LUNDY v R [2018] NZCA 410

LUNDY v R [2018] NZCA 410 [9 October 2018]IN THE COURT OF APPEAL OF NEW ZEALANDI TE KŌTI PĪRA O AOTEAROACA232/2015[2018] NZCA 410BETWEEN MARK EDWARD LUNDYAppellantAND THE QUEENRespondentHearing: 17, 18 and 19 October 2017 (further submissions received27 October, 6 November, 10 November and 11 December 2017)Court: Cooper, Winkelmann and Asher JJCounsel: J H M Eaton QC, J Oliver-Hood, J-A Kincade and H C Coutts forAppellantP J Morgan QC, B D Vanderkolk and M L Jepsonfor RespondentJudgment: 9 October 2018 at 11.30 amJUDGMENT OF THE COURTA The appeal is dismissed.B The applications to adduce further evidence for the purposes of the appealare granted or declined in accordance with the attached schedule.____________________________________________________________________REASONS OF THE COURT(Given by Cooper J)Table of ContentsPara NoIntroduction [1]The murders [5]The first trial and appeals [10]Trial [10]Appeal to this Court [16]Appeal to the Privy Council [22]Time of death [25]CNS tissue [27]The computer issue [38]The appeal allowed [41]Pre-trial applications and appeal to this Court [44]Dr Miller's slides [49]IHC analysis [52]The mRNA evidence [62]The second trial — overview [90]The issues on appeal — overview [110]Abuse of process [113]Admissibility of the IHC evidence [132]The evidence at trial [132]Appellant's submissions [183]Analysis [200]Admissibility of the mRNA evidence [205]Fuel consumption [259]Failure to give a demeanour direction [288]Failure to give a lies direction [313]Miscarriage and fair trial [322]Result [394]ScheduleIntroduction[1] Mr Lundy appeals against his conviction for the murders of his wife Christineand his seven-year-old daughter Amber at the family home in Palmerston North earlyin the morning of Wednesday 30 August 2000. He had previously been convicted, butthe convictions were set aside on appeal to the Privy Council and a retrial wasordered.1[2] The principal issues advanced on appeal relate to the probative value andprejudicial effect of scientific evidence relied on by the Crown to link Mr Lundy tothe murders. But it is also said that there were significant omissions from the Judge's1 Lundy v R [2013] UKPC 28, [2014] 2 NZLR 273 [Privy Council judgment].summing-up, and that the retrial was an abuse of process because of substantialchanges to the Crown case compared with that advanced at the first trial.[3] This judgment deals with each of the issues raised in support of the appeal.For reasons that we explain, we have decided that evidence relied on by the Crownbased on messenger RNA (mRNA) tracing was inadmissible, notwithstanding that ithad been ruled admissible pre-trial by the High Court, a ruling confirmed by this Courtand the trial Judge. Every other ground of appeal is rejected.[4] The exclusion of the mRNA evidence has required this Court to considerwhether Mr Lundy would have been convicted notwithstanding the evidence whichwe have ruled inadmissible. Because Mr Lundy was retried on an indictment firstpresented in February 2002, the consideration of that issue takes place underthe proviso to s 385(1) of the Crimes Act 1961, as if it had not been repealed. We arerequired by its terms to consider whether, notwithstanding the determination thatthe mRNA evidence was inadmissible, the appeal should nevertheless be dismissed onthe basis that no substantial miscarriage of justice has actually occurred. We are alsorequired to consider whether the trial was fair. Having considered those issues wehave decided that the proviso should be applied and the appeal dismissed.The murders[5] The bodies of the deceased were discovered around 9 am on Wednesday30 August. They had been hacked to death, likely with an axe or tomahawk used toattack their heads. The murder weapon was never found.[6] The bodies were discovered by Mrs Lundy's brother Glenn Weggery.Mr Weggery was an owner/driver for a freight company, and he had gone tothe property to inquire after progress being made by Mrs Lundy with his tax returnswith which she regularly helped him. He entered the house through a single slidingdoor at the rear which he found half open. When he called out there was no response.He commenced to walk down a hallway, when he observed Amber lying face down atits far end. He then rang for the emergency services. Having been put through tothe ambulance service, he said he needed to report a murder. Shortly afterwards,ambulance officers arrived, followed by the police. Mr Weggery was at one timeconsidered a possible suspect by the police but eliminated from their inquiries on thebasis that there was no evidential foundation showing his involvement.[7] Dr James Pang, a forensic pathologist, attended the scene at about 5 pm, whenhe examined both bodies. The next day, 31 August, he carried out a detailedpost-mortem examination of Amber's body at the Palmerston North mortuary.He carried out a similarly detailed post-mortem on Mrs Lundy's body on 2 September.In each case he was able to describe multiple and very severe wounds to the head, andin Mrs Lundy's case to the face. She had been attacked as she lay in bed. Dr Panggave evidence of various injuries to her arms and hands consistent with her trying todefend herself from the attack. The wounds sustained by Amber and Mrs Lundy wereconsistent with having been inflicted by the same weapon.[8] Mr Lundy was a travelling salesman, whose work often took him away fromhome to various cities and towns in the lower half of the North Island. On the nightof the murders he had been staying in the Foreshore Motor Lodge in Petone. He hadutilised the services of a prostitute who came to the motel and was present for aboutan hour, between 11.50 pm on 29 and 12.50 am on 30 August. Later on the morningof 30 August he was in Johnsonville when he was telephoned by a friend who told himabout the presence of police at his house and a police cordon that had been established.Mr Lundy then drove quickly back to Palmerston North. He was stopped by the policeat an intersection near his home. His car was seized.[9] The car was searched on 3 September. Mr Lundy told the police that apolo shirt folded inside out in a suitcase in the car was a shirt he had worn on the nightof 29 August. The Crown was to claim that central nervous system tissue (CNS tissue)was found within stains on the chest pocket and sleeve of the shirt when, after a delay,the shirt was forensically examined. This became crucial evidence against Mr Lundy,especially since Mrs Lundy's DNA was found on the shirt at the locations of the stains.The first trial and appealsTrial[10] Mr Lundy was tried before Ellis J and a jury at Palmerston North, commencingon 5 February 2002. He was convicted on both counts on 20 March.[11] The Crown's case was that the murders took place against a background ofdisharmony between Mr and Mrs Lundy over the issue of money and the deterioratingstate of the couple's finances. Pathological evidence was called to establish thatMrs Lundy and Amber were killed around 7 pm on 29 August 2000. Cell towerevidence showed that Mr Lundy had an opportunity to commit the murders, althoughbarely so. The Crown conceded he would need to have made a very fast car tripbetween Wellington and Palmerston North, but claimed it was not impossible for himto have done that. The Crown called evidence of the distance involved, the fact hewas driving a large and powerful car and was used to driving at high speeds. This wassupplemented by evidence that Mr Lundy had filled his car with petrol onthe afternoon of 29 August and of fuel consumption when the police drove Mr Lundy'scar the distance he claimed he had travelled after filling up. The Crown contendedthat if he had only travelled the distance he claimed he would have used fuel at therate of 27 litres per 100 km, approximately twice the normal rate that might have beenanticipated. On the other hand, his actual fuel consumption was consistent withthe Crown's case of him travelling at speed to and from Palmerston North.[12] The most significant evidence for the Crown was given by Dr Rodney Miller,an expert pathologist and Director of Immunohistochemistry (IHC) atProPath Laboratory in Dallas, Texas. He had taken slides from stains observed onthe shirt which Mr Lundy admitted wearing on the night of the murders, found in hiscar. The stains were on the front sleeve and the chest pocket. Dr Miller analysedthe stains using a technique based on IHC testing.2 His evidence was that tissue foundon Mr Lundy's shirt was CNS tissue. There was also evidence that Mrs Lundy's DNAhad been found on the shirt where the CNS tissue was located. This led the Crown toallege the CNS tissue must have also come from Mrs Lundy, a proposition not2 We discuss the IHC technique employed in more detail below.challenged by the defence at the first trial. In addition, DNA from Amber was foundon Mr Lundy's shirt. The most probable source was her blood. This was consistentwith the Crown's case that Mr Lundy had murdered Mrs Lundy while she lay in bedand had then killed Amber because she had witnessed the murder of her mother.[13] Bone fragments from Mrs Lundy's skull were found to have flakes of orangeand blue paint on them, justifying an inference that the murder weapon had beenmarked with orange and blue paint. Other evidence established that Mr Lundy hadbeen in the habit of keeping tools and painting them in those colours.[14] The Crown also alleged that while in Palmerston North Mr Lundy hadmanipulated the family computer to create the false impression it had been turned offat 10.52 pm on the night of the murders, for the purpose of giving him a false alibi.By that time Mr Lundy had returned to a motel in Petone, from where he telephoneda prostitute at 11.26 pm in whose company he was until about 12.50 am the followingday. The Crown also called evidence from an eye witness, Mrs Dance, who said shehad seen a person apparently wearing a blond curly wig running along the footpathaway from the scene at about 7.12 pm, shortly after the time the Crown alleged themurders had been committed. Mrs Dance said this person looked "absolutely terrifiedor frightened" and was "fat, quite a fat sort of person", and wearing a "horror struckexpression".3[15] The defence case was that Mrs Lundy's brain tissue must have got ontoMr Lundy's shirt as a result of police conduct during which, either deliberately oraccidentally, the shirt came into contact with the brain tissue. The DNA from Amberwhich was also on the shirt could have got there innocently on a previous occasion.The evidence could not justify an inference that the tissue or DNA was on the shirt asa result of Mr Lundy's presence at the murder scene since it was physically impossiblefor him to have travelled to and from Palmerston North and committed the murdersbetween 5.30 pm and 8.28 pm, a period including peak traffic volumes, when3 We have taken this account of Mrs Dance's evidence from this Court's judgment on the appealfrom the conviction at the first trial: R v Lundy (2002) 19 CRNZ 574 [First Court of Appealjudgment] at [13]. Mrs Dance was not called at the second trial.the evidence established he was in Petone.4 It was physically impossible for him tohave committed the murders within the narrow time frame asserted by the Crown.In support of this contention the defence relied on evidence given by a licensedprivate investigator, Mr Bass, who reported on three journeys he had made betweenPetone and Palmerston North a year after the murders. But the jury must have rejectedthis evidence. Mr Lundy was convicted.Appeal to this Court[16] The appeal against the 2002 conviction was advanced on the ground, set out ins 385(1)(a) of the Crimes Act, that the verdicts of the jury were unreasonable or couldnot be supported having regard to the evidence. However, this Court held that the jurywas entitled to accept Dr Miller's evidence that the brain tissue on the shirt could nothave been outside Mrs Lundy's body for very long (seconds or minutes).5 There wasthen an available inference that the brain tissue must have become attached to the shirtwhile Mr Lundy was killing her.6 While it was possible there was an innocentexplanation for the presence of Amber's DNA on the shirt, the most probable sourcewas her blood.7 The jury was entitled to regard this evidence as significant, and thepresence of the blood was a strange coincidence if Mr Lundy was innocent.8[17] This Court also noted the jury had been entitled to accept the evidence thatthe murders had taken place at or just after 7 pm, and that it was more than acoincidence that the time of death would have been almost exactly halfway throughthe available time period from 5.30 pm to 8.28 pm.9 The jury had also been entitledto accept the evidence that there was time for Mr Lundy to have completed the roundtrip and committed the murders in the time available and to conclude on the basis ofall the other evidence that he must have done so.10 The other evidence included notonly the evidence about the brain tissue and DNA, but also the evidence about the4 There was evidence that Mrs Lundy rang Mr Lundy on her cell phone at 5.30 pm, the call lastingfor eight minutes. Cell phone site data established that at that time he was near the motel in whichhe was staying. He claimed that he had stayed in Petone after the call ended.5 First Court of Appeal judgment, above n 3, at [2].6 At [2].7 At [10].8 At [10].9 At [3] and [6]. The defence called no evidence to challenge the Crown's case about the time ofdeath, based on the evidence of Dr Pang, supported by two other medical experts.10 At [19]–[20].paint flakes from the murder weapon, about the computer having been manipulated tocreate a false record of the time it was shut down, and the evidence of the eye witness.[18] The Court discussed briefly other issues raised by the defence. These includedevidence from a witness seeing lights on in the Lundy house at about 10 pm, whichwere not on in the morning, and evidence concerning Mr Lundy's relationship withhis daughter.11 These points, not related to the physical impossibility argument thatformed the main thrust of the appeal, did not cause the Court to doubt thereasonableness of the jury's verdict.12[19] The Court had allowed fresh evidence to be called on appeal, from a witnesswho saw a man he thought was Mr Lundy sitting in a car on the Petone foreshorereading a book between 6.30 to 7 pm one evening.13 The Court concluded thisevidence did not have sufficient cogency to justify a new trial, set against the otherevidence suggesting Mr Lundy was the murderer.14[20] The Court expressed its conclusion in the following terms:[20] In summary, the Crown case was a strong one. When viewed incombination, the presence of Mrs Lundy's brain tissue on Mr Lundy's shirt,the paint fragments found on her skull, and the forensically established timeof death at about the mid point of Mr Lundy's suggested absence from Petone,constituted powerful evidence of his guilt. The jury were also entitled to takethe view the Crown had established that it was physically possible forMr Lundy to have committed the crimes. For these reasons we are notpersuaded that the jury's verdicts were unreasonable or unable to be supportedhaving regard to the evidence.[21] The appeal was consequently dismissed.15Appeal to the Privy Council[22] The appeal to the Privy Council that followed proceeded on a completelydifferent basis and it is appropriate to deal with it in some detail.11 At [16].12 At [16].13 At [17].14 At [17].15 At [20].[23] Rather than contending that the jury's verdict was unreasonable, the appellantmounted an attack on the convictions based on a substantial amount of expert evidencethat had not been called at the trial, which was obtained after this Court's decisiondismissing the first appeal. The Privy Council's decision allowing that evidence to becalled remains the leading decision in New Zealand law about the basis on which freshevidence may be permitted on an appeal, and it will be necessary to address that aspectof the decision later in this judgment.[24] For present purposes it will be sufficient to note that the further evidence onwhich Mr Lundy sought to rely before the Privy Council addressed key aspects ofthe Crown case concerning the time of death, the identification of the CNS tissue onMr Lundy's shirt as being CNS tissue of Mrs Lundy (a matter that had been concededby the defence at the trial) and the evidence about the computer having been switchedoff at 10.52 pm on 29 August 2000.Time of death[25] As to time of death, the Privy Council concluded there was:16 now a welter of evidence available from a number of highly reputableconsultants which, if accepted, would nullify the claimed scientific supportfor the time of death which was so central to the Crown case.That meant that the interests of justice required the new evidence to be admitted.17Further, based on the various affidavits that had been filed it was appropriate to reachthese conclusions:18(i) Examination of stomach contents alone cannot provide guidance as tothe precise time of death;(ii) There is nothing in reputable medical literature to support the claimthat the absence of smell from stomach contents is an indication of thetime of death;[19](iii) The preponderance of the evidence established that gastric emptyingcan take place several hours after food has been ingested and that awide variation in duration is possible;16 Privy Council judgment, above n 1, at [129].17 At [129].18 At [103].19 The Crown's witness, Dr Pang, had relied on smell to draw an inference as to the time of death.(iv) The lag and the emptying phases may be prolonged by specificexternal variables such as stress, or by intrinsic factors such as the sizeand nature of the meal;(v) Authoritative evidence is now available to the appellant that gastricemptying can be used to give, at best, a very rough estimate of timeof death, with a margin of at least six hours.[26] Although defence counsel, Mr Behrens, had been made aware of some ofthe essential matters now covered in the affidavits relied on in the Privy Council, andhe had been able to make some progress in cross-examination, the Crown's witnesseshad not been "subjected to the critical onslaught" represented by the evidence ofthe four consultants now relied on by the appellant.20 If their opinions were accepted,the time of death evidence based on stomach contents would be "whollydiscredited".21 This was "highly significant" because of the centrality of the time ofdeath to the case mounted by the prosecution.22 While the murders could haveoccurred in the narrow timeframe asserted by the Crown, the new evidence eliminatedscientific support for that.23CNS tissue[27] As to the CNS tissue, the Privy Council considered the identification ofthe CNS tissue as that of Mrs Lundy had been of "overwhelming significance" atthe trial.24 No doubt for that reason it gave detailed consideration to the CNS tissueevidence given at the trial; advice that had been given on that issue to defence counselpreparing for the trial, on the basis of which the concession was made thatthe CNS tissue on the shirt came from Mrs Lundy; and the attack made on the evidencefor the purposes of the Privy Council appeal. What follows is a brief summary ofthose aspects of the judgment.25[28] The Crown had failed to disclose prior to the trial an opinion obtained bythe police from a neuropathologist, Dr Heng Teoh, calling into question the robustness20 At [104].21 At [104].22 At [104].23 At [104].24 At [131].25 The summary is based on the Privy Council's account of the affidavits before it, not the affidavitsthemselves.of the analysis carried out by the Crown's experts to conclude cells in the specimenobtained from Mr Lundy's shirt were from CNS tissue of Mrs Lundy.26 Dr Teoh couldconclude only that the cells were tissue cells.27 The Privy Council considered it wasinconceivable that the defence would not have been alerted to the need to obtain areport on the matter from a suitably qualified neuropathologist had it been made awareof Dr Teoh's views.28[29] Other material not disclosed before the trial was advice given byDr David Doyle, head of the department of neuropathology in a hospital in Glasgow,who the police had consulted about the work of Dr Miller.29 While generallysupportive of Dr Miller's conclusions, Dr Doyle observed that the use of IHC byDr Miller to establish the nature of the specimen was "a novel approach".30The Privy Council said:134. The real significance of Dr Doyle's letter lies in the observation thatuse of IHC by Dr Miller to establish the nature of the specimen was a "novelapproach". The novelty of the use of IHC in the forensic context is relevantfor two reasons. Firstly, it is relevant as to the impact that new evidence,challenging the validity of its use in a criminal trial without having beensubjected to laboratory or empirical testing, has on the safety of the conviction.Secondly, it raises questions about the admissibility of the evidence.[30] The Privy Council then referred to other correspondence between the policeand a pathologist in the United States Armed Forces Institute of Pathology which itregarded as raising questions, on the other evidence discussed, about:31 the use of IHC in the forensic setting of a criminal trial. Its widespread andsuccessful use as a diagnostic tool is undisputed but its acceptance as a meansof establishing a scientific proposition as an element of proof of guilt remainsuntested by any experimental or empirical means.[31] The fresh evidence relied upon by the defence included an affidavit fromProfessor Sheard, Associate Professor of Physiology at Otago University. He said thatIHC is known to be potentially inconsistent and unreliable, and the techniquesemployed by Dr Miller were inadequate to reach a reliable conclusion as to the nature26 At [131].27 At [131].28 At [131].29 At [132].30 At [134].31 At [137].of the material taken from the shirt.32 Among other criticisms,Associate Professor Sheard said the specimens were "poorly fixed [and] necrotic".33Professor Kevin Gatter, Professor of Pathology at the University of Oxford, swore anaffidavit endorsing Professor Sheard's report and conclusions.34[32] The evidence adduced before the Privy Council also included an affidavit fromProfessor Helen Whitwell, previously Professor of Forensic Pathology atthe University of Sheffield, and a Home Office pathologist, whose evidence wascritical of expert advice given to defence counsel before the first trial to the effect thatthe tissue on Dr Miller's slides was CNS tissue.35 That advice, by a Dr Beth Synek,had been based on photographs of Dr Miller's slides.36 Professor Whitwell consideredthat it was not possible on the basis of the photographs to reach "any meaningfulopinion on the nature of the material".37 Professor Whitwell also considered evidencebased on a "dab slide" taken from the shirt stain on the front sleeve of Mr Lundy'sshirt by the forensic scientist who examined the shirt 59 days after the murder,Mr Sutherland. The slide was referred to by the Privy Council as the ESR slide.38After examining the ESR slide, Professor Whitwell was of the opinion it comprised"possible cellular material", but it was impossible to determine the nature of the cells.39Having also examined the actual fragments from the shirt, she concluded that theminute amount of tissue on the slide was probably cellular material; she thought it waseither human or animal, but could not reach any more definite conclusion.40[33] Dr Synek also swore an affidavit for the Privy Council appeal. In it sheexplained that her involvement in the case had originally been confined to the questionwhether brain tissue found on a shirt would be recognisable pathologically after30 hours from the time it was removed from the body.41 She had not been informed32 At [52].33 At [53]. Other criticisms by Associate Professor Sheard were described by the Privy Council at[52] and [54].34 At [60].35 At [57].36 At [57].37 At [57].38 At [23].39 At [56].40 At [56].41 At [58].that the ESR slide had not been taken until 59 days after the deaths.42 She would nothave expected detail to be detectable after 59 days without fixation by air-drying inthe meantime.43 She confirmed she had advised defence counsel before the trial thatshe thought the photographs accompanying Dr Miller's report showed the presence ofCNS tissue, but she had not been offered any tissue sections from Dr Miller forreview.44 Nor had she been aware that the photographs presented were only a selectionof those taken by Dr Miller.45[34] The Crown obtained an affidavit from Dr Miller responding to the variousissues that had been raised by the defence witnesses. The Crown also relied onevidence from Professor Allen Gown, Clinical Professor of Pathology at theUniversity of British Columbia and a Medical Director and Chief Pathologist atPhenoPath Laboratories in Seattle. Professor Gown was asked to review Dr Miller'swork on the case. He considered Dr Miller's methodology was appropriate.The controls were "more than adequate to permit definitive interpretation ofthe specimen in question" and "[w]ithout any doubt whatsoever" the studiesconclusively demonstrated that the tissue fragments represented CNS tissue.46 Otherevidence was called by the Crown from Dr Cynric Temple-Camp who had examinedDr Miller's slides before the trial and again on 29 March 2013, concluding thatthe "best preserved fragments are unequivocally and unmistakably those of brain".47[35] The Privy Council noted that the affidavits of Dr Miller, Professor Gown andDr Temple-Camp prompted responses from the appellant's experts, but did notconsider it necessary to discuss them in full.48 For present purposes, however, it isrelevant to note the summary of further evidence given by Professor Gatter andDr Squier (another neuropathologist, associated with the University of Oxford andJohn Radcliffe Hospital):4942 At [58].43 At [58].44 At [59].45 At [59].46 At [72].47 At [78].48 At [83]. Some aspects of these further affidavits were nevertheless addressed at [84]–[91].49 As noted earlier, Professor Whitwell had also said it was not possible to say the tissue had a humansource.91. The two experts accepted that the staining on the slides was consistentwith brain tissue. But it was also consistent with other animal tissue. Even ifthe substance was brain, it was impossible to say that it was human brain oreven mammalian brain. The stain on the shirt, if caused by a processed meatproduct, could produce an appearance similar to that which they observed onthe slides. Finally, they said that if the substance was brain tissue depositedon the shirt following a traumatic injury, they would have expected to find redblood cells and none was present.[36] At [92], the Board listed seven separate areas of controversy which it said werethe principal matters either in dispute or remaining unresolved:(i) Is IHC known to be inconsistent and unreliable — or is this a criticismbased solely on a theoretical approach to its use? Is IHC only to beregarded as capable of indicating likely origin and identity of cellularmaterial, if examined under what Professor Sheard described as"strictly controlled experimental conditions"? An associated questionis whether the experience of its use in diagnostic pathology providesa dependable foundation on which to draw, in order to sustainDr Miller's conclusions.(ii) Should the circumstance that IHC had not been previously used in aforensic context affect how it should be regarded as an element ofproof that the substance retrieved from Mr Lundy's shirt wasCNS tissue?(iii) What is the extent and significance of the variation of staining onDr Miller's slides? Is it legitimate to select those slides which appearto show good preservation and produce results that are indicative ofthe presence of CNS tissue and to disregard those which produceambivalent or inconsistent results? Does the variation of stainingindicate the possibility of the presence of artefacts?(iv) What is the state of preservation of the tissue on (a) the ESR slide; and(b) Dr Miller's slides? Is it possible that parts of the fragment fromthe shirt were well preserved and parts not? What is the likelymechanism of air-drying in this instance? Is it possible that the tissuesmeared on to the ESR slide was poorly preserved while samplestaken by Dr Miller from the same stain were well preserved?(v) Are glial cells and blood vessels detectable on the ESR slide? Whatis the significance of the presence (or absence) of these features?(vi) Can the shrinking and darkness of the cell nuclei be explained by thecompression of the tiny fragment when it was smeared on the shirt?(vii) Is it possible to deduce that parts (at least) of the specimen taken fromthe shirt were not necrotic because histological andimmunohistochemical examination of the specimen was possible or isthis a circular argument?[37] The Board concluded that the new evidence of the consultants called onthe CNS tissue debate should be admitted for two reasons. The first was based onconsiderations stated by the United States Supreme Court in Daubert v Merrell DowPharmaceuticals Inc for evaluating the soundness of novel science: whether thetheory or technique can be and has been tested; whether the theory or technique hasbeen subject to peer review and publication; the known or potential rate of error or theexistence of standards; and whether the theory or technique used has been generallyaccepted.50 The Privy Council said these considerations were a useful template forconsidering whether evidence based on a technique such as IHC, which it regarded asnovel in a criminal trial, should be admissible.51 The admissibility debate had not yetbeen engaged with sufficiently to enable it to be resolved, but the need for such adebate showed that it might well have had an impact on whether there had been amiscarriage of justice and an unsafe conviction.52 Secondly, there were nowsignificant disputes and disagreements about the robustness of Dr Miller'smethodology and results.53 If resolved in Mr Lundy's favour, the issues could strikeat the heart of the case against him.54The computer issue[38] If the computer in the Lundy home had been switched off at 10.52 pm theevidence established it could not have been by Mr Lundy, since he had been shown tobe in Petone at that time. At the trial the Crown relied on the evidence ofMr Maarten Kleintjes, chief technical investigator of the New Zealand police, whocloned the Lundys' computer. He explained in evidence how it was possible to changethe time setting on the computer clock during operating and start-up modes, or by theuse of a floppy disk, with no record of the change being left. On his examination ofthe Lundys' computer he had discovered that its "registry files" were "all out of order",which he attributed to the date and the time of the computer clock having been changedbackwards and forwards.55 In his police interview Mr Lundy had said that the timehis wife had gone to bed could be inferred from the fact that she would have turned50 Daubert v Merrell Dow Pharmaceuticals Inc 509 US 579 (1993).51 At [139].52 At [139].53 At [140].54 At [140].55 At [40].the computer off before retiring, as was her normal practice. The evidence ofMr Kleintjes was that the record of the computer having been turned off at 10.52 pmcould be false. This enabled the Crown to suggest Mr Lundy might have manipulatedthe time of the shutdown recorded in the computer.[39] The defence had briefed an expert able to give evidence on these issues but hewas not called. Asked for an explanation of this, Mr Behrens advised the PrivyCouncil registrar that he could not be certain why the expert was not called.In the Privy Council, the defence sought to rely on new evidence that would establishthat the reason for the registry files being out of order was a virus known asJS KAK worm.56 Further, it was said that if a floppy disk had been used there wouldbe a record that it had been. This meant that the only way to change the time and leaveno trace was to use a highly sophisticated technique of which even many computerexperts would be unaware.[40] This evidence, if accepted, directly challenged the plausibility of the Crowncase by suggesting that Mrs Lundy was still alive at 10.52 pm.57 It had "an obviousand significant potential impact on the safety of the conviction and the possibility of amiscarriage of justice".58The appeal allowed[41] The Privy Council then referred to various New Zealand, Australian andUnited Kingdom authorities as to the test to be applied in assessing whether newevidence which was admitted established that there had been a miscarriage of justice.In the end it approved the approach taken by the New Zealand Supreme Courtin R v Matenga59 which it found to be in accord with United Kingdom andAustralian authorities.60 The Privy Council expressed the law in these terms:150. In light of these authorities, the Board is satisfied that the proper testto be applied by an appellate court in deciding whether a verdict is unsafe ora miscarriage of justice has occurred, where new evidence has been presented,is whether that evidence might reasonably have led to an acquittal. 56 At [109].57 At [142].58 At [142].59 R v Matenga [2009] NZSC 18, [2009] 3 NZLR 145.60 At [149].To amount to a miscarriage of justice the Supreme Court of New Zealandconsidered that the error in the earlier proceedings must have been capable ofaffecting the verdict. The Board detects no difference in this approach fromits formulation of the test as being whether the fresh evidence (or the error attrial) might reasonably have led to an acquittal.[42] Applying that law, the Privy Council concluded the verdict was unsafe.61Further, it considered this was not an appropriate case for the application of the provisoto s 385(1) of the Crimes Act under which an appeal could be dismissed if the courtconsidered no substantial miscarriage of justice had actually occurred.62 In thiscontext the Privy Council referred again to R v Matenga in which Blanchard J said fora unanimous Court:63The Court may exercise its discretion to dismiss the appeal only if, havingreviewed all the admissible evidence, it considers that, notwithstanding therehas been a miscarriage, the guilty verdict was inevitable, in the sense of beingthe only reasonably possible verdict, on that evidence. Importantly, the Courtshould not apply the proviso simply because it considers there was enoughevidence to enable a reasonable jury to convict. In order to come to the viewthat the verdict of guilty was inevitable the Court must itself feel sure of theguilt of the accused.[43] The Privy Council recorded that in this case it could not be sure ofthe appellant's guilt nor that a guilty verdict was inevitable.64 Its advice was thatthe appeal should be allowed, the convictions quashed and there should be a retrial.65Pre-trial applications and appeal to this Court[44] Before the second trial commenced the Crown applied under s 344A ofthe Crimes Act for orders on the admissibility of aspects of the intendedprosecution evidence.[45] There had been an important change in the Crown's case. It no longer intendedto allege that the deaths had occurred in the timeframe of approximately 7 to 7.15 pm.Rather, as Kós J recorded in his judgment on the s 344A application, the Crown didnot know when the deaths occurred other than at some time between 6.56 pm on61 At [151].62 At [159].63 R v Matenga, above n 59, at [31] (footnotes omitted).64 At [162].65 At [165].29 August 2000, when Mrs Lundy was known to have been alive, and 9 am on thefollowing day when their bodies were found.66[46] In its application the Crown asked the High Court to hold admissible:67[t]he evidence tending to prove that tissue (cellular material), found onthe accused's shirt worn by him on 29 August 2000 and located in his caron 30 August 2000, after the murders of Christine and Amber Lundy, wascentral nervous system tissue (brain or spinal cord) and was probable humanCNS.[47] As Kós J explained, this wording referred to four categories of the scientificopinion evidence on which the Crown intended to rely at the retrial.68 They were:(a) The DNA analysis for the stains found on Mr Lundy's shirt sleeve andshirt chest pocket. DNA extracted from the stained cut-out pieces offabric showed in the case of each sample that it was 450,000,000 timesmore likely to be the DNA of Mrs Lundy than another unrelated femalemember of the New Zealand public chosen at random.69(b) The IHC analysis claimed to establish that cellular tissue taken fromspecimens tested by Dr Miller was CNS (brain or spinal cord) tissue.70The analysis could not show whose CNS tissue it was or indeed whetherit was human in origin.(c) Analysis based on RNA, described by the Judge as a molecular materialfound in human and animal tissue cells, which differs according to celltype.71 This analysis, described in the evidence of Dr Laetitia Sijen ofthe Netherlands Forensic Institute (the NFI), was said to show that thetissue in the sleeve specimen was CNS tissue and was probably human66 R v Lundy [2014] NZHC 2527 [High Court pre-trial judgment] at [11]. The judgment wasdelivered on 15 October 2014.67 At [14].68 At [15].69 This likelihood ratio was greatly increased as a result of further analysis referred to atthe second trial.70 These were specimens 3003/3 and 3003/4.71 High Court pre-trial judgment, above n 66, at [99].in origin, as opposed to being from certain animal species.This evidence had not been called at the first trial.(d) Fluorescent in situ hybridisation (FISH) analysis, on which the Crownsought to rely, said to show that the sleeve specimen containedfemale human tissue fragments.[48] In the event the defence abandoned the challenge to the DNA evidence exceptin one respect rejected by the Judge. The Crown conceded that the FISH analysisshould not be called. This left as the main issues for consideration the IHC andmRNA evidence.Dr Miller's slides[49] Slides made by Dr Miller were common to both the IHC and mRNA analyses.[50] Specimens referred to as 3003/3 and 3003/4 (also called C3003/3 and C3003/4by some witnesses) were strips cut from Mr Lundy's shirt where it was visibly stained.They relate respectively to the left shirt sleeve and chest pocket stains. The strips werefirst used on 31 October 2000 by Dr SusanVintiner in a beaker soak to facilitate DNAanalysis, and then dried. The strips were subsequently taken by the police toDr Miller's laboratory in Dallas. They were handed to Dr Miller by DetectiveSergeant Grantham on 4 February 2001. They were then photographed. Dr Millerhad prepared paraffin blocks in which the strips were embedded. Slices were takenfrom the blocks and slides made for IHC analysis.72 The technique used enabled slicesto be successively taken of consistent quality. Kós J noted that slides made from theparaffin blocks had subsequently been used by all the IHC experts and theRNA analysis experts.73 All of the IHC experts had reported reliable results from theuse of the slides.7472 The process was described in the High Court pre-trial judgment, above n 66, at [70], and need notbe repeated here. The slices were described as being four microns thick.73 At [71].74 At [71].[51] The defence however raised various issues concerning the slides, whichthe Judge rejected.75 He found that the process of fixing the fabric samples inparaffin blocks as carried out by Dr Miller was an orthodox scientific procedure at thetime.76 Further there were no chain of custody or potential contamination issuesjustifying ruling the evidence inadmissible.77 While testing in 2014 had revealed thepresence of DNA from an unidentified person, this was apparently an isolated caseprobably attributable to subsequently introduced contamination of that particularslide.78 Finally, deterioration of the dab slide (as referred to by the Privy Council asthe ESR slide) taken from the same area as the stain contained in specimen 3003/3 didnot have implications for the reliability of specimens 3003/3 and 3003/4.79IHC analysis[52] Turning then to the IHC evidence itself, Kós J noted that while the Crown hadrelied only on IHC analysis from Dr Miller at the first trial it now intended to calladditional evidence, from Dr Daniel du Plessis, a consultant neuropathologist andClinical Lead of the Department of Cellular Pathology at Salford Royal Hospital NHSFoundation Trust. His evidence was supportive of Dr Miller's methodology and hehad been able to replicate the results obtained. He concluded there was "compellingevidence in support of central nervous system tissue".80 Two other Crown witnessesreached the same conclusion: Professor Daniel Brat who is Professor of Pathology atEmory University School of Medicine in Atlanta, Georgia and Professor Allen Gown,Professor of Pathology at the University of British Columbia in Vancouver.[53] Two of Professor Gown's statements were quoted by Kós J.81 First, referringto his own work carried out for the purpose of the case:75 At [82].76 At [77].77 At [78].78 At [79].79 At [80]. What Kós J called the "dab slide" is what the Privy Council called "the ESR slide".It was taken from the front left sleeve of Mr Lundy's shirt, before the strip was cut from the shirton 27 October 2000. The material in the ESR slide was also referred to as specimen 3003/2, andhad become degraded by the time of the hearing before Kós J.80 At [85].81 At [86].The immunohistochemical studies I performed confirm, without anyequivocation or doubt, the presence of [CNS] tissue in the fragmentsinterspersed amongst the shirt fibres [54] Second, as to the reliability of the IHC analysis:IHC is a reliable technique. Decades of experience with IHC has proven it tobe robust and extremely reliable and it is currently a cornerstone of pathologydiagnostic practice.[55] Two witnesses were called to give evidence for the defence. One wasDr Colin Smith, Reader in Pathology at the University of Edinburgh. Kós J recordedhis conclusions as follows:82 specimen 3003/3's immunoprofile "is convincingly that of CNS tissue",without any false positive concern. Similarly, in the case of specimen 3003/4.There could be "no doubt" that the tissue in the shirt was CNS tissue.[56] The second defence expert on this issue was Professor James Ironside,Professor of Clinical Neuropathology, also at the University of Edinburgh andHonorary Consultant in Neuropathology at NHS Lothian. Kós J recorded:83Professor Ironside was in full agreement with Drs du Plessis and Smith thatspecimens 3003/3 and 3003/4 represented CNS tissue. Professor Ironsideconsidered there was no doubt that it was. But he could not say whether itwas human CNS tissue, in contrast to diagnostic neuropathology where thesource is not in doubt.[57] The Judge noted that none of these expert witnesses called had any concernabout the use of IHC to identify tissue from an unknown source.84 Although the useof IHC analysis to identify tissue material from an unknown source was relativelyunusual, it was not novel.85 None of the experts had been left in any doubt that theIHC analysis of specimens 3000/3 and 3003/4 showed that it was CNS tissue.86Moreover, their conclusions were independently supported by electron microscopyundertaken by Dr du Plessis. As Kós J pointed out, electron microscopy is a technique82 At [88].83 At [88].84 At [94].85 At [93].86 At [95].distinct from IHC, involving extremely high magnification of intracellularcomponents.87[58] Also before the Court was the affidavit sworn by Associate Professor Sheardfor the purposes of the Privy Council appeal, attaching a report dated 26 January 2010which had also been before the Board. Kós J referred to that affidavit and also to asummary of Associate Professor Sheard's views that the Judge said had been providedon the eve of the hearing, which he described as accepting that the tissue in 3003/3and 3003/4 was CNS tissue.88[59] The defence objection had been advanced on the basis that IHC had been usedfor the purpose of identifying an unknown substance on fabric that had been stored inunknown conditions for over five months, the period that had elapsed since the stripshad been cut from the shirt by Mr Sutherland on 27 October 2000.89 This was said tobe "some distance away" from accepted uses of IHC, which involve its deployment incontrolled circumstances to identify the presence of disease in known tissue.90However, the Judge noted that in closing, Mr Hislop QC (appearing for Mr Lundy)maintained the defence objection as a matter of formality and did not seek to advancefurther submissions on the issue.91 This was described by the Judge as a realisticapproach, in light of the evidence.92[60] The Crown's application was therefore successful on the admissibility ofthe IHC evidence. This was subject, however, to alteration of some of the evidence toremove references to the CNS tissue observed being brain (as opposed to spinal cord)and of human origin. The IHC evidence was to be given on the basis that it wasCNS tissue and of unknown origin.[61] As has been seen, the issues raised about the IHC evidence in the Privy Councilwere of central importance to the Board's conclusion that there should be a retrial.87 At [96].88 At [78]. We refer again later to the summary of Associate Professor Sheard's views and anaffidavit of Mr Sean Doyle which was also before Kós J although not discussed in his judgment.89 At [90].90 At [90].91 At [91].92 At [91].The judgement of Kós J, which was not appealed on this point prior to the second trial,appeared to have dealt with those issues for the purposes of the retrial, especially inthe circumstance that both experts who gave evidence for the defence at the pre-trialhearing had expressed themselves as being sure that the tissue was CNS tissue.The mRNA evidence[62] The second substantive issue dealt by Kós J concerned the Crown's evidenceabout RNA typing. This evidence was not called at the first trial. It was procured bythe Crown in response to submissions that had been made to the Privy Council thatsuggested the CNS tissue found may have had a non-human source. As noted above,some of the defence experts who provided affidavits for that hearing had said it wasnot possible to determine whether the cellular material Dr Miller and others hadidentified as CNS tissue had a human or non-human source. The mRNA evidence wasto address that issue.[63] The Crown's primary witness on the subject of mRNA was Dr Laetitia Sijen,and she was relied on at the pre-trial hearing before Kós J. She is a forensic scientistemployed by the NFI. The NFI is the largest forensic institute in the Netherlands,primarily engaged in work for the police in that country, but also working on projectsfor foreign governments and in the field of war crimes. Dr Sijen is in the NFI's HumanBiological Traces Department, heading its Research and Development team. She hadpreviously done post-doctoral work in the field of RNA at Universities in Amsterdamand Utrecht. She has published in the prestigious journals Nature, Cell and Science,and is plainly an eminent expert in the field relevant to her evidence in this case.[64] Dr Sijen explained in her evidence the difference between DNA and RNA,addressing this under a heading referring to "the difference between 'Whose is it?' and'What is it?'". As she put it, DNA contains all the genetic information of a person.It is the same in all cells of an individual, while differing from person to person.This enables DNA analysis to link biological traces to particular individuals.[65] She continued:The body consists of different types of cells, each with their own specificfunction. This variety in function is possible because each cell type usesdifferent parts of the same DNA. The parts of DNA used by a specific cell areconverted into RNA. RNA can therefore be seen as the translation of theinformation in the DNA that is required by the relevant cell to perform itsfunction within the body. This means that each cell type has a specific set ofRNAs. Blood cells therefore have an RNA set that is different from that ofskin cells. The RNA set that is specific for a certain cell type is the same foreach person. In order to determine the nature of cell material in a trace, anexamination is undertaken, denoted RNA analysis, to determine which(combinations of) RNA markers reside in a trace, from which the cell typesthat are present in a trace are inferred.[66] At the trial, Dr Sijen explained that certain portions of DNA are transcribedinto messenger RNA (abbreviated to mRNA) used in the process of making differentkinds of cells. An expert retained by the defence, Dr Marielle Vennemann, explainedthat in the process of building proteins, gene information is transcribed in mRNA,which then transports that information to the ribosomes within cells. That informationis used to then synthesise the protein. She stated that mRNA represents approximately2–3 per cent of the total RNA amount within cells. The mRNA is specific to each celltype. We infer, broadly, that mRNA analysis is the process of looking for thecell specific markers.[67] Dr Sijen said that the NFI currently applies two methods of RNA analysis.The first she referred to as "RNA cell typing", designed to detect the presence of blood,semen, saliva, skin cells, nasal mucosa, vaginal cells and/or menstrual secretion in atrace. This is typically carried out in the case of sexual assaults where the presence orabsence of some of these kinds of cells may be decisive. The second kind of analysiscarried out is called "RNA organ typing". This is RNA analysis for the presence ofskin cells, blood, brain tissue, lung, liver, muscle, heart and/or kidney in a trace.Dr Sijen said this kind of analysis is applied in the case of traces from violent incidentssuch as shootings.[68] However, for the purposes of this case, the NFI developed an"additional assay" which the NFI called "brain-plex". This tested the presence orabsence of mRNA markers expressed in human brain. Dr Sijen explained that thebrain-plex targets types of human brain cells which are different to those present inthe existing NFI organ typing test. The new brain-plex was developed (in consultationwith a neuropathologist employed by the NFI) because histological results obtainedby Dr Miller indicated that brain markers used in the existing organ typing test mightnot apply: photomicrographs of the samples from the shirt (C3003/3 and C3003/4)showed that astrocyte cells were present and possibly oligodendrocytes and microgliawhich would not be detected in the organ typing system. It was Dr Sijen's evidencethat during development of the brain-plex assay there was validation of thebrain-plex's "sensitivity, specificity considering other human tissues and body fluidsand specificity regarding brain tissue of human origin".[69] Dr Vennemann was provided with information concerning the development ofthe brain-plex assay and was present on occasions when examinations involving RNAextraction and proofing were carried out. Although she was not involved in thedevelopment of the brain-plex, appropriate non-human species for validation testingof RNA were specified in consultation with her: these were "bovine, cat, chicken,dog, guinea pig, pig, rabbit and sheep".[70] Because the brain specific markers used in the existing organ typing test werenot applicable, suitable alternative candidates were identified on the basis of aliterature search for central nervous system neural cell types. Four gene candidateswere chosen, comprising three astrocyte markers (GFAP, ACSBG1 and S100B) andone oligodendrocyte marker (OPALIN). A fifth marker was also used, described as18S-rRNA, "a sensitive housekeeping marker". Primers for these gene candidateswere designed using gene sequences on online genome databases. Dr Sijen said thathuman specificity was "initially affirmed by in silico NCBI primer-BLAST".Specificity for brain tissue was assessed by subjecting the brain-plex to RNAs of21 known source human tissues. In addition, performance was assessed for bodyfluids generally encountered in a forensic context: blood, menstrual secretion,nasal mucosa, saliva, semen and vaginal mucosa.[71] Dr Sijen said that except for S100B, the developed brain markers did not shownon-specific signals in any of the tested human tissues. The result for S100B was notconsidered significant, however, because under her interpretation guidelines, aresponse of at least half the markers was required.[72] Specificity of the brain-plex to human brain was tested by subjecting thebrain-plex to a selection of eight animal brain RNAs from the species discussed withDr Vennemann. False positive signals were observed at what was said to be thestandard annealing temperature of 60 degrees Celsius. However, "[w]ith optimizedPCR settings" (an annealing temperature increased to 64 degrees Celsius),the brain-plex was found to be specific for brain material of human origin. Dr Sijenalso described a process in which the ability of the brain-plex to identify brain tissuein formalin-fixed, paraffin-embedded (FFPE) tissue sections (such as those made byDr Miller) was confirmed. This process was carried out using human brain tissuesmeared on fabric to mimic the samples from Mr Lundy's shirt. Dr Sijen claimed thatinformative mRNA profiles were successfully generated from the tissue sections usingthe brain-plex. She said that brain cells had been successfully identified inFFPE tissue sections stored for over 10 years.[73] Dr Sijen went on to describe the process adopted to test the samples on slidesmade from specimens 3003/3 (sleeve) and 3003/4 (chest), taken from Mr Lundy'sshirt. Based on the microscopic photographs it was considered the amount of cellmaterial would be very limited. It was decided to combine the material on all10 sleeve slides in one extraction (similarly with the 10 slides of material from thechest) and to concentrate the RNA by ethanol precipitation, an approach agreed withDr Vennemann.[74] Applying the new brain-plex methodology to the slides from specimen 3003/4resulted in no indication of the presence of human brain tissue, but that was not thecase with respect to the slides containing the sleeve material, 3003/3. These werepooled together (with defence agreement) in what the NFI called sample 22.Sample 22 was given three replicate tests with the result summarised in this table:93S22(3003/3) ACSBG1 GFAP S100B OPALINS22 + +S22 rep 1 + + +S22 rep 2 + +93 We have taken the table from the High Court pre-trial judgment, above n 66, at [102].[75] In summary, of the 12 results obtained over the three tests using the fourbrain markers, only seven (or 58 per cent) were positive. In contrast, as Kós Jobserved, the reference testing of two parallel samples from Dr Miller using knownbrain tissue smeared on a t-shirt showed all brain specific markers except ACSBG1.94[76] On this basis Dr Sijen felt able to express this conclusion:In my opinion, the RNA typing results are more probable if the slides from thesleeve area [specimen 3003/3] contain human brain tissue than if they containbrain tissue of the animal species examined. It is not possible to determinehow much more probable these results are (i.e. to assign the exact weight ofthe evidence to the results).[77] Pre-trial, the defence opposition to the admissibility of this evidence was basedon the evidence of Dr Vennemann. She was employed as head of section in theForensic Molecular Biology Institute of Legal Medicine at the University of Münsterin Germany and her expertise includes the forensic application of RNA technologies.Kós J noted she had significant reservations about the use of RNA profiling forforensic purposes generally and in this case.95 He summarised her criticisms underfour headings.[78] It is sufficient at this stage to record first that Dr Vennemann noted thatthe brain-plex was novel, had not been used in other cases and had only beenself-validated by the NFI. There had been no publication in peer-reviewed scientificjournals. Second, Dr Vennemann was concerned that the 50 per cent minimumthreshold applied by the NFI was arbitrary and not properly validated. Third, she hadserious doubts that the four markers adopted had definitive specificity for bothCNS tissue and human source CNS tissue only. Finally, for reasons she explained, shehad rejected, in her own work, the use of RNA (or, specifically, mRNA) testing forforensic application on the basis that RNA is less stable than DNA: when taken fromnon-living tissue its degradation rates are poorly understood; negative results can bedifficult to interpret; and mRNA is not necessarily tissue specific.94 At [103].95 At [106].[79] The defence opposed the Crown's application on the basis of these criticisms,emphasising the absence of objective and generally accepted scientific evidencesupporting the claim to human specificity and adding that the potential error rate ofthe technique is unknown; the evidence is of little probative value; and that the disputeinvolved a "highly technical area of science that a jury would find impossible toevaluate if faced with competing scientific opinion".96 It would in the circumstancesbe unsafe for the claim to human specificity to go before the jury.97[80] Kós J concluded ("by a relatively narrow margin") that the Crown haddemonstrated a sufficient foundation of reliability for the RNA analysis to go tothe jury.98 His analysis followed the approach in Daubert.99 Although the brain-plexmethodology was novel, the absence of publication and peer review was notdeterminative.100 The NFI's RNA analysis had been tested and was "hard science"and capable of objective proof or disproof by experimentation.101 It was an extensionof the existing NFI multiplex organ-typer, which had been the subject of publicationand peer review.102 The new markers had been chosen on a rational basis and theincreased annealing temperature had been shown to eliminate false positives.103[81] Next, the constituent tissue in specimen 3003/3 was known to be CNS tissue,on the basis of the IHC analysis.104 In addition, the significance of the RNA typingwas at the level of species specificity, showing in the case of specimen 3003/3 that itwas more likely to be human CNS tissue than CNS tissue from the eight animal speciesselected.105 Further, the required standard of proof as to species specificity is notbeyond reasonable doubt.106 Here, Dr Sijen had not claimed to meet that standard.Her conclusion was simply that the substance was more probably of human origin thanfrom the other species examined. She was also not able to say how much more96 At [114].97 At [115].98 At [117].99 Daubert v Merrell Dow Pharmaceuticals Inc, above n 50.100 At [118]–[119].101 At [119].102 At [120].103 At [120].104 At [121].105 At [122].106 At [123].probable. Kós J stated that the "proximity of the results to the 50 per cent thresholdreinforces that no compelling conclusion from RNA analysis is possible".107[82] Finally, the role of the jury would not be to resolve the scientific debatebetween Drs Sijen and Vennemann, but rather to decide whether the Crown had provedMr Lundy's guilt beyond reasonable doubt.108 While the RNA analysis would be veryimportant, alongside the other circumstantial evidence including the DNA andIHC evidence, the jury would be assisted in not overweighting the RNA evidence byhearing from Dr Vennemann and by appropriate jury directions.109 For these reasonsKós J concluded the NFI methodology was "not so apparently unreliable that it shouldbe withdrawn from the jury", and "the scientific contest can sensibly be put to the juryfor its consideration".110[83] Mr Lundy sought leave to appeal to this Court on the admissibility ofthe mRNA evidence. Leave was granted and the judgment of Kós J was upheld by amajority (Harrison and French JJ), Ellen France P dissenting.111[84] The majority considered the admissibility of the mRNA evidence in terms ofs 25(1) of the Evidence Act 2006, concluding that it was highly relevant and veryprobative of the issue as to its likely source.112 As to its reliability, they reasoned thatthere was no dispute that the NFI is a reputable organisation and that the scientists whoundertook the testing were credible and eminent in their field.113 The work had beencarried out in the presence of a defence expert (Dr Vennemann), had been fullyexplained and the results interpreted conservatively.114 Although some aspects ofthe work were novel, they were logical and rational extensions of established scientifictheory and techniques.115 Use of the brain-plex for species specificity was anextension of the existing multiplex analysis used for tissue specificity.116107 At [123].108 At [124].109 At [124].110 At [125].111 Lundy v R [2014] NZCA 576 [Court of Appeal pre-trial judgment].112 At [78].113 At [80].114 At [80].115 At [81].116 At [83]. As we understand it, Dr Sijen regarded it as an extension of the existing multiplex analysisused for organ typing.[85] What was novel was the particular combination of markers, but that wasnecessitated by the fact that only certain cell types were present in the samples.117There was nothing novel in the fact that markers were selected which were specific tounique strands of RNA and using these in a polymerase chain reaction process to testfor particular cell types.118 In the circumstances, adopting the approach of Tipping Jin R v Calder, the underlying technique involved and the selection of the markers forthe brain-plex properly rested on "perfectly intelligible scientific reasoning".119[86] The majority acknowledged there was a dispute between the experts about thereliability of the brain-plex, mentioning in particular the annealing temperature used,but the testing method used to verify the brain-plex for tissue and species specificitywas "thorough, diligent and transparent. It was clearly explained and involvedthe application of applied scientific method to eliminate false positives."120 Correctlyanalysed, the contest between the experts was about reliability and that could properlybe determined by the jury in deciding whether to accept the evidence.121 This was"not a point that contradicts the underlying reliability of the science in a thresholdadmissibility sense".122 Similarly, the majority thought the 50 per cent scoring systemadopted was conservative.123 In any event the jury could decide whether the scoringthreshold should have been higher.124 Overall, whether enough had been done bythe NFI scientists could properly be left to the jury.125 The jury would not be decidingthe legitimacy of the science as such but rather whether aspects of the testing had beensufficiently rigorous to justify placing weight on the results.126[87] Ellen France P considered the admissibility issues were best analysed in termsof s 8 of the Evidence Act.127 She concluded that the unfair prejudice of the evidencewould outweigh its probative value (because of the absence of both validation of the117 At [83].118 At [84].119 At [85], applying R v Calder HC Christchurch T154/94, 12 April 1995.120 At [86]–[87].121 At [90].122 At [90].123 At [92].124 At [92].125 At [93].126 At [93].127 At [49].brain-plex and international standards in relation to its use in a forensic setting), andit should be excluded for that reason.128[88] We note finally that the defence sought to introduce further expert evidence insupport of the appeal from Professor Stephen Bustin. Without resolving ajurisdictional issue raised by the Crown to the effect that further evidence may not becalled on a pre-trial evidence admissibility appeal, the Court rejectedthe defence application, essentially on the basis that the affidavit did not appear toraise anything new.129 Professor Bustin was later to give evidence at the second trialand he has also filed further affidavits which we are now asked to consider forthe purposes of the appeal.[89] In accordance with the view of the majority, the Court granted leave to appeal,but ruled the mRNA evidence admissible and dismissed the appeal.130The second trial — overview[90] The second trial commenced on 9 February 2015 and was completed on1 April 2015. Although some aspects of the Crown case on the retrial were different,important elements of the case remained. The mRNA evidence was new.[91] Importantly, the Crown relied on the evidence that CNS tissue was found intwo spots on the shirt the appellant acknowledged he had been wearing on the nightbefore the bodies were discovered. Notwithstanding the strong attack that had beenmounted in the Privy Council on this aspect of the Crown's case, the defence did notcontest at the second trial the fact that CNS tissue had indeed been found onMr Lundy's shirt.[92] In addition, chemical tests that had been carried out on the two spots indicatedthe probable presence of blood in the stains. Pieces of the fabric with the CNS tissueon them had been subjected to DNA analysis. That analysis showed the presence ofDNA of Mrs Lundy. The finding was a strong one in terms of the quantity and quality128 At [69].129 At [103].130 At [104].of her DNA. The Crown's case was that because of the presence of Mrs Lundy's DNAon the fabric, then in the context of the overall evidence in what was a circumstantialcase the CNS tissue would have been Mrs Lundy's brain tissue which had becomeattached to the shirt in the course of her murder.[93] The mRNA analysis was then relied on to demonstrate that the CNS tissue wasmore likely to be human CNS tissue than CNS tissue of any of the animals that hadbeen tested by the NFI.[94] Other strands of the Crown's case included petrol usage and an apparentdiscrepancy in the kilometres travelled by the appellant's car between relevant dates.The petrol usage and kilometre discrepancy were said to be consistent with Mr Lundyhaving made another trip from Petone where the appellant was staying toPalmerston North and back on the night of the killings as opposed to Mr Lundy'sclaims in police interviews in 2000 and 2001 that he had stayed in Petone for the wholenight.[95] The Crown's case also rested on the nature of the injuries sustained bythe deceased. It was said these were of such an extent as to be inconsistent with arandom burglar unknown to the victim. The injuries were rather consistent withsomebody who had developed an intense antipathy towards Mrs Lundy, who hadgained access to the home deliberately, attacked her as she slept in bed and then stagedthe scene so as to make it appear it was a burglary by a random burglar. This includedthe fact that a jewellery box had apparently been stolen by the burglar. A bracelet fromthe jewellery box was found in Mr Lundy's car when it was searched.[96] Part of the Crown's case was that the murder weapon, which must have been aheavy and sharp implement, had left traces of orange and blue paint near the points ofimpact and embedded into a skull fragment of Mrs Lundy. Mr Lundy painted his toolswith orange and blue paint and some of the paint found near the point of impact wasindistinguishable from the paint Mr Lundy used.[97] In addition, there were tiny spots consistent with dried blood on Mr Lundy'sshirt which, after DNA testing, were found to contain DNA of Amber Lundy.The Crown also called evidence from a witness X who claimed that when he andMr Lundy were in custody together in March 2002, Mr Lundy told him that he wouldnot be there if his daughter had not seen what he was doing to his wife.[98] As to motive, in addition to the antipathy point earlier referred to,the Crown case was that in the days immediately prior to and including 29 August2000 a series of telephone calls between Mr and Mrs Lundy indicated that theirfinancial situation was difficult and a subject of dispute between them.[99] It should be noted that in one important respect, the Crown case atthe second trial was significantly different. At the first trial, as has been seen, a keyallegation made by the Crown was that the murders occurred around 7 pm on29 August 2000. At the second trial, the Crown abandoned that part of its case.The allegation made, as the second trial approached, was that the murders occurredsome time after 6.56 pm on 29 August and before their bodies were found at 9 am on30 August. Mr Eaton QC for the appellant complains that notice of this change wasnot given on a timely basis. The Crown also no longer persisted in the allegation thatthe computer had been manipulated so as to show incorrectly that it had been turnedoff at 10.52 pm on 29 August. On the contrary, the Crown now relied on the computerhaving been turned off at 10.52 pm, and the implication that Mrs Lundy was still aliveat that time. Crown counsel advised the defence on 21 January 2015 that the Crowndid not contend for a time of death prior to the computer being shut down.[100] The time of death in the Crown's case would still have to accommodateMr Lundy's meeting with a prostitute from about 11.50 pm on 29 August until12.50 am on 30 August and the indictment was in fact amended on 26 January 2015(a week before the second trial commenced) to allege that the murders were committedon 30 August. But the important issue at the first trial about Mr Lundy's alleged veryfast trip (partly during the rush hour) between Petone, Palmerston North and back, fellaway.[101] The change also meant that the Crown did not need to rely on Dr Pang'sevidence purporting to establish the time of death based on the stomach contents ofthe deceased, his opinion on that issue having been discredited by the wide-rangingcriticism in the affidavits of experts filed for Mr Lundy in the Privy Council.As mentioned earlier, Dr Pang had carried out the post-mortem examination of boththe deceased and was called at the first trial to describe what he observed inthe post-mortems. He expressed the view at the first trial that the stomach contentsand lack of gastric smell suggested the deaths had occurred about an hour afterthe deceased ate an evening meal. Since the Crown could establish that Mrs Lundyhad purchased a McDonald's meal at 5.43 pm on 29 August Dr Pang's evidence wasconsistent with death at around 7 pm.[102] Dr Pang confirmed at the second trial that his examination showed thatAmber's stomach contained a large meal consisting of potato chips and "probably"fish; there was "identifiable food material", with "no definite signs of digestion" orgastric juices. And in the case of Mrs Lundy, he gave a similar description, explainingthe chips observed were long potato chips.[103] In evidence-in-chief he conceded on the basis of all he had read sincethe first trial (including the evidence given to the Privy Council) that it was notpossible to be precise about the time of death. In fact all he could say with certaintywas that death had occurred between the time the deceased were last known to be aliveand the time the bodies were discovered. In cross-examination, however, heacknowledged that for the majority of individuals it takes at the most six hours forthe stomach to go from full to empty. The Crown also relied on evidence fromDr Martin Sage, an experienced forensic pathologist, who expressed the opinion thatestimating the time of death on the basis of stomach contents is inherently unreliable.[104] The Crown's abandonment of the early evening time of death also meant thatit could not rely on the evidence of the supposed eye-witness, Mrs Dance, claiming tohave seen a person matching Mr Lundy's appearance running down the footpathwearing a wig, and she was not called at the second trial.[105] As noted above, there was no challenge by the defence to the fact thatCNS tissue had been found on Mr Lundy's shirt. The cornerstone of the defence wasthat it was physically impossible for Mr Lundy to have committed the murders becauseof his absence in Petone and that, as a consequence, there must be some otherexplanation for the finding of the CNS tissue on his shirt. Possibilities relied on werecontamination during the investigatory process, or an alternative source of CNS tissuesuch as food. There was a fresh challenge to the admissibility of the mRNA evidence,based on a brief provided by Professor Bustin, but Simon France J rejected it.131He acknowledged that the brief of Professor Bustin amounted to a more sustainedassault on the mRNA methodology than had previously occurred.132 However, he didnot consider that the evidence advanced the case against admissibility as opposed tothe weight of the evidence.133 In the circumstances, he thought a case had not beenmade out to revisit the issue of admissibility that had been determined by the majorityin the second Court of Appeal judgment.134[106] Other aspects of the defence relied on Dr Pang's post-mortem evidence aboutthe contents of the deceaseds' stomachs. It was said that had they eaten earlier inthe evening, the state of their stomachs was inconsistent with their having beenmurdered as late as 2–2.30 am. Prior to that, Mr Lundy could not have committed themurders as he would have been unable to be in Palmerston North prior to 2.30 am.Evidence was called from Professor Michael Horowitz of the University of Adelaide,and Director of the Endocrine and Metabolic Unit at Royal Adelaide Hospital.Professor Horowitz gave evidence based on the fact that Dr Pang had said as a resultof his post-mortem that there was no evidence that gastric emptying had commenced.Professor Horowitz said that meant that, for death to have occurred after 2 am theycould not have eaten any earlier than midnight.[107] The defence also relied on the possibility of Mr Lundy's shirt beingcontaminated during the investigation, a lack of integrity in other aspects of the initialinvestigation and mistakes evidently made by the computer expert initially relied onby the Crown (Mr Kleintjes) as well as Dr Pang's discredited evidence about time ofdeath. The defence claimed that the mRNA evidence of Dr Sijen was unreliable andshould be discounted. It was claimed in addition that the silver bracelet found inMr Lundy's car did not belong to either of the deceased. The defence also relied on131 R v Lundy CRI-2001-054-832244, 19 February 2015. The objection was, of course, dealt with inthe absence of the jury.132 At [32].133 At [33].134 At [34].the absence of proof that any of Mr Lundy's tools were missing, or that he ever owneda tool capable of inflicting the fatal injuries (referred to in the evidence as being like atomahawk). Added to this, the defence pointed to the absence of other forensicevidence such as blood on Mr Lundy, or any other items of clothing that he wore onthe night, and in his car. It was claimed that witness X was unreliable and that thedeceased Christine Lundy's brother, Mr Glenn Weggery, may have been responsiblefor the killings. The defence contended that the alleged motive of financial pressurewas overstated and could not in any event have led to the murder.[108] The defence also claimed that there was insufficient petrol in Mr Lundy's carfor him to have made an extra trip from Petone to Palmerston North and back in theearly hours of the morning of 30 August. In this respect, the defence sought to rely ona Crown witness, Mr Johanson, who had attempted to replicate those journeys in 2000.[109] Reliance was also placed on the evidence of Mr Tupai, a neighbour ofthe Lundys, who gave evidence of having observed an open door and lights on at10.59 pm on the evening of 29 August 2000. It was said that Mr Tupai's evidence wasconsistent with the killing occurring at about that time and up to midnight whenMr Lundy was definitely in Petone.The issues on appeal — overview[110] The principal issues now raised on appeal were as follows:(a) Mr Eaton submitted that the changes made to the Crown's case at thesecond trial, resulting in the abandonment of any theory of a particulartime of death and the appellant's alleged rapid car journey, amountedto an abuse of process. Once the changed nature of the Crown's casebecame apparent, the proceeding ought to have been stayed on thatbasis.(b) The IHC evidence on which the Crown relied for the purpose ofestablishing the presence of CNS tissue in the two stains on Mr Lundy'sshirt was unsatisfactory and should not have been admitted. Reliancewas placed on the Privy Council judgment's description ofthe IHC technique as being untested by any experimental or empiricalmeans as a scientific basis of proving guilt in a criminal trial. It wassubmitted that the IHC evidence should not have been admitted atthe trial.(c) The mRNA evidence was not scientifically valid, and thereforeunreliable. In addition, it could not be said to have been substantiallyhelpful and its low probative value was outweighed by its unfairlyprejudicial effect. The Judge erred by failing to give a reliabilitydirection as required by s 122 of the Evidence Act in respect of thatevidence.(d) The appellant seeks to bolster with new evidence the case made at thetrial that there was insufficient petrol for Mr Lundy to have made theso-called "secret journey" from Petone to Palmerston North and backto commit the murders. The new evidence is designed to establish amore than threefold increase in fuel consumption if a car such as thatMr Lundy drove is pushed hard, as was said to be the case whenMr Lundy drove from Johnsonville to Palmerston North after beingtold about the police presence at his home on the morning of 30 August.(e) There was a failure to give a demeanour direction. This point rests onthe behaviour of Mr Lundy at the funeral of the deceased involving adisplay of uncontrolled grief which would have been observed by manypeople on television, and possibly regarded as acting. The appellantalso claims there should have been a demeanour direction as a result ofthe jury's request for a replay of part of Mr Lundy's video interview.(f) There was a failure to give a lies direction. This point is based uponthe suggestion made in their addresses by Crown counsel thatMr Lundy had lied to the police and/or misled the investigation into themurders.[111] Because of the conclusion that we reach on the admissibility ofthe mRNA evidence we then need to consider whether this is a case in whichthe proviso to s 385(1) of the Crimes Act should be applied on the basis that nosubstantial miscarriage of justice has actually occurred, and the trial was fair.[112] We deal with each of these issues in turn.Abuse of process[113] The appellant's argument is that the changes the Crown made to its case atthe second trial, including the abandonment of any theory of a time of death and ofthe appellant's alleged rapid car journey were radical, and amounted to an abuse ofprocess. Once the Crown's case was revealed as having been "so transformed"the proceeding ought to have been stayed as an abuse of process.[114] This argument has to be considered now in the context that no application wasmade at any stage to the trial Court for an order staying the proceeding.[115] Mr Eaton emphasised that the evidence at the first trial included:(a) a time of death for Mrs Lundy and Amber of 7.15 pm;(b) evidence of a person looking like the appellant being seen nearthe property at about that time, running down the road wearing a curlywig;(c) the successful completion by the appellant of a remarkably fast drivefrom Wellington to the family home and back in order to concoct analibi; and(d) the claim that the appellant had falsified the computer registry data inthe family computer, again for the purposes of an alibi.[116] These were all strands of the Crown case that had been abandoned atthe second trial. Mr Eaton claimed the defence team were notified ofthe Crown's change in position in January 2015, only five weeks before the trialstarting. Up until that point, the Crown had not told the defence that it intended tochange its position as to the time of death.[117] Mr Eaton submitted that the Privy Council would not have ordered a retrialhad it known that these changes were to be made. He relied on Australian authoritiesto reinforce that it would have been improper for a retrial to take place in thosecircumstances. He claimed the fact the Crown told the defence prior to the trial that itintended to change its case so significantly did not mean the relevant principles do notapply. The appropriate remedy in the present case was for the trial to be stayed.The circumstances were relied on, as we understand it, to claim that proceeding withthe trial was in the circumstances itself a miscarriage of justice.[118] Mr Eaton relied first on this Court's decision in Banks v R.135 Mr Eaton drewattention in particular to the following summary of the law:[17] In R v Samuels, this Court explained that it had not sought to evolvehard-and-fast rules about exercise of this discretion. Rather, it had followedthe "flexible" approach recommended in Reid v R, a decision ofthe Privy Council on appeal from the Court of Appeal of Jamaica.The Jamaican legislation was in similar terms to s 385(2). Their Lordshipsheld that although the principal verb (in the Jamaican legislation, "shall";in New Zealand's, "must") was mandatory, the interests of justice mightrequire balancing a "whole variety" of factors, not all of them confined tothe interests of the defendant and the prosecution in the particular case.[18] Reid was an identification case in which, the appellate courts agreed,the jury verdict was unreasonable having regard to the unsatisfactory state ofthe evidence at trial. The Privy Council held that a retrial ought not to havebeen ordered, because to do so was to give the Crown a chance to fill the gapsthat the first trial had revealed in its case. Their Lordships acknowledged thepublic interest in bringing the guilty to justice. That consideration wouldprevail where an appeal had succeeded from some error of the trial judge. Butanother consideration was the principle that it was for the prosecution to makeout its case at trial. It would conflict with this basic principle "if a new trialwere to be ordered in cases where at the original trial the evidence which theprosecution had chosen to adduce was insufficient to justify a conviction byany reasonable jury which had been properly directed." In such a case a retrialshould not be ordered, save in exceptional circumstances, for to do so wouldbe to give the prosecution a second chance to make out its case.135 Banks v R [2015] NZCA 182.It is not in the interests of justice as administered underthe common law system of criminal procedure that the prosecutionshould be given another chance to cure evidential deficiencies inits case against the defendant.(Footnotes omitted.)[119] Mr Eaton relied on four decisions of the High Court of Australia, the first beingKing v R, where Dawson J, writing for the majority, said:136It is well established that the discretion to order a new trial should not beexercised when the evidence in the court below was not sufficiently cogent tojustify a conviction or to allow the Crown to supplement a case which hasproved to be defective. In particular, the Crown should not be given anopportunity to make a new case which was not made at the first trial:R v Wilkes. If the verdict against King in this case was inconsistent with theverdict in favour of Matthews, then the Crown could properly succeed againstKing upon a retrial only by putting a new case.In that case, two men had been jointly charged as principals with the murder ofthe wife of one of them. The Crown alleged that the husband, as an accessory beforethe fact, had procured the other man to do the killing. Summing up, the trial Judgedirected the jury that it was not open to them to bring in different verdicts in relationto the two men. The prosecutor objected, the Judge withdrew that direction andinstead told the jury that they could convict the husband if they were satisfied that hehad arranged that someone should kill his wife, even if it was not the other mancharged. The husband was convicted and the other defendant was acquitted.[120] The husband's conviction was set aside on appeal, and a retrial ordered.The majority held that the redirection by the Judge at the request of the Crown hadinvolved such a change in the course of the trial at a late stage that the husband'sconviction could not stand and that a retrial had been properly ordered rather thanentry of a verdict of acquittal. The retrial was appropriate, because the two verdictsgiven by the jury were not inconsistent. For present purposes, it is also relevant toconsider the reasoning of the minority in King. Deane J wrote for a minority ofhimself, Mason and Murphy JJ. They concluded that a new trial should not have beenordered: the verdicts of the jury showed they had convicted the husband on the basisof the case which the trial Judge left to them, but which the Crown had not sought to136 King v R (1986) 161 CLR 423 at 433.make against him and which he had no proper opportunity of meeting. Deane Jwrote:137To order a new trial in these circumstances would be to give the Crown asecond chance to secure a conviction of the applicant for the murder of hiswife, after its election to present its case in a particular way at the trial shouldhave led to an acquittal. The Crown could not have challenged such anacquittal by seeking to re-try the applicant and present its case in a differentway. It should not be permitted now to achieve a similar result when the firsttrial would have resulted in an acquittal, had it not been for the Crown'swrongful conduct in persuading the trial judge to leave a different case tothe jury. Put differently, the jury's verdict of guilty against the applicant onthe basis of a Crown case which was not raised against him until the evidencehad been completed, addresses had been made and the summing up was inprogress was a miscarriage of justice.[121] The majority referred to R v Wilkes.138 In that case, Dixon J said that he wouldhave taken the same course as the Court of Criminal Appeal in refusing to order anew trial in the circumstances of that case. He said:139I would have done so because it would necessitate the presentation by theCrown either of the case on which the accused had substantially been acquittedor of a new case which had not been made at the first trial, a case moreoverwhich, I should have thought, was highly improbable and a desertion ofthe assumptions which the jury's previous verdict seems to require.[122] Mr Eaton also drew our attention to Jiminez v R, where it was held:140 as a general rule, a new trial should not be ordered to enable the Crown tomake a new case at a second trial. In the present case, a second trial wouldallow the Crown to make a case different from that which it put to the jury atthe first trial.[123] In Jiminez the appellant was charged with a driving offence. His vehicle failedto take a curve and collided with a tree, killing a passenger in the car. He told apoliceman who attended the scene that he had fallen asleep. The High Court held thatthe trial Judge had failed to direct the jury that if the appellant fell asleep, his actionswhile he was asleep were not voluntary and could not amount to driving in a dangerousmanner. It allowed the appeal and quashed the conviction. On the question of whetherthere should be a new trial the Court decided that would be inappropriate. McHugh J,137 At 429–430.138 R v Wilkes (1948) 77 CLR 511.139 At 518.140 Jiminez v R (1992) 173 CLR 572 at 590 (footnotes omitted).in a separate judgment relied on the principle that a new trial should not be ordered toenable the Crown to make a new case at a second trial. However, the other six Judgesrested their decision that there should be no new trial on the time that had elapsed sincethe accident occurred, the fact that the case against the defendant was not a compellingone, and there was evidence to the effect that he had had adequate sleep beforecommencing to drive and had no prior warning that he was about to fall asleep, as wellas other considerations.[124] Finally, the appellant relies on Parker v R, where the High Court held thatthe Court of Criminal Appeal's decision to order a retrial was wrong because:141 it would be unfair to the appellant to order a new trial in which he wouldhave to meet a significantly different case to that the jury were asked toconsider.[125] In Parker, a new trial had been ordered by the Court of Criminal Appeal inresponse to what was held to be a serious misdirection by the trial Judge concerningthe operation of a statutory provision providing that when a person received moneywith a direction that it be applied to any particular purpose the money was deemed tobe the property of the person from whom the money was received until the directionhad been complied with. The Court of Criminal Appeal directed a new trial butthe High Court of Australia allowed an appeal against that order. One of the groundsfor doing so was that on a retrial, the appellant would be called upon to meet a quitedifferent case to that presented against him at the trial. In essence, that turned onthe ability of the Crown in the circumstances to assert that the relevant statute hadbeen breached on different grounds from those relied on at the trial.[126] We consider that Banks, Reid and the Australian authorities on which Mr Eatonrelied, are very different from the present case. First, they are cases aboutthe circumstances in which a retrial should or should not be ordered. In the presentcase, Mr Lundy was retried because the Privy Council directed that should occur, andthe experienced counsel who acted for him in respect of the second trial at no stagesought a stay. We do not accept the cases articulate principles to be applied once aretrial has been ordered. Once that order has been made the questions to be confronted141 Parker v R (1997) 186 CLR 494 at 520.are those raised by the statutory grounds of appeal: in this case, the issue is whetherthere has been a miscarriage.[127] Second, the basis upon which the appeal was allowed was the new evidencethat the Privy Council allowed to be called before it, which it was satisfied called intoquestion key aspects of the Crown case. As has been seen, those concerned the timeof death, the origin of CNS tissue on Mr Lundy's shirt and the evidence concerningtampering with the computer. We consider it is implausible to suggest, as Mr Eatonappeared to argue, that in ordering a retrial the Privy Council would not haveanticipated that the Crown case presented at the retrial would endeavour to take intoaccount the criticisms that had brought the Privy Council to the view that the originalconviction was unsafe. We know that is the case of course with respect tothe CNS tissue because the Privy Council noted in the judgment that the admissibilityof the CNS tissue evidence could not at that stage be resolved.142 But in respect of theother issues that led to the Privy Council allowing the appeal, it must havecontemplated the Crown giving further consideration to and possibly not advancingaspects of its case which had been criticised in evidence called beforethe Privy Council. The retrial cannot properly be described as unfair or an abuse ofprocess solely on the basis that the Crown abandoned unsatisfactory aspects of its caseafter the Privy Council judgment.[128] Next, and importantly, although some aspects of the Crown case changed, theessential elements of it remained. The Crown's case on the CNS tissue did not change;it was strengthened by further expert evidence, but that is not the point for presentpurposes. It remained the Crown case that Mr Lundy killed his wife and daughter,travelling from Wellington to Palmerston North and back to do so. It remainedthe Crown case that the murder weapon was a heavy and sharp implement that lefttraces of orange and blue paint near the points of impact and embedded into a fragmentof Mrs Lundy's skull. The fact that the Crown no longer asserted a particular time ofdeath was also not in our view a change that could be said to give rise to any abuse ofprocess. While we accept that the Crown case changed and the defence was requiredto confront new scientific evidence it cannot be said the defendant on the retrial had142 Privy Council judgment, above n 1, at [139].to confront a different set of allegations, requiring him to mount a significantlydifferent defence.[129] Further, unlike Reid and King, this is not a case where the Crown had notproduced sufficient cogent evidence to secure a conviction at the first trial, and a retrialwould enable it to fill gaps. Unlike Parker, a retrial would not enable the Crown toadvance a different interpretation of the statutory basis of guilt. And unlikethe situation addressed by Dixon J in Wilkes, the retrial did not require the Crown topresent a case on which the defendant had been substantially acquitted, or a new casewhich had not been presented at the first trial.[130] Finally, the seriousness of the crimes alleged in this case militates against aclaim of abuse of process based upon the time that has elapsed since the murders werecommitted and the period for which Mr Lundy has already served. There was a verystrong public interest in the retrial proceeding and, in the circumstances, the absenceof any application for a stay is both understandable and telling.[131] We reject this ground of appeal.Admissibility of the IHC evidenceThe evidence at trial[132] As has been seen, the admissibility of evidence about IHC testing on whichthe Crown sought to rely in establishing that CNS tissue was present on Mr Lundy'sshirt was not challenged at the first trial, which proceeded on the basis of adefence concession that it was CNS tissue. It was only in the Privy Council thatthe validity of the technique was called into question. The issues then raised wereconsidered to be of such significance that they formed part of the basis on whichthe Privy Council allowed the appeal.[133] The defence was therefore fully alive to any criticisms that could be made ofthe IHC evidence. However, except for issues raised concerning Dr Miller's slidesand the submission that IHC was being used in different circumstances from its usualapplication, there was no direct challenge to the results obtained by Dr Miller atthe pre-trial stage. Indeed, as is apparent from the summary of Kós J's judgment givenabove, experts called for the defence at that stage agreed with the witnesses on whomthe Crown relied that the IHC process had established the presence of CNS tissue onthe slides. As we have noted, there was no appeal from Kós J's ruling that the IHCevidence was admissible.[134] Further, the same defence witnesses, when they gave evidence at the trial,reaffirmed their opinions that CNS tissue was present on the slides. Before addressingtheir evidence at the trial, and the issues now sought to be raised on appeal, it will behelpful to review the evidence of the witnesses called by the Crown on this issue.The principal witness was Dr Miller, who had given evidence at the first trial. In hisevidence at the second trial, Dr Miller explained his use of IHC and how it had beenapplied to the slides that he made after processing the strips of material taken from MrLundy's shirt and fixing them in paraffin blocks from which slices were subsequentlyable to be taken for analysis by him and other experts involved in the case.[135] He explained that the slides were first subjected to two stains calledhematoxylin and eosin (H & E) dyes, a technique used before the advent of IHC.The dyes stain different areas of a cell with different colours. He explained that bythis method, pathologists were able to make the majority of their diagnoses bystudying tissues stained with H & E under the microscope. However, that techniquewill not detect the presence or absence of specific different types of proteins found indifferent parts of the body. Various proteins and other components of cells areascertained in the IHC process by the use of probes or antibodies that are specific toparticular targets, that is, they enable the pathologist to stain for the presence orabsence of different types of target.[136] In this case, after examining slides stained with H & E a number of other stainswere applied in the IHC process so as to determine the type of tissue that was presenton specimens 3003/3 and 3003/4. The process involved first applying stains specificto skin, kidney and liver: a positive result would indicate that the slides did not containbrain tissue. In this case, negative results were obtained. Next, brain specific markerswere employed, called GFAP, synaptophysin, S100 and neurofilament.The application of these proteins was consistent with the material on the slide beingCNS tissue. Dr Miller's conclusion for reasons which he fully explained in hisevidence-in-chief was that the tissue on both specimens 3003/3 and 3003/4 was brainor spinal cord tissue.[137] That was essentially to repeat the evidence that Dr Miller had given at the firsttrial. However, his evidence at the second trial also covered further work that he hadcarried out in 2014. That work involved smearing brain tissue obtained fromthe Dallas Medical School on to a shirt (the Miller shirt) and taking strips fromthe shirt which were subjected to exactly the same process as had been carried out withthe strips of material taken from Mr Lundy's shirt after it had been allowed to age fora period of 28 days. Mr Eaton was highly critical of the circumstances in whichDr Miller had obtained the brain tissue used in this further work and sought to rely atthe hearing of the appeal on an affidavit provided by Mr Mike Ware. Mr Eaton claimedon the basis of the affidavit that the brain tissue had been procured in breach of relevantethical and regulatory rules by which Dr Miller was bound in the United States.The suggestion was made that this, together with other material which suggestedDr Miller was partisan, should cause this Court to doubt the objectivity with which hehad approached the case. However, the steps he took to obtain the brain tissue do nothave a logical bearing on the results that Dr Miller achieved and, as we willdemonstrate there was broad agreement at the trial that the analytical methods headopted had shown the presence of CNS tissue on the strips taken from Mr Lundy'sshirt. In the circumstances we do not find Mr Ware's affidavit cogent and declineleave to adduce it.[138] Dr Miller also took slices from the paraffin blocks for C3003/3 and C3003/4which had been brought back to Dallas by Detective Senior Sergeant Hughes for thepurpose of the further work Dr Miller intended to carry out. The IHC process wasthen repeated in respect of slices from C3003/3, an exercise carried out in the presenceof an expert retained by the defence, Dr Anna Sandiford. In respect of the Lundy shirtmaterial the results obtained were consistent with those previously obtained and theresults using the Miller shirt were identical except for one result that involvedapplication of a new stain. As Dr Miller summarised the position, he obtained thesame negatives that he would have expected when testing normal CNS tissue, and thesame positives from the four markers used back in 2001.[139] He repeated this exercise after the shirt had been left for periods of 95 days,when he obtained an identical outcome. He again repeated the exercise after 159 days,a period selected because it approximated the time between the deaths of Mrs Lundyand Amber and Detective Grantham's provision of the Lundy shirt to Dr Miller on4 February 2001. The stains apparent after 159 days were "virtually identical" to whatwas shown after 95 days. Finally, Dr Miller repeated the process after one year hadelapsed, and the results were again "identical".[140] The cross-examination of Dr Miller focused mainly on the potential forcontamination of the original samples from Mr Lundy's shirt during the original IHCprocess. Mr Hislop also put to Dr Miller that the standards observed in his laboratorywould be less rigorous than those applicable for forensic as opposed to diagnosticpathology. Dr Miller responded by noting that while forensic pathology was directedat different matters than diagnostic pathology, in this case the main issue presentedwas whether there was tissue on Mr Lundy's shirt, and if so, tissue of what type.He said that diagnostic pathology and IHC was "perfectly suited to that task", andgave the opinion that his laboratory would have much more experience dealing withthat kind of problem than many forensic laboratories. He did, however, acknowledgethat more stringent standards might apply in the forensic pathology setting in terms ofissues such as cleanliness and sterilisation. But he expressed the opinion that it wasclear from looking at the "Lundy slide" that there was no way it had been contaminatedfrom anything that occurred in his laboratory.143 He emphasised in answer to anotherquestion which had been based on the presence of DNA on C3003/3 which couldpossibly be sourced from two people:The immunohistochemistry shows that it's brain tissue. Even if there's otherRNA, DNA, whatever there, that doesn't change the fact that it's brain tissue.Someone could have coughed on the specimen and had their DNA there butthat does not change the fact that there is brain tissue on that shirt which isabsolutely diagnostic by [immunohistochemistry].[141] The Crown also called evidence from Dr Gown, the Medical Director andChief Pathologist at PhenoPath Laboratories in Seattle. He said that his laboratoryhad been in the forefront of developing methodologies and new reagents for143 Dr Miller's reference to the Lundy slide was to the slide referred to by the Privy Council as"the ESR slide" and referred to elsewhere in this judgment as the "dab slide".IHC applications to clinical specimens. It was his evidence that from the 1990s, IHChad developed from being a research technique to a standard clinical laboratorytechnique now used in virtually every pathology laboratory around the world. He hadcarried out his own IHC testing on slides provided to him in February 2014 (byDetective Hughes and Dr Sandiford). These were slides cut from the paraffin blockmade by Dr Miller, which he referred to as the "Rod Miller slides". Dr Gowndescribed the methodology he used, employing proteins which he knew would beexpressed exclusively or at high levels in the central nervous system as well as controlantibodies for proteins which ought not to be present in the central nervous system.Using the latter, he obtained no reaction from any of the slides. However, usingneurofilament, GFAP, synaptophysin and S100, all yielded a positive signal. It washis evidence that the tissue was cellular material: and, furthermore, that these cells were almost certainly from the centralnervous system since those glial fibrillary acidic protein and neurofilamentstogether are present only in the central nervous system.[142] He continued:Immunohistochemistry lets you look inside the cell and based upon the patternof the reactivity you can often tell what subset of the structures the targetantigen is in. So, for example, some proteins are only found in the nucleusand if that's the case one would expect to see structures that look like nuclei.In the case of these, or at least two of them, the neurofilaments and the GFAP.[143] Asked to explain what he had observed to conclude that he was looking atthe GFAP protein, he said:It's specifically that, that protein, again it localises to these, these normalstructures to filaments. I could actually see filamentous structures to whichthis antibody was binding, they were inside the cells that I was looking atunder the microscope.[144] Again, in the case of synaptophysin, he gave evidence of having seen astructure within the cell which was consistent with what he would expect for acentral nervous system cell. Based on his own work, the tissue on the slidesrepresented CNS tissue "unequivocally".[145] In cross-examination, Mr Hislop put to Dr Gown a statement by adefence expert, Dr Smith, with which he agreed. The statement was:While there can be no doubt that the tissue on the t-shirt is CNS tissue I cannotcomment on the species or sex of the material as this is out [of] the realm ofhistopathology and immunohistochemistry.[146] Mr Hislop also put to Dr Gown a statement by another defence expert,Professor Ironside. He agreed with Professor Ironside's statement that:The small size of the tissue fragments make it impossible to determine whetherthis CNS tissue is from brain or spinal cord. Both are possibilities and hencethe generic term CNS tissue is preferred.[147] Dr Gown also agreed with Professor Ironside's statement that none ofthe antibodies used was specific for human tissue, and so the tissue could not beassigned to human origin.[148] A further witness called by the Crown on this issue was Professor Daniel Brat,Professor and Vice Chair of Pathology and Laboratory Medicine at Emory University,in Atlanta, Georgia. Professor Brat is a neuropathologist. He has published textbookson diseases of the brain and has also published on the appearance of normalbrain tissue. He reviewed slides sent to him by Dr Miller in March 2014. There weretwo sets of slides, one labelled as having come from Mr Lundy's shirt (C3003/3) andanother set of slides which came to him labelled as from the "Miller shirt at 28 days".There was an H & E stain slide, and a series of IHC stain slides.[149] Having examined the slides, Professor Brat concluded that they displayedCNS tissue, a conclusion as to which he was in no doubt. Only one question was askedin cross-examination, to clarify that when he had referred in his evidence to "brain"he meant to say it was "brain or spinal cord".[150] Another witness who gave relevant evidence on this issue was Dr du Plessis,consultant neuropathologist and Clinical Lead of the Department of CellularPathology at Salford Royal Hospital, Manchester. Dr du Plessis was contacted by theNew Zealand police in the latter part of 2013 to advise on the nature of the materialsmeared on Mr Lundy's shirt.[151] The work carried out by Dr du Plessis had three strands. First, he reviewedDr Miller's IHC work carried out in 2001, the affidavits of defence experts filed inthe Privy Council and in addition the further work carried out by Dr Miller in 2014.Second, Dr du Plessis carried out his own IHC work using a series of slides preparedfrom the slices off the paraffin blocks in which the strips from Mr Lundy's shirt sleeveand pocket had been preserved. Third, Dr du Plessis used electron microscopy, aprocess in which he also had expertise.[152] Dr du Plessis said he was impressed by the quality of the IHC work carried outby Dr Miller in 2001 and did not have any quarrel with the quality of the results.He said he could interpret the slides with ease and considered Dr Miller's observationswere correct and in accordance with his own. In one respect, his analysis went furtherthan that of Dr Miller. Dr du Plessis described Dr Miller's approach as involving ananalysis of what stains had produced a positive and which had produced a negative forthe presence of CNS tissue. As a neuropathologist, Dr du Plessis said he was able tolook at: whether the actual discrete quality of the staining tells me more aboutwhether this is central nervous [system tissue] or not, and in that respect I feltthat these stains were even more informative, it's not just that it was positiveas you would expect it but the quality of the staining was entirely typical ofcentral nervous system tissue.[153] Dr du Plessis said that the 2001 work could possibly be criticised on the basisthat there had been no attempt made at the time to age known brain tissue and test thestaining on that. That would not normally be done in a routine diagnostic environmentbut in this case, given the delay between when the tissue might have been transferredonto the shirt and the carrying out of the tests, some effort needed to be made to obtain"control tissue" subject to degradation, to see whether that would adversely affectthe quality of the staining. That was done in the further work carried out in 2014, bothby Dr Miller and by Dr du Plessis himself.[154] As to Dr Miller's 2014 work, Dr du Plessis confirmed that the same results hadbeen obtained as in 2001, and they were "as persuasive as before, only difference beingthat [Dr Miller] has now added controls subject to degradation".[155] As to his own IHC work, he used some stains that were the same and some thatwere different from those employed by Dr Miller. Two of the antibodies, CD31 andCD34 were different. Dr du Plessis employed them to test objections that had beenmade in the Privy Council that the tissue analysed appeared not to be supplied withblood vessels. Both CD31 and CD34 are proteins associated with blood vessels. Heconsidered that these clearly demonstrated that there were blood vessels within thetissue. This took the Crown's case further than the conclusion of "probable blood"which it had previously been in a position to claim and which had been consideredproblematic by experts who gave evidence at the Privy Council.144[156] In summarising the results he achieved, Dr du Plessis said: those markers that corresponded to markers that Dr Miller used gaveexactly the same result. Some additional markers gave a result, as you wouldexpect in central nervous system tissue. So the collective result was the samesupplemented by additional features confirming what we call a phenotype orappearance, collective appearance confirming that this is central nervoussystem tissue.[157] Dr du Plessis also carried out tests to ascertain what happened to human brainleft at air temperature over various periods of time. He did this withsurgical specimens of human brain which he was able to obtain, taking care to ensurethat he used viable tissue unaffected by necrosis or dying off. These specimens ofbrain were smeared on to various surfaces: some on glass slides, others on fabric andthese were air dried and then left at room temperature without the use of any otherpreservation techniques. Further testing was then done at intervals of a month, twomonths and six months. IHC and electron microscopy were carried out on the air-driedbrain specimens. He summarised the result of the IHC work as follows:Consistently the same results were obtained for immunohistochemistry so itkept on being stained like normal brain or central nervous system tissue. Allof the proteins or markers that we've just discussed gave positive results orappropriate — some of them negative but were appropriate. And the qualityof staining was discrete, anatomically discrete quality of staining wasmaintained throughout [158] He explained that the lack of deterioration of the brain tissue was due tothe air-drying process.144 As noted in [35] above, the Privy Council recorded that Professor Gatter and Dr Squier had saidthat if the tissue was brain tissue they would have expected to find red blood cells and none werepresent.[159] Dr du Plessis' work also included electron microscopy. For this purpose, heobtained from Dr Miller further sections of tissue from C3003/3 which had not beenput on to slides, and were five microns thick. He described these as "spare sections"or "shavings". He processed these spare sections so as to make them suitable forelectron microscopy. He was anticipating that, because of the processing involved inembedding material into a paraffin block, he might not be able to discover anything inthe electron microscopy process. He referred to it as "just about as un-ideal a specimenas possible". Despite that, he was able to find structures that were consistent withmyelin in regularly stacked layers of membrane. Myelin is a fatty sheath produced byoligodendroglial cells in the central nervous system that envelopes nerve fibres andfacilitates electrical transmission along the fibres.145[160] Overall, Dr du Plessis considered that Dr Miller's work and his own workconstituted "overwhelming, incontrovertible, incontestable" evidence that the tissueon Mr Lundy's shirt was CNS tissue. There was "exceptionally robust evidence insupport of that despite the effects of degradation".[161] Dr du Plessis also carried out work for the purpose of ascertaining a possibleexplanation for the degradation of the dab slide taken from the sleeve (C3003/2)compared to the well-preserved counterpart tissue in Dr Miller's paraffin block.He was satisfied as a result of the work that he described in evidence thatthe explanation lay in the methodology that had been employed by Mr Sutherland inmaking the dab slide. He largely followed the process adopted by Mr Sutherland andsaid that it was remarkably easy to obtain material on the dab preparation which wasunrecognisable as brain. The brain looked either like dead or necrotic tissue or wasjust non-descript "muck" similar to that seen on the original dab slide.[162] In cross-examination, Mr Hislop referred to Professor Whitwell's affidavitfiled for the purposes of the Privy Council appeal. Mr Hislop quoted her observationthat the dab slide comprised possible cellular material, but it was not possible todetermine the nature of the cells. Dr du Plessis agreed.145 Dr du Plessis' IHC work was also able to demonstrate the presence of myelin, which had not beendemonstrated in the work carried out by Dr Miller.[163] Cross-examination of Dr Miller and Dr du Plessis on the results ofthe IHC work was very confined. In the course of asking one of two questions onthe subject, Mr Hislop told Dr du Plessis that the defence did not dispute that therewas CNS tissue on the sleeve and chest of Mr Lundy's shirt. The two questions weredesigned to procure agreement that it was not possible to say whether the tissue washuman or non-human, or whether it had come from a male or female source.Dr du Plessis agreed that was so.[164] The main emphasis of the cross-examination was on the comparison betweenthe dab slide made 58 days after the murders and the strips of material taken fromthe shirt 169 days after them. Dr du Plessis agreed that the difference between thedab slide and the tissue preserved in the paraffin block was stark; the former could notbe recognised as brain, the other could be recognised as CNS tissue. He agreed thatone possible explanation was that C3003/2 was a different substance to C3003/3 andC3003/4. He said that water exposure was not sufficient to explain the difference.The better explanation was the mechanical disruption to the tissue caused by makingthe slide itself.[165] Mr Hislop then proceeded to put the defence case, based on the evidence to becalled from Dr Colin Smith and Professor Ironside, that questioned Dr du Plessis'ability to replicate the process followed by Mr Sutherland in making the dab slide;that the tissue in the dab slide was necrotic and that would mean it was a differentsubstance from that analysed in C3003/3 and C3003/4. While Dr du Plessis acceptedit might be necrotic he also reiterated it could be post-mortem degraded brain,degraded by the dab slide process. Further, there was nothing necrotic found in whathad been embedded in the fabric; this meant that what was on the dab slide would bea "unique part of the specimen". Also, it would imply that if the source was one ofthe deceased she would have had to have "some problem with their brain".The relevance of this cross-examination was the defence theory that Dr Miller andDr du Plessis may have been analysing tissue that had become contaminated at somepoint after the dab slide was taken and prior to the completion of the paraffin blocks.It did not go to the question of whether the tissue placed in the paraffin blocks was infact CNS tissue.[166] Mr Hislop also questioned Dr du Plessis about his evidence concerning myelin,and the possibility that what he had observed was rather "myelin figures", whichresemble or mimic myelin. He conceded he could not say beyond a reasonable doubthe had observed myelin, but he "strongly favoured" that being the case and listedvarious reasons why that was so. He rejected the opinion of Dr Smith that what wasshown in the image could not be said to be myelin but apparently acceptedProfessor Ironside's opinion that more would be required for publication of that viewin a peer-reviewed scientific journal. He added, however, that he had not previouslyseen "myelin figures" of a similar quality.[167] In re-examination, Dr du Plessis confirmed that apart from skin flakes all thathad been found amongst the shirt fibres was CNS tissue, not showing any alteredappearance indicating necrosis. He continued: it would be quite remarkable if there's a mixture of material but just onetype of substance lay on this surface, was perfectly separated from the restenmeshed in the shirt fibres because you get this perfect separation. I meanthat must be unusual that you get that. Secondly, that Mr Sutherland managedto perfectly and cleanly scrape off this other substance leaving no residuewhatsoever in the fabric. I suppose not utterly impossible but it's, I thinkthe odds of that must be pretty low.[168] He repeated his view that it was much more likely that the dab slide contained"artifactually altered" CNS tissue and the alteration was due to the way the dab slidehad been prepared.[169] The defence witness Dr Colin Smith was employed as a Reader in Pathologyand an Honorary Consultant in Neuropathology at the University of Edinburgh. Hewas called as the next witness following Dr du Plessis, out of turn. He had listened toDr du Plessis' evidence. The main purpose of Dr Smith's evidence was to underlinethe difference between the level of preservation in the appearance of the dab slideC3003/2 and the tissue in Dr Miller's slides C3003/3 and C3003/4. He said that hecould not offer a biological explanation for the difference and did not accept that ithad been adequately explained by Dr du Plessis. He was also critical of Dr du Plessis'finding of the presence of myelin.[170] In cross-examination he confirmed that the state of C3003/2 was such thatnothing could be said about it other than the fact that it was cellular tissue. He alsoconfirmed however that he was satisfied that the IHC work he had carried out on slidesmade from Dr Miller's paraffin blocks established without question that the slices fromC3003/3 and C3003/4 were CNS tissue. He confirmed that the dab slide addednothing to his opinion that what was on the polo shirt was CNS tissue. He alsoconfirmed his agreement with Dr du Plessis that if the polo shirt stain was the resultof contamination with a foodstuff, it was odd that the substance analysed containedonly CNS tissue. There was this exchange:Q. And the notion, and that fact that only central nervous system tissueis present just is completely against the notion that this is some foodcontaminant, is that right?A. Yes, to my mind, yes, I'd agree.[171] Questioned by the Judge, Dr Smith confirmed that because it was impossibleto tell what substance was on the dab slide, it was of little utility. There was thisexchange:Q. The dab slide, to tell us anything, you need to make an assumptionotherwise it doesn't tell you anything?A. Well, it, it doesn't, which is why I've made the point, to my mind it'sacademic, but I fully accept that — we seem to have spent a lot oftime trying to say well there's CNS tissue on the shirt therefore thisdab slide must represent CNS tissue because that's where it's comefrom. That may well be true, but equally if it's suggested there'ssomething else on the shirt at the time the dab preparation is made thatmay well be true. So the dab preparation is not really advancing eitherof these theories because it is so degenerate.[172] Professor Ironside was the next witness called. As we have noted, he is theProfessor of Clinical Neuropathology in the University of Edinburgh, the editor of aleading textbook on neuropathology and clearly eminent in his field. Near the outsetof his evidence, he recorded his agreement that C3003/3 and C3003/4 comprisedCNS tissue. That opinion was based on IHC performed by Dr Smith, which he theninterpreted independently, coming to the same view as Dr Smith. He described thedab slide as containing "very degenerate cellular material" which he interpreted asbeing necrotic cells. The appearance suggested to him that it contained cellularconstituents that had been dead at the time of sampling. This tissue had a very differentappearance microscopically to the tissue in C3003/3 and C3003/4 which was notnecrotic. He said it was not possible to say with any degree of certainty what the tissueon the dab slide originally came from.[173] Professor Ironside also confirmed that the tissue in C3003/3 and C3003/4 waspossibly human, but it could also originate from other mammalian species. He referredto the possibility that the tissue could have originated from food, giving as an examplemeat products that from time to time contain CNS tissue.[174] While he considered that Dr du Plessis had done his best to recreate thecircumstances and methodology of the preparation of the dab slide, the details givenby Mr Sutherland were insufficient to allow an accurate recreation of thosecircumstances. He raised the possibility that what was on the dab slide had notoriginated from the brain or central nervous system: it was impossible to tell. He didnot consider that the electron microscopy or other work carried out by Dr du Plessiswas sufficient to establish that the necrotic material on the dab slide was of centralnervous system origin.[175] Mr Morgan QC, counsel for the Crown at both the trial and the present appeal,emphasised in cross-examination the matters on which there was agreement betweenthe experts called on both sides. Professor Ironside first confirmed that the IHC workdone with the paraffin embedded slides samples in C3003/3 and C3003/4 establishedthey contained CNS tissue, whether it be brain or spinal cord, and nothing else.Second, the dab slide showed degraded tissue, degraded to the point that the materialon the dab slide could not be identified. Third, although Dr du Plessis had establishedby carrying out electron microscopy there might be myelin on the dab slide, a definiteconclusion to that effect could not be made.[176] In one part of the cross-examination Mr Morgan obtained Professor Ironside'sagreement that one explanation for the condition of the dab slide was that tissue otherthan CNS tissue had become placed over the CNS tissue initially on the shirt and hadthen been cleanly removed during the dab slide process. In this scenario, the necrotictissue on the dab slide would have come from the same area on the shirt as theCNS tissue, but removed entirely in the dab slide process leaving only CNS tissuebehind on the shirt. Another possibility was that identified by Dr du Plessis, that afterremoval from the shirt the tissue was somehow rendered necrotic by the processesused in the dab preparation. This would mean it was the same tissue as that preservedin the paraffin block. Professor Ironside considered both of these possibleexplanations were less than satisfactory.[177] In re-examination by Mr Hislop, Professor Ironside agreed that athird possibility was that something happened to the material cut from the shirt andplaced in an envelope in the period between the making of the dab preparation andthe time when Dr Miller made the paraffin blocks.[178] In the result, both Crown and defence experts agreed that the tissue in C3003/3and C3003/4 was CNS tissue. There was unanimity on this point, and that it had beendemonstrated to a high level of certainty by the IHC analysis carried out by Dr Millerand Dr du Plessis, and replicated from the material in Dr Miller's paraffin blocks byDr Smith performing his own IHC analysis. There was uncertainty about the reasonsfor the absence of recognisable CNS tissue on the dab slide, and the condition of thatslide compared with the tissue preserved in the paraffin blocks. It was this aspect ofthe evidence that the defence pursued in closing to the jury, not any alleged limitationsof the IHC evidence or its inherent frailty as a means of establishing the presence ofCNS tissue.[179] Thus Mr Hislop, addressing the jury, emphasised that the neuropathologistshad agreed "the difference between the dab slide and the Miller slides [was]remarkable". This was important, because it raised the issue of possiblecontamination, at some point after the dab slide was taken and prior to the completionof the paraffin blocks. Mr Hislop reminded the jury that Dr du Plessis had acceptedthat one possible explanation for the difference was contamination, as hadProfessor Ironside. The latter had also discounted the possibility that, as Dr du Plessissuggested, the method by which the dab slide had been made could explainthe difference in preservation rates. So, "what happened after the dab slide", whichhad been taken at a time when "nobody knew what was on that shirt":What happened? Well we know that those fabric pieces sat inOfficer Grantham's drawers, not in the exhibits room No one's told us why,why it wasn't kept with the other exhibits but it sat there, didn't it? It sat therefor days, it sat there for weeks, it sat there for months and we don't know why.[180] To this were added further rhetorical questions concerning what happed inDr Miller's laboratory. The fabric pieces had been left to sit on a bench, and there wasno evidence about the cleanliness of the bench. What happened when the specimenswere taken to St Paul's hospital — when they were put in a machine with otherunknown specimens? What happened when they were returned to Dr Miller's lab?Mr Hislop told the jurors that when asking these questions they should ask: how is it that the states of the two are so different yet one is so much olderthan the other You're entitled to ask yourself about the issue of integrity ofexhibits in this case.[181] This was a far cry from any suggestion that there was something about IHCthat made it unsuitable to determine that what was on the shirt was CNS tissue, or thatthere was any question that was what had been demonstrated. Simon France Jsummed up accordingly. He told the jury that the tissue on the shirt had been shownto be CNS tissue, but noted the defence case that there was something not right thatneeds explaining:The tissue on the glass slide made by Mr Sutherland from the stain looks verydifferent from the tissue in the shirt. The defence submission is that it maywell be a mistake by Mr Miller that provides the answer to why they aredifferent and what has happened is that CNS was mistakenly added at the timethe wax block was made. This mistake then just keeps getting repeatedbecause everyone is using slices from that block.[182] And later, the Judge referred to the defence reliance on evidence about poorpractices at Dr Miller's laboratory which were not up to the standards that would berequired of a forensic laboratory. It will be recalled that Mr Hislop hadcross-examined Dr Miller about these matters which had also been the subject ofevidence from the defence witness Dr Vennemann who had attended Dr Miller'slaboratory over two days in July 2014. This was counter-balanced by reference tothe Crown's response on these issues, which need not be referred to for presentpurposes.Appellant's submissions[183] We have given this rather lengthy treatment of the IHC issue so as to place inproper context the argument which Mr Lundy seeks to advance on appeal that we mustnow address. Notwithstanding the way the trial was run, and notwithstanding theagreement of the defence witnesses about what had been demonstrated bythe IHC evidence called by both the Crown and defence, it is now claimed thatthe IHC evidence of Dr Miller, Dr du Plessis, Dr Brat and Dr Gown should not havebeen admitted in evidence. The appellant seeks to challenge the admissibility of theopinion evidence that the IHC established the presence of CNS tissue in the two shirtstains.[184] Mr Eaton notes that despite the significant challenge mounted to Dr Miller'sevidence in the Privy Council, the further testing that he carried out in 2014 forthe purpose of addressing the concerns expressed by the Privy Council had ledthe defence not to challenge the presence of CNS tissue in both specimens atthe retrial. He acknowledged that at all times the defence had been guided by externaladvice but it was now sought to challenge the evidence. While IHC is awell-recognised diagnostic tool in a clinical setting, he claims there has been noprevious case before or since this where IHC has been admitted in a criminal case toprove the source of tissue. Nor has there been any research as to the "foundationalscientific reliability of IHC" which has nevertheless been relied on as a key plank ofthe Crown case.[185] Reliance is now placed on what was said by the Privy Council in the presentcase:146It is important not to assume that well established techniques which aretraditionally deployed for the purpose of diagnosis can be transported, withoutmodification or further verification, to the forensic arena where the use towhich scientific evidence is put is quite different from that involved in makinga clinical judgment.[186] In another passage quoted by Mr Eaton, the Privy Council also said:147146 Privy Council judgment, above n 1, at [81].147 At [82].Scientific proof such as fingerprint or DNA evidence is customarily givenagainst the background of its having been theoretically tested in, if notlaboratory conditions, at least empirical survey. The novelty of using, in acriminal trial, the type of evidence offered by Dr Miller, especially when itsreliability has not been subjected to such laboratory or empirical research,does not necessarily make it inadmissible but it prompts caution as to its rolein establishing guilt beyond reasonable doubt.[187] Mr Eaton submits that these observations mean that a criminal court must treadcarefully and with scepticism before accepting the admissibility ofIHC opinion evidence in a forensic setting. It must be recognised that evidence thatmay confidently be relied on to reach a medical verdict does not measure up to themore stringent requirements that arise in the setting of a criminal trial. Mr Eatonargues that the most fundamental argument against using IHC in forensics is that onecannot take a technique designed for a different purpose and expect meaningful resultswithout extensive testing and peer review. This testing and review had not occurred.[188] On these premises, admission of the evidence and/or the manner in which itwas dealt with at trial had given rise to a miscarriage of justice, because:(a) The evidence was inadmissible by reference to ss 7 and 25 ofthe Evidence Act 2006.(b) Dr Miller, Dr du Plessis and others were permitted to overstate thereliability of the IHC expressing the results as a 100 per cent certainty.(c) The Crown was permitted to rely on the IHC evidence without anyreservations as to its validity and therefore overstated its significanceto the jury.(d) The manner in which the evidence was given and dealt with at trialmeant the unfairly prejudicial effect of the IHC evidence outweighedits limited probative value.[189] Mr Eaton sought to justify these claims by reference to issues additional to theclaim of novelty. These included general reliability issues surrounding IHCrecognised in scientific and medical literature, the absence of validation of IHC, thesubjectivity inherent in analysing the results of IHC, the non-specific nature ofantibodies used, and the possibility of cognitive bias. Reliance was also placed on asubstantial new report published in the United States, a report to the President datedSeptember 2016, "Forensic Science in Criminal Courts: Ensuring Scientific Validityof Feature-Comparison Methods" provided by the President's Council of Advisors onScience and Technology, which we will refer to as the PCAST report.148[190] These concerns were addressed and explained in affidavits on whichthe appellant sought to rely as fresh evidence for the purposes of the appeal. One, byAssociate Professor Sheard explained that IHC is not a standardised technique anddwelt on various problems arising from inconsistency and variability of outcome.Although much of his evidence was general in nature, Associate Professor Sheard didgive specific evidence, said to be illustrative of the difficulties inherent instandardisation of IHC protocols, about anti-synaptophysin, one of the nerve-specificantigens used on tissue in the present case. He went on to explain why any two of theantibodies would provide markedly different results from one-another when processedon the same piece of tissue under the same conditions, the difficulty then being todetermine which if any is correct. Associate Professor Sheard also commented onvarious passages of the PCAST report which he considered to be of relevance to thepresent case. These included passages emphasising the need to ensure that testimonyabout the accuracy of a foundationally-valid feature-comparison method is limited towhat can validly be claimed on the basis of empirical evidence; and other passagesstressing the need for caution where a subjective analytical method is involved.[191] As to foundational validity, according to the passage in the PCAST report onwhich Associate Professor Sheard relied, this requires that a method has beensubjected to empirical testing by multiple groups, under conditions appropriate to themethod's intended use. The studies must demonstrate that the method is repeatableand reproducible, provide valid estimates of its accuracy (that is, how often it reachesan incorrect conclusion) that indicate the method is appropriate to the intendedapplication. Associate Professor Sheard also addressed the possibility of subjective148 President's Council of Advisors on Science and Technology Forensic Science in Criminal Courts:Ensuring Scientific Validity of Feature-Comparison Methods (September 2016).bias where a witness such as Dr Miller was called on to give evidence about theaccuracy and reliability of his own analysis.[192] Much of this evidence is in our view couched in terms that are too general tobe cogent. Some of it however criticises Dr Miller's approach and conclusions moredirectly, accusing him effectively of setting out to look for evidence of CNS tissue,and failing to look for or take account of counter-indications. The difficulty of courseis that none of the more particular criticisms now raised were put to Dr Miller, andAssociate Professor Sheard, known to and available to the defence since before thehearing in the Privy Council and called as a witness pre-trial was not called at the trial.Tellingly, Associate Professor Sheard's evidence provided for the pre-trial hearingincluded the following statement:Personally, I have reason to doubt the objectivity of several of the experts.However, looking at the weight and consistency of evidence I find myselflargely in agreement with most of the experts who have examined either slidesor images of the outcomes of these rounds of IHC testing. That is, it seemslikely that the tissue on the Lundy shirt is of CNS origin.[193] Another affidavit was by Mr Sean Doyle. He describes himself as aconsultant forensic scientist with a degree in applied chemistry and a certificate ofadvanced analytical chemistry. He also reviewed the IHC evidence, concluding thatthe evidence based on it did not meet "the required standards" and is consequentlyunreliable. This is apparently because the IHC method used had not been validated toa forensic standard, there is no indication of the quality of the reported result andthe evidence had been "weakened by cognitive bias". Mr Doyle referred to standardsin the PCAST report and also guidelines published by the Forensic Science Regulatorin England and Wales in 2016.149[194] As with Associate Professor Sheard, Mr Doyle had provided evidence for thepre-trial hearing before Kós J, but he was not called at the trial. Apart from additionalmaterial relating in particular to the PCAST report and the Forensic ScienceRegulator's guidelines, the substance of Mr Doyle's criticisms was before Kós J forthe purposes of the pre-trial hearing. It should be noted that Mr Doyle makes no149 Forensic Science Regulator Codes of Practice and Conduct for forensic science providers andpractitioners in the Criminal Justice System (3rd ed, February 2016).reference to the evidence of the forensic neuropathologists Dr du Plessis and Dr Smithnor to Professor Ironside. We add that neither Mr Doyle nor Associate ProfessorSheard said that Dr Miller or any of the other IHC witnesses gave evidence that isincorrect.150 In the end we did not find their evidence cogent.[195] It is most unusual that we should be presented on appeal with applications toadduce further evidence from witnesses previously engaged by the defence, but notcalled at the trial, to give evidence about matters that they have previously addressedand which, with two exceptions, cannot be described as fresh. The two exceptions arethe commentary on the PCAST report and the Forensic Science Regulator'sguidelines.[196] The Privy Council judgment in this case remains authoritative as to thecircumstances in which fresh evidence may be adduced on appeal. It summarisedthe law as follows:151120. The Board considers that the proper basis on which admission offresh evidence should be decided is by the application of a sequential series oftests. If the evidence is not credible, it should not be admitted. If it is credible,the question then arises whether it is fresh in the sense that it is evidence whichcould not have been obtained for the trial with reasonable diligence.If the evidence is both credible and fresh, it should generally be admittedunless the court is satisfied at that stage that, if admitted, it would have noeffect on the safety of the conviction. If the evidence is credible but not fresh,the court should assess its strength and its potential impact on the safety of theconviction. If it considers that there is a risk of a miscarriage of justice if theevidence is excluded, it should be admitted, notwithstanding that the evidenceis not fresh.[197] In the particular circumstances of this case there should also be reference tothe Privy Council's discussion of this Court's judgment in Wallace v R, where it wasacknowledged that a critical reliance on what proved to be "bad science" could leadto an unsafe or wrong conviction.152 The Board said that in cases involvingscientific evidence the requirement that evidence be fresh can be of "less critical150 Another affidavit on which the defence sought to rely was provided by Dr Anna Sandiford, whoexpressed reservations about the use of IHC in a criminal case but accepted thatAssociate Professor Sheard and Mr Doyle were more qualified to address that issue. In thecircumstances we do not need to deal with that aspect of her affidavit.151 Privy Council judgment, above n 1.152 Wallace v R [2010] NZCA 46 at [48].importance".153 As we understand it the reasoning is that the powerful effect on a juryof expert scientific evidence may make it necessary to receive evidence questioningits validity without punctilious adherence to a requirement that it be fresh.[198] Applying the Privy Council judgment to the evidence presently beingconsidered there is no question that it is credible. But only in so far as it refers tothe PCAST report and the Forensic Science Regulator's guidelines is it fresh. Indeed,as we have just noted, the substance of much of the evidence had previously beenplaced before the High Court for the purposes of the pre-trial hearing. It is necessarytherefore to consider the strength of the evidence and the question of what impact itsexclusion might have on the safety of the conviction.[199] Having done so we do not consider the evidence is sufficiently cogent to admitit on this appeal. The issues sought to be raised concerning the reliability ofthe IHC evidence were raised pre-trial and comprehensively addressed inthe High Court. To the extent it seeks to advance argument that the IHC evidence mayafter all be bad science, the answer that can be given is that all of those with the mostrelevant expertise agreed that the IHC evidence had established that the tissue onthe shirt worn by Mr Lundy was CNS tissue. Even Associate Professor Sheard thoughtthat had likely been demonstrated and neither he, Mr Doyle or Dr Sandiford can giveevidence to the contrary.154 In this case, the freshness criterion can be applied becauseof the full canvassing of the relevant issues that took place pre-trial. Accordingly,while we have read the affidavits for the purposes of considering whether they shouldbe admitted, we have decided not to admit them.Analysis[200] That leaves for consideration the challenge to the IHC evidence itself. Onceagain, the extent of the common ground among the experts for both the Crown anddefence as to what the IHC evidence showed is decisive. The appellant's submissionsthat IHC applied in a criminal setting is novel and that it has not been properly153 Privy Council judgment, above n 1, at [121].154 We do not consider the reference to the PCAST report is in itself a reason to admit Mr Doyle'sevidence; it is a report that focuses generally on the reception of novel scientific evidence and notparticularly on IHC. In any event, we are prepared to allow reference to the PCAST report withoutrequiring it to be produced by affidavit.validated for such use, that there are general issues affecting its reliability, that itsresults are not shown to be repeatable, reproducible and accurate so as to have proper"foundational validity", all run up against what the experts have in fact said about theIHC carried out in this case. We have already described above the relevant evidencegiven at both the pre-trial hearing and the trial. Those issues were fully ventilatedbefore the jury. Issues of contamination aside, there was common ground among theexperts called at the trial that the methodology employed by Dr Miller was sound, andthe results he achieved were able to be replicated by Dr du Plessis and Dr Smithperforming their own tests. None of the experts was in any doubt that the tissue inC3003/3 and C3003/4 was CNS tissue.[201] As matters have developed, further work has been carried out and a greaterrange of expertise has been brought to bear, we consider that the seven separate areasof controversy identified by the Privy Council about the IHC evidence andthe dab slide have been resolved, at least to the extent possible.155 There was evidenceon the consistency issue and the relevant experts were satisfied IHC was sufficientlyreliable to demonstrate the presence of CNS tissue. None of the experts who gaveevidence at the trial suggested the fact that IHC had not previously been used in aforensic context prevented a firm conclusion being reached about the presence ofCNS tissue. The issues concerning differences in staining of Dr Miller's slides werenot considered sufficient to detract from the conclusions reached about the presenceof CNS tissue and were not pursued further. The degraded condition of the tissue onthe ESR or dab slide was a matter canvassed in the evidence at the trial; it was reliedon by the defence to argue that the presence of the CNS tissue on the materialsubsequently analysed by Dr Miller and the other experts was due to contamination.It is clear that it was not possible to say anything of any real value about the materialon the dab slide and the defence has not apparently pursued the sixth and seventhissues raised by the Privy Council.156[202] In the circumstances we do not think it appropriate now to approachthe admissibility issue on the basis that the IHC was a novel technique. As notedabove, the Privy Council judgment referred to the principles set out by155 These were set out at [36] above.156 See [36] above.the United States Supreme Court in Daubert as part of its reasoning for allowingfurther evidence to be called from consultants instructed by the defence after the firsttrial. As we recorded earlier, the particular considerations derived from Daubertmentioned by the Privy Council were "whether the theory or technique can be and hasbeen tested" and "whether the theory or technique has been subjected to peer reviewand publication", known or potential error rates or the existence of standards and"whether the theory or technique used has been generally accepted".157 These weresaid to be a useful template for considering whether evidence based on IHC, whichwas described as "novel, at least, in the forensic setting of a criminal trial" should beadmissible.158 It is important, however, to note that the judgment continued:159But the debate as to whether the listed factors should operate to renderinadmissible such evidence has not been engaged — at least, not to the extentthat it can be resolved. For present purposes, it is sufficient to say that theneed for such a debate signifies the impact that it might well have on anassessment of whether there has been a miscarriage of justice and an unsafeconviction.[203] We are now well past the point at which the Privy Council made thoseobservations. The debate, such as it was, took place pre-trial and the experts qualifiedto address the issue plainly had no doubt about the efficacy of IHC to establish thepresence of CNS tissue and that is what Dr Miller's work, and the IHC carried out byDrs du Plessis and Smith had achieved. This necessarily involved the witnesses whoaddressed the issue in a consideration of the methodology employed by Dr Miller incarrying out the IHC and the robustness and reliability of the results he achieved.Drs du Plessis and Smith were also able to replicate the results he obtained. In thesecircumstances it has not been shown that the IHC evidence was unreliable. Indeed,the reverse is true: following testing and peer review the IHC methodology has beenaccepted and is able to be relied on.[204] For these reasons we consider the IHC evidence was clearly admissiblewhether addressed under ss 7, 8 or 25 of the Evidence Act. This ground of appealaccordingly fails.157 Privy Council judgement, above n 1, at [138].158 At [139].159 At [139].Admissibility of the mRNA evidence[205] We have already described the nature of the mRNA evidence on whichthe Crown relied in the context of discussing the High Court pre-trial judgment andthe subsequent appeal to this Court. The evidence was called for the purpose ofrebutting the possibility that the CNS tissue on Mr Lundy's shirt might have had anon-human source. The appellant now mounts a fresh challenge to the admissibilityof the evidence.[206] As has been seen, the analysis on which the Crown relied had been carried outby Dr Sijen, a forensic scientist employed by the NFI. She was the Crown's principalwitness on this subject. Her evidence at the trial was largely consistent with what shehad said at the pre-trial hearing. We have summarised that above and need not repeatit here.[207] The principal defence witness at the trial was Professor Bustin. As earliernoted, the defence had attempted unsuccessfully to call him for the purposes of theappeal from the High Court pre-trial judgment.160 Professor Bustin is a Professor ofMolecular Medicine at Anglia Ruskin University in Chelmsford inthe United Kingdom, having previously held various academic positions at theUniversity of Middlesex, and at Queen Mary University of London. Among his manyqualifications to give expert evidence is his co-authorship of a book on polymerasechain reaction (PCR), "A-Z of Quantitative PCR", as well as four other books on thesame subject. He has published numerous academic articles in peer reviewed journals.He is clearly a well-qualified expert in this field. The defence also again relied onDr Vennemann.[208] Their evidence was foreshadowed when under cross-examination Dr Sijenconfirmed that the brain-plex developed by the NFI for this case had not previouslybeen used, nor was it subjected to independent peer review. It was her evidence thatthe methodology or strategy had been peer reviewed, but not the exact brain-plexactually used. When used, the brain-plex was not accredited by the NFI.The sole oligodendrocyte marker, OPALIN, had provided only two positive results for160 See [88] above.known human CNS tissue out of nine tests. One of the astrocyte markers, S100B,could show positives for both brain and non-brain tissue, yet sample 3003/03 showedas consistently negative for both brain and non-brain tissue. The reactivity forthe brain-plex markers differed significantly when tested against known human brain(sourced from Dr Miller) and sample 3003/03.161[209] In addition, Dr Sijen accepted that false positives were possible betweenhuman CNS tissue and pig tissue for the ACSBG1 marker at standard annealingtemperatures. Also, if the same tests were repeated, despite the fact that the testsshowed seven positive signals out of 12, she could not guarantee that the result wouldbecome 14 positive signals out of 24. She agreed with Dr Vennemann's evidence thatthere were no internationally set guidelines for RNA typing, and no guidelines inliterature on the design of RNA assays or the interpretation of the data. The Crown inclosing was subsequently to concede that the GFAP marker was not human specific,as Professor Bustin said in evidence.162[210] It was Professor Bustin's evidence that the process of reverse transcriptionPCR (RT-PCR) involved in applying the brain-plex developed by the NFI is a highlyvariable process, providing variable results which are often unreproducible.The difficulty is exacerbated when working with small amounts of tissue, as istypically the case in forensic science and as is the case here. He endeavoured toexplain the process to the jury, but his explanation was inevitably complex.[211] As to the reverse transcription step, Professor Bustin said that the enzyme usedfor the PCR reaction is very inefficient at transcribing RNA. A way had to be foundto present the enzyme with DNA that it could recognise and amplify; the process ofreverse transcription involved taking an RNA molecule and converting it into aDNA molecule. It was his evidence that: of all the techniques that we use in molecular biology the reversetranscription step is the least successful in that it is the most variable. It is avery, very unsatisfactory method. Unfortunately it's all we have and that'swhy we use it.161 As noted above at [75], in testing the Miller t-shirt smeared with known brain tissue showed allthe brain specific markers except ACSBG1.162 Of two other possibilities he identified it would respond to were horse and golden hamster.He referred to scientific papers detailing "substantial performance discrepancies" incommercially available kits for reverse transcription. While a PCR reaction is anextremely stable and reproducible method, the same does not apply with RT-PCR.[212] He also referred to an experiment he had undertaken concerning the stabilityof RNA itself, highlighting the possibility of discordant results analysing the samemarkers. The instability cannot be predicted without extensive validation experimentson the chosen markers and in his opinion: the likelihood that you hit on four markers that are totally stable just byselecting four at random is highly, highly unlikely. So you'd have to choosea selection of markers and then spend a long time validating their use.[213] He emphasised that there were no recognised standards for the choice ofmarkers for mRNA testing, and consequently no uniformity or reproducibility oftechniques. Further, the results shown by the mRNA brain-plex markers utilised bythe NFI could not be reproduced, even across the same samples.[214] He was particularly critical of the fact the OPALIN marker used had resultedin three negative results when tested against a known brain sample. In his view, thatshould have invalidated the test from the outset. He said: to anyone familiar with RNA and RNA based work, this rings huge alarmbells this result and let me explain to you why. We have freshly cut, a freshlycut slice of brain tissue. So this is the best possible scenario for detecting anyRNA. Your assay must detect a target in what is after all a positive controlsample. But look at the result. All three replicates are negative so theirpositive control sample gives them a negative result in all three replicates.Now that alone, in my opinion, invalidates this assay.[215] Another issue raised by Professor Bustin concerned the "housekeeping gene"used in the brain-plex. As noted earlier, Dr Sijen had used a fifth marker, 18S-rRNA,to confirm the presence of RNA. Professor Bustin was critical of the fact that thehousekeeping gene did not consistently indicate the presence of RNA. These resultshad been disregarded. We infer from his evidence that he thought the gene wasinappropriate to detect small amounts of RNA, the task required of it. Professor Bustinreferred to a paper published by Dr Sijen which he claimed in fact indicated that the18S-rRNA marker was not in fact suitable for this purpose.[216] Professor Bustin complained of the false analogy drawn between mRNA andDNA for the purposes of Dr Sijen's "joint interpretation" exercise, because "RNAanalysis is considerably more difficult than that of DNA since the molecule'schemistry, stability and abundance are dissimilar". The joint interpretationmethodology did not make up for a lack of consistency and reproducibility in the NFItesting method. Here, the brain-plex had resulted in successive discordant results. Theappropriate response was not: that you can lump them together and come up with a formula that explainswhy this might be real but the question you ask is, why am I getting thisdiscordance and what do I need to do to improve my assay, my test, so that Imake concordant results?[217] In Professor Bustin's view it made no sense to lump together the results fromthree or four different markers and then allocate a 50 per cent threshold. This wasconsistent with the evidence given by Dr Vennemann.[218] Dr Vennemann described the 50 per cent threshold used by Dr Sijen as"completely random". She was asked whether she could think of any reason why50 per cent was used as a cut-off point. She answered:No, the only reason that I could think of is that they adopted this from theDNA interpretation of the use of this, of such a guideline, but for RNA, frommy point of view it is a completely random threshold.[219] Asked whether it worked for RNA in fact, she said it did not:You can decide that it works but if you want to convince others that it worksthen you have to validate it and show that it works and this is what is missing.[220] In Dr Vennnemann's opinion, the use of a joint interpretation or scoring systemtransported from DNA interpretation could not be assumed to be valid when appliedto mRNA in the absence of testing, evaluation and validation.[221] She confirmed that there are no internationally accepted standards orguidelines for the design of forensic mRNA assays, or the interpretation of data norwere there any commercially available testing systems available for forensic use.[222] She agreed with Professor Bustin's criticism of the brain-plex saying therewere a number of issues around the testing procedure. It started with the decision touse an "end-point PCR technology" and a "multiplex assay" for the purpose ofRNA detection. This is problematic because: RNA is fundamentally [different] to DNA when it comes to its stability andhow it is built within the cell and how it is destroyed within the cell, and alsoregarding issues of its abundance in different cells. So it is not necessarily avery good idea to apply a method that is very well able to deal with anyDNA issues in forensic science and simply take this and apply it to RNAdetection.[223] Other issues she raised included the efficiency of the reverse transcription step,the use of the "housekeeping gene" 18S-rRNA, the PCR conditions and the specificityof the primer sequences.[224] Dr Vennemann also noted that taking the same sample from the same sourceand testing it via the same mRNA analysis did not guarantee reproducible or identicalresults. She said: if you take the same sample from exactly the same source and test it severaltimes you might end up getting different results and that is of particularconcern. This is something that we notice with RNA only We know that inextreme situations in DNA we can have slightly different results but wefully understand why this happens. In RNA, on the other hand, we do notunderstand at all why replicates don't behave as they should, and that is amajor problem here.[225] We were told at the hearing of the appeal that Professor Bustin was carryingout further work. Because of the possibility that work might shed further light on thereliability of the RNA testing process adopted by the NFI we granted leave for a furtheraffidavit to be filed, and for the Crown to respond. We indicated that after consideringthe affidavits and any submissions that accompanied them, we would decide whetherit would be necessary to hear counsel at a further oral hearing. Both parties filedfurther affidavits since the hearing, one is by Professor Bustin, the other by Dr Sijen.We did not think it necessary to reconvene the hearing of the appeal to hear furtherfrom counsel.[226] We accept that the evidence in both affidavits is credible and cogent.The novelty of the scientific issues in dispute makes it appropriate to admitthe evidence. Because of the developing evidence of this science this is properlydescribed as fresh evidence. Leave is granted to the respective parties to adduce theaffidavits for the purposes of the appeal. In broad terms, Professor Bustin claims thatfurther work he has carried out confirms his previous conclusion that the brain-plexassay is significantly flawed. Dr Sijen claims that Professor Bustin has carried out aseries of experiments without replicating all of the techniques employed by the NFIand he has done so in a way that the sensitivity of the assay is increased substantially,together with the risk of spurious results. She claims that the way mRNA is targetedin the NFI's assay is designed to ensure that the sort of reactions referred to byProfessor Bustin do not appear in their results. The debate between thesetwo scientists takes place at a technical level which it is frankly difficult to follow, stillless resolve.[227] And that is the difficulty here. The NFI developed the brain-plex for thepurposes of the work it was asked to carry out in this case. Dr Sijen explained thatwas necessary because the RNA organ typing methodology normally used by the NFIwas not considered appropriate. In particular it was considered histological resultsobtained by Dr Miller indicated the brain markers normally used by the NFI wouldnot detect cell types apparently present. So alternative markers had to be found andtheir utility assessed in this one-off context.[228] There is no doubt about Dr Sijen's expertise, and that of her colleagues atthe NFI who were engaged in creating the brain-plex, and attempting to ensure it wasfit for purpose. Yet eminent scientists called by the defence raise various issues as toits appropriateness, difficulties associated with the methodology involved in carryingout the testing and issues concerning the validity of the approach taken to interpretingthe results. It is clear there is no agreed external standard that can be applied.Further, much of the evidence that the jury heard on both sides of the argument washighly technical in nature and we have inevitable misgivings about how much of itwas able to be understood.[229] In our judgment the defence experts have raised these issues of substance:(a) The brain-plex used had not previously been used, but was developedfor the purposes of this case. While it might be described as adevelopment of the NFI organ typing methodology, our impression isthat rather it was the creation of a new test, albeit that it relied on sometechniques previously used. The variability of the results achievedacross the chosen matters might be thought to reflect its status assomething previously untried.(b) It was not the subject of peer review by persons independent of the NFI.(c) There is the absence of recognised standards for the selection ofbrain markers for a process such as that undertaken by the NFI.(d) One of the markers, OPALIN, had provided only two positive resultsfor known human CNS tissue out of nine tests. This must, in our view,raise real issues concerning its appropriateness.(e) The marker S100B gave negative results for sample 3003/03 for bothbrain and non-brain tissue.(f) The reactivity for brain-plex markers differed significantly when testedagainst known human brain tissue.(g) There was no guarantee that if the same tests were repeated the sameoverall result would be achieved: seven positive markers out of 12(58 per cent) might not become 14 positive markers out of 24.(h) No body of scientific knowledge or learning justified the conclusionthat it was safe to proceed on the basis that 58 per cent was anappropriate indicator for establishing the tissue was more likely humanthan that of the other species tested.(i) There is questionable logic of applying a percentage score to the overallresults achieved when some markers had apparently not worked at all,or markers had produced inconsistent results.(j) The necessary use of the reverse transcription technique is liable toproduce variable results.(k) Professor Bustin criticised the fifth marker as being potentially unfit forits intended purpose of showing the presence of RNA.[230] We accept that Dr Sijen has provided answers to the issues raised. She says,for example, that there is nothing remarkable about one of the markers not givingpositive results and that a negative response does not mean that brain tissue is notpresent. So too, the variation in the responses of markers is accommodated by the useof multiple markers, and the responses of all of them analysed as a group.[231] A question we have to consider, however, is whether the jury could properlybe regarded as competent to resolve these issues given that required the considerationof extremely technical subject matter. That leads to the related questions of thereliability of the evidence and the impact of its admission on fair trial rights.[232] On the basis of the matters raised by the defence experts Mr Eaton argues thateven though this Court has previously ruled the mRNA evidence admissible that issueshould be revisited. He advances three submissions:(a) The mRNA evidence was not scientifically valid and was thereforeunreliable. The Court should not permit evidence that makes a claimof scientific validity to be led at a criminal trial without the necessaryconditions for its validity being established.(b) The mRNA evidence was not substantially helpful. Even if it wereconsidered reliable, Dr Sijen's opinion based on the evidence was notsubstantially helpful as required under s 25 of the Evidence Act. Thisbecause the defence criticisms of the evidence, and its lack of afoundation in scientific research, meant it was not substantially helpfulfor the purposes of the test in s 25(1) of the Evidence Act.(c) The low probative value of the mRNA evidence was outweighed by itsunfairly prejudicial effect. This was the approach taken byEllen France P in her dissenting judgment on the pre-trial appeal.163[233] Mr Morgan relied on the law as expressed in the High Court andCourt of Appeal when the issue was examined pre-trial. He submitted the relevantprinciples are not in doubt, noting the majority of the Court of Appeal acceptedthe summary given by Ellen France P.164 He submitted the evidence was clearlyrelevant under s 7 of the Evidence Act, and could not be excluded under s 8 becauseits probative value was not outweighed by the risk it was unfairly prejudicial. In termsof s 25 of the Evidence Act, he maintained the jury would have "unquestionably"obtained substantial help from the evidence in ascertaining facts of consequence to thedetermination of the proceeding.[234] While accepting that the brain-plex had not been the subject of studies inpeer reviewed journals, the use of mRNA for tissue and organ typing had been. Whilstarticles relied on by the defence experts referred to various difficulties, they fell shortof demonstrating "unreliability of the principle or the technique"; rather they were a"call for standardized techniques".[235] Mr Morgan accepted that the evidence was complex and not easy for anon-specialist to understand and evaluate, but submitted that counsel andthe trial Judge had been at pains to ensure it was presented in a way that it could beunderstood. He also submitted there was significance in the fact thatDr Sijen's evidence was not the only evidence that the CNS tissue on the shirt washuman CNS tissue, suggesting that the IHC evidence was relevant to establishingthe admissibility the mRNA evidence, a proposition he also extended to other aspectsof the evidence called by the Crown about the paint flakes, DNA, the distancediscrepancy and so on. In other words, there could be confidence aboutthe mRNA evidence because of the other aspects of the Crown's circumstantial case.163 Court of Appeal pre-trial judgment, above n 111, at [69].164 At [71].[236] Mr Morgan did not raise any issue about this Court's ability to revisit theadmissibility issue on the basis that it had already been decided pre-trial. That wasappropriate, because while revisiting such an issue is unusual it is acknowledged thatmay occur in certain circumstances, such as for example a change in the law or newevidence becoming available making that course requisite in the interests of justice.165What justified the course in this case is that the Crown's evidence was subject to amuch more substantial challenge at the trial than had occurred pre-trial. Thatchallenge cast fresh light upon both the scientific controversy and, we believe, theability of the jury to deal with the difficult issues raised by this evidence. This has ledus to admit the fresh evidence. Our conclusion will therefore be based on differentfacts to the decisions already made.[237] We have already referred to the Privy Council's adoption of what was said bythe United States Supreme Court in Daubert as a useful template to be applied for theadmissibility of evidence based on novel science in a criminal trial.166 It was thesource of the analytical approach followed by Ellen France P in this case. We havealready summarised the relevant principles, but it will assist if we now quote themfrom the Privy Council judgment:167(1) whether the theory or technique can be and has been tested:Scientific methodology today is based on generating hypotheses and testingthem to see if they can be falsified; indeed, this methodology is whatdistinguishes science from other fields of human inquiry.(2) whether the theory or technique has been subjected to peer review andpublication:[S]ubmission to the scrutiny of the scientific community is a component of"good science," in part because it increases the likelihood that substantiveflaws in methodology will be detected.(3) the known or potential rate of error or the existence of standards; and,(4) whether the theory or technique used has been generally accepted.165 See R v Gallagher [1993] 1 NZLR 659 (CA) at 661–662 and the other cases discussed in thisCourt's recent decision in Winders v R [2018] NZCA 277 at [45].166 Daubert v Merrell Dow Phamaceuticals Inc, above n 50, quoted in the Privy Council judgment,above n 1, at [138]. Daubert has also been approved by the Supreme Court of Canada in R v J-LJ[2000] 2 SCR 600 at [33].167 Privy Council judgment, above n 1, at [138].[238] Before considering the application of this approach here it is useful to analysethe nature of the issue to which statements such as those in Daubert are addressed.The approach concerns the nature of expert evidence and the law's recognition thatassistance is legitimately to be derived from expert opinion to resolve factual issuesthat must be decided in the course of a trial, whether it be civil or criminal in nature.Under s 25(1) of the Evidence Act, an expert opinion is admissible if the fact-finder islikely to obtain substantial help from the opinion in understanding other evidence orascertaining any fact that is of consequence to the determination of the proceeding.Clothing this rule in the facts of this case the admissibility issue becomes whetherthe jury would be likely to obtain substantial help from Dr Sijen's opinion thatthe tissue she tested was more likely to be human in origin than that of any of the otherpotential sources analysed. Clearly, if accepted, her evidence would help the jury toascertain a fact of consequence to the determination of the proceeding, that is whetherthe tissue on the shirt was of human origin.[239] What then constitutes substantial help? We consider it is axiomatic that ifthe fact-finder is to be helped to ascertain facts, expert opinion evidence must meet athreshold of reliability. Otherwise the evidence will hinder, and potentially misleadrather than help. So the majority of this Court was clearly right when in the pre-trialappeal it identified that one purpose of what it called (with reference to Daubert)the "superadded admissibility requirements" is to protect the jury from "what issometimes colloquially called pseudo-science, meaning idiosyncratic and plainlyunsatisfactory theories".168 It contrasted this with the concept of "evidence from areputable source which is robustly and carefully researched and analysed", andobserved that the fact of disagreement among scientific experts about the degree ofreliability of evidence would not be in itself a reason for withholding it from the jury,provided it is substantially helpful and not unfairly prejudicial.[240] The expression "pseudo-science" may be apt to describe evidence which isplainly not helpful because it pretends to have a value it cannot command. This Courtdealt with a category of such evidence in AM (CA351/2017) v R to which we were168 Court of Appeal pre-trial judgment, above n 111, at [72].referred by Mr Eaton.169 The phrase "idiosyncratic and plainly unsatisfactorytheories" is similar to language used by Tipping J in R v Calder and has its origin inthe report of the Law Commission Evidence Law: Expert Evidence and OpinionEvidence.170 But the limited gatekeeping role envisaged in Calder must be reassessedhaving regard to the Privy Council judgment in this case.[241] We consider the Daubert considerations are clearly intended to reject a widercategory of evidence than idiosyncratic and plainly unsatisfactory theories. This isbecause in the scientific field whether a methodology is satisfactory or unsatisfactorymust depend ultimately on the response that is given to it by the relevant scientificcommunity. The robustness of a methodology cannot legitimately be established byan inexpert judge or jury. The essential work of validation must occur beforethe courtroom is entered. That is why the Daubert considerations require testing ofthe technique, peer review and publication, known or potential rate of error andwhether the theory or technique has been generally accepted. Those considerationsare not satisfied by accepting novel science because it has come from an apparentlyreputable source, such as the NFI in this case. What is required is a track record ofacceptance by a body of scientific opinion. This will be demonstrated when analysedin accordance with the Daubert considerations.[242] The result should be that what is proffered as evidence is no longer theoreticalin nature but can be shown by reference to a number of factors to have passed musterin the scientific community as something worthy of acceptance in a court of law. It isthat kind of generalised support that enables courts to admit with confidenceDNA evidence and the kinds of likelihood ratios which form part of it.The complications in the science have been resolved, the techniques broadly acceptedand the probative value of their application agreed in the scientific community.[243] We consider that inviting the jury to resolve the issues presented on both sidesof the mRNA issue in this case was to ask them to carry out a task for which theycannot have been equipped. They were effectively required to resolve a complicated169 AM (CA351/2017) v R [2017] NZCA 345. In that case, the expert claimed the ability accuratelyto assess credibility by reference to sentence structure and speech patterns.170 Law Commission Evidence Law: Expert Evidence and Opinion Evidence (NZLC PP18, 1991)at [63].scientific debate about whether the methodology employed was robust in the absenceof the general acceptance of the methodology envisaged by Daubert.[244] We accept as Mr Morgan submitted that efforts were made to try to ease theirtask, and the Judge in particular was most helpful in the approach he took inthe summing-up. In the end, however, we think it would be surprising if the juryunderstood much of the evidence that was called on the subject.[245] As to reliability there was no independent verification of the brain-plex. As wehave seen, various criticisms were made of it, and Dr Sijen defended its robustness,but in doing so she was not able to fall back on any known or potential rate of error oraccepted standards by which the methodology could be judged. Her conclusion thatthe results (with the varying responses of the markers) were satisfactory to establishthe opinion she proffered could not be verified by reference to any externally generatedstandard.[246] We cannot accept Mr Morgan's suggestion that the mRNA evidence could bestrengthened by reference to other evidence called by the Crown. In our viewthe admissibility of complex and novel scientific evidence must be assessed in its ownterms. We do not see how any other evidence in the case can assist in such anassessment. The other evidence does not relate to the reliability of the process in whichthis scientific evidence has been brought into being. Nor does it assist to establishpropositions said to be based on the scientific evidence. Mr Morgan's submissionultimately invites circular reasoning, which distracts from a proper consideration ofthe fundamental reliability of the science.[247] There is another aspect to this issue which is inherent in the concept of afair trial. The right to a fair trial is one of our basic rights, confirmed in s 25(a) ofthe New Zealand Bill of Rights Act 1990, but before that a right "which has been partof the common law inheritance of generations of New Zealanders".171 In the presentcase, Mr Lundy has been able to command the resources to engage the necessaryexpert assistance to question the reliability of the mRNA evidence. If he had not beenable to do so evidence that is problematic for the reasons we have addressed would171 R v Howse [2005] UKPC 30, [2006] 1 NZLR 433 at [43] per Lord Rodger and Sir Andrew Leggatt.have gone before the jury without challenge. And even then, the process of mountingthe challenge was beset by the difficulty caused by the absence of generally acceptedstandards applicable to establishing the validity of the conclusions purportedly drawn.The notion that the robustness of cutting edge scientific techniques can be establishedbefore juries creates a clear tension with the right to a fair trial in the many cases thatmight arise where the defence would simply be unable, through lack of resources, tomount soundly based challenges to the science relied on by the Crown. This is afurther consideration which underlines the need for caution in deciding to admit suchevidence.[248] For the reasons discussed, application of the Daubert factors leads tothe conclusion that the Crown's mRNA evidence should not have been admitted atthe trial. The further evidence to which we have referred, filed since the trial, has beenrelevant to this assessment. We consider that the evidence cannot have beensubstantially helpful to the jury, for the following related reasons. First, the subjectmatter was inherently complex, and it had to be covered by both the Crown anddefence experts to a level of detail commensurate with the fact that the Crown werenot able to point to widespread acceptance of the methodology employed. In a realsense the conflict was about the validity of the NFI methodology. The jury cannotrealistically have been in a position to resolve the highly specialised competing claimsmade by reputable scientists on both sides of the argument. Second, the evidencecould not cross the reliability threshold in the absence of peer review, known orpotential rate of error, standards, and general acceptance in the scientific community.In our view the evidence was not admissible under s 25(1) of the Evidence Act forthese reasons.[249] Considered in terms of s 8, the uncertainties surrounding it were such that itsprobative value was outweighed by the risk that it would have an unfairly prejudicialeffect on the hearing.[250] The consequences of these conclusions will be addressed in the final sectionof the judgment.[251] Before leaving the subject of the mRNA evidence, we mention a related issueraised by Mr Eaton that can be dealt with discretely. He noted that at one pointDr Sijen had stated the conclusion she had reached in terms different to those usedpre-trial. It will be recalled that pre-trial she had expressed her conclusion in theseterms: the RNA typing results are more probable if the slides from [specimen3003/3] contain human brain tissue than if they contain brain tissue ofthe animal species examined.172 It is not possible to determine how muchmore probable these results are (i.e. to assign the exact weight of the evidenceto the results).[252] At the trial in her evidence-in-chief she said: from these results we inferred that it's more probable that human CNStissue is present than tissue of the other animals we tested.[253] And a little later there was this exchange with the Judge:Q. Just to make sure I've got that. On the one on the pocket you didn'tobserve brain and the results for the one on the sleeve mean that yousay it is more probable than not that it is human central nervous systemtissue is that right?[173]A. Exactly.[254] Mr Morgan immediately followed up on that, asking Dr Sijen why she hadused the term "more probable than not". She answered:A. Well we cannot say how much more probable but it's depending ofcourse on what you put next to it. If you ask me, is it more probablethat it's human brain or orangutan brain material, CNS tissue I willsay that's equally possible.Q. Right.A. But if we compare it to the eight species we have tested we think, weinfer the presence of human CNS tissue.[255] In closing Mr Morgan addressed the jury on the basis that Dr Sijen, after aseries of tests on certain animals, had been able to conclude that it was "more probablethan not that it is human brain as opposed to the brain of the animal species tested".172 It will be recalled that the non-human species used for validation testing were bovine, cat, chicken,dog, guinea pig, pig, rabbit and sheep.173 The Judge is clearly referring here to 3003/4 and 3003/3 in this question (in that order).He added "[i]t's expressed extremely conservatively," language about which Mr Eatonmakes a separate complaint. But Mr Morgan's submission reflected the more limitedconclusion that Dr Sijen had otherwise consistently maintained. Mr Hislop, in hisclosing address, did not directly address Dr Sijen's actual conclusions, preferring toask the jury to reject her evidence because of the points made in the evidence ofProfessor Bustin and Dr Vennemann about the untested and unreliable nature ofthe brain-plex methodology.[256] The Judge however, probably influenced by his exchange with the witnessreferred to above, in his summing-up said that Dr Sijen's evidence was to the effectthat it was more probable than not that the analysed tissue was human tissue. Thus hereferred to "Dr Sijen's evidence that the tissue is probably human brain" and, in aspecial written handout he gave the jury and took them through, summarisingthe mRNA evidence and the defence attack on it, he referred to the jury decidingwhether they were satisfied they could rely on "Dr Sijen's opinion that the tissue isbrain observed, probably human".[257] Although this was inconsistent with the way Dr Sijen had generally put herconclusion we do not think anything turns on this issue. What the Judge said reflectedthe practical reality having regard to the breadth of the animal species tested byDr Sijen and the inherent unlikelihood of the tissue having a potential origin outsidethose.[258] Because of our conclusion that the mRNA evidence is inadmissible, we havenot found it necessary to consider the argument that the Judge failed to give areliability direction.Fuel consumption[259] The Crown's case at the second trial, as at the first, was that Mr Lundy travelledfrom Petone to Palmerston North to commit the murders before returning to the motel.The defence argued that this "secret journey" could not have occurred because therewas not enough petrol to have completed it having regard to what was known aboutMr Lundy's movements in the car in the hours leading up to the evening ofTuesday 29 August 2000 and the petrol remaining in the tank when the car was seizedon Mr Lundy's return to Palmerston North on 30 August.[260] It will be recalled that the Crown's case at the first trial was that the murdersoccurred around 7 pm on 29 August. This required Mr Lundy to have completed aremarkably fast drive from Wellington during peak evening traffic toPalmerston North and back having regard to cell phone and cell tower records placinghim in Petone at 5.30 pm and 8.28 pm. In fact, a principal element of the defence casewas that it was impossible for him to have made this journey within the time available.[261] At the first trial the Crown called evidence establishing the likely distanceinvolved in Mr Lundy's journey, the absence of roadworks and temporary speedrestrictions, as well as Mr Lundy having driven at high speeds on occasions.Mr Lundy's car was a 1998 model Ford EL Fairmont. It had a four litre six cylinderengine, and a 68 litre capacity petrol tank. One of the police officers involved in theinvestigation of this aspect of the case was Detective Danny Johanson, who gaveevidence at both trials. He arranged to borrow a vehicle equivalent to that driven byMr Lundy from a Palmerston North Ford dealer, Courtesy Ford, and he made a numberof trips to and from Palmerston North with a view to establishing distance travelledand likely fuel usage having regard to Mr Lundy's known movements on 29 and30 August 2000.[262] At the second trial the Crown also relied on evidence from Mary Ellwood, anintelligence analyst employed by the police. She was tasked with investigating thedistance travelled by Mr Lundy's vehicle between 21 August 2000, when it wasserviced by Courtesy Ford in Palmerston North and 30 August when it was seized bythe police on Mr Lundy's return to Palmerston North that morning. For the purposeof her investigation Ms Ellwood reviewed and analysed various documents that hadbeen generated by the investigation, Mr Lundy's stated movements as advised by himto the police, and telecommunication records associated with his cell phone.The results of her investigation and calculations based on it were presented to the juryin a booklet marked "Distance/Petrol Analysis".[263] The information so presented included:(a) the car's odometer reading of 80,589 km at the time the car wasserviced on 21 August;(b) the odometer reading of 81,859 km at the time the car was seized,meaning it had travelled 1,270 km since service;(c) travel of an unknown extent by Mr Lundy in Palmerston North between21 and 23 August;(d) travel of 235 km by Mr Lundy from his home to a motel inNew Plymouth on 23 August (the distance was based on a Google Mapsestimate);(e) an unknown extent of travel in New Plymouth on 23 and 24 August;(f) travel from New Plymouth to commercial premises inPalmerston North on 24 August, estimated at 235 km;(g) an unknown extent of travel in Palmerston North between 24 and29 August;(h) travel from his home on 29 August to commercial premises inPalmerston North, Petone and Lower Hutt, culminating in arrival atthe BP service station at Naenae, a total of 175 km; and(i) post-BP Naenae travel of 202 km.[264] This accounted for a total of 847 km travelled, leaving a balance of 423 kmunaccounted for between 21 and 30 August. Ms Ellwood also referred to a receiptfrom BP Eltham (in Taranaki) on 23 August at 11.55 am for 64.28 litres of fuel and areceipt for the petrol purchased at BP Naenae for 62.54 litres. She gave evidence thatshe examined 11 receipts for petrol purchases using Mr Lundy's credit cards between28 June and 29 August 2000, showing an average purchase of 57.26 litres.[265] Ms Ellwood also prepared a table showing how the travel of 202 km afterrefuelling at the BP Naenae service station had been calculated. The information aboutMr Lundy's travel had been provided by Mr Lundy and was able to be verified byreference to telecommunication records, his cellphone, financial records and otherwitness statements. The direction of travel from premise to premise was in particularbased on the account Mr Lundy had given as to his movements. Distances betweenthe various premises were again based on Google Maps estimates. The premisesvisited were successively in Naenae, Petone, Lower Hutt and in Lambton Quay incentral Wellington, involving a total of 55.6 km. Then, there was the trip frompremises in central Wellington to where Mr Lundy was stopped by the police inPalmerston North, involving a further distance travelled of 146.16 km.[266] Mr Lundy had indicated in a statement that the petrol warning indicator hadactivated by the time the vehicle was stopped, indicating that the petrol tank was nearlyempty. Police records showed that the vehicle was refuelled on 5 September 2000,when 57.9 litres were placed in it. From this, Ms Ellwood inferred, given the tankcapacity of 68 litres, that there would have been 10.1 litres remaining when the carwas seized. This meant that Mr Lundy had travelled 201.76 km consuming 57.9 litresof fuel, a consumption rate of 28.74 litres per 100 km travelled. This was contrastedwith Ford factory figures for optimal fuel consumption of 13 litres per 100 km in a"city cycle" and 8 litres per 100 km in a "highway cycle" for this type of motor vehicle.It was further contrasted with Australian Department of Primary Industries figures of15.08 litres per 100 km in a "city cycle" and 10.81 litres per 100 km in a"highway cycle".[267] Another table prepared by Ms Ellwood showed travel after the purchase ofpetrol at BP Eltham on 23 August down to Mr Lundy's arrival at BP Naenae.The distance travelled was 461 km, but the table included reference to the "unknown"travel in New Plymouth on 23 and 24 August, and in Palmerston North between 24 and29 August. The extent of travel was "unknown", because Ms Ellwood had noinformation to go on about the premises that might have been visited on theseoccasions.[268] Ms Ellwood was however able to make a calculation based on Mr Lundy'sstated practice of habitually filling the vehicle when he refuelled.174 On this basis,Ms Ellwood calculated 62.54 litres (the input at BP Naenae) had been used to travel461 km, representing a fuel consumption of 13.56 litres per 100 km.Telecommunication records for the relevant days established that Mr Lundy had notleft New Plymouth and Palmerston North. If the jury accepted Ms Ellwood's approachthey were entitled to infer that whatever travel had taken place in New Plymouth andPalmerston North, the overall petrol consumption between BP Eltham and BP Naenaewould have been at approximately the 13.56 litres per 100 km rate.[269] Ms Ellwood was asked to add notional figures for distance travelled inNew Plymouth and Palmerston North totalling 75 km and 125 km. The resultantrecalculation of fuel consumption produced figures for the period between BP Elthamand BP Naenae of 11.67 litres per 100 km and 12.87 litres per 100 km. Finally, shecalculated that if 125 km were added to the total travelled from both cities, fuelconsumption for the period would have been at 10.67 litres per 100 km.[270] A final calculation made by Ms Ellwood calculated the distance travelled byMr Lundy on State Highway 1 from Johnsonville until his vehicle was stopped bythe police. Mr Lundy, in the interview that he gave to the police on 30 August 2000described having received a telephone call from a friend advising him that there wasa police caravan parked outside his house and that he should "get home fast".Mr Lundy received that call at about 11.52 am. He said that when he took the call, hewas in Johnsonville on the main road. Ms Ellwood estimated the distance travelledfrom that point back to Palmerston North was 139 km. Since Mr Lundy had beenstopped at 1.15 pm, she was able to calculate an average speed for that portion of thejourney of 100.48 km per hour. During that period, Mr Lundy described himself ashaving averaged "140–160 kmhr", although he said that there was "a lot of slowtraffic".174 In his statement to the police on 30 August 2000 he said "I just drive until the fuel warning beepsand then I find a petrol station. I always say 'fill it up' to the attendant and I pay with my creditcard."[271] Ms Kincade cross-examined Ms Ellwood extensively: emphasisingthe unknown extent of travel in Palmerston North between 21 and 23 August, and24 and 29 August; the unknown extent of his travel in New Plymouth between 23 and24 August; suggested omissions in the record of commercial premises that he visitedafter refuelling at BP Naenae; challenging the assumption that Mr Lundy had filledthe car's petrol tank to capacity and challenging also the fact she had relied onFord factory figures for optimal fuel consumption. It was suggested that her analysishad not taken into account Mr Lundy's very fast drive from Johnsonville toPalmerston North on 30 August.[272] The defence did not call its own evidence on these issues for reasons that wewill discuss below. At the trial, however, it was sought to rely on Detective Johanson'sevidence in October 2000 in the car borrowed from Courtesy Ford.Detective Johanson's evidence described journeys from BP Naenae around Wellingtonand then to Palmerston North comprising a distance travelled of 215 km using 35.35litres, giving a fuel consumption of 16.44 litres per 100 km.175[273] Detective Johanson also described a round trip that he had made fromPalmerston North to the Foreshore Motor Lodge via Haywards Hill and return toPalmerston North via the Ngauranga Gorge, a total distance of 297.9 km at a rate of13.6 litres per 100 km.[274] As Mr Morgan was at pains to point out, the Crown's case relied not only onthe evidence about fuel consumption, but also about distance travelled, and the knownfacts that in the period from 21 to 30 August, Mr Lundy's car travelled 1,270 km.Ms Ellwood had been able to demonstrate and account for 847 km, leaving a balanceunaccounted for of 423 km. It was the Crown's case that Mr Lundy must have madea return trip in the early hours of 30 August from the Foreshore Motor Lodge in Petoneto the Lundy home in Palmerston North and back, a distance of 300 km. Theremaining approximately 123 km could then be accounted for as being unknown travel175 Detective Johanson conceded that he had missed some commercial premises visited by Mr Lundy(whose movements he was intending to replicate). The result might have been an increase inthe distance travelled, although the figure of 215 km that he reported was not significantlydifferent from the 202 km referred to by Ms Ellwood.in Palmerston North and New Plymouth on the days where Ms Ellwood had simplyreported the travel as "unknown".[275] In closing, Mr Hislop put a counter-proposition. Noting that Ms Ellwood hadreported "eight unknown days of travel", Mr Hislop submitted that the car need onlyhave travelled an average of 37 km on each of those days for the suggested discrepancyof more than 300 km to disappear. In summing up on this issue, Simon France J notedthat no one knew what driving Mr Lundy had done on the days when he was drivingaround New Plymouth and when he was based at home in Palmerston North. Thequestion for the jury was whether, notwithstanding the gaps, it was satisfied thatthe Crown had established there was mileage that was unaccounted for but whichcould be explained by the alleged extra trip.[276] On the issue of petrol consumption, the rival arguments were addressed bySimon France J in his summing-up as follows:(a) The defence focusThe defence say there is not enough capacity in the car's fuel tank forMr Lundy to do what we know he did on 29/30 August, plus makethis alleged secret journey. We know the tank is 68 litres, and weknow there was about 10 litres left. The defence says the proper focusis Sergeant [Johanson's] trips back in 2001. The defence say that isthe best figure because it is actual, not based on manufacturer theories.This is the D4 chart.[176]Journey one was meant to be a pretty exact replica ofthe Johnsonville-Palmerston trip — hard driving. It used 35 [l]itres.Mr Hislop submits if you then assume the same driving on the secretjourney, you need 84 litres. The car only has a capacity of 68, even ifhe used it all which he did not.The same exercise is then done in relation to the other trips withdifferent figures reached but always the same outcome — you needmore than 68 litres.So this is the first of Mr Hislop's three impossibilities.The Crown response is twofold:• It is wrong to assume on the alleged secret journey that he wouldbe driving like Sergeant [Johanson] did or like Mr Lundy himself176 The D4 chart to which the Judge referred was a defence exhibit summarising evidence aboutdistances between commercial premises visited by Mr Lundy after the BP Naenae refuelling on29 August 2000.did from Johnsonville to Palmerston North. The night timejourney would have been a much quieter trip, not trying to drawattention and on the Crown case with plenty of time to get backfor morning. So it submits the 300 km extra trip would have usedfar less fuel than the flat out trip, and it draws support for this fromhow much mileage Mr Lundy got on the last full tank — tab 8,461 km plus whatever he did on the seven days marked in yellow,and tab 9 showing that graphically.[177]• The second point it raises is that it is not exact science and we donot know if Sergeant [Johanson's] use was the same asMr Lundy's.So that is the defence's positive point — if you accept its analysis thatMr Lundy could not do it because he did not have enough fuel, thenyou should stop there and acquit. If you accept the Crown's point thatit is invalid to use [Johanson's] figures for the alleged late nightjourney because the driving would be so different, then you move on.(b) The Crown focusThe prosecution has two limbs to its argument:• First, it focuses on tabs 8 and 9 — these relate to the precedingfull tank. Look it submits at how much driving Mr Lundyachieved, and ask yourself whether it is believable that on thisAugust 29 tank he only got the 220–230 km. It submits theanswer must be there is another trip that has used up the petrol.Mr Hislop queries tab 9 — on none of the red trip did he drivelike he did that morning of the 30th so he suggests to you it ismisleading.• The second Crown point is that if you look at that previous tank,and look at the various figures for road usage, and accept he wouldhave been driving normally on the alleged night time journey,Ms [Ellwood's] calculation shows it can be done. It would betight, but that is why the warning light is on.In response Mr Hislop submits, as he said before, the best usagefigures are Sergeant [Johanson's] and he queried the validity ofMs [Ellwood's] working, emphasising the unknown days that he sayshave not been account for by her.When discussing how much fuel the car might use, Mr Hislop alsoplaced some reliance on figures suggested by a man calledKevin Priest. I need to talk to you about this. We have not heard fromthis person.177 Tab 8 was a reference to a table included in Ms Ellwood's "Distance/Petrol Analysis", the daysmarked in yellow being "unknown" travel days. Tab 9 referred to a marked-up map showingMr Lundy's movements in the same document.[277] The Judge then discussed what Mr Priest had told the police about possiblefuel usage, giving a warning about the care needed with hearsay evidence beforeconcluding:That is all I want to say on the car, fuel, mileage debate. You heard itwell put by Mr Morgan and Mr Hislop. Each claims it is verysignificant evidence, and Mr Hislop goes further and says it isdecisive.[278] It has not been suggested that the Judge's summing-up on these issues wasinadequate or in error. The defence case on appeal is rather that the jury did not heardirect evidence as to the effect of fast driving on fuel consumption. It seeks now toadduce as fresh evidence a report which it has obtained from Mr Bruce Robertson, amechanical engineer employed as a Design Engineer and Technical Services Managerin the Mechanical Engineering Department at the University of Canterbury.[279] In his report dated 3 October 2016, Mr Robertson described fuel consumptiontesting he had carried out to demonstrate that standards based fuel consumption valuesare inappropriate to show the consequences of driving conducted in a fast and erraticmanner as opposed to normal driving. The testing he carried out was performed on aclosed road circuit, Mike Pero Motorsport Park, which he said allowed "consistencyacross the tests and the ability to drive at speeds and in a manner which [is] not safeor legal on public roads". The tests were carried out using a vehicle similar to thatused by Mr Lundy (a 1998 Ford Falcon EL 4.0) and a different vehicle chosen as anindependent reference point (a 2007 model BMW M5 5.0). For present purposes, wedo not need to be concerned with the second vehicle.[280] Mr Robertson carried out two tests, one designed to accommodate what mightbe called normal driving and the second fast driving. The latter involved "applyingfull throttle to accelerate as quickly as possible and achieve the highest possible speedbetween corners". In the normal driving, the Ford Falcon used fuel at a rate of 11 litresper 100 km. In the fast driving, the rate increased to 36 litres per 100 km. Accordingto Mr Robertson's calculations, to accommodate the Crown theory of an additionalreturn trip between Petone and Palmerston North (assuming normal driving at aconsumption rate of about 9.3 litres per 100 km), for the fast trip between Johnsonvilleand Palmerston North on 30 August, fuel consumption must have been limited to16.1 litres per 100 km in order for the total fuel consumption post-BP Naenae to beaccommodated within the 58 litre figure. Mr Robertson naturally contrasted that withthe hard driving test result which he achieved of 36 litres per 100 km.[281] The application to adduce Mr Robertson's report as fresh evidence is advancedin the context of defence complaints of late notice that the Crown no longer intendedto allege that Mr Lundy had undertaken a very fast trip between Wellington andPalmerston North between 5.30 and 8 pm on 29 August. Although Ms Ellwood'sreport had been disclosed in December 2015, it was not accompanied by anyexplanation of a shift in the Crown case. By the time the defence was advisedthe Crown had abandoned the 29 August fast drive scenario the trial was onlyeight weeks distant, including the period of the Christmas vacation. It was not untilreceipt of an email on 21 January 2015, in which senior Crown counsel advised thatthe Crown would not contend for a time of death prior to the computer being shutdown that the defence realised the Crown had abandoned the time of deathsuccessfully prosecuted at the first trial. In the circumstances, there had been littletime to properly prepare the defence case to accommodate the Crown's reliance on amore leisurely late night round trip from Petone to Palmerston North and back.[282] As against that, Mr Morgan pointed out that the defence had been notified byletter dated 4 June 2014 that the Crown would not specify at trial a time of death andthat one of the periods of time in which it was maintained the deaths of the deceasedoccurred was between 12.48 am on 30 August 2000 and 8.09 am on 30 August.That stance had been recorded in the High Court pre-trial judgment. The defence infact did brief a witness to deal with the relevant issues, Mr Christopher Wall.However, Simon France J ruled his evidence inadmissible. In his ruling, the Judgenoted the defence wished to call Mr Wall to give evidence of a test he carried out usinga dynamometer. The application was opposed by the Crown, and after somediscussion the Judge indicated his view that the proposed evidence in its current statewas "some distance" from being admissible. He said that a fuller explanation ofthe process followed by Mr Wall was required, together with a better articulation ofhow the results of what he did were relevant to Mr Lundy's journey. The Judgereferred to other criticisms that had been made of Mr Wall's evidence, which he neededto answer.178 Although permission was given for the application to call the evidenceto be renewed with a modified brief, that did not occur.[283] Mr Eaton acknowledged in this Court that the issue giving rise to the Crown'sobjection could not in fact be remedied, and the Crown closed its case on the day afterthe ruling. Mr Eaton noted that Mr Wall's brief of evidence had been disclosed on18 February 2015 prior to any Crown evidence being called as to fuel consumption,but the Crown did not give notice that it objected to the evidence until 20 March, overa month later. However, it is not said that exclusion of the evidence as occurred gaverise to a miscarriage and there has been no attempt to resurrect Mr Wall's evidence forthe purposes of the appeal.[284] Given the circumstances, we might have been prepared to overlook the factthat the evidence now sought to be adduced is not fresh in the sense that there was noreason why it could not have been obtained prior to the trial and called at theappropriate time. It is not without significance the Crown did not comply withRonald Young J's order that the evidence to be relied on by the Crown atthe second trial was to be disclosed by mid-August 2014. However, while we aresatisfied that Mr Robertson's evidence is reliable, we do not consider it cogent.[285] First, and obviously, he reports the result of tests carried out on a closed circuit.The circumstances of the tests were therefore very different from those faced byMr Lundy when he drove back to Palmerston North at speed from Johnsonville onthe morning of 30 August. They were achieved by driving hard on a race track, a verydifferent context from driving on the open road. The consumption is approximatelythree times higher than normal consumption. Second, the results of Mr Robertson'stests insofar as the reported figure of 36 litres per 100 km is concerned do not fit withthe figures achieved by Detective Johanson in the road trips which he recorded inevidence and on which he achieved, hard driving on each occasion, figures of16.44 litres per 100 km and 13.6 litres per 100 km. These figures were unchallengedin the defence closing. Unlike a race track, these conditions were no doubt similar tothose faced by Mr Lundy. The distance between those figures, achieved while driving178 R v Lundy [2015] NZHC 542.hard in real on-road conditions, and the closed track result achieved by Mr Robertsonare too great.[286] We note finally that the figure of 16.1 litres per 100 km, to which Mr Robertsonsaid his hard driving figure would need to be reduced to fit the fuel available for theJohnsonville to Palmerston North journey, is in the vicinity of the figure achieved byDetective Johanson in 17 October 2000.[287] For these reasons, we are of the view that the application to adduce the furtherevidence of Mr Robertson should be declined and we will order accordingly.Otherwise, our conclusion is that there was evidence before the jury on which it wasentitled to conclude that Mr Lundy could have made the late night trip between Petoneand Palmerston North for the purposes of the murders and return as contended bythe Crown. They could also have concluded he would have had sufficient fuel forthe travel that he made the following day, including the fast trip back toPalmerston North from Johnsonville, although as the Crown conceded all along, itwould have been tight. That is consistent with the fact that as Mr Lundy said tothe police, by the time he had reached Palmerston North the petrol warning device onhis car was sounding.Failure to give a demeanour direction[288] Mr Eaton submitted that two issues arose in the trial that required a demeanourdirection to be given.[289] First, he noted that the trial had engaged extensive media comment and publicdebate focussing closely on Mr Lundy's credibility in large part what he referred to asthe "infamous funeral scene". At the funeral of the deceased Mr Lundy had beenfilmed while engaged in an outpouring of grief. Mr Eaton drew attention to atelevision programme in which the possibility he was acting had been raised.He argued it was inevitable the jury had been exposed to those images andcommentary.[290] Mr Eaton justified that submission by reference to an affidavit the appellantobtained from Ms Kloe Palmer, a consultant with a public relations and marketingbusiness based in Christchurch. She reported work she had commissioned by Isentia,a media monitoring organisation, establishing that between the start of the second trialon 9 February and the verdict date of 1 April 2015 the trial had been mentioned intelevision news bulletins 327 times. She also referred to the funeral footage havingbeen broadcast on 3News in the 6 pm bulletins on 9 February and 30 March; it waspossible similar images had been broadcast on One News in the 6 pm bulletin on31 March. She noted the images may have been shown in other broadcasts, but thiscould not be ascertained because not every story is uploaded to the relevant network'swebsite.[291] The second issue raised under this heading is that, during its deliberations,the jury asked and were permitted to view for a second time a selected portion ofMr Lundy's second police interview including passages where he had become highlyemotive after being forced to look at photographs of the deceased. Mr Eaton referredto an affidavit of Mr Burns, one of defence counsel for Mr Lundy at the second trial,who provided an affirmation for the purposes of the appeal including the followingparagraphs:3 The Lundy jury asked only one question, which was to have footageof the scene and a portion of the interview replayed to them. Thisportion was where the interviewing officer confronted Lundy withphotographs of the corpses of his wife and daughter at the crime scene.I am not going to attempt to describe his reactions, which can best beconveyed by viewing the portion of interview.4 When the jury came back in, one of the jurors positioned himself sohe was looking directly at Lundy. It was more than just turning hishead; he seemed to have moved his seating position. He remained thesame throughout the replays, and his attention to Lundy was soobvious that I believe I watched him almost continuously. I did notsee him look once at the replayed footage or interview. It was obviousto me that he was watching to see how Lundy reacted. I glanced atLundy from time to time myself to check how he was appearing tothat juror.[292] Mr Burns also stated that when the jury returned to its deliberations, heimmediately had a discussion with Mr Hislop, who was also concerned about what hehad just observed. Mr Burns referred to the appearance conveyed that the jury wasplacing weight on the behaviour of Mr Lundy, both in the interview and at the trial.[293] The appellant also seeks to rely on an affidavit provided by Mr Michael White,who is a journalist and currently a senior writer at North & South magazine. Mr Whitereported on the retrial and sat at the press bench in the courtroom for all but two daysof the trial. His observations were consistent with those of Mr Burns:17. On Day 36 of the trial, Tuesday March 31, the jury was deliberatingits verdict.18. At around 2.30 that afternoon, it requested to see two videos that hadpreviously been shown during evidence at the trial.19. The first video was of the police walk-through of the crime scene.20. The second video was a segment of the interview of Mark Lundy onFebruary 23, 2001, conducted by Detective Sergeant Steve Kelly.This was the final police interview with Mark Lundy before he wasarrested for the murder of his wife and daughter.21. Much of this section of the interview is taken up withDetective Sergeant Kelly showing Mark Lundy photos of his wife anddaughter as they were found after being murdered, and the injuriesthey sustained.22. From the beginning of this video being played in court, I noticedseveral jurors watching Mark Lundy as he sat in the dock.23. One juror in particular, paid virtually no attention to the video whichwas on a large screen directly opposite the jury box, or to the transcriptof the interview all jurors had access to.24. Instead, this juror turned his head to his left and stared atthe defendant.[294] In later paragraphs of the affidavit, Mr White recorded his reaction at the timethat there was a possibility the juror had been asked by other members of the jury toconcentrate on observing Mr Lundy's reactions during the replaying ofthe video interview.[295] The appellant seeks leave to rely on the affidavits of Ms Palmer, Mr Burns andMr White. We did not understand the Crown to oppose leave being granted. As theyestablish a factual context for the arguments addressed in this part of the case, and theyhave a subject matter that obviously could not have been addressed prior to trial wegrant leave accordingly.[296] On the first issue, Mr Eaton referred to the judgment of the Supreme Court inTaniwha v R, a judgment delivered after the conclusion of Mr Lundy's second trial.179Mr Eaton noted the Supreme Court's acceptance that where there is a real risk that ajury might place inappropriate weight on the demeanour of a witness in making acredibility or reliability assessment, it is the duty of the trial Judge to give anappropriate warning. This issue had of course been addressed in earlier judgments ofthis Court, notably E (CA799/2012) v R.180 The general approach taken in that casewas effectively endorsed and further explained in Taniwha.[297] As the passages in Taniwha on which Mr Eaton relied show, the concernaddressed is inappropriate reliance on the demeanour of a witness in the jury'sassessment of veracity or reliability of witnesses. For example, the Supreme Courtobserved:181The key consideration for the Judge will be whether there is a real risk thatwitness demeanour will feature illegitimately in the jury's assessment ofwitness veracity or reliability.[298] Later, after referring to the risk that a jury might interpret a "tailored" directionas an expression of doubt by the Judge as to the veracity of a particular witness orwitnesses, the Court observed:182 we consider that trial judges could usefully, as a matter of course wherecredibility is likely to be a major issue at trial, include in their opening remarksto the jury a brief statement about the approach the jury should take toassessing competing accounts from witnesses, as the Judge in the present casedid.[299] And the example direction that the Court gave in the judgment was couched inlanguage which warned against using demeanour in the witness box as a way ofassessing the truth or falsity of the evidence.183 Consequently, we considerthe discussion in Taniwha contemplates a demeanour warning where there is a contestabout the credibility or reliability of witnesses giving evidence. The language usedwould not be apt for a situation where the real concern is not what a witness has said179 Taniwha v R [2016] NZSC 121, [2017] 1 NZLR 116.180 E (CA799/2012) v R [2013] NZCA 678.181 Taniwha v R, above n 179, at [43].182 At [44].183 At [46].in the trial itself, but where, as in this case, there is an assumption that jury membersmight react in a prejudicial way as a consequence of their observation of televisioncoverage of the defendant prior to the trial. Giving a demeanour direction in relationto matters of that nature would conflict with the direction invariably given that a juryis to try the case only on the basis of the evidence, and is to exclude from considerationmatters of which they may be aware that have taken place outside the courtroomwhether before or during the trial.[300] Mr Eaton referred to other cases which have acknowledged that incidentsarising in the courtroom may require the Judge to determine whether the ability ofthe jury to consider the case impartially has been compromised.184 However, thosecases discuss events occurring in the courtroom and their reasoning does not extendto events that have occurred prior to the trial.185[301] We accept, on the basis of the evidence of Ms Palmer, that members of the jurymay have been exposed to repeated showings of the footage depicting Mr Lundy atthe funeral during the ongoing trial. However, that fact alone would not in our viewgive rise to a need for a warning in terms of Taniwha because it is not obviously relatedto the credibility of witness testimony during the trial. And because it would haveoccurred outside the courtroom, it would not constitute any error in the trial processof the kind referred to in the other cases to which Mr Eaton referred.[302] The way in which the risk of prejudice to the trial which may arise from eventsoutside the courtroom is traditionally dealt with by way of judicial direction to putsuch events out of mind. In this case, the Judge relevantly told the jury in hisopening directions:No research/inquiries10. The fairness of a trial, and the validity of the verdict, comes from12 people (you) deciding the case together, all possessing exactlythe same information.184 Simon France (ed) Adams on Criminal Law (online ed, Thomson Reuters) at [TP17.04].185 Cases referred to include R v Accused (CA89/88) [1989] 3 NZLR 27 (CA), in which a person inthe public gallery had stood and shouted that the lawyer knew the defendant was guilty; and R vBurney [1989] 1 NZLR 732 (CA), where the issue concerned facial expressions of the Clerk ofthe Court during the trial and their potential influence on the jury.11. That information can only come to you during the trial — first fromwitnesses, then from lawyers by way of arguments they will make toyou, and then from me who will give you directions and assistance.That is the only information that can inform your decision.12. The fairness of a trial is threatened if any of you in any way seek toobtain other information. Do not do so. Do not internet search or useany other means. Trust the process. Trust that you will hear all thereis to hear that is relevant. Be fair and decide the case only onthe information that comes to you in this courtroom. If you receiveother information about the case, or become aware that other jurorshave, your duty is to tell me about it.[303] This clearly told the jury that their decision must be made only on informationprovided in the courtroom. In the summing-up, the Judge underlined this, instructingas follows:Rules[2] Your task is to decide whether the prosecution has satisfied youbeyond reasonable doubt that Mr Lundy was the person who committed thesecrimes.[3] It is important you reach your verdict on the basis of the evidence thatyou have heard. Has that evidence left you sure Mr Lundy committed theseoffences? If not, the verdict will be not guilty.[4] There is quite a lot that underlies this idea of deciding the case onlyon the evidence. It is probably the most important thing for me to commenton, so I am starting with it:(a) it means putting aside anything you thought you knew aboutthe case before you started, and anything you have read orheard outside these four walls;[304] This instruction was wide enough to cover any pre-trial familiarity withthe footage of Mr Lundy at the funeral, and any experience of the repetition of thatfootage during the trial. Proceeding, as we do, on the assumption that clearinstructions given by the Judge will be complied with, we must conclude that ademeanour direction was not required in relation to Mr Lundy's conduct at the funeralof the deceased.[305] Mr Eaton's argument about the second issue has two foundations. The first isthe fact that the jury asked and were allowed to see a replay of Mr Lundy's secondpolice interview, including passages where he had become highly emotive having beenforced to look at photographs of the deceased. It is suggested that this separately gaverise to the need for a demeanour direction, because of Mr Lundy's conduct shown onthe video. This does not sit well with an addendum made by the Judge to the recordof his summing-up in which he records:The jury retired around 1.00 pm Monday. There was no communication untilaround 2.00 pm Tuesday when a request was made to view the scene videoand Mr Lundy's second DVD interview — p110 to the end. It was agreed thatthe video would start at p 109, and that no matters needed to be drawn tothe jury's attention.[306] The second foundation is the suggestion that the request and the conduct ofthe jury in relation to it emphasised the need for the direction. For the purposes of thispart of the argument, Mr Lundy's conduct at the funeral is again relied on as part ofthe relevant context in which a demeanour direction should have been considered inrelation to the jury's request for replaying of the relevant part of the video interview.[307] Mr Eaton contends that the jury question should have been seen as elevatingthe risk of impermissible reasoning by the jury. He submitted the trial Judge ought tohave inquired why the transcript was not adequate to address any issue of interest; thiswould likely have drawn the response that the jury were interested in observingMr Lundy's demeanour. Mr Eaton contends that in that event, the Judge would havebeen compelled to give a very full and clear demeanour direction.[308] As we have recorded, Mr Eaton further argued that the evidence of Mr Burnsand Mr White suggests that the jury were engaging in a form of experiment duringtheir deliberations with one juror in particular tasked with observing Mr Lundy'sdemeanour as he himself watched the replay.[309] One of the difficulties we have with the appellant's argument on this issue isthat the theory the jury was conducting some kind of experiment is essentiallyspeculative. The fact that one of the jurors was looking at Mr Lundy while otherswatched the repeated video interview may simply reflect that that juror did not wishto see the video, for whatever reason. Secondly, if the jury considered Mr Lundy'sdemeanour in the courtroom was significant, that might be thought inconsistent withonly one of the jurors watching him as the video was played. We note in addition, thatthere is no suggestion in the affidavits that there was any behaviour by Mr Lundywhilst the video was being replayed from which it is argued any juror or jurors coulddraw any conclusions.[310] We think it is significant that there was no objection made at the time ofthe jury's request to view the video interview or suggestion anything needed to be saidat that point; counsel were, after all, familiar with its content. More importantly, wethink it is significant that, as Mr Morgan emphasised, the potential prejudice arisingfrom what happened in Court when the jury viewed the video interview was not raisedby counsel with the Judge. Mr Burns said that he and Mr Hislop did not considerasking for a direction about Mr Lundy's behaviour in the interview or in Court and norwas it raised by Crown counsel. If there was real concern about what had happenedas the video was played, the Judge could have been asked to give a direction atthe time. In the absence of that, it is difficult to conclude that there has been adeficiency in the trial process.[311] We also consider there is merit in Mr Morgan's submission that had the Judgechosen to give the jury a direction about the danger of reasoning based on demeanourin conjunction with the jury's request to see the video for a second time, the contentof the warning would have been difficult to construct. We say that in the absence ofany submission that there was something unnatural in Mr Lundy's reaction to seeingthe mutilated state of his wife and daughter in the video interview, the Judge could notbe sure that the direction would not be prejudicial. And referring to the funeral sceneat that point might have been to draw attention to an issue and give it an emphasis itmight not otherwise have had.[312] For these reasons, we do not consider the Judge erred by not giving ademeanour direction.Failure to give a lies direction[313] The final issue raised by the appellant rests on the fact that the Judge did notgive a lies direction. Mr Eaton noted that in his opening address Crown counselhighlighted the appellant's alleged attempts to mislead the police inquiry as being oneof a handful of critical factors which the Crown said collectively proved Mr Lundy'sguilt. The matters mentioned were an attempt to conceal the poor state of his financesand the false explanation of the circumstances in which he came to move his car havinginitially parked it outside the unit in which he was staying at the Foreshore MotorLodge on 29 August: it was suggested that he had lied in saying that he drove toThe Esplanade and started reading a book. Mr Eaton also referred to Mr Morgan'sclosing address when those two points were reiterated.[314] It was also suggested that Mr Lundy had lied or misled the police in relation tohow a silver bracelet came to be in his vehicle, an apparent attempt to make the sceneappear as if a burglary had occurred, claiming to be the victim of petrol theft and othermatters.[315] Mr Eaton acknowledged that the Judge had asked counsel whether alies direction should be given, prior to his summing-up. Neither the Crown nor thedefence had sought such a direction. Nevertheless, Mr Eaton submitted the trial Judgeshould have given a direction, because there was in fact a risk that the jury might placeundue weight on evidence of lies told. He argued that in a case such as the present,when the defence is denial, allegations by the prosecution that the defendant had liedto the police or misled the investigation should be seen as invitations to the jury toinfer that the defendant lied or misled the inquiry because he was guilty. A liesdirection was essential to prevent the jury reasoning in that way. The absence of a liesdirection resulted in a miscarriage of justice.[316] In response, Mr Morgan referred to the note made by the Judge recordingmatters that he had discussed with counsel following the closing addresses. Includedamongst them was the following:[L]ies direction stemming from Crown's "misleading the inquiry" submission— not seen by me to be needed, and not wanted by Mr Hislop.[317] Section 124 of the Evidence Act provides:(1) This section applies if evidence offered in a criminal proceedingsuggests that a defendant has lied either before or duringthe proceeding.(2) If evidence of a defendant's lie is offered in a criminal proceedingtried with a jury, the Judge is not obliged to give a specific directionas to what inference the jury may draw from that evidence.(3) Despite subsection (2), if, in a criminal proceeding tried with a jury,the Judge is of the opinion that the jury may place undue weight onevidence of a defendant's lie, or if the defendant so requests, the Judgemust warn the jury that—(a) the jury must be satisfied before using the evidence thatthe defendant did lie; and(b) people lie for various reasons; and(c) the jury should not necessarily conclude that, just becausethe defendant lied, the defendant is guilty of the offence forwhich the defendant is being tried.[318] As can be seen, s 124(2) provides that if evidence of a defendant's lie is offeredin a criminal proceeding tried with a jury, the Judge is not obliged to give a specificdirection as to what inference the jury may draw from that evidence. The directionbecomes obligatory under subs (3) if the Judge is of the opinion the jury might givethe lies undue weight, or the defendant seeks the direction. The result of givingthe direction is necessarily to invite the jury to focus on the lie. In this case, the Judgerecorded his view that a direction was not required. Given that counsel for Mr Lundyhad been asked whether the defence sought such a direction and said it was not wantedwe see no proper basis for holding the Judge should have given it.[319] As Mr Morgan pointed out, a lies direction can simply have the effect ofemphasising the fact that lies have likely been told, and Mr Hislop's closing to the juryproceeded on the basis that Mr Lundy had not lied. For the Judge to invite specificattention to this issue by giving the direction contemplated by s 124(3) of the Act maywell have been counter-productive.[320] As will be apparent from the judgment to this point, Mr Lundy was representedat the trial by able and experienced counsel, including senior counsel fromthe English Bar and Mr Burns, a very experienced former prosecutor. The presentissue essentially invites this Court to decide that the view of the trial Judge and that ofMr Lundy's counsel on this issue was wrong. We are not prepared to do so.[321] We reject the submission that there was a miscarriage on this ground.Miscarriage and fair trial[322] Section 385 of the Crimes Act provides that the Court must allow an appeal if,amongst other things, it is of the opinion that there has been a miscarriage of justice.186However, that is subject to the proviso, which provides: provided that the Court of Appeal or the Supreme Court may,notwithstanding that it is of opinion that the point raised in the appeal mightbe decided in favour of the appellant, dismiss the appeal if it considers that nosubstantial miscarriage of justice has actually occurred.[323] We have now considered and rejected all of the grounds of appeal save inrelation to the admissibility of the mRNA evidence. Convicting Mr Lundy onevidence that includes the mRNA evidence would be a miscarriage of justice unlessthis Court were to conclude, in accordance with the proviso, that notwithstanding ourconclusion on the admissibility issue, "no substantial miscarriage of justice hasactually occurred".[324] The question of whether the proviso should be applied involvestwo considerations. First, the Court must feel sure of the guilt of the accused in thesense that, on a review of all the admissible evidence, conviction was inevitable.Secondly, the Court must be satisfied that the trial was fair, with the result there hadbeen no breach of the right to a fair trial guaranteed by s 25(a) of the New ZealandBill of Rights Act. The appropriate approach was that set out in Matenga, to whichwe have earlier referred. Blanchard J wrote:187[31] Proceeding in this way and having identified a true miscarriage, thatis, something which has gone wrong and which was capable of affectingthe result of the trial, the task of the Court of Appeal under the proviso is thento consider whether that potentially adverse effect on the result may actually,that is, in reality, have occurred. The Court may exercise its discretion todismiss the appeal only if, having reviewed all the admissible evidence, itconsiders that, notwithstanding there has been a miscarriage, the guilty verdictwas inevitable, in the sense of being the only reasonably possible verdict, onthat evidence. Importantly, the Court should not apply the proviso simplybecause it considers there was enough evidence to enable a reasonable jury toconvict. In order to come to the view that the verdict of guilty was inevitable186 Section 385(1)(c).187 R v Matenga, above n 59.the Court must itself feel sure of the guilt of the accused. Before applyingthe proviso the Court must also be satisfied that the trial was fair and thus thatthere was no breach of the right guaranteed to the accused by s 25(a) of the Billof Rights Act.(Footnotes omitted.)[325] It is worth noting also what was said at an earlier stage in the same judgment:188While the jury is in general terms the arbiter of guilt in our system of criminaljustice, the very existence of the proviso demonstrates that Parliamentintended the judges sitting on the appeal to be the ultimate arbiters of guilt incircumstances in which the proviso applies. The general rule that guilt isdetermined by a jury rather than by judges does, however, mean that theproviso should be applied only if there is no room for doubt about the guilt ofthe appellant; and, as we will mention again below, considerable caution isnecessary before resorting to the proviso when the ultimate issues depend, asthey frequently will, on the assessment of witnesses.[326] In a case such as the present where the miscarriage arises from the wrongfuladmission of evidence, this Court on appeal must feel sure that a guilty verdict wasinevitable had the trial proceeded without the inadmissible evidence being called.The Crown submits we can properly reach that conclusion. The appellant submits tothe contrary.[327] We have earlier summarised the prosecution and defence cases atthe second trial. We have also discussed the principal issues pursued on appeal. Oneconsequence of adopting that structure has inevitably been a focus on the potentialproblems with the Crown case as opposed to its strengths. As to that, it is clear thatthe Crown had a very strong circumstantial case against Mr Lundy. Its twinfoundations were the CNS tissue found on Mr Lundy's shirt, and the DNA evidencethat was associated with it.[328] It is clear that there was CNS tissue on Mr Lundy's shirt, a shirt that headmitted wearing on the night of the murders. The tissue was found in two separatespots, one on the chest and one on the sleeve. The spots were small and the CNS tissuewas embedded in the fabric. Both spots tested positive for blood. If, as was suggested,the CNS tissue was only initially in one place, and could have been smeared on to theother place on the shirt, that needed to have occurred when the tissue was fresh or188 At [29]."unfixed". It should be emphasised that CNS tissue was the only tissue observed onthe shirt, and blood and blood vessels were also observed. As noted in our discussionof the CNS evidence above, leaving issues of contamination to one side, there wasessentially no challenge at the second trial to the fact that the tissue analysed byDr Miller was CNS tissue, and only CNS tissue and it had been smeared into the fabricwhich must have occurred when it was fresh.[329] One defence explanation for the presence of the CNS tissue was that it was infact Mrs Lundy's brain tissue, but was on the shirt as a result of contamination inthe course of the police investigation. The shirt however had been located on3 September 2000 in Mr Lundy's suit bag, in his car. It was immediately placed intoan exhibit bag and only examined by Mr Sutherland who cut out the sections ofthe shirt embedded with tissue on 27 October 2000. This issue was raised beforethe jury and no real narrative explaining the contamination emerged.[330] The second possibility raised by the defence was that the substance on the shirtcame from the food chain. However, if the material on the shirt came from thefood chain it would be expected there would be tissue other than CNS tissue in thesample. The defence case required therefore that the other tissue had been removedfrom the shirt, leaving only the CNS tissue. The defence postulated that this othertissue had been removed in the dab slide process, which would also account for whythe dab slide had behaved differently to C3003/3 and C3003/4. This was very unlikelyas Dr du Plessis put it in evidence:So it would be quite remarkable if there's a mixture of material but just onetype of substance lay on this surface, was perfectly separated from the restenmeshed in the shirt fibres because you get this perfect separation. I meanthat must be unusual that you get that. Secondly, that Mr Sutherland managedto perfectly and cleanly scrape off this other substance leaving no residuewhatsoever in the fabric. So that's my, this is only first observation I can makewhich is a common sense one. I suppose not utterly impossible but I thinkthe odds of that must be pretty low.[331] The fact that the tissue observed was only CNS tissue was strong evidencerebutting the possibility that the tissue was the consequence of contamination whetherby food or some other source. In this respect, the following cross-examination ofthe defence witness Dr Smith deserves some emphasis:Q. Let me just rephrase. So the point being is that as a consequence ofyour IHC, Dr du Plessis' IHC, and Dr du Plessis' electron microscopyabout which you and he have collaborated, it is the notion that it's veryodd that only central nervous system tissue is present, is that right?A. Yes, that's correct.Q. And the notion, and that fact that only central nervous system tissueis present just is completely against the notion that this is somefood contaminant, is that right?A. Yes, to my mind, yes, I'd agree.[332] Mr Eaton noted in this context that Professor Ironside referred to the possibilityof contamination when addressing the difference between the dab slide and Dr Miller'sslides.189 In that context he referred to the fact that fragments of CNS tissue fromanimals can enter the food chain, and that could have been a potential source ofcontamination. This was largely speculative, although Professor Ironside had giventhe example of neck chops that might contain parts of the spinal cord. CNS tissuemight also find its way into sausage. In either case however, it would be remarkableif the only part of the chop or sausage ended up spilled onto clothing was CNS tissue,as in the case of Mr Lundy's shirt. Like other experts, Professor Ironside readilyaccepted that the CNS tissue observed in this case was "certainly central nervoussystem tissue and nothing else".[333] Another speculative possibility that Professor Ironside was prepared tocountenance was that the material cut from Mr Lundy's shirt may have becomecontaminated between its removal from the shirt and its provision to Dr Miller byDetective Grantham. During this period it had been in a sealed envelope in a drawer,when it was removed for specific purposes that were the subject of evidence. Thispossibility of course owed nothing to any expert opinion. It was raised as a possibility,but we are unable to give it any weight in the absence of any appropriate evidentiaryfoundation in other facts. Again, it would be quite extraordinary if the samples cutfrom the shirt had accidentally become associated with tissue that was exclusivelyCNS tissue.189 The dab slide was the slide made by Mr Sutherland when examining Mr Lundy's polo shirt(C3003) which was subsequently found to be in a degraded condition. The dab slide was takenfrom the stain on the left sleeve of the shirt. The Privy Council referred to this as the ESR slide,see [51] above.[334] We refer next to the DNA evidence. The two pieces of material cut fromthe shirt were subject to a beaker soak at ESR which extracts about 20 per cent ofthe biological material and in each case DNA testing of the eluted material was thatthe only human DNA found was that of the deceased Mrs Lundy.190 There was nochallenge to this aspect of the Crown's case. The relevant evidence was given byMs Vintiner, who in her evidence explained that since the evidence was given atthe first trial more sensitive tests had become available permitting a morefine-grained analysis of the DNA. Her key conclusion was encapsulated inthe following exchange:Q. So this time you've expressed it at one million, million, million, socan you just explain what that means?A. Well this is a very large number and it's one times 10 to the 18.Another way of expressing it is one billion, billion. So by that I meanthat this DNA evidence is one million, million, million times morelikely to occur if the DNA in these two samples from the polo shirtoriginated from Ms Christine Lundy, rather than from someone elseunrelated to her and chosen at random from the general New Zealandpopulation.[335] In other evidence Ms Vintiner responded to a question about the quantity andquality of DNA extracted from both C3003/3 and C3003/4 by saying:Well I've mentioned from the quantitation test that in forensic speak theamount of DNA that was recovered I would regard as a high level of DNA.And the profiling results themselves were both of good quality so thatindicates that the DNA in each sample was of good quality there was noevidence of degradation of the DNA breaking down.[336] She also said the quantity and quality of the DNA meant it could not bedescribed as trace DNA, that is DNA that had been transferred by simply touching anitem. Trace DNA, in contrast to what had been found, would be detectable in verylow amounts. The absence of any sign of degradation was another indication it wasnot trace DNA. It must be remembered that there is no narrative that suggestsMrs Lundy's DNA got on the shirt as a result of contamination.[337] We consider this evidence led to the inescapable conclusion that the presenceof Mrs Lundy's DNA on the shirt was not the result of any normal domestic190 Although a trace amount of unidentified DNA was found in relation to C3003/4, which was likelypresent in the sample only after the testing in 2001.occurrence. Much more likely was the fact it was there because of Mr Lundy havingworn the shirt at the murder scene.[338] On the basis of the evidence about the DNA and the CNS tissue, the Crownsubmitted there was an irresistible inference that the CNS tissue on Mr Lundy's shirtwas CNS tissue that had come from Mrs Lundy. As Mr Morgan put it in this Court, ithad been demonstrated that Mr Lundy, apparently going about his ordinary activitiesand staying overnight in a motel, eating routine food, had ended up with freshCNS tissue (and nothing else) embedded into his shirt in two separate places.The conclusion that it was Mrs Lundy's CNS tissue can then quite easily be reachedin two small steps. First, on the night that the appellant was wearing the shirt,Mrs Lundy was murdered by having her head attacked with an axe with theconsequence established in Dr Pang's evidence that a significant portion of her brainwas exposed and indeed missing from the skull cavity. Second, the only human DNAfound in association with the CNS tissue was that of Mrs Lundy. It was of good qualityand found in substantial amounts.[339] There was a defence suggestion at one stage that Mrs Lundy's DNA might havebecome placed on her husband's shirt as a result of mucous expelled by a sneeze.Ms Vintiner accepted that mucous expelled by a sneeze could be a source of largeamounts of DNA. She contrasted this with a sneeze into the air spraying air droplets,which she did not consider would be a likely source of large amounts of DNA. In anyevent neither eventuality would explain why the CNS tissue had arrived at exactly thesame places on the shirt, a remarkably unhappy coincidence for Mr Lundy.[340] In approaching the issue of whether the Crown's evidence would inevitablyhave resulted in Mr Lundy's conviction if the mRNA evidence had not been called, itis relevant to note that the mRNA evidence was called so as to meet a possible defencecontention that the CNS tissue on Mr Lundy's shirt was from an animal in thefood chain. The Crown now submits there was in fact no evidence of that,the possibility merely being raised as a result of the condition of the dab slide.[341] However, Mr Eaton has referred to evidence called by the Crown fromMs Elizabeth Wictum, a scientist attached to the Forensic Unit of the School ofVeterinary Medicine at UCLA, Davis. He claimed her evidence supported a possiblenarrative that the CNS tissue found on Mr Lundy's shirt was from the food chain.Ms Wictum had conducted tests on elutions prepared by ESR from the pocket andsleeve on Mr Lundy's shirt and forwarded to her in California. She referred to the testsas the "meat ID", the "quantitative PCR" and the "amplified mitochondrial DNA"tests. The elutions were subjected to each test.[342] The first test was "weakly positive" for both pork and beef in both samples,and in one case weakly positive for sheep. The results were very low, indicating therewas very little mitochondrial DNA in both samples. It was "very weak" and "barelydetectable". The quantity of DNA present was "very low". The quantitative PCR testwas used to test the presence of "cattle nuclear DNA". It gave negative results. Thethird test was inconclusive. In relation to the results achieved in the first test,Ms Wictum later explained: because it's very low-level obviously it's not meat itself So this is moreconsistent with, for example, food spatter, if you were frying say a sausage orsomething that had those species in it, it's again very low-level backgroundtype of mitochondrial DNA.[343] Mr Eaton submits this was enough to found the defence that the CNS tissue onthe shirt could be animal CNS tissue, but that plank of the case had been effectivelydenied Mr Lundy because of the mRNA evidence. However, the amount ofanimal DNA detected by Ms Wictum was very small, and we think insignificant whencompared with the substantial quantity of Mrs Lundy's DNA located on the shirt inconjunction with the CNS tissue. Quite apart from the mRNA evidence we considerit likely the jury would have discounted the possibility of the CNS tissue being animalin origin because of the vast distance between the small traces of animal DNA foundand that of Mrs Lundy. It is not without significance that Mr Hislop, while submittingin his closing address to the jury that the CNS tissue could be animal in origin, did notattempt to rely on Ms Wictum's evidence as supporting that proposition. The Judgenevertheless reminded the jury of the evidence.[344] We consider the CNS tissue and DNA evidence together justified the Crown'ssubmission to the jury that Mr Lundy had Mrs Lundy's brain tissue on his shirt. Anyconclusion derived from the mRNA testing was unnecessary to justify that conclusion.[345] We have discussed above at some length the other aspects of the Crown casewhich were challenged on the appeal. The discussion of the fuel consumption anddistance travelled discrepancy leads us to the view that a jury would find thatMr Lundy would have been able to carry out the murders having regard to the amountof petrol available, and the time available. Added to this may be the followingconsiderations which were in themselves powerful contributors to the Crown'scircumstantial case:(a) The traces of orange and blue paint near the points of impact ofthe murder weapon (found on Mrs Lundy's bedding and embedded intoa skull fragment of hers and other traces associated with Amber Lundy),enabled the Crown to submit the murder weapon had left those traces.That justified an inference that one of Mr Lundy's tools had been usedas the murder weapon. The defence argued that such a tool might havebeen wielded by a random burglar entering the house or Mrs Lundymay have got an amount of paint in her hair as a result of being inthe garage where the paint was stored. These were matters for the jury,but they were entitled to accept the inference relied on by the Crown.(b) Red particles were observed on Mr Lundy's shirt which when testedyielded Amber's DNA. The defence argued this may have had aninnocent explanation but it was further circumstantial evidence the jurycould have relied on in assessing the Crown's case.(c) The notion that the deceased were the victims of a random burglar wasunlikely, for a combination of reasons. First, the savagery of the attackon both Mrs Lundy and her daughter. There were in each case a highnumber of very hard blows to the head. In the case of Mrs Lundy theyhad the effect of destroying much of her face and head. Further,Amber Lundy could not have been regarded as any kind of threat to arandom burglar, as opposed to somebody she recognised in the courseof the attack on her mother. Further, a blood stain able to be linked tothe deceased Mrs Lundy was found on the interior lip of the frame ofan open window to the conservatory. The Crown submitted that thiswas manufactured evidence of a forced entry. Again, the taking ofthe jewellery box from the bedroom was contrasted with the fact thatMrs Lundy's purse and wallet remained. All of these considerationsrebutted the idea that Mrs Lundy and her daughter had been the subjectof a random attack by an intruder.(d) Although the defence sought to suggest that Mrs Lundy's brotherMr Weggery was responsible, there was no real evidential foundationfor that. He had been eliminated by the police from their inquiry.(e) A silver bracelet was found on the seat of the car. Mr Lundy confirmedthat it belonged to Mrs Lundy. The Crown said the bracelet was nolonger used by her (being too small) but it was likely to have been inthe jewellery box supposedly stolen by the intruder.(f) The Crown's case included reference to what it said were attempts byMr Lundy to mislead the police during the course of the inquiry.First, he had claimed to the police not to be under financial stress.Second, he had at one stage attempted to explain the empty petrol tankon 30 August 2000 by suggesting he may have been the object ofthe theft of petrol. Third, the Crown invited the jury to reject hisexplanation that he had not parked his car outside his motel unit onthe night of 29 August having left the motel in the early evening to goto the Petone foreshore to read a book.(g) There was an insurance policy over Mrs Lundy's life from whichMr Lundy would benefit on her death. Given the financial difficulties,he had a motive.(h) There was some evidence of conflict over financial matters betweenMr and Mrs Lundy.[346] Mr Eaton noted what was said by the Privy Council in allowing Mr Lundy'sappeal.191 The Privy Council rejected the application of the proviso on the basis thatthe Crown's contentions about DNA from Amber's blood, the areas of smeared bloodon the inside of the window frame, the suggestion of theft of petrol and the braceletwere matters which were not sufficient to establish guilt. We accept that is so, but weare also of the view that they were matters capable of supporting a strongcircumstantial case once the CNS tissue and DNA evidence is accepted. The positionis different to that which applied before the Privy Council.[347] Mr Eaton also emphasised two aspects of the evidence that he said pointedaway from Mr Lundy being the offender and should make this Court unsure ofMr Lundy's guilt in the absence of Dr Sijen's evidence. First, he relied onDr Pang's evidence that the deceased when examined at the post-mortems had fullstomachs. Assuming they ate a meal that was purchased from McDonald's soon afterits purchase, between 5.43 and 6.30 pm, the stomach contents should have beencleared before Mr Lundy's arrival in Palmerston North, at the earliest at 2.30 am.192Mr Eaton noted evidence that Amber normally went to bed at 8 pm. He contendedthat on the Crown case she would have to have kept eating long after that. The defencecase at the trial was it was far more plausible that the murders occurred at a muchearlier stage in the evening, shortly after the computer was turned off at 10.52 pm.[348] The evidence that Mrs Lundy's and Amber's stomachs were full came fromDr Pang who had carried out the post-mortems. It is fair to say that his evidence atthe first trial had been substantially discredited by experts who gave evidence tothe Privy Council, rejecting in particular his claim to be able to estimate time of deathon the basis of absence of smell from the stomach contents. That did not of courseaffect what he purported to have observed physically. In the case of Mrs Lundy, hereported that the stomach contained "a large meal consisting of potato chips andprobably fish which showed no obvious signs of digestion and no obvious smell ofgastric juices". The report was similar for Amber. These observations were apparentlybased on an examination of the duodenum.191 Privy Council judgment, above n 1, at [154]–[159].192 The meal purchased was reflected by a McDonald's receipt produced in evidence. It consisted ofnine nuggets, one filet-o-fish, one chicken burger, one medium fries, one large fries and twoapple pies.[349] He conceded in his evidence-in-chief at the second trial that he had been wrongto rely on the absence of smell. He now said it could not be said with certainty whenthe deceased died over the period until their bodies were found.[350] Notwithstanding that position, Dr Pang was reluctant to concede that he haddramatically altered his evidence from that given at the first trial when he purportedto be able to say on the basis of both smell and stomach contents that the meals musthave been consumed within one and a quarter hours of death. On that issue, it is clearDr Pang prevaricated and appeared reluctant to accept obvious propositions that wereput to him by Mr Hislop in cross-examination. This would have affected his credibilitywith the jury. He was eventually brought to accept that in the majority of individuals,it takes at the outside about six hours for the stomach to go from full to empty.[351] The Crown's case on this issue was substantially based on the evidence ofanother forensic pathologist called Dr Sage. Dr Sage effectively challenged Dr Pang'sability to conclude that there had been emptying into the duodenum. He noted that thenormal state of a duodenum in any autopsy is more or less empty in appearancebecause, the body lying flat, there is no hindrance to it emptying into the rest of thesmall bowel. Consequently, it is not possible to reliably say whether or not there hasbeen significant emptying into the duodenum. In Dr Sage's opinion it could notreliably be said whether or not there had been significant emptying into the duodenum.He also emphasised the difficulty of describing a stomach as full. He accepted that ifthere is still what appears to be undigested food in the stomach, it suggests the stomachhas not been emptied completely. However, a number of factors could influence thatand in order to be certain about the implications of the stomach contents for the timeof death, it was necessary to be certain about when the last meal was eaten.[352] The defence expert, Professor Horowitz, is a Professor in the Department ofMedicine at the University of Adelaide and a Director of the Endocrine andMetabolic Unit of the Royal Adelaide Hospital. He gave evidence that was based onthe calculated calorie content of the meal that Mrs Lundy had purchased onthe evening of 29 August. On the assumption that Dr Pang was right and he hadobserved the duodenum as being empty their last meal would have been consumed amaximum of two hours before they died. If Dr Pang's observation that there wasnothing in the duodenum was in fact incorrect, the stomachs would nevertheless havebeen completely empty within six to eight hours but probably less. The defence reliedon this evidence as establishing that the murders must have occurred well before thetime when Mr Lundy could have returned to Palmerston North. But there wascommon ground between the experts that using stomach contents to estimate time ofdeath was fraught with difficulty.[353] Mr Morgan noted that there had been some evidence, by the witnessDeborah Malcom (a friend of the Lundys), that Mrs Lundy did have a habit ofpurchasing takeaway food and not eating it immediately, heating it up and eating itlater. Mr Morgan conceded, however, that it is unlikely that Amber would have beenprepared to wait. But he claimed that in the end it was not possible to discount thepossibility that they had eaten later than a time proximate to the purchase of theMcDonald's meal or indeed that they had eaten further food after the McDonald's mealwas consumed.[354] In our view Dr Pang's evidence about the stomach contents is too uncertain toenable any firm conclusion to be formed about the time of death.Professor Horowitz's evidence took as a starting point, based on Dr Pang's evidence,that there was no evidence of solid or liquid meal contents distal to the stomach, sothat gastric emptying had not commenced. But the points made by Dr Sagesubstantially negate the significance of this observation. Ultimately, these issues werefairly put by Simon France J in his summing-up, and the jury was not persuaded bythe defence theory of impossibility based on the stomach contents of the deceased.Having considered the evidence, we are left in the same position. In our assessmentthe stomach contents evidence is too uncertain to raise a reasonable doubt aboutMr Lundy's guilt.[355] This uncertainty is reinforced by the fact that Mrs Lundy must have been alivewhen her computer was turned off at 10.52 pm. The post-mortem evidence indicatedthat there was food in both Mrs Lundy's and Amber's stomachs of a type that is notavailable at McDonald's, and therefore likely to have been consumed at a timeconsiderably later than around 6 pm.[356] The second counter indication to Mr Lundy's guilt emphasised by Mr Eatonwas said to be a defect in the approach taken by Ms Ellwood on whose analysisthe Crown had relied on the petrol consumption aspect of the case. It was said thatshe had no particular expertise on the subject of petrol consumption rates and she hadnot made any allowance for the speed and manner of Mr Lundy's driving during hisfast return to Palmerston North after learning of the presence of the police at his home.These issues were thoroughly canvassed before the jury. We have already recordedour view that the evidence on which the Crown relied was sufficient to establishMr Lundy had the opportunity and fuel needed to make the journey to and fromPalmerston North on the morning of the murders. Again, this issue is not related tothe inadmissible evidence. In any event, this point does not persuade us that Mr Lundywould not be convicted for the reasons already addressed in discussing the fuelconsumption and distance travelled aspect of the case.[357] Mr Eaton also referred to other matters that had been raised by the defence atthe trial, including:(a) There were a number of unidentified fingerprints at the scene as wellas an unidentified footprint.(b) There was an absence of blood in Mr Lundy's car, on his glasses andring and the motel unit where he stayed in Petone. The defence hadrelied on the evidence of an expert Ms Gillian Leak, a biologist andprofessional member of the Chartered Society of Forensic Science withexpertise in forensic investigation of crime scenes, to emphasise thedifficulties Mr Lundy would have had to confront in avoiding any traceof blood other than that found on his shirt in what was a very bloodymurder scene. Mrs Leak also referred to the risk that a garment wornover the polo shirt would have left fibres on the shirt.(c) There was the possibility, again referred to by Ms Leak, that the bloodsmear on which the Crown relied for its staged burglary theory was theresult of contamination by police and others present at the scene afterthe bodies were found.(d) Scrapings taken from under the fingernails of both the deceased, afterbeing analysed in 2014, were shown to contain the DNA of unknownmales.(e) A number of fibres were found under the fingernails of both deceased.These were analysed and could not be attributed to Mr Lundy'spolo shirt.(f) There were 11 hairs in the upturned palm of Mrs Lundy's left hand and10 hairs in her right hand.(g) There was evidence given by a neighbour, Mr Tupai who said he sawthe conservatory door of the Lundy house open while standing outsideon his property talking on the telephone. He was talking to his fatherin Samoa, and telephone records established the call took place overseven minutes and 25 seconds after 10.59 pm on 29 August.As mentioned above there was evidence the computer was turned off at10.52 pm. The defence argued at the trial that this was evidence themurder took place at a time when Mr Lundy was in Petone.[358] All of these matters must have been rejected by the jury as not raising areasonable doubt about Mr Lundy's guilt. None of them are particularly cogent as anindicator of another person being responsible for the crime. In our own view, they donot raise a credible basis on which we could go behind the implications ofthe CNS tissue and DNA found on Mr Lundy's shirt.[359] With particular reference to the fingernail scrapings we note that the analysis,carried out by the Crown's witness Ms Heidi Baker, involved what is called "Y-STRDNA testing". Ms Baker described this as "highly sensitive" with the consequencethat it is not always possible to determine when any DNA was deposited or how it wasdeposited, or what its source was: fluid, blood, or another tissue such as skin.Ms Baker observed: it's very important to consider what legitimate opportunities people mayhave had to transfer their DNA to those samples because it is such a highlysensitive technique and will detect incredibly low levels of DNA.She gave as an example that the DNA transfer could be effected by a simplehandshake.[360] In the case of Amber, the right fingernail scrapings showed there was DNAfrom at least two and possibly more unrelated males in this sample. In the case of MrsLundy scrapings from her left fingernail indicated there was DNA from at least oneand possibly more males in the sample. The DNA was in very small amounts, possiblythe result of scratching but possibly also the result of normal day to day contact. Wedo not consider the evidence about the finger nail scrapings to be of significance tothe outcome of the case.[361] As to the hairs, these were uplifted at the mortuary on 2 September 2000 byDetective Jonathan Oram who produced a sketch he made, showing the location of thehairs, in his notebook . We were not referred to (and have not been able to find in therecord) any further evidence concerning these hairs, or analysis undertaken concerningthem. This evidence does not assist in resolving the issues we have to decide.[362] Mr Eaton further submitted that the Court should be reluctant to applythe proviso given the fact that the Crown case was entirely circumstantial and involvedthe drawing of inferences from a variety of different pieces of physical evidence.He emphasised the length of the trial, the extensive cross-examination and the fact that140 witnesses had been called. He also reminded us about the Supreme Court'sobservations in Matenga that in coming to a conclusion concerning the inevitabilityof a verdict, the appeal court:193 must of course take full account of the disadvantage it may well have inmaking an assessment of the honesty and reliability of witnesses on the solebasis of the transcript of the oral evidence.[363] However this is not a case that turned on the credibility or honesty of witnesses.Nor is it a case like Gassy v R, an Australian case to which we were referred by193 R v Matenga, above n 59, at [32].Mr Eaton, which turned on what the jury made of observations made about actions ofthe defendant by witnesses who did not know the defendant but purported to identifyhim.194 While not seeking to downplay advantages that must accrue to the trial courtfrom seeing expert witnesses give their evidence, we consider this case in the end turnson the assessment of scientific evidence and its implications. It is not in the categoryof requiring more than the usual degree of care in applying the proviso.[364] In the end we have been left sure of Mr Lundy's guilt, and that aspect ofthe proviso has been satisfied.[365] This brings us to the final issue that must be addressed. The question iswhether the wrongful admission of the mRNA evidence meant that the trial was unfair.The issue has to be addressed because even though this Court on appeal might be surea defendant is guilty, he or she is entitled to a fair trial. As was said in Matenga, theCourt must also be satisfied that there has been no breach of the right guaranteed bys 25(a) of the New Zealand Bill of Rights Act.[366] Mr Eaton submitted that the wrongful admission of the mRNA evidence meantthat the trial was unfair. In the circumstances, the proviso could not be applied.In support of the proposition that the trial was unfair, Mr Eaton referred to thejudgment of the majority of the High Court of Australia in Wilde v R, in which it wassaid:195It is one thing to apply the proviso to prevent the administration of the criminallaw from being "plunged into outworn technicality" it is another to upholda conviction after a proceeding which is fundamentally flawed, merelybecause the appeal court is of the opinion that on a proper trial the appellantwould inevitably have been convicted. The proviso has no application wherean irregularity has occurred which is such a departure from the essentialrequirements of the law that it goes to the root of the proceedings. If that hasoccurred, then it can be said, without considering the effect of the irregularityupon the jury's verdict, that the accused has not had a proper trial and thatthere has been a substantial miscarriage of justice. Errors of that kind may beso radical or fundamental that by their very nature they exclude the applicationof the proviso 194 Gassy v R [2008] HCA 18, (2008) 236 CLR 293.195 Wilde v R (1988) 164 CLR 365 at 373.[367] Mr Eaton submitted that the Crown's reliance on Dr Sijen's mRNA evidencein this case was an irregularity that constituted such a departure from the essentialrequirements of the law that it went to the root of the proceedings. He noted that inWilde the High Court of Australia held that the wrongful admission of inadmissibleevidence can render a trial unfair and that "[i]t is the significance of the evidencewrongly admitted, in the context of the trial, which must determine whether the errorwas of a fundamental kind."196[368] Mr Eaton also noted that Wilde had been considered and approved bythe Privy Council in R v Howse, in which reference had also been made to anotherstatement of the law by the Privy Council in Randall v R. In that case Lord Bingham,writing for the majority of the Board, stated:197 it is not every departure from good practice which renders a trial unfair.Inevitably, in the course of a long trial, things are done or said which shouldnot be done or said. Most occurrences of that kind do not underminethe integrity of the trial, particularly if they are isolated and particularly if,where appropriate, they are the subject of a clear judicial direction. It wouldemasculate the trial process, and undermine public confidence inthe administration of criminal justice, if a standard of perfection were imposedthat was incapable of attainment in practice. But the right of acriminal defendant to a fair trial is absolute. There will come a point whenthe departure from good practice is so gross, or so persistent, or so prejudicial,or so irremediable that an appellate court will have no choice but to condemna trial as unfair and quash a conviction as unsafe, however strong the groundsfor believing the defendant to be guilty. The right to a fair trial is one to beenjoyed by the guilty as well as the innocent, for a defendant is presumed tobe innocent until proved to be otherwise in a fairly conducted trial.[369] In Randall the issue was misconduct by prosecuting counsel in circumstanceswhere the trial Judge had failed to enforce proper standards of conduct. By contrast,Wilde and Howse were both cases in which the issue said to make the trial unfair waswrongly admitted evidence. Mr Eaton submits that properly applied, those authoritiesshould have the consequence that what we have held to be the wrongful admission ofDr Sijen's evidence meant that the trial was unfair.[370] Mr Eaton advances that proposition on the basis that the inadmissible evidencewas at the heart of the Crown's most important submission to the jury, namely that196 At 374.197 Randall v R [2002] UKPC 19, [2002] 1 WLR 2237 at [28].Mr Lundy had his wife's brain on his shirt. Mr Eaton quoted in particularthe following passage from Mr Morgan's closing address:The Crown say the central nervous system tissue on his shirt demonstratesclearly and unequivocally he's the killer. And actually, the use of this cardemonstrates he's the killer as well. So those will be two of the critical partsof my address to you over the next little while.Now, let's just put it out there, members of the jury. My submission to you is,the Crown case to you is, that Mark Lundy has his wife, Christine Lundy'sbrain on his shirt. That is the effect of the evidence. All — everybody sort ofdanced around the edges of it in this trial didn't they, becausethe neuropathologists all say, "Oh, no all we can say is that it's central nervoussystem tissue." And the DNA lady says, "Oh, oh, all I can tell you isthe diluted biological material out of those, out of those fabric scraps and it'sChristine Lundy's DNA." And all the lady from the NFl can say is, "Well,I've conducted a test and the conclusion of my test is brain observed andit's more probable than not that it's human brain than the brain ofthe animal species tested."And then there's all this business about contamination lurking in there as welland Professor Ironside comes along and speaks about necrotic tissue on theone hand and degraded tissue on the other. I'm going to address you on all ofthose topics members of the jury because I can. I'm not constrained bythe fact that I'm a neuropathologist or just a DNA expert. I'm making asubmission to you, based on the whole of the evidence, and if you put itall together it demonstrates, clearly and unequivocally, Mark Lundy hadChristine Lundy's brain on his shirt.(Emphasis added by Mr Eaton.)[371] Mr Eaton contended that this submission placed Dr Sijen's evidence at the coreof the Crown's case that the tissue on the shirt was Mrs Lundy's brain. He claimedthat Mr Morgan had referred to it as "the most important piece of evidence in thiscase".[372] So this was not a case where the inadmissible evidence related to some unfairlyprejudicial detail which might have influenced the jury against the defendant.Rather, the inadmissible evidence had been deployed as part of the answer to the mostimportant evidential question in the case, namely the origin of the CNS tissue.He submitted that the evidence could only have had "additional significant weight" inthe sense envisaged in Wilde and Howse.[373] In response to these submissions Mr Morgan noted that Mr Lundy had beengiven the opportunity of having his own expert advisor, Dr Vennemann, attend atthe NFI laboratory to enable her to assess the process carried out and she had givenevidence on Mr Lundy's behalf at the pre-trial hearing. There had been afull opportunity to contest the inadmissibility of the evidence at that stage and afull opportunity to challenge the evidence at the trial on the basis of the expertevidence called by the defence. In addition, the Judge had provided the jury with acomprehensive set of directions on the NFI issue. Mr Morgan contended that therecould be no doubt in these circumstances that the appellant in fact had a fair trial.[374] Those submissions only take the Crown so far. Given the single error that wehave identified relating to the mRNA evidence, it is not possible to bring this trialwithin Lord Bingham's category of a case where the departure from good practice hasbeen "so gross", or "so persistent" as to result in an unfair trial. But the authorities doshow that the wrongful admission of important evidence can of itself have theconsequence that the trial was unfair. That was the view of the Privy Council inHowse, and the possibility was also recognised by the High Court of Australia inWilde.198 The question then is whether this is that kind of case.[375] The passage quoted above from Wilde was referred to with apparent approvalby Elias CJ and Glazebrook J in Guy v R.199 They went on to observe:[36] The threshold on which it may be concluded that a trial is unfair is setat a high level; the operation of the proviso is "not to be stultified". But, inconsidering whether a trial is indeed fair, the inquiry is on the right to fair trialitself, not the proviso question whether the appellate court is satisfied ofthe guilt of the accused on the basis of the evidence.(Footnotes omitted.)[376] That the threshold is high is also apparent from the fact that the majority inWilde were referring to circumstances in which the proceedings before the pre-trialcourt had "so far miscarried as hardly to be a trial at all".200 The high threshold is alsoapparent from the language employed by Lord Bingham in Randall. We concludehowever, that in a case where the issue is whether wrongly admitted evidence hasmade the trial unfair, the answer depends on an assessment of the significance of198 Wilde v R, above n 195, at 373.199 Guy v R [2014] NZSC 165, [2015] 1 NZLR 315 at [35].200 Wilde v R, above n 195, at 373.the evidence wrongly admitted, in the context of the trial. That is the test applied inWilde,201 and we apprehend it is the test that we must apply having regard to itsadoption by both the majority and minority judgments in Howse.202[377] Notwithstanding its adoption by both the majority and minority in Howse therewas disagreement over the application of the test. The majority concluded that whenviewed in context the evidence wrongly admitted did not constitute a fundamentalerror that made the trial unfair.203 The context addressed for that purpose includedevidence that was considered to be an overwhelming prosecution case againstthe defendant, so that there was no realistic possibility that the jury would have felt itnecessary to have recourse to the inadmissible evidence to be satisfied that the accusedhad murdered the deceased girls.204 It concluded:205No doubt the jury took the inadmissible evidence into account in coming totheir verdict, and this will often be the position where inadmissible evidencepointing to guilt is admitted. But where the other evidence properly admittedproves with overwhelming force that the accused is guilty, their Lordshipsconsider that it cannot be said that the admission of the improper evidenceconstituted a fundamental error which made the trial unfair.[378] It is clear that in this context the strength of the prosecution case is to beconsidered not for the purpose of asking whether, had the impugned evidence not beencalled, there would nevertheless have been a conviction. Rather, the other prosecutionevidence is relevant to the question of whether the evidence would have assumed suchimportance its wrongful admission could be said to have made the trial unfair. As wassaid in Wilde in resolving the issue presented in that case:206When viewed in context, it does not appear that the evidence wronglyadmitted in relation to the counts upon which the applicant was convicted canhave carried any significant additional weight having regard to the otherevidence. This is, of course, to take into account the strength ofthe prosecution case upon those counts and the weakness of the defence, butit is to do so for the purpose of determining the gravity and significance ofthe error and not for the purpose of determining whether the jury wouldinevitably have convicted notwithstanding the error. The two questions areobviously intertwined where the error is one of the wrongful admission ofevidence, but they must be considered separately.201 At 374.202 R v Howse, above n 171, at [36] and [55].203 At [39].204 At [39].205 At [39].206 Wilde v R, above n 195, at 374.[379] In Wilde the defendant had been tried on a number of counts in relation tooffending committed in three separate incidents. The Court of Criminal Appeal heldthat the evidence relating to the first two counts alleged in the first incident shouldhave been severed. At the trial, the jury acquitted the defendant on one of those counts.It convicted him on the remaining offences, committed in the other two incidents.The High Court held there had not been a fundamental error making the trial unfairbecause the evidence on counts one and two would not have added significantly tothe prosecution case on the charges based on the subsequent events.[380] Lord Rodger and Sir Andrew Leggatt, who wrote the minority judgment inHowse, after analysing the approach taken in Wilde, said:[56] Applying the same approach in the present case, we can, of course,take account of the strength of the other evidence available to the Crown onthe murder charges and the apparent weakness of the defence, but only as oneof the factors to be considered when deciding how significant the admissionof the inadmissible evidence was in the context of the trial. By contrast, touse it at this stage, as the majority of the Board propose, to decide that "noreasonable jury would have acquitted [the appellant] of the murder of thetwo girls" is to do precisely what the High Court say, at p 374, an appeal Courtshould not do when considering whether the trial was fair.[381] In the balance of the judgment the minority dealt with the very differentcircumstances of Howse, discussing the "catalogue of defects" which had beenidentified in the trial by the Court of Appeal.207 These included: the admission of alarge amount of hearsay evidence which should have been excluded; the admission ofan undesirable amount of evidence that was more prejudicial than probative;the failure to give a clear and firm direction to the jury about the limited proper use ofa particular item of evidence, and the inclusion in the summing-up of an unnecessaryand emotive explanation of why manslaughter was not a live issue.208 The lengthydiscussion of the errors, in particular the repetitive hearsay evidence designed toestablish the defendant's motive on the basis he had sexually abused the victims,the misuse of that evidence as establishing the truth of its contents, and the waythe Judge had summed up had in combination resulted in an unfair trial.209 In fact, theminority stated, it was:210207 R v Howse, above n 171, at [57].208 At [57].209 See the lengthy treatment of these issues at [61]–[68].210 At [69]. impossible to imagine a clearer example of a trial that has gone off the railsby the admission of evidence which, the law provides, should not be admittedprecisely because it is dangerous for a jury to rely on it.[382] This is a very different case. We have no doubt that the Crown called Dr Sijen'sevidence because it considered it would add significant weight to its case that the tissueon Mr Lundy's shirt was likely to be Mrs Lundy's. But it was only one strand ofthe evidence on which it was able to rely for that purpose. Significantly,the conclusions that Dr Sijen was able to express were couched in much less certainterms than the conclusions reached on the CNS tissue and DNA evidence which wereon the preponderance of the evidence of far more compelling indicators that theCNS tissue identified was that of Mrs Lundy. In that respect, the association ofMrs Lundy's DNA with the tissue analysed must have been compelling. By contrast,Dr Sijen's conclusions were expressed on the basis that the tissue was "more probable"to be human than the other animal species examined, but she was unable to state howmuch more probable.[383] We think it likely in addition that the force of Dr Sijen's evidence must havebeen substantially reduced by the attack made on it by the defence witnesses.Ideally, we accept that the defence ought not to have been required to mount thatattack. Confronted with the evidence, however, a vigorous assault on it was mounted.That cannot have had an effect other than to reduce its impact.[384] Nor do we accept Mr Eaton's claim that the mRNA evidence prevented thedefence from running an argument that the source of the CNS tissue was from the foodchain. First, that is the argument that was run under the guise of the variouscontamination suggestions. The mRNA evidence did not in any sense prevent thatdefence being run. It may have meant that it was less convincing, but it could also besaid that the defence did not call any significant evidence to strengthen the practicalpossibility that Mr Lundy had got food on his shirt and that such food was the sourceof the CNS tissue.[385] Mr Hislop, in closing the case for the defence, endeavoured to present such anargument to the jury. He effectively invited the jury to set aside Dr Sijen's evidenceon the basis it had been shown to be unreliable and submitted that the tissue on the shirtwas either the result of contamination, or it was CNS tissue which was not human:"[i]t could be food stuffs". He said:Can I make this clear? He doesn't have to have been a hunter. He doesn'thave to have been an offal eater. Let's ignore those extreme examples.Hyperbole of example is a well known forensic tool. It's designed to persuadeyou from a point of view. It's not going to help your task. Put it to one sideand let's deal with the hard evidence. He was not a hunter but he didn't needto be. He was a cook, that's the hard evidence. He was the cook in the house.He didn't need to be [an] offal eater. You know and heard Professor Ironsidetell you about neck chops and indeed any type of chops. So we try cases onhard evidence not hyperbole. To steal a favourite phrase from my learnedfriend, forget the notion of a hunter, forget the notion of an offal eater and deal,we suggest, with the hard evidence, he was a cook and we know what's inneck chops and many other chops in this country, especially way back then.[386] There was nothing to prevent that evidence being put on a firmer footing as apossible narrative. Dr Sijen's evidence did not stand in the way of that being done.[387] We accept Mr Eaton's submission that the Crown sought to rely onthe mRNA evidence, but we do not accept his submission that Mr Morgan describedit as "the most important piece of evidence in this case". That submission waspurportedly made in reliance on what Mr Morgan said near the end of hisclosing address:I've made a submission to you that I started yesterday afternoon, which wasthis is Christine Lundy's brain and I accept that is the most important piece ofevidence in this case, but it does not stand alone.[388] The mRNA evidence however did not establish that the tissue was Mrs Lundy'sbrain. Fairly read, we consider Mr Morgan's submission was based on the evidenceas a whole. In establishing that the tissue was Mrs Lundy's brain, the most powerfulevidence in fact was the evidence that it was solely CNS tissue, that there was a highquantity of Mrs Lundy's DNA and the very fact that Mrs Lundy's brain had beenexposed and partly scattered in the attack that ended her life. We do not considerMr Morgan claimed or asserted that the mRNA evidence was the most powerfulevidence in the case, and we do not think it was. We do not consider the jury wouldhave seen it in that light either.[389] This is consistent with the way the Judge summed up. Under the heading"[t]issue on the shirt" Simon France J described the Crown's starting point ("and whatthe Crown would submit is the end point as well") as two facts: first, the presence ofCNS tissue on the shirt that Mr Lundy had been wearing the night his wife was killedin a manner that exposed her brain, and second, the fact that the two pieces of tissuecut from the shirt yielded significant amounts of Mrs Lundy's DNA. He continued:[75] I set these two facts out in this manner because I think it capturesthe essence of Mr Morgan's argument on this part:(a) Mrs Lundy was killed in a way that exposed large portions of her brainand flung them around the room;(b) CNS tissue was found smeared on the shirt her husband was wearingthat same night;(c) on the same bits of shirt, in good quality amounts, Mrs Lundy's DNAwas found.[76] Leaving to one side Dr Sijen and all that debate, the Crown says toyou that you can be satisfied based on these facts it is her brain. It submittedthe facts irresistibly point that way, and any other explanation is just unreal.[390] When addressing Dr Sijen's evidence the Judge prefaced it by saying"[t]he Crown case on this tissue issue is at its strongest if you accept her evidence".Then, after thoroughly addressing the arguments on both sides about its reliability, hesaid:As I say if you accept Dr Sijen's evidence, then the Crown's argumentbecomes stronger. If, however, you accept the criticism of Professor Bustinand Dr [Vennemann], and put it to one side, then it is still necessary to considerthe validity of the conclusion the Crown is seeking to draw from those basicfacts, namely the combination of the way she was killed, the shirt he wore onthe night, the presence of central nervous system tissue, and the presence inthe same area of her DNA establish it is Mrs Lundy's brain.[391] Plainly the Judge in summarising the Crown case was not saying that themRNA evidence was the most important evidence in the case. He was anticipatingthat this evidence, because it was so controversial, might be put to one side by the juryand that the jury would consider the Crown case in relation to the tissue on the shirtby reference to the CNS tissue and DNA evidence only.[392] In the end we have not been persuaded looking at the evidence as a whole andthe course of the trial overall that the admission of Dr Sijen's evidence had the effectof making the trial unfair.[393] Consequently, nothing stands in the way of applying the proviso.For the reasons we have set out we are sure of Mr Lundy's guilt while placing noweight on the mRNA evidence. We do not think that the admission of that evidenceled to an unfair trial. We have decided that the proviso should be applied.Result[394] The appeal is dismissed.[395] The applications to adduce further evidence for the purposes of the appeal aregranted or declined in accordance with the attached schedule.Solicitors:Crown Solicitor, Palmerston North for RespondentScheduleA. Leave is granted to adduce the following affidavits for the purposes of theappeal:(a) the affidavit of Allan Burns dated 18 October 2016;(b) the affidavit of Michael White dated 8 June 2017;(c) the affidavit of Kloe Palmer dated 16 June 2017;(d) the affidavit of Anna Sandiford dated 16 June 2017;(e) the affidavit of Stephen Bustin dated 31 July 2017;(f) the affidavit of Sally Harbison dated 22 September 2017;(g) the affidavit of Laetitia Sijen dated 5 October 2017;(h) the affidavit of Daniel du Plessis dated 5 October 2017;(i) the affidavit of Stephen Bustin dated 17 November 2017; and(j) the affidavit of Laetitia Sijen dated 11 December 2017.B. Leave to adduce the following affidavits for the purposes of the appeal isdeclined:(a) the affidavit of Philip Sheard dated 14 June 2017;(b) the affidavit of Bruce Robertson dated 14 June 2017;(c) the affidavit of Mike Ware dated 15 June 2017; and(d) the affidavit of Sean Doyle dated 10 July 2017.