IN THE HIGH COURT OF NEW ZEALAND                            M.22-PL/03

AUCKLAND REGISTRY

 

 

IN THE MATTER of the Judicature Amendment Act 1972

 

AND IN THE MATTER of the Hazardous Substances and New Organisms Act 1996

 

 

BETWEEN

 

MOTHERS AGAINST GENETIC ENGINEERING INCORPORATED

 

                   Applicant

 

 

AND

 

THE MINISTER FOR THE ENVIRONMENT

 

                   First Respondent

 

 

AND

 

THE ENVIRONMENTAL RISK MANAGEMENT AUTHORITY

 

                   Second Respondent

 

 

AND

 

AgRESEARCH LIMITED

 

                    Third Respondent

 

 

______________________________________________________________________________

 

AFFIDAVIT OF JUDITH ANNE CARMAN ON BEHALF OF THE APPLICANT

Dated:         April 2003

______________________________________________________________________________

 

 

Solicitor Acting:

J P Ferguson

Walters Williams & Co

Level 7 Axon House

1 Willeston Street

P O Box 1654

Wellington

Telephone:   (04) 495 9999

Facsimile:     (04) 495 9990

 

Barrister Acting:

Peter Andrew

Level 3, The Annex, 41 Shortland Street

P O Box 31

Auckland

Telephone:  (09) 300 7215

Facsimile:    (09) 300 7219

 

I, Judith Anne Carman, swear as follows:

 

1.                   I am an Affiliate Senior Lecturer in the Department of Public Health at the University of Adelaide, South Australia.  I am the spokesperson on GM foods for the Public Health Association of Australia.  I am a Director of the Institute of Health and Environmental Research Inc. 

 

2.                   In 1989 I graduated with a PhD dDegree in Medicine (University of Adelaide) in the field of Metabolic Regulation/Nutritional Biochemistry, particularly in relation to cancer.  In 1998 I obtained a Master of Public Health from the University of Sydney specialising in Epidemiology and Biostatistics.  I have worked in a number of positions with those relevant to this affidavit being: I have taught at Roseworthy Agricultural College (now part of the University of Adelaide); I worked at the CSIRO (Commonwealth Scientific and Industrial Research Organisation) Division of Human Nutrition (this is now the Division of Health Sciences and Nutrition); worked in HIV/AIDS in Sydney and was the Senior Epidemiologist in the Communicable Disease Control Branch of the South Australian Department of Human Services investigating outbreaks of disease.  While in this position, I was the chief investigator into the multi-institutional investigation in Australia into whether Rabbit Calicivirus (Viral Haemorrhagic Disease of Rabbits) was able to infect people. 

 

3.                   I am making this affidavit on behalf of myself and my fellow Directors of the Institute of Health and Environmental Research Inc.  This Institute is a private institution for health and environmental research, whose principal current focus is on GE foods.  My fellow Directors of the Instituteinclude are Dr Catherine Anne Clinch-Jones and Dr Phillip Andrew Davies.  Dr Clinch-Jones is a Medical Practitioner who obtained a medical degree (BMBS) from Flinders University in 1988 and has practised as a General Practitioner for 12 years.  Dr Davies obtained an Honours Degree in Agricultural Science in 1984 from the University of Adelaide, and a PhD Ddegree in genetics in 1989 from the Australian National University, Canberra.  He is a plant geneticist, specialising in the application of biotechnology to plant breeding.

 

4.                   We have read the decision of the second respondent, the Environmental Risk Management Authority (“ERMA”), Application No. GMD02028, the subject of these proceedings.  We have also read the affidavits of Dr Neil Macgregor, Professor Adrian Gibbs and Associate Professor Peter Wills.

 

5.                   This affidavit addresses some serious concerns we have about the adequacy of ERMA’s risk assessment in GMD02028 decision, and the controls it has imposed on AgResearch, the third respondent.  In particular, we contend that the possible adverse effects of horizontal gene transfer (HGT) have not been adequately assessed or quantified and that both the AgResearch application and the description of the approved organism (as contained at Annex 1 of the ERMA decision) are too broad and general to adequately and responsibly determine the level of risk.

 

6.                   We also have very serious misgivings about the containment capability of the field site and control 1.3 imposed by ERMA (page 54 – Annex 2 – of the ERMA decision).  We conclude by recommending, amongst other things, that transgenic animal containment should be a purpose-built building in which nothing of potential transgenic origin is to leave, and that incarceration of all unwanted material should become mandatory.

 

ERMA’s GENERAL APPROACH

 

7.                   At page 12 of its decision ERMA complains about the lack of information contained in the AgResearch application by stating “a bare minimum level of information had been provided in some respects”.  ERMA also notes that the broadness and lack of specificity of the organism description, and the fact that the organisms were to be outside rather than in a contained laboratory, made assessment of all possible adverse effects on the environment very difficult.  (Page 12).  However, despite registering these concerns, ERMA then decided that they had sufficient information, expertise and knowledge, not only to actually to assess the risks but to conclude that the benefits outweighed the risks and the costs.  (Page 2).

 

8.                   Our understanding is that an applicant for approval to develop a genetically modified organism (in this case, of course, AgResearch) is required to provide information on all the possible adverse effects of the organism on the environment. 

 

9.                   From a scientific perspective, we cannot understand how ERMA can have concluded that AgResearch had met this burden in circumstances where it was complaining that a bare minimum level of information had been provided and where the broadness and lack of specificity of the organism description essentially meant that an assessment of all possible adverse effects was not really achievable.  In our view, this application required a very clear understanding of the risks arising from HGT and that ERMA’s consideration of the possible risks arising from this issue is seriously wanting.  We believe that ERMA has not properly informed itself on this issue.  We believe it was incumbent on the applicant, AgResearch, to have provided full information on this important issue in the first instance.

 

RISK ASSESSMENT

 

10.               In our view, the difficulties associated with the breadth and lack of specificity of the organism description has meant that many of the risks identified by ERMA have been determined by conjecture and assumption, and that in the process, ERMA has under-estimated the risks.  We contend that many of these risks are in fact currently unquantifiable and that ERMA has erred in believing that it can quantify the unquantifiable.  ERMA has taken many of these unquantifiable risk forecasts and tied them into one cumulative measure that the risks are low.  If ERMA is wrong on any one of these many risks, the cumulative effect on humans, animals and the environment could be very significant.  This is particularly important as the application by AgResearch deals with self-replicating genetic material, which when incorporated into living things (animal, plant or microbioligical) may persist for thousands of years.


 

(a)               Decomposition and disposal

 

11.               ERMA contends that the decomposition of carcasses on site in unlined offal pits will ensure “that any genetically modified animals are likely to remain on-site until they decompose” and that the disposal of GM cattle in unlined offal pits does not pose any significant additional likelihood of adverse effects attributable to horizontal gene transfer” (page 17 of the Decision).  In our view this is hard to justify given that decomposition will actually be effected by billions of micro-organisms.  This can only increase the possibility of horizontal gene transfer. Furthermore, some leaching of decomposing body fluids is inevitable from an unlined pit, even with run-off to ground water apparently minimised.  We also note that ERMA did not specify any design requirements for these pits in order to minimise run- off to ground water but has left the design to AgResearch.  Cremation would be a safer method of disposal.  We believe that if a safe method of disposal cannot be permitted due to cultural sensitivity, then the entire project needs to be reconsidered.

 

12.               ERMA believes that conventional cattle that fail to implant a GM embryo carry so little risk that they may be disposed of off-site.  This may also include being sold (page 18 of the Decision), presumably for breeding, milking and consumption purposes.  AgResearch has made it clear that being unable to sell these animals would affect their profits.  We are concerned about the implication that profits may have been placed before safety.   We also fail to see how ERMA can maintain this position in the face of known scientific facts.  That is, for many years it has been accepted medical practice to administer anti-D immunoglobulin to Rhesus negative women (to prevent their development of anti-D antibodies, which may affect subsequent pregnancies) who bleed or miscarry even very early in pregnancy, on the grounds that any bleeding or pregnancy loss exposes women to antigens (proteins) from the foetus.  This is clear evidence that proteins from the embryo/foetus can cross into the tissues of the parent animal and it is therefore important that cattle that reject or fail to implant a GM embryo be disposed-of on site and not be permitted to be sold.  Additionally, consideration should be given to whether failure to implant the embryo may result from a transmissible or toxic effect of the transformation.  That is, some of the genetic elements such as the SV40 promoter in combination with other genetic elements, may have killed the embryos, and if this happens, it is possible that this toxic element could be transmitted to the cow and hence to humans consuming products from that cow.

 

13.               ERMA contends that disposal of milk via treatment and spraying onto pasture cannot be a route for the escape of heritable material, principally because autoclaving, while it does not completely destroy DNA, is likely to break it down into small random fragments and the risk from horizontal gene transfer of such fragments is considered to be “negligible” (page 19 of the Decision).  However, autoclaving also does not destroy prions and it appears that small naked DNA fragments may be potentially more available to horizontal gene transfer than intact DNA.  In addition, it has long been known that sprayed substances (e.g. herbicides) can drift for kilometres.  Furthermore, the milk that remains on site will come into contact with micro-organisms in the soil and faeces on site and can move off-site through wind (after desiccation), surface water and ground water. ERMA also suggested that the digestion of the milk in an effluent digester was reasonable and that it may be done off-site.  However, this would require the transportation of milk and this would pose risks in the event of a spill or vehicle accident.  In addition, the milk carrying compartment(s) in the transportation vehicle could be a site of movement of GM DNA out of containment.  That is, there is a history that milk compartments that were not sufficiently rigorously cleaned from one load to the next contaminated subsequent loads of milk with salmonella, creating an outbreak of disease.  As GM DNA is expected to be present in milk (at least in the white cells which milk normally contains but also in other cells in milk such as adipocytes or red blood cells, or in extracellular form), and the above example shows how hard it is to sufficiently remove similar material through cleaning, milk transport containers should be used only for the transportation of milk from these cows.  For similar reasons, the water and detergent used for cleaning would all need to be disposed-of on-site. Furthermore, it is our understanding that an effluent digester would bring huge numbers of micro-organisms into contact with GM DNA present in the milk, thereby posing the same problems as an offal pit.  This is likely to increase the probability of horizontal gene transfer.

 

14.               ERMA has determined that effluent from cattle could be collected and sprayed onto pasture, and faeces and urine from grazing animals would be left to decompose in the field, on the basis that “the likelihood of HGT and unintended production of toxins to both [is] very low” (page 20 of the Decision).  However, it has long been known that sprayed substances (e.g. herbicides) can drift for kilometres and that faecal material on pastures can pollute both above-ground and below-ground water systems.  Outbreaks of disease have occurred in this manner.  In addition, it is also well known that faecal material can desiccate and be blown off-site resulting in the microbes infecting people.  Outbreaks of Q-fever in people have occurred in this manner.  This is of particular concern as horizontal gene transfer involving transgenic DNA has recently been demonstrated to occur in the human gut.  That is, HGT in the mammalian gut is a real phenomenon, not a hypothetical one.  For these reasons, staff working with these animals should also be prevented from coming into contact with these animals, and their products.  This is because outbreaks of disease show that people are very good at transmitting disease through the oral-anal route.  This includes the transmission of microbes from animals to humans via human mouths, and the subsequent spread from human to human.

 

15.               ERMA has determined that the likelihood of HGT could be substantially reduced by controls on the nature of the DNA to be used.  In particular, without the use of “inherently mobile genetic elements (such as (transposable elements) and viruses) (page 24 of the Decision), the escape of infectious genetic elements is very unlikely to occur” (page 24 of the Decision).  This assumes that we know enough about HGT to be able to assess these risks.  However, this area is in its infancy and accurate assessments cannot be made based on so little knowledge.  It has recently been determined, for example, that GM DNA can transfer into bacteria in the gut of people after just one meal containing GM soy.  This was an unexpected finding.

 

16.               ERMA states that without a selective advantage occurring, HGT is unlikely to lead to the establishment of a new trait in the population.  This is not necessarily true: organisms carry many genes which are not required in most circumstances, but can become active in times of unforeseen stress or different environmental conditions. ERMA also states (page 25) that “given the broad scope of the organism description in the application it is possible that some sequences to be introduced could provide selective advantages”.  In spite of this, ERMA still assesses the HGT risk to be low, even with the uncertainty involved (page 25).

 

17.               ERMA considered that “the risk of a new prion disease forming was negligible because genes associated with prion-diseases are excluded from the approval” and “pathogenic proteins that develop from mis-folding are generally well-known” (page 28 of the Decision).  Such apparent certainty ignores the fact that prion diseases (and their pathogenic proteins) are a recent discovery and our knowledge about them is in its infancy.  In fact many prion diseases may not even have been recognised at this time.  Furthermore, it is recognised that we all contain prion-like proteins in our bodies.  In order to be activated they only need to have their structure folded in a different way.  If this ERMA-sanctioned research were to extend beyond simple protein expression in milk, to altering the physical structure of the proteins by folding them differently, for example to ensure they would be active in a human subject, then the chance of causing prion diseases would likely increase, even if no prion proteins are used in the transformation.

 

18.               ERMA’s view is that it is unlikely that the proposed experiments would result in any unknown toxins developing, and “if any unknown mammalian toxins were produced the effects would be manifested in cattle first, and so very unlikely to pose a human health risk” (page 28 of the Decision).  This assumes that the toxins would not be preferentially expressed in milk.  It also runs counter to common experience that some toxins have very slow mechanisms of action and may take many years before showing their effects, for example, the substances in tobacco smoke.  In addition, if there is a high turnover of cattle and few are kept into middle or old age, some toxic effects are likely to be missed.  Furthermore, thousands of genetic permutations are allowed under this application.  Simple mathematics informs us that a small risk, multiplied by several thousand, can become a significant risk.  For example, if the overall risk is low at 1 in 10,000, and 1,000 experiments are undertaken, then the overall risk becomes 1 in 10, which is a significant risk.

 

(b)               Viral Cell Receptors

 

19.               ERMA believes that their requirement that all introduced sequences must be characterised reduces the likelihood of viral cell receptors being introduced and that well-characterised sequences, particularly those derived from a common bacterium will reduce uncertainty associated with potential consequences of HGT.  However, knowing the sequence of these elements is not the same as knowing the effect of such sequences on mammalian cells (including cattle cells), thousands of different types of bacteria and the environment.  Such a restriction is intended to reduce the risk of HGT.  That is, ERMA recognises the risk of HGT, and presumably viral recombination.  However, without objective scientific evidence, it has been assumed that restricting sequences will control this risk.  Given the similarity between many genes in different species, this is unlikely to be the case.  In addition, excluding known human and animal viral cell receptors indicates that ERMA recognises the risks of development of interspecies and new viral pathogens being facilitated by the GM process.  However, current scientific knowledge has not identified many, and perhaps most, viral receptors.  New viruses may be able to use new proteins as receptors.  Therefore, this control is unlikely to have a significant effect on the risk of developing new viral pathogens.  Furthermore, requiring characterisation of genetic sequences so that only elements of known function are inserted similarly relies on the current level of scientific knowledge.  Given the scope of potential genes and elements permitted by this application, and the current deficiencies in understanding of gene function, this is likely to be an ineffective control.  In addition, knowing the sequence of these elements is not the same as knowing the effect of such sequences on mammalian cells (including cattle cells), thousands of different types of bacteria and the environment.

 

20.               Under “viral cell receptors” (pages 23- 24 of the Decision), ERMA recognises that incorporation of viral cell receptors from other species has the potential to create a new disease in cattle, or a new reservoir for an existing virus.  It is also aware of the potential to include genes for currently unknown viral receptors, and that biting insects could move that viruses off-site.  Then there is an assumption  They then assume that other cattle would not be affected, as they would not contain the necessary receptors.  Unfortunately, this may not be the case.  There is widespread awareness that inter-species transfer of viruses can result in new disease outbreaks, and that viruses can mutate in animal hosts to broaden their range of hosts.  In addition, animals from the specified gene donor species (cattle, mice, deer, goats, sheep and humans) are all endemic to New Zealand, and could be susceptible to a mutated virus that could bind to their receptors. 

 

1.ERMA states that without a selective advantage occurring, HGT is unlikely to lead to the establishment of a new trait in the population.  This is not necessarily true: organisms carry many genes which are not required in most circumstances, but can become active in times of unforeseen stress or different environmental conditions. ERMA also states (page 25) that “given the broad scope of the organism description in the application it is possible that some sequences to be introduced could provide selective advantage”.  In spite of this, ERMA still assesses the HGT risk to be low, even with the uncertainty involved (page 25).

(c)       Promoters

 

 

21.               ERMA states that the SV40 “virus as a whole rather than this [SV40] promoter can cause pathogenic effects” and “the promoter itself has no inherent ability for independent movement or infection” (page 26 of the Decision).  However, viral fragments have been shown to be able to recombine with other viruses in the host or the environment, including inactivated viruses, regaining their pathogenesis, and potentially changing their host range. It is also important to note that simian virus 40 already infects monkeys, our close genetic relatives.

 

(d)      Antibiotic resistance

 

22.               ERMA believes that risk will be reduced by only using antibiotic resistance marker genes that are not of clinical significance in human medicine (page 51 of the Decision).  However, in our view, the specific antibiotics that can be used should have been of no clinical significance to human or veterinary medicine and should have been determined by medical and veterinary practitioners and then listed by ERMA and not left to the discretion of AgResearch.  Moreover, two specific antibiotics are mentioned, in the context of already having resistance associated with them: kanamycin and ampicillin.  The way in which they are discussed suggests that these have been identified as suitable antibiotics to use.  In fact, there would be implications for human medicine if either of these were used.  Kanamycin resistance is, as stated, conferred by the gene nptII.  This gene also confers resistance to neomycin, which has occasional use as a topical antibiotic, but more importantly, to gentamicin.  Gentamicin is used as an intra-venous antibiotic to fight serious gram-negative infections in hospitalised patients with septicaemia, burns and kidney infections, for example.  Ampicillin is closely related to amoxycillin, and ampicillin resistance would most likely also confer amoxicillin resistance.  Amoxycillin is increasingly being used in human medical practice, in an effort to restrict exposure to wider spectrum antibiotics and minimise the development of resistance to newer antibiotics.  Clearly neither of these antibiotics should be jeopardised by the use of marker genes. 

 

23.               In addition, ERMA admitted that “there is little information available on the occurrence of antibiotic resistance in New Zealand” and that “mechanisms for resistance have not been studied” in soils, thereby admitting to the scientific ignorance about antibiotic resistance (page 27 of the Decision).  Yet on the basis of this ignorance, ERMA still concluded that the risk of increased levels of antibiotic resistance associated with this application was “negligible” (page 27 of the Decision).  However, in our view, ERMA needs to understand that there is some evidence that once a bacterium has acquired one gene for antibiotic resistance it is more likely to be able to acquire others.  The significance of this finding is that even using a gene for resistance to an obscure antibiotic, may ultimately increase resistance to clinically significant antibiotics.  ERMA also seems to be of the view that as antibiotic resistance already occurs, there is less reason to be concerned about it. This runs counter to recent moves to limit practices that promote antibiotic resistance in medicine and agriculture in order to prevent further antibiotic resistance and thereby prevent a looming crisis.

 

FIELD TRIAL –v- DEVELOPMENT

 

24.               We understand that ERMA has determined that all of the six steps proposed for the AgResearch work constitute a development (at page 9 of the dDecision).  In order to comply with the legal development of “development”, it appears that the AgResearch project is to be limited to “preclude statistically valid research on the effects of the organism”.  (Page 10 of the decision).  This appears to me us to mean that in an effort to prove that this is application is not a field test, any assessment of environmental impact is forbidden, except for monitoring the HGT.  This is of particular concern.  From a scientific perspective, we contend that this application is in fact a field trial because the GE cattle will be released into a field to graze like normal cattle, rather than be kept in secure laboratory facilities.  Therefore, in our view, it is clear that, as in a field trial, one of the tests being undertaken, whether explicitly stated or not, is whether the GE cattle will be able to survive, grow and be viable in the open, away from the intensive animal husbandry of the laboratory.  In addition, the cattle and their wastes will be exposed to the air, soil and water systems, and the creatures which inhabit them, such as micro-organisms, birds, rabbits, mice, rats, mosquitoes, ticks, flies, insects which feed on manure and detritus, earthworms, unintended predators, etc.  It should be remembered that in Australia, Rabbit Calicivirus (Viral Haemorrhagic Disease of Rabbits) was spread for hundreds of kilometres at a time by the action of mosquitoes and flies.  Each and all of these creatures must be considered in a risk assessment, and this application should be considered as a field trial.

 

MONITORING

 

25.               Even though ERMA believed that “monitoring should be as extensive as possible”, it approved the application without thorough independent monitoring or auditing of the processes and outputs.  For example, there appear to be no procedures or capability to monitor insertion sites or the genetic material.  There were also no controls requiring monitoring of insertion sites (page 21).  Therefore, there is no monitoring of insertion sites, the genetic insert to determine if it stays in place (or moves) or of movement from unlined offal pits.  There is also a need for monitoring to determine if any cattle have received any unintended gene sequences (e.g. those from EBV), on the basis that there are examples of unintended gene sequences ending up in transgenic plants.  There are also no studies on HGT in cattle intestinal bacteria and parasites (page 21) or soil biota.  ERMA stated that these controls were “impractical” (page 21).  We contend that they are not impracticable impractical.  Rather, they are necessary controls that simply require a little time and effort to undertake.  If they are deemed to be unreasonable due to cost or practicality, we consider that it is unreasonable to approve the application.  In addition, monitoring for HGT at the disposal sites will be done by AgResearch.  Self monitoring has rarely resulted in full, accurate and timely disclosure by groups with a vested interest in being quiet.  All monitoring needs to be done by independent experts on a monthly basis.  ERMA requires that if HGT is detected, the project will be halted and a remediation plan developed.  However, a remediation plan should be created before the experiments begin, covering all possible adverse eventualities.  Such a plan would also highlight any impracticalities in remediation which may in turn limit the project.  It would also prevent a lag time of possibly months between finding a problem and determining a remediation plan, precious time during which GM micro-organisms could be growing exponentially and the problem could become seriously worse irretrievably worsen.

 

26.               ERMA has given a seven and a half year approval to this broad application.  A year-by-year re-approval process should have been instigated, after monthly monitoring and a thorough yearly inspection and audit, so that if AgResearch was found to not be following controls, the operation could be shut-down.

 

27.               When  considering the “establishment of an undesirable self-sustaining population and ease of eradication” (page23 22 of the Decision), ERMA limits its context to feral cattle populations only.  This is a serious oversight, as management strategies for dealing with micro-organisms having undergone HGT should also be is also mandatory.  Furthermore, medical experience, such as hospitals trying to eradicate multi-resistant staphylococcus aureus should be drawn upon.  Such experience shows that even closed indoor facilities have problems in managing microbial populations, and hence it is ill advised to even consider managing GM animals in open environments, such as fields.

 

CONCLUSION

 

28.               On the basis of the above, we believe that AgResearch’s application should be reconsidered, and that at this time, relevant medical, scientific and public health information and advice should be used to determine the outcome of the application.  If the application goes ahead, we recommend that more substantial containment be undertaken in order to reduce the risks.  This has the advantages of permitting scientific research whilst substantially reducing the risks of exposure of GM DNA and organisms to humans and the environment. In particular, we recommend:

 

(a)               A secure containment building(s) be built to hold these animals and their products.

 

(b)               Once in the facility, all animals and their products stay in and are not permitted to leave.  This includes all cows, bulls, semen, ova, embryos, foetuses, faeces, urine, skin, meat, milk etc.

 

(c)               All dead animals and their products listed above, when no longer required are to be incinerated at temperatures high enough to eliminate all prions and DNA.  As a result of their recent experiences with Mad Cow Disease, the UK has procedures for this.  This incinerator would be built on-site to prevent the escape of any DNA or prions from the site.

 

(d)              All humans when in contact with these animals and their products wear gowns, gloves and masks and are not permitted to eat or drink.

 

(e)               All windows have fine mesh screens on them to prevent the entry of flying insects such as flies, mosquitoes, and midges.

 

(f)                Measures be taken to prevent the entry and breeding of mice, rats, cockroaches, etc, and these measures to include the laying of poisons to eradicate them from the site.

 

(g)               Ideally, no antibody resistant marker genes be used.  If this is not possible, then only those coding for antibiotics defined to be of no clinical importance by a panel of medical and veterinary clinicians would be used.

 

(h)              Monitoring of insertion sites, genetic material, unintended gene sequences etc, as well as processes and procedures by a team of independent experts on a monthly basis and heavy penalties be imposed for breaches.  The licence would be reviewed every year, and any breaches result in the removal of the licence.

 

(i)                 A remediation plan would be developed by a team of independent experts before the facility is permitted to begin operation.

 

29.               I have read the Code of Conduct for Expert Witnesses contained in Schedule 4 of the High Court Rules and agree to comply with the Code.

 

 

SWORN at Adelaide                                   )

this          day of April 2003                         )

 before me:                                                    )                      ______________________________

                                                                                              Judith Anne Carman

 

 

 

____________________________________

A Solicitor of the High Court of New Zealand/Notary Public