

ure

Deg

rees

Cel

sius

Central HonshuNorthern HonshuWestern HonshuKyushuShikoku

Figure 1. The average monthly temperature from 5 regions of Japan The UK has less extreme variation in temperature compared to Japan. The warmest and sunniest parts of the UK are in the south of the country. The UK climate is relatively mild for its latitude, due mainly to the Gulf Stream currents. The UK’s changeable climate is a result of air mass convergence between the warm tropical airflow and the cooler polar airflow. The mean annual temperatures range from 8.5 оC to 11 оC, with the average precipitation ranging

1

from 800 to 2000 mm per year. Figure 2 shows the average temperature (2006 and 2007) of four regions of the UK and the average UK temperature overall. Comparing figure 1 and figure 2 it is evident the UK is less prone to climatic extremes than Japan

2007

0

2

4

6

8

10

12

14

16

18

20

JAN

FEB

MARCH

APRILMAY

JUNE

JULY

AUGSEPT

OCTNOV

DEC

Month

Tem

pert

ure

(Deg

rees

Cel

sius

)

UK

England

Wales

Scotland

N Ireland

2006

0

2

4

6

8

10

12

14

16

18

20

JAN

FEB

MARCH

APRILMAY

JUNE

JULY

AUGSEPT

OCTNOV

DEC

MonthTe

mpe

rtur

e (D

egre

es C

elsi

us)

UK

England

Wales

Scotland

N Ireland

Figure 2. The average monthly temperature for four regions of the UK, and the UK overall in 2006 and 2007.

Japan’s topography is also very different to that of the UK; almost 75% of Japan is mountainous. The steep undulating landscape adds to climatic variation throughout Japan. Temperature decreases with altitude or increases with decent, at a rate of 0.0065

оC per metre

(0.65 оC /100m - the Standard Atmosphere Temperature Gradient SATG). Figure 3 shows the

temperature variation with altitude on two mountains in Japan, with the average monthly temperature of England in 2006 and 2007. The graphs show that to match the climate of the UK’s growing season to that of Japan, altitude must increase as latitude decreases. In Central Honshu (Mt. Shirane) an altitude of between 800m-1700 metres above sea level provides the closest match to England whereas on the island of Kyushu (Mt. Aso), where the current culture was collected the temperature is warmer in the summer months and colder in the winter, at higher altitude areas. Mt Aso Mt Shirane

-5

0

5

10

15

20

25

30

35

Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.

Month

Tem

pert

ure

(Deg

rees

Cel

sius

)

0m200m400m600m800m1000m1142m1300m1592mEngland 2007England 2006

-15

-10

-5

0

5

10

15

20

25

Jan

Feb

MarAp

rilMay

June Ju

lyAu

gSe

pt

Octobe

rNov Dec

Month

Tem

pert

ure

(Deg

rees

Cel

sius

) 1400

1600

1800

2000

2200

2400

2600

UK 2007

UK 2006

Figure 3 Temperature decrease with altitude for Mt. Aso, Kyushu and Mt. Shirane, Central Honshu. UK average temperature (2006, 2007) overlayed.

2

Temperature controls the development of an organism throughout its life cycle- effecting the time of development, fecundity, ovioposition and mortality. The upper and lower development threshold of an organism determines when the development of an organism stops. Following the method of Olsen et. al., (2003), we determined the lower development threshold of A. itadori under controlled temperature quarantine conditions. We recorded the development times of individual psyllids under 7 constant temperatures (10oC, 12 oC, 17 oC, 21 oC, 23 oC and 28 oC). Figure 4 shows the relationship between the development rates (1/day) at the 7 temperatures. Extrapolation of the line indicates the lower development threshold is 8oC. However there was no development at 10oC and very limited development at 12oC. The fitted line includes the zero observations for 10oC to reflect the absolute lack of development at this temperature, and identify a slope for the linear model. Given the poor survivorship at 12oC , the insect's response is likely non linear at low temperatures, however additional experimentation is required to fit this more explicit model.

R2= 0.9328 Dev Rate per day = 0.01921+0.002162 Temp

Figure 4. The relationship between development rates at seven constant temperatures Using the graph and the equation line in figure 4 we were able to calculate the day-degrees requirement for the psyllid by calculating the reciprocal of the slope of the regression line

3

(1/y). This gave us a day-degrees requirement of 462.5 from egg to adult. To define, day-degrees are the total amount of heat required, above the development threshold, for an organism to develop from one stage to another stage of its life cycle.

Figure 5. A map of the accumulated day-degrees in the UK (2007). The legend refers to the number of day degrees. Using the calculated day-degrees figure of 462.5 and referring to the day degrees map of the UK (Figure 5), we can envisage the possible distribution of the psyllid throughout the UK, based on temperature. In the south east of the UK, the psyllid could establish two generations, whereas in the south west and central England the psyllid would establish one-possibly two generations. The darker areas of the map, excluding the blue area of the Lake District and the Scottish highlands, mainly Scotland and higher altitude areas of Wales, would

4

establish one generation of the psyllid. The blue areas, the areas showing the lowest day-degrees, are free from or have a low abundance of Japanese knotweed (Figure 6). Obviously, temperature is not the only determinant with regard to the spread of the psyllid in the UK. The availability of the host plant will also be highly significant in determining the area covered by the agent (Figure 6).

Figure 6 The area covered by Fallopia japonica in the UK. We would like to acknowledge and thank Dr Rob Bourchier from the Agriculture and Agri-food Canada-Lethbridge Research Centre for the work he conducted on the climatic matching and statistical analysis. The climatic matching is part of an ongoing research programme and this assessment will be updated and published.

5

Appendix 5 – The test plant list Introduction and approach The determination of host range is a critical consideration in the development of a biological agent for weed control and to this end a comprehensive test plant list has been drawn up, following the centrifugal-phylogenetic system for specificity proposed in 1974. This approach continues to serve as the basis of current host-range testing protocols as recognized by the IPPC Code of Conduct for the Import and Release of Exotic Biological Control Agents (ISPM No.3). Essentially an initial small group of taxonomically related plants, with similar morphological and biochemical characteristics to the target weed are tested, gradually expanding the scope to include more distantly related plants until specificity is established. Established protocols also include ornamental species as well as economically important plants and crops and those closely related plants found in neighbouring countries. Japanese knotweed (Fallopia japonica) belongs to the Class Dicotyledonae, Order Caryophyllales and Family Polygonaceae. The family Polygonaceae is composed of two sub families (1) Eriogonideae and (2) Polygonoideae with Fallopia japonica belonging to the latter. There are approximately 44 genera in the sub family Polygonoideae, but only seven are found in the UK, Fagopyrum, Persicaria, Polygonum, Fallopia (includes Fallopia japonica), Rumex, Rheum, Oxyria and Muehlenbeckia. In compiling the test plant list, plant species were initially selected from within the Family Polygonaceae, sub-families Polygonoideae and Eriogonideae. As it was unfeasible to include all native and ornamental species from the UK from this Family, careful consideration was given to the status of the species. Native species took priority, along with commonly available ornamentals and economically important genera e.g. Rheum. After selecting 47 species in the Family Polygonaceae, following the centrifugal-phylogenetic system, we moved out of the Family to select more distantly related species. This involved selecting representative species, with UK relevance, from the other Families in the Order Caryophyllales (Figure 1). We then selected species with similar morphology, similar biochemical characteristics and plants of economical importance. Lastly, as a result of interactions with regulators for plant pathogens, a further three plant species were added to the proposed test plant list namely Lemma minor, Larix decidua and Populus tremula. The taxonomy and phylogeny of the species selected was taken from Stace (2003) and Mabberley (1987). The status of each species was evaluated using Preston et. al. (2002) and the JNCC Red Data List (Cheffings et. al., 2005). Ornamental species were selected using the Royal Horticultural Society’s Plant Finder (2003). The complete test plant list can be found in Table 1 and a further introduction to each species can be found in the following section. In all, 73 species from 19 families have been included in the test plant list, consisting of 41 species native to the UK, 10 introduced species, 2 native to Europe, 6 ornamentals and 14 economically important species. Some species are a combination of some of the above. The evolution of the test plant list has been ongoing throughout the project and as a result a number of species with a high degree of phylogenetic separation to the target weed were removed from the test plant list in 2006, namely, Coccoloba uvifera, Drosera intermedia, Nepenthes fusca, Phytolacca americana, Bougainvillea species, Cactus species, Delosperma cooperi, Lewisia columbiana, Chenopodium giganteum, Celosia argente var. cristata, Polygonatum verticillatum, P. multiflorum and Solanum melongena. This allowed more time to concentrate on the more closely related species.

1

Figure 1 - A cladogram of the phylogenetic relationships between families in the Caryophyllids (Figure courtesy of University of Berkley) In contrast, as a result of suggestions with regulators for plant pathogens, a further three plant species were added to the proposed test plant list namely Lemna minor, Larix decidua and Populus tremula. In addition it was decided to add Frankenia laevis to the test plant list due to its close phylogenetic relatedness to Polygonaceae. Sourcing the test plants on the proposed test plant list has been a challenge and all but three species have been sourced either as plant or seed stock. Our suppliers have included numerous garden centres and nurseries and national collections, especially from Rowden Gardens in Cornwall- the UK’s national Polygonum collection. We have sourced seed stock from throughout the UK and in some cases abroad where UK stock was limited. Kew Seed Bank has supplied us with many hard to find seeds from closely related plant species like Polygonum maritimum and Fallopia dumetorum. Three plant species have proved very difficult to obtain, namely Polygonum boreale, Persicaria mitis and Persicaria minor. However, recently we have managed to source, what is thought to be, Polygonum boreale. In the latter part of 2007, we secured information on the location of Persicaria mitis and Persicaria minor in the UK and in spring 2009 we plan to make field collections of these species.

2

Table 1 - The Test Plant List Sorted by relatedness and then by selection criteria (E=Economically important; NE= native to Europe; O= Ornamental; N= Native; EV=Environmental value; I=Introduced). Conservation designations for UK taxa according to JNCC Red Data list (Cheffings et. al. 2005): VU=Vulnerable; EN= Endangered; NT= Near threatened; LC=Least concern. ** = Protected under Schedule 8 of the Wildlife and Countryside Act, 1981. Information on the occurrence of species in or near stands of Japanese knotweed were taken from Beerling et. al., (1994), Preston et. al., (2002) and Blamey et. al., (2003). Species Common name UK status Grows near JK Order Caryophyllalles Family Polygonaceae Subfamily Poligonoideae Fallopia japonica var. japonica Japanese knotweed I baldschuanica Russian vine I Yes convolvulus Black bindweed N/EV/LC Yes dumetorum Copse bindweed N/EV/VU No sachalinensis Giant knotweed I Yes japonica var. compacta Dwarf variety I No bohemica hybrid. I No conollyana hybrid - Yes Polygonum arenastrum Equal leaved knotgrass N/EV/LC Yes aviculare Knotgrass N/EV/LC Yes maritimum Sea knotgrass N/EV/VU** No oxyspermum Ray’s knotgrass N/EV/LC No rurivagum Cornfield knotgrass N/EV/LC No boreale Northern knotgrass N/LC No Persicaria amphibia Amphibious bistort N/EV/LC Yes bistorta Common bistort N/EV/LC Yes campanulata Lesser knotweed I/EV Yes hydropiper Water pepper N/EV/LC Yes lapathifolia Pale persicaria N/EV/LC Yes capitata Himalayan persicaria I No mollis Soft knotweed NE/EV No affinis Himalayan fleece flower O No maculosa Redshank N/EV/LC Yes vivipara Alpine bistort N/EV/LC No tinctoria Dyer’s knotgrass E No polystachum Himalayan knotweed I No mitis Tasteless Waterpepper N/EV/VU No minor Small Waterpepper N/EV/VU No

3

Species Common name UK status Grows near JK Fagopyrum dibotrys Tall buckwheat I No esculentum Buckwheat I No Rumex acetosa Common sorrel N/EV/LC Yes acetosella Sheep’s sorrel N/EV/LC Yes aquaticus Scottish Dock N/EV/VU No conglomeratus Clustered Dock N/EV/LC No hydrolapathum Water Dock N/EV/LC No palustris Marsh Dock N/EV/LC No pulcher Fiddle Dock N/EV//LC No longifolius Northern dock N/EV/LC No maritimus Golden dock N/EV/LC No obtusifolius Broad-leaved dock N/EV/LC Yes rupestris Shore dock N/EV/EN** No sanguineus Wood dock N/EV/LC No crispus Curled dock N/EV/LC Yes scutatus French sorrel NE No Oxyria digyna Mountain sorrel N/EV/LC No Rheum x hybrid Rhubarb E/I No palmatum Rhubarb O/I No Muehlenbeckia complexa Wire plant O/I No Subfamily Eriogonoideae Eriogonum umbellatum Sulphur buckwheat O No Other representative species within Caryophyllales Limonium bellidifolium Matted sea lavender N/EV/LC No binervosum Rock sea lavender N/EV/LC No Cerastium glomeratum Sticky mouse-ear N/EV/LC No fontanum Common mouse-ear N/EV/LC No Carpobrotus edulis Kaffir Fig O No

4

Species Common name UK status Grows near JK Chenopodium album Fat hen N/EV/LC Yes Beta vulgaris Beetroot E No Frankenia laevis sea-heath N/EV/NT No Tamarix gallica French Tamarisk I No Biochemically Similar Plants Vitis vignifera Grape E No Rubus fruticosus Blackberry N/EV/LC Yes Morphologically Similar Plants Calystegia sepium Hedge bindweed N/EV/LC Yes Houttuynia cordata Chameleon plant O No

Economically Important Plants Malus spp. Crab-apple E Yes Rosa City of Leeds Rose E Yes Phaseolus vulgaris French bean E Yes Broad bean E Yes Lycopersicon esculentum Tomato E Yes Solanum tuberosum Potato E Yes

5

Species Common name UK status Grows near JK Triticum aestivum Wheat E No Zea mays Corn E Yes Hordeum vulgaris Barley E No Brassica napus Rape seed E No Recent Additions Lemma minor Duckweed N/EV/LC Yes Larix decidua European Larch N/EV Yes Populus tremula Aspen N/EV Yes

6

Justification for the inclusion of plant species Fallopia baldschuanica (Russian vine) Included in the test plant list due to its close relatedness to the target species. Russian vine is a non-native species and often considered invasive just as it is in the USA where it is sold under its synonym Polygonum aubertii or Japanese fleeceflower. Still available in garden centres and grown as a garden covering vine, Russian vine has the potential to grow in close proximity to Fallopia japonica especially in urban areas and on brown field sites. Fallopia convolvulus (Black bindweed) A very common annual native species found throughout the UK and Ireland. Regarded as an archaeophyte, this species was formerly regarded as a weed of cultivated sites. Included in the test plant list due to its close relatedness to the target. As a weed of waste ground and roadsides Fallopia convolvulus is often found in close proximity to the target species. Fallopia dumetorum (Copse bindweed) Fallopia dumetorum is the only native Fallopia in the UK and listed as a Red Data book species (classification: vulnerable). This species has a very local distribution around the south east of the UK – as far down as the south coast, though its distribution has declined over recent years. Preston et. al. (2002) regards the plant as having an erratic appearance in its preferred habitat of hedgerows and woodland margins. Included in the test plant list due to its close relatedness to the target. Fallopia sachalinensis (Giant knotweed) Giant knotweed became commercially available to gardeners in the UK in 1869 and since has become distributed throughout the UK countryside. This non-native species is found on waste ground, roadsides and river banks, mainly in lowland areas where it forms dense monocultures, sometimes mixed in with the target. It can also be a pollen partner producing the highly invasive F x bohemica. The species has no environmental benefits. Included in the test plant list due to its relatedness to the target. Fallopia japonica var. compacta (Dwarf variety) Fallopia japonica var compacta has limited availability in the UK as a non-native garden ornamental though due to its namesake ‘knotweed’ the plant is not commonly planted. Four suppliers are listed as supplying this species on the Royal Horticultural plant finder website, though few are sold (John Carter, National Polygonum Collection pers. com.). A further 3 suppliers are listed for Fallopia japonica var compacta ‘Milky Boy’- and again, this variegated variety is uncommon as an ornamental. Two further F. compacta varieties are listed on the RHS site, namely F. compacta ‘Midas', a variety wrongly identified, it is actually a variety of F. japonica, and no longer available (last listed in 2004) and F. japonica var compacta f. rosea- which again is not commonly available. Fallopia bohemica (hybrid) This hybrid of Fallopia japonica x F. sachalinensis is regarded as a rampant non-native weed and is distributed throughout the UK. Found mainly on waste ground and along roadsides and river banks in lowland areas. Cultivated in the UK since 1872, F. bohemica was not collected from the wild until 1954. Again, included in the test plant list due to it being closely related to the target species.

7

Fallopia conollyana (hybrid) A non-native hybrid of Fallopia japonica x Fallopia baldschuanica which can persist from time to time in the UK countryside.

Polygonum The genus Polygonum contains a group of species which are the most closely related to Fallopia and this is represented by both Fallopia and Polygonum being classified together under the tribe Polygoneae. All Polygonums which are native in the UK are included in the list.

Polygonum arenastrum (Equal leaved knotgrass) An native archaeophyte which is widespread throughout the UK and Ireland, often found on waste ground, trampled ground and growing out of gaps in concrete. Polygonum arenastrum is a low-growing mat-forming annual species. It is feasible that the plant will grow in similar habitats to that of the target. Polygonum aviculare (Knotgrass) A native annual species, commonly found throughout the UK in gardens, arable fields, pavements, tracks and waste ground. This plant has a more sprawling and erect form to that of Polygonum arenastrum. It is feasible that the plant will grow in similar habitats to that of the target. Polygonum maritimum (Sea knotgrass) A native perennial herb of sand and shingle along the south coast of the UK. Listed under the JNCC Red Data list as a vulnerable species and protected under schedule 8 of the Wildlife and Countryside Act 1981. Japanese knotweed does not favour this habitat so is unlikely to grow in any of the locations where Polygonum maritimum occurs. Polygonum oxyspermum (Ray’s knotgrass) Found predominately along the west and southern coast of the UK, this native annual herb has fluctuating populations from one year to the next. Although Fallopia japonica does not prefer sandy habitats near the coast it is occasionally found in such habitats. Polygonum rurivagum (Cornfield knotgrass) Polygonum rurivagum is classed by Preston et. al. (2002) as a native archaeophyte found mainly in southern England. Found mainly on arable land and chalk soils this species has considerably extended its range in the UK. Unlikely to grow in similar habitats to that of the target. Polygonum boreale (Northern knotgrass) A rare native species with a very localised distribution in Scotland and the surrounding islands of Orkney and Shetland. Polygonum Boreale is similar in form and structure to Polygonum aviculare, and therefore it is difficult to tell the two species apart. This species has proved difficult to obtain, however, a botanist in the Orkney Islands is due to send samples this summer with permissions.

8

Persicaria The genus Persicaria is a large group of native and ornamental species in the UK. Due to their economic value 14 species (8 natives and 6 introduced/ornamental) represent this group in the proposed test plant list. Two of the species in this group Persicaria minor and P. mitis are national rare and have proved difficult to obtain. Persicaria amphibia (Amphibious bistort) A native species to the UK, with two distinct forms- an aquatic free floating form found on slow moving water bodies and a terrestrial erect form common in grassland and damp areas throughout the UK. Persicaria bistorta (Common bistort) A native perennial herb commonly found throughout the UK though absent from the higher altitude areas of the Scottish Highlands. Common along road sides and river banks. It is therefore feasible that Polygonum bistorta will grow near stands of Japanese knotweed. Persicaria campanulata (Lesser knotweed) A tall attractive non-native species often grown as a garden ornamental in the UK. Persicaria campanulata was first introduced into the UK in 1909 and was recorded in the UK countryside in 1933. Found scattered throughout the UK in localised patches along roadsides hedge banks and streams, it is feasible this species can grow close to the target species. Persicaria hydropiper (Water pepper) An attractive UK native commonly found throughout the UK on damp mud. This species is known to grow in and near Japanese knotweed patches along river banks and roadsides. Persicaria lapathifolia (Pale persicaria) Very similar to the above species, Persicaria lapathifolia is a native found throughout the UK, absent only from high altitude areas of the Scottish Highlands. Found along streams, ditches and cultivated land this species has the potential to grow around stands of Japanese knotweed. Persicaria capitata (Himalayan persicaria) Often a garden escapee in southern and central UK. This mat forming non-native annual species can be found along roadsides, paths and in urban areas. As an introduced species this plant is not renowned for its environmental value to the UK. Persicaria mollis (Soft knotweed) A garden escapee from Europe found in only a few sites in the UK countryside around the south of the UK. Where it does grow it has the potential to form dense clumps. As an introduced species this plant is not renowned for its environmental value to the UK. Persicaria affinis (Himalayan fleece flower) A mat-forming perennial ornamental species, not known to occur naturally in the UK countryside. However, as a species grown in urban areas, Persicaria affinis has the potential to grow near stands of Japanese knotweed.

9

Persicaria maculosa (Redshank) A native and widespread species similar to Persicaria hydropiper. Found along river banks, roadsides and waste ground this species is likely to grow in and near stands of Japanese knotweed. Persicaria vivipara (Alpine bistort) A short tufted native perennial herb growing on wet rocks and grassland, often abundant in montane grassland in Scotland. Due to the high altitude distribution of this species, it is unlikely Persicaria vivipara will grow in or near stands of Japanese knotweed. Persicaria tinctoria (Dyer’s knotgrass) Present in the UK in limited populations, Persicaria tinctoria is a non-native species grown for the natural blue dye the plant’s leaves produce when crushed. Persicaria polystachum (Himalayan knotweed) A tall, rhizomatous perennial herb found along streams and hedge banks in the UK. Non-native and considered invasive in some areas, Persicaria polystachum, also known as Persicaria wallichii, has been observed growing in close proximity to Japanese knotweed. Persicaria mitis (Tasteless Waterpepper) A rare annual native species of wet soils. Found alongside ponds, lakes and rivers and in damp meadow and cattle trampled pastures. Unlikely to be found growing near the target species. Currently, this species is unobtainable though it is hoped field collections will be made in the summer of 2008. Red Data book status = Vulnerable. Persicaria minor (Small Waterpepper) A rare annual species of wet soils. Found alongside ponds, lakes and rivers and in damp meadow and cattle trampled pastures. Unlikely to be found growing near the target species. Currently, this species is unobtainable though it is hoped field collections will be made in the summer of 2008. Red Data book status = Vulnerable. Fagopyrum The genus Fagopyrum is represented on the test plant list by two non-native species Fagopyrum esculentum and F. dibotrys. Fagopyrum dibotrys (Tall buckwheat) An ornamental non-native buckwheat species only recorded as a garden escapee in Wales. This species has long bamboo-like stems and unlike its close relative, Fagopyrum esculentum, it is a rhizomous species with vigorous growth and spread. Unlikely to be found growing near the target species. Fagopyrum esculentum (Buckwheat) Fagopyrum esculentum is an economically important crop species in many countries throughout the world, though to a lesser extent in the UK. However, the seeds can be crushed to produce flour which is free of gluten making it an alternative to wheat flour. The plant is recorded in the UK as non-native appearing erratically on waste ground, rarely persisting long at any one site.

10

Rumex and Rheum Rumex and Rheum genera belong to the tribe Rumiceae. Rumex, also known as docks and sorrels are represented in the proposed test plant list by all native species found in the UK and one species Rumex scutatus native to mainland Europe. Rumex acetosa (Common sorrel) A widespread native dock species found throughout all altitudinal ranges of the UK. Found in a multitude of habitats including, riparian, coastal and urban. This species will grow in and near stands of Japanese knotweed. Rumex acetosella (Sheep’s sorrel) A widespread native dock species present throughout all altitudinal ranges of the UK. Found in a multitude of habitats including, riparian, coastal and urban. This species will grow in and near stands of Japanese knotweed. Rumex aquaticus (Scottish Dock) Found in very isolated patches in 3 sites in Scotland. This native species of damp marshy land can hybridise with the close relative Rumex obtusifolius. There are no known records of Japanese knotweed growing in close proximity to this species. Red Data book status = Vulnerable. Rumex conglomeratus (Clustered Dock) A widespread native and short-lived native species of wet meadows and streams. This species potentially can grow alongside Japanese knotweed. Rumex crispus (Curled Dock) A widespread native dock species found throughout all altitudinal ranges of the UK. Covered by the 1959 Weeds Act. Found in a multitude of habitats including, waste ground, roadside, arable, coastal and urban. This species will grow in and near stands of Japanese knotweed. Rumex hydrolapathum (Water Dock) A tall, native, tufted perennial species found along slow moving rivers and lakes. Rumex hydrolapathum does not survive well in closed vegetation. This species can grow alongside stands of Japanese knotweed. Rumex palustris (Marsh Dock) A native species found throughout southern and central England. Growing mainly on marsh land and river banks. This species can grow alongside stands of Japanese knotweed. Rumex pulcher (Fiddle Dock) Found commonly in the south of the UK, this native species is a biennial or short lived perennial herb of dry coastal pastures and disturbed grassland. This species can potentially grow alongside of Japanese knotweed. Rumex longifolius (Northern dock) As the name suggests this native species is common in the north of the UK on open ground, roadsides, riverbanks and lake shores. This species can potentially grow alongside of Japanese knotweed.

11

Rumex maritimus (Golden dock) A native annual species growing near lakes, rivers and ditches throughout central England. It is feasible this species can grow near stands of Japanese knotweed- though records are not available. Rumex obtusifolius (Broad-leaved dock) A widespread native dock species found throughout all altitudinal ranges of the UK. Covered by the 1959 Weeds Act. Found in a multitude of habitats including, riparian, coastal and urban. This species will grow in and near stands of Japanese knotweed. Rumex rupestris (Shore dock) A perennial herb of sand and shingle beaches. This native species is found in a few locations along the Devon and Cornwall coast and in three locations in Wales. Unlikely to grow near stands of Japanese knotweed. Protected under Schedule 8 of the Wildlife and Countryside Act, 1981. Red Data book status = Endangered. Rumex sanguineus (Wood dock) A native species found in wooded margins, hedgerows, roadsides and waste ground. Found throughout the UK though mainly in lowland areas. It is feasible Wood dock will grow around stands of Japanese knotweed. Rumex crispus (Curled dock) A widespread native dock species found throughout all altitudinal ranges of the UK. Found in a multitude of habitats including, riparian, coastal and urban. This native species will grow in and near stands of Japanese knotweed. Rumex scutatus (French sorrel) A species native to mainland Europe and occasionally grown as a garden ornamental in the UK.

The economic importance of Rheum (Rhubarb) warrants the inclusion of both an ornamental species Rheum palmatum and a food plant Rheum x hybridum. Rheum x hybrid (Rhubarb) A rhizomatous non-native perennial herb found in gardens, urban areas, railway banks, rivers and stream and on waste ground. Preston et al. (2002) regard the presence of this species in the UK countryside as relics or outcasts of cultivation. Found throughout the UK and can potentially be found growing near stands of Japanese knotweed. Rheum palmatum (Rhubarb) Found in scattered isolated populations throughout central England. An introduced species occurring in the countryside as a result being an outcast of cultivation. Muehlenbeckia complexa (Wire plant) Muehlenbeckia complexa has been included in the proposed test plant list due to its close relatedness to Fallopia. There are no native species of Muehlenbeckia in the UK but there are 6 species listed on the RHS plant finder web site. We decided to test one species from this group Muehlenbeckia complexa due to it being easily obtainable.

12

Oxyria digyna (Mountain sorrel) The genus Oxyria is represented in the test plant list by one species, Oxyria digyna is the only native species in this group of two species. Found in the higher altitude areas of the UK, including Wales and the Scottish Highlands, O. digyna is a species of ungrazed mountains and streams. Unlikely to grow near stands of Japanese knotweed due to its high altitudinal distribution. Eriogonum umbellatum (Sulphur buckwheat) A non-native species grown as a garden ornamental. Other representative species within Caryophyllales Limonium bellidifolium (Matted sea lavender) A native member of the Plumbaginaceae Limonium bellidifolium has a restricted distribution in the UK occurring only on the Norfolk coast. Unlikely to grow near stands of Japanese knotweed due to its high coastal distribution. Limonium binervosum (Rock sea lavender) A group of apomictic perennial herbs comprising of nine species (sub-species). Coastal and native. Unlikely to grow near stands of Japanese knotweed due to its high coastal distribution. Cerastium glomeratum (Sticky mouse-ear) A native species found in disturbed areas, often in places where there is nutrient-enriched soil. Likely to grow in close proximately to Japanese knotweed stands. Cerastium fontanum (Common mouse-ear) A native mat-forming perennial species of grasslands. Found throughout the UK. Cerastium fontanum could potentially be found growing near stands of Japanese knotweed. Carpobrotus edulis (Kaffir Fig) A non-native ornamental species with the tendency to become invasive along coastal areas and cliffs. Highly invasive in the coastal Mediterranean and unlikely to grow near stands of Japanese knotweed. Chenopodium album (Fat hen) Found throughout the UK, this annual species grows on disturbed nutrient rich soils, cultivated field and urban areas. Chenopodium album can potentially be found growing near stands of Japanese knotweed. Beta vulgaris (Beetroot) A commonly grown garden species. As beetroot can be found in urban areas it has the potential to grow near stands of Japanese knotweed. Frankenia laevis (Sea-heath) A native coastal, low growing mat forming species of salt marshes and sand dunes. The distribution of this species has declined in the UK over the last few decades. Red Data book status= Near Threatened.

13

Tamarix gallica (French Tamarisk) A non-native tree species found in coastal areas of the UK often as part of hedging. Biochemically similar species The following two species share a common chemical with the target species, namely resveratrol. Vitis vignifera (Grape) A commonly grown garden and greenhouse species included in the test plant list due to the high concentrations of resveratrol found in the plants fruits. Rubus fruticosus (Blackberry) A native species found throughout the UK and included in the test plant list due to the high concentrations of resveratrol found in the plants fruits. The species grows within and near stands of Japanese knotweed. Morphologically similar species The following two species were included in the test plant list due to their morphological similarities to Japanese knotweed. Calystegia sepium (Hedge bindweed) A bindweed species. A sprawling native species common in hedgerows and in urban areas. Recorded as growing within stands of Japanese knotweed. Leaf morphology is similar to that of Japanese knotweed. Houttuynia cordata (Chameleon plant) It is often hard to tell the leaves of this species apart from Japanese knotweed. Non-native species grown as a garden plant.

Economically important species Malus spp. (Crab apple) A commonly-grown garden tree found throughout the UK. Malus has the potential to be grown near the target species. This was included because at the time a sawfly Amatastegia polygoni was under consideration and another sawfly Ametastegia glabrata which specialises on dock can also feed on apples. Rosa City of Leeds (Rose) A commonly-grown garden species found throughout the UK. Rose species have the potential to grow near the target species Phaseolus vulgaris (French bean) A commonly grown garden and allotment species found throughout the UK. French bean has the potential to grow near the target species. Phaseolus-vulgaris (Broad bean) A commonly grown garden and allotment species found throughout the UK. Broad bean has the potential to grow near the target species.

14

Lycopersicon esculentum (Tomato) A commonly grown garden and allotment species found throughout the UK. Tomato has the potential to grow near the target species Solanum tuberosum (Potato) A commonly grown garden and allotment species found throughout the UK. Potato has the potential to grow near the target species. Triticum aestivum (Wheat) A commonly grown arable species throughout the UK. Likely to be close to roadside knotweed. Zea mays (Corn) A commonly-grown garden, arable and allotment species found throughout the UK. Corn has the potential to grow near the target species. Hordeum vulgaris (Barley) A commonly grown arable species throughout the UK. Brassica napus (Rape seed) A commonly grown arable species of cultivated land. Found along roadsides and waste ground. Rape has the potential to grow near the target species Recent additions As a result of interactions with regulators focussing on plant pathogens, the following three plant species were added to the test plant list as a result of them being host to other Mycosphaerella spp. Lemma minor (Duckweed) An aquatic mat-forming native species present through the UK. Larix decidua (European Larch) A popular fast growing and widely planted tree species Populus tremula (Aspen) A small tree species found throughout the UK.

References

Beerling, D.J., Baliley, J.P. and Conolly A.P. (1994) Fallopia japonica (Houtt.) Ronse Decraene. Journal of Ecology. 2: 959-979

Blamey, M., Fitter, R. and Fitter, A. (2003) Wild flowers of Britain and Ireland. A & C Black Publishers Ltd. London

Cheffings, C.M & Farrell, L. (Eds.), Dines, T.D., Jones, R.A., Leach, S.J., McKean, D.R., Pearman, D.A., Preston, C.D., Rumsey, F.J., Taylor, I. (2005) The vascular plant Red Data list for Great Britain. Species status 7: 1-116. Joint Nature Conservation Committee, Peterborough

15

Mabberley, D.J. (1987) The plant book: a portable dictionary of the higher plants. Cambridge University Press, Cambridge.

Preston, C.D., Pearman, D.A. and Dines. (2002) New atlas of the British and Irish flora. Oxford University Press.

Stace, C.A., Ellis, R.G., Kent, D.H. and McCosh, D.J. (2003) Vice-County Census Catalogue of the Vascular Plants of Great Britain, the Isle of Man and the Channel Islands. Botanical Society of the British Isles, London.

16

Appendix 6 - Proposed Monitoring and Contingency Plan Introduction Whilst much of the emphasis of biocontrol programs has historically been on finding, screening, releasing and distributing control agents, it has long been recognised by biocontrol practitioners that post-release monitoring, though neglected, must form part of the biocontrol process. The collection of quantitative data is vital if the effect of biocontrol agents on target plant performance, their spread through the plant population, their safety in relation to non-target plant species and the response of associated plant communities are to be recorded and monitored. At the very least, well executed long-term monitoring programs offer exciting opportunities for ecological and applied work and will provide valuable “lessons-learnt” for future biocontrol programmes. Ideally, such a monitoring programme should be set-up and in place before any agent is released in the countryside. It is also important to evaluate any non-target impacts immediately after release, should they occur in the release area, and have a mitigation or contingency plan in place which can be brought into action at short notice. In this document a three tiered structure is proposed to monitor the release of the psyllid (Aphalara itadori) against Japanese knotweed looking at local, regional and national scale studies (see later section). Management of the monitoring programme Based on the detailed specialist knowledge about the biocontrol organism, Aphalara itadori, it is proposed that CABI will provide overall project management for the post-release monitoring programme to Prince2 standards. As the distribution of Japanese knotweed is nationwide it would be impractical for CABI to collect data throughout the UK and therefore nationwide collaboration is essential. CABI will enlist the help of impartial governmental organisation like Defra, Natural England and the Environment Agency to assist with the nationwide monitoring programme and will also include universities to conduct the research into the impacts, direct and indirect, the spread and the establishment of the BCA in the UK. Using students enlisted on MSc and PhD programmes would provide a cost-effective method for data collection and allow for detailed, high quality research on aspects of the release, e.g. modelling the performance of the control agent and community dynamics subsequent to release. Furthermore, this strategy will ensure the publication of research results in peer-reviewed scientific journals. Discussions are already underway with leading London institutions. We understand that Network Rail would be willing, in principle, to provide monitoring sites on their land. The advantage of using Network Rail property is that their land (railways) covers a linear network of sites which do not have any public access, hence minimising the confounding impacts of human interference. However, there are numerous safety restrictions that may make this impractical

1

Release site selection Release sites will be selected in consultation with the current Project Board and take into account any special conditions required by the UK regulators. Currently, the number of sites for the initial release(s) and the location of these sites is being discussed. Depending on the final discussions the monitoring programme may initially be implemented on a local scale (small number of localised releases), or on a national scale (multiple release sites throughout the country). It is likely that sites in Cornwall and South Wales will be favoured as possible release sites based on the favourable climatic conditions of these regions for establishment of the psyllid and the availability of previous GIS survey data of knotweed distribution. There are tentative plans for a monitoring programme to be established on the Isle of Man whose island status might make it an excellent monitoring site. When selecting a potential site for release consideration should be given to the following points:

• Size of site and level of target weed infestation • Location and potential disturbance • Previous management practices • Landowners permission for continued access and monitoring on long-term basis • Presence of rare or protected species

Ideally, sites would be selected throughout the country so as to enable a comparison of impact and performance of the psyllid against climatic variation and habitat type in the UK. If at all possible pre-release surveys of each release site should be conducted to determine the vegetation diversity and structure of the floral community. Population monitoring of those identified “high-risk” non-target species based on host-range studies and vulnerable species highlighted in the test plant list could also be carried out. This would provide baseline data for post-release evaluation. In addition, surveys outside the target release site to quantify the range and distribution of closely related native species (if any) can also be carried out for the same reason. Furthermore, sampling of the invertebrate community at the release site would be advantageous but it must be acknowledged that sorting and analysis of these data can be time consuming and the benefit versus cost of such an activity needs to be considered. Permanent markers and GPS reading should be established at each release site for future location by different teams of recorders. Release Fowler et al (2008) reviewed the various developments that have been made with regards to release strategies of arthropod biocontrol agents, and the many factors which can have dramatic effects on the probability of their establishment and subsequent success in controlling the target weeds. Memmott et al (2005) carried out a field experiment in which they manipulated release sizes prior to the critical first stages of the invasion of a psyllid weed biocontrol agent, Arytainilla spartiophila, against Scotch broom Cytisus scoparius, in New Zealand. They were able to follow the psyllid’s progress over 6 years for different release densities. Analysis showed that the probability of establishment was significantly and positively related to initial release size, but that this effect was important only during the psyllids' first year in the field. Releases that survived their first year had a 96% chance of surviving thereafter, as long as the sites remained secure. Colonies underwent a period of establishment or a lag-phase during the first year and population

2

size decreased initially but was followed by a period of exponential growth. Smaller populations had longer lag-phases. These factors will be taken into consideration before a release is made but it is certain that for the knotweed psyllid, the seasonal timing of release is critical in determining if it will establish in the field. Adult psyllids would be the most appropriate life stage of the species to release into the UK and to maximise establishment potential given the day-degree requirement of the agent, the release adults as early in the season as possible and certainly before mid-June would be advisable. We anticipate using mesh sleeves to restrict the adults on to individual plants or branches to encourage localised oviposition and provide a protected environment. The Monitoring Programme The monitoring programme will follow the tiered approach as often practiced in releases in other parts of the world. Level I will be implemented on a local scale around the initial release sites and levels II and III will be implemented on a national/regional scale. Level I will include the post-release evaluation of the BCA’s performance- to include abundance of the BCA, life-cycle studies in the field, impact of the BCA on the target weed and monitoring of direct and indirect non-target effects. Level II will measure the BCA and target-weed population dynamics and will look for any non-target effects on a regional and national scale. Level III will monitor the levels of establishment if the BCA and its spread on a national scale. Appropriate spatial, temporal and seasonal scales will be incorporated and techniques for sampling and monitoring will be standardised. Consideration could be given to the development and testing of predictive models of impact which would be of value to practitioners and regulators for future biocontrol programmes. The ultimate goal of the biological control programme against Japanese knotweed is to reduce the impact of the target plant and restore the invaded ecosystem. As with most BCA introductions, a significant impact on the target is not expected in the short term and an impact may not be seen until many years after the agent has been introduced. With this in mind CABI recommends that any monitoring programme should have a duration of ten years, though the intensity of the monitoring will reduce over time. For example, in the first year of release, based on a spring release, safety monitoring (non-target impacts) will be paramount and the impact monitoring of the BCA on the target may be reduced as a significant impact would not be expected at such an early stage in the monitoring programme. Level 1- detail Measuring the impact of the BCA on the target species Key outcomes:

• Monitor any potential non-target effects and if any are detected, implement contingency plan. Sample to determine the effect, if any, on plant community dynamics relative to baseline records (see below for non-target monitoring)

• Measure the impact of the BCA on the target species and establish quantifiable easily-obtained parameters to measure impact for use in Level II

• Monitor the performance (number of generations and fecundity, any potential parasitism of the agent in a natural situation

3

One outcome of Level I will be to establish a robust and standardised measure of the impact of the psyllid on the target species to allow scientific evaluation. Establishing the impact of the BCA on a target weed is often time consuming and technically challenging. However, once key relationships have been identified (plant parameter relationships/ in the absence and presence of the BCA) these can be used to simplify monitoring on a regional scale. As Japanese knotweed does not reproduce by seed in the UK, this can be ignored though seed production is often a useful indication of plant performance. Plant performance relationships could include:

• Ratio of leaf area to height • Stem girth to height • Leaf surface area/ rhizome production per unit area

When factoring in the BCA key measurements the following could be added:

• Number of damaged plants per unit area • Number of eggs/nymphs per plant • Presence/absence of wax • Leaf area and leaf count • Height of plant

To establish the impact of the psyllid on the knotweed populations, similar sites with and without the psyllid are needed. It may not be suitable to compare sites with psyllids present to sites outside the current distribution of the psyllid as these may differ in a number of environmental parameters (for example soil nutrients, rainfall, climate and grazing) which may affect the performance of Japanese knotweed. The preferred method of testing the impact of a BCA is agent exclusion – using either mesh cages or pesticides. Pesticides offer the most effective method, applied to discrete areas, close to the release site, since this reduces spatial variation and assures that no psyllids are present in the sample site. Photopoints are a relatively simple method to demonstrate the impact of a biocontrol agent on the target plant. Photos will be taken at regular intervals at each release site. During the monitoring of the initial release sites in Level I, the spread up to a 100m radius from the release site will be monitored. The monitoring of impact should also be closely linked to the consequential effect this will have on the floral community as a whole. Since we anticipate that the BCA will reduce the competitive advantage of the knotweed, it will be important to monitor the changes in floral species composition at the site. Monitoring of non-target effects Non-target monitoring efforts will be directed at the most closely related native members of the Polygonaceae, as well as at Russian vine, Fallopia baldschuanica, that exists close to release sites, or knotweed stands that are subsequently infested with the control agent. Species such as Fallopia dumetorum which are classed as “vulnerable” by JNCC’s red data book criteria, and have fewer than 1000 individuals recorded from five or fewer locations, are not considered to be at risk since the sites where they occur are likely to be under constant vigilance, well-monitored

4

and free of Japanese knotweed. However, land managers and recorders in such areas will be contacted and asked to report any suspicious damage as soon as possible. One option which could be put in place to detect any non-target effects would be to plant out the most important plant species of concern at the release site(s) as “trap plants” so as to provide an early warning system of any non-target impacts. Thus, F. dumetorum (if available) will be planted out at the initial release sites and closely monitored for psyllid eggs and developing nymphs. If unanticipated negative impacts are observed, immediate action (see below) could then be taken to eradicate the insect and prevent further spread. Whilst uninvaded reference sites and/or long-term documentation of plant and arthropod communities would provide useful benchmarks pre-release, such biological inventories are rarely available. Consequently, any assessment of indirect impacts on food webs and communities will be extremely difficult. Even where some baseline data are available and communities have been studied extensively and manipulated to test hypotheses, the sheer complexity of natural ecosystems and the biotic and abiotic factors interacting in the community dynamics make it very difficult to draw definite conclusions. Invaded ecosystems may have lost some degree of complexity, but it is important that any impacts attributed to the psyllid can be reliably distinguished from natural oscillations and plant succession and that any lag-effects are taken into account. Over-wintering Repeated surveys for hibernating psyllids will be made on evergreen trees near to release sites between the time of annual dieback of knotweed and the recolonisation of knotweed in the following spring. Surveys will also be made on leaf litter and dead knotweed canes. Monitoring the agent performance Measures of the agent’s performance including rate of development and population increase would be recorded at multiple sites. Predation and parasitism will also be monitored visually Level II (regional/national scale) Key outcomes:

• Measure the impact of the BCA on the target species • Monitor any potential non-target effects on selected members of the Polygonaceae

As the psyllid becomes established in the UK countryside, natural dispersal will occur from the initial release site(s) to other areas where Japanese knotweed is present. As a result the monitoring programme will be broadened to a regional and eventually a national scale so as to monitor the spread of the psyllid along with its impact. The monitoring of non-target effects will be the same as set out in Level I non-target monitoring, although it will not involve transplanted plants. Prior to release CABI will compile factsheets which will include information on the priority plant species (most closely related) for safety monitoring. The factsheets will also contain photographic information on the type of damage associated with psyllid feeding on Japanese knotweed. Public awareness raising efforts linking with the various knotweed groups, nature groups and schools could further improve the volume of information available and this could also be coordinated through the project website.

5

Level III (regional/national scale) Key outcomes:

• Monitor the establishment and spread of the BCA It is important to monitor the establishment and spread of the BCA as it moves from the initial release sites throughout the UK. Thus a network of monitoring sites will be established aimed at detecting the first arrival and establishment of the psyllid. See Table 1 for a proposed timetable, assuming a spring release Funding Securing funding for a long-term monitoring programme when there is no certainty that release will take place is unlikely and the associated Life+ project was unsuccessful. Fortunately, discussions with the Japanese knotweed Project Board are at a late stage and a firm commitment for around 50% of the estimated cost has been made. This significant contribution from one sponsor’s budget could be used as match funding towards the required total making funding easier to come by. It is expected that CABI would collaborate with UK universities to seek further funds from government organisations, like NERC, to fund PhD studentships within the monitoring programme. The scope of research is vast and should, in theory, be of interest to the research councils and wider scientific community. Other sources of funding for this work could come from the beneficiaries of the biocontrol program who might be keen to monitor post-release impacts on the target plant where it poses a problem for them and where additional monitoring of non-target species could be done concurrently and in a cost-effective way. Contingency Should unacceptable, unintended effects occur at the initial release sites, then appropriate control and mitigation responses are required. In order for eradication of the psyllid to be successful, there needs to be:

• Early detection and reporting of any non-target impacts • A rapid assessment of the significance and implications of the observed effects • An immediate response to contain and eradicate the psyllid

If it is decided that the psyllid poses an unacceptable risk to native species or beneficial and economic plants post-release, it is essential that a pre-ordained rapid response eradication plan is implemented before the psyllid has entered its expansive phase. This would involve the application of an appropriate chemical insecticide, followed up by repeated monitoring and spot treatments as required. This should minimise the likelihood of establishment in the early stages. A shop-bought systemic insecticide (Provado ultimate bug killer (Bayer), aerosol, 0.25g/l imidacloprid & 0.5g/l methiocarb) was found to be 100% effective in the laboratory on a mixed population of psyllids on knotweed, though replication was low and follow-up monitoring of any eggs. A brief study using a Pemethrin-based aerosol contact insecticide only gave 87% mortality and the surviving psyllids (late nymph instars) were able to continue feeding on the plant. Insecticides such as deltamethrin and bifenthrin are also likely to be effective against the psyllid, and are registered for use on a wide range of crops (particularly deltamethrin). However, there may be problems in applying such products in certain non-crop situations, or sites, and this

6

should be investigated further (in consultation with PSD). Specific off-label approvals (SOLAs) may be necessary and will be subject to case-by-case risk assessments (i.e. of the potential impacts of the pesticide) and this should be in place before release. Pesticide products approved for non-crop areas are mainly limited to herbicides for weed control, although applications are likely to be prohibited in some situations (SSSIs). Natural England should also be consulted with regard to this issue. The application of pesticides may also be restricted or prohibited in certain riparian situations (buffer zones exist around watercourses), and the Environment Agency should be consulted. Low risk release option An additional option would be for a trial release of mated adults in the autumn (early September) prior to a full release in the following year. With an autumn release, adults would still be capable of laying eggs though the emerging nymphs would not be able to develop into adults due to the lack of remaining degree days in the season. The released adults would not be likely to survive long enough to overwinter. If, as suggested above, important plant species were planted out within stands of Japanese knotweed- any non-target effects could be recorded quickly and there would be little risk of the psyllid population establishing. In theory, this approach should allows biocontrol to proceed almost on an experimental basis as a temporary and reversible release before full-scale implementation. A similar precautionary approach has been reported by Cuda et al (2008) where temporary and reversible releases of two biocontrol agents were proposed through single sex releases of the defoliating sawfly and artificial sterilization of a leafroller. References Cuda, J.P., Moeri, O.E., Overholt, W.A., Manrique, V., Bloem, S., Carpenter, J.E., Medal, J.C.,

Pedrosa-Macedo, J.H. 2008. Novel approaches for risk assessment: Feasibility studies on temporary reversible releases of biocontrol agents In: Proceedings of the XII International Symposium on Biological Control of Weeds, April 22-27, 2007, La Grande Motte, France, p. 102.

Fowler, S.V. Harman, H.M. Memmott, J. Peterson P.G. and Smith L. 2008 Release strategies in weed biocontrol: how well are we doing and is there room for improvement? In: Proceedings of the XII International Symposium on Biological Control of Weeds, April 22-27, 2007, La Grande Motte, France, p. 495-502.

Memmott, J., Craze, P.G., Harman, H.M., Syrett, P., Fowler, S.V., 2005. The effect of propagule size on the invasion of an alien insect. J. Anim. Ecol. 74, 50–62.

7

Table 1 - Monitoring timetable

Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year 10 Q1 2 3 4 1 2 3 4 1 2 3 4 4 3 44 1 2 3 1 2 3 4 1 2 3 1 2 3 4 1 2 1 2 3 4 1 2 3 4 Initial Release Further releases (if needed)

Level I monitoring (Local release sites) Non-target impacts Agent performance monitoring Impact of BCA monitoring Overwintering monitoring Level II Monitoring (nation-wide) Non-target impacts Agent performance monitoring Impact of BCA monitoring Level III Monitoring the establishment and spread

9

application for licence to release a non-native species ... · 3 see rebeca work package 5 –...

Documents