I am satisfied beyond reasonable doubt that the DNA extracted by Ms Scibetta was, as she firmly stated, "true to the bone". I am also satisfied beyond reasonable doubt that the results obtained by Ms Scibetta and Ms Lee are valid and true reflections of the bones found at Tregear. I am also satisfied beyond reasonable doubt that the bones are those of a female, that their mitochondrial DNA is consistent with an offspring of Christine Strachan and that their nuclear DNA is consistent with an offspring of Christine Strachan and Gaspar Baan.
731 These results do not, of course, establish that the bones are those of Jean Keir. As Mr Hanley put, to consider what, if anything, they contribute to the establishment of the identity of the bones it is necessary to consider both the statistical evidence proffered by Mr Goetz and, also, the other evidence on the case.
732 Mr Goetz is a very experienced forensic biologist employed by the Division of Analytical Laboratories in Lidcombe. There is no issue as to his qualifications to express opinions on questions relating to DNA including the use of data bases and statistical principles to develop, amongst other things, likelihood ratios.
733 Provided with the nuclear DNA profiles obtained by Ms Lee of the two bones examined and of Gaspar Baan and Christine Strachan Mr Goetz prepared a table which posed the following question:
"How many times more likely to find the profile from the bone if it originated from a child of Baan and Strachan than from a child of an unknown couple in the population. The calculation also includes a correct (sic) for sample size (95 percent confidence limits)".
734 The table then provided likelihood ratios for a considerable number of different populations with three different levels of statistical correction.
735 Referring to figures taken from New South Wales and Queensland data bases Mr Goetz gave the following evidence:
"Q. In other words, in effect using the heading on page 1 are you saying that it is 447,000 times more likely to find the profile that Demris Lee came up with from the bone if it originated from a child of Baan and Strachan than from a child of an unknown couple in the population.
A. That's correct."
736 Referring to figures taken from New South Wales and New Zealand data bases Mr Goetz gave the following evidence:
"Q. And there what did you do?
A. I again used the New South Wales data for profiler plus, six loci, and this time I used, rather than three loci from Queensland I used three loci from New Zealand. And again their population was slightly less. They only had 206 people in their database and I calculated the whole calculation for sample size determined on a sample size of 206 and then came up with a result for the 3 percent FST an upper confidence level of 358,000"
737 Referring to data bases from a wide range of other countries Mr Goetz gave the following evidence:
"Q. And the fourth page as well. Looking at the figures you got for the three at the top of page 3 and all the others as well, what figure did you report to this court as the appropriate figure in terms of the question you posited at the top of page 1, how many times more likely et cetera?
A. I have given the court, I suppose, a range of figures that they could select. When you look at the table I presented, that table there (indicated), you can see the range in a tabular form, I suppose, or graphical form. It does show that there is, although it looks wide it really, scientifically is not that significantly different from each other. You probably could draw a line, I would estimate around the 400,000 mark that would cover the majority of the population above that line and the majority of population below that line, and estimate as I did with New South Wales that 447,000 is fairly a good estimate, of the likely ratio in this case.."
738 In cross-examination Mr Goetz gave the following evidence:
"Q. I think his Honour may have asked you about whether there was a medium or contra graph, and I think you pointed to 400,000 level. Does that take into account assessing the likelihood ratio in relation to the world populations?
A. The likelihood ratio I am referring to here is really independent of world population or any population size. It is looking at two hypotheses and comparing one hypothesis to the other. It is not comparing population sizes at all.
….
Q. What about if you added up the number of people in each of the databases and used the same calculations would that have any differing effect on your likelihood ratio? Or would you expect to get a mean of about 400,000?
A. That's an interesting idea that I never thought about. Because then I would have to add up all the allele frequencies for each of the different locations and then I have ( sic ) would have to work out an average for those and base my calculation on that. I never thought of doing that because it's not really a scientifically valid way of doing it.
I believe if you look at the number of values that have been obtained below 400,000 and above 400,000 and as mentioned here I have included ( obviously should be 'exclude' ) two way above it, it was in 4 millions from memory, you end up with a mid point of about 400,000. But I am not saying that is the one that is definitely the one should be taken. By looking at the graph I think it's an estimate that is reasonable."
739 Whilst it is not entirely clear that Mr Goetz would regard the figure of 400,000 as a "reasonable" alternative to his figure of 447,000 I think the better course is for me to adopt the lower figure of 400,000 as an appropriately conservative estimate of the likelihood ration. I should mention that the 447,000 was itself taken from the most conservative level of statistical correction.
740 Having regard to the absence of any scientific evidence to challenge Mr Goetz's view and the limited nature of cross-examination it is not necessary for me to go to the detail of how Mr Goetz arrived at his figures. I should mention that he provided for a difference in the kits used in America (Profiler) with the kit used in New South Wales (Profiler Plus) by methods which were not challenged. I should also note that I have directed myself that I am not bound to accept the expert evidence or to act on it unless I consider it appropriate to do so.
741 I consider that the evidence established that it is 400,000 times more likely that the bones tested in America originated from a child of Baan and Strachan than from a child of an unknown couple in the population.
742 I should note that it was established that Gaspar Baan and Christine Strachan were the parents of Jean Keir and that they did not jointly have another child.
743 At this point I should give myself a warning to avoid the "prosecutor's fallacy" (See Doheny and Adams [1997] 1 Cr App Rep 369 per Phillips LJ at 372, R v Keir [2002] NSWCCA 30, R v GK [2001] NSWCCA 413.) That fallacy arises from a course of reasoning that concludes from the likelihood discussed above that there is a 400,000 to one chance that the bones are those of Jean Keir as distinct from any other person.
744 Having regard to the view taken by Counsel as to the summing up of Kirby J I direct myself in accordance with the summing up, as amended to delete irrelevant issues and the finding as to the likelihood ratio which I have made.
745 The relevant section of the summing up, amended as I have said, would read:
"Now, there are, of course, many more than four hundred thousand people in Australia. So, statistically, just in Australia, leaving aside the rest of the world, there may be 50 or more people in Australia that you would expect to have the same DNA profile which appears in exhibit (-), that is 20 million divided by four hundred thousand.
….
So you can immediately see that the profile, exhibit (-), and the likelihood ratio does not identify the bones as those of Jean Keir. It is consistent with the bones being those of Jean Keir, if you accept Gaspar Baan and Christine Strachan are her parents. But there would be a number of individuals who might, by chance, have such a DNA profile and yet not be offspring of Gaspar Baan and Christine Strachan.
….
So, the likelihood ratio tells you how common the DNA profile is in the general population. It does not give you odds as to whether the bones belong to Jean Keir. That is the likelihood ratio does not tell you that it is four hundred thousand to one ….. that these bones belong to Jean Keir. All it gives you is some indication of how rare the profile is. That is, to repeat what I said: That statistically perhaps 50 individuals or more in Australia, leaving aside the rest of the world …. might be expected, as a matter of chance, to have that profile …. ."
746 Kirby J then went on to the use to be made of the determined ratio in combination with the other evidence. I shall come to that later. I should note that Kirby J later pointed out to the jury that the fact that the bones were of a female and an adult, as opposed to a child, would reduce the broad figures, that is, the number of such individuals. There may be other factors that should reduce that figure, however, as the trial has been conducted I do not consider that I should have regard to them.
747 I should take account of the further warning expressed in the summing up as follows:
"When you approach this issue you should be aware that the use of statistics throws up what looks like a very precise figure; four hundred thousand to one; …. depending upon which evidence you accept. However, the figures are based upon databases and generalisation and there can never be an exactitude in this area.
The task you have is of combining the likelihood ratio with other evidence, it is one that can never be reduced to a mathematical computation."
748 As I have sought to explain I consider that the procedures adopted in America and the results obtained exclude any effect from earlier contamination should that have occurred. However, I should deal with Mr Hanley's submission that contamination may have occurred and, further, that the tests conduced in Adelaide, particularly having regard to missing documents, raised issues which put the profiles established in America in doubt.
749 On 16 December 1991 Dr Atchison received six bones in separate plastic bags and two tubes of blood labelled "Gaspar Baan" and two tubes of blood labelled "Christine Strachan".
750 At that time the doctor was the Scientist-in-Charge of Clinical Sciences at the Victorian Institute of Forensic Pathology. He is now Manager of the Victorian Institute of Forensic Medicine Molecular Biology Unit. For present purposes the changes may be regarded simply as name changes.
751 Attempts to extract nuclear DNA from all the bones and to carry out PCR analysis produced some partial results. Dr Atchison is now not sure "if those reports were actually due to DNA from the bone or just a spurious result".
752 In the circumstances it is unnecessary to go to the details of the tests. Dr Atchison did say that he "felt that the DNA had degraded to the extent it was not reliable". He accepted that DNA in bones degraded with age and that smaller bones may degrade more rapidly.
753 Dr Atchison's laboratory did not receive at that time the hairbrush or curling wands. Analysis was carried out of the blood samples.
754 On 19 March 1992 the bones and blood were returned to Westmead.
755 On 4 August 1992 the six bones, blood, a hairbrush and two curling wands were delivered to Dr Van Daal. The bags containing the bones were in an Esky.
756 Dr Van Daal was at the time a senior forensic scientist at the State Forensic Science Service in Adelaide. She was well qualified and had been working in the field of molecular biology using DNA techniques since 1982.
757 Bone number 4 was selected for testing and was numbered JK4 by an analyst Ms Harrington for the purposes of the Centre. No other bone was tested, nor were the brush or curling wands tested, I accept, although there had been a request that this be done.
758 Dr Van Daal supervised the work, however, testing on JK4 between October 1992 and December 1992 was carried out by Ms Harrington.
759 In a Court Report dated 8 April 1993 Dr Van Daal reported the typing of the blood of Christine Strachan and Gaspar Baan and went on:
"DNA extraction was performed on a bone sample and then subject to HLA. DQAI PCR analysis. No result was obtained."
760 I should observe that at that time the Centre was using a monoplex rather than the later multiplex DNA testing system.
761 A covering letter with the report advised Detective Sergeant Merkel that further developments, including STR PCR systems, gender testing and mitochondrial technologies might assist as a later time, gave costings for further work and indicated that the Centre would not have time to do further work for another six months or so.
762 In the event nothing further was done with any of the items at the Centre although they remained there until 14 July 1994 when Detective Chief Inspector Dayment collected the bones and blood which were in an Esky. On the following day Detective Kinsman picked up the brush and wands, which had been overlooked the day before. It does not emerge whether they were stored in the same room, however, it is clear that the brush and wands were not in the Esky.
763 On 15 July 1994 Detective Chief Inspector Dayment delivered all the items to Dr Atchison in Melbourne.
764 There was a good deal of evidence as to the extent to which Dr Atchison opened the various bags. I do not need to go to the detail. I think it likely that he opened, or had opened, the bag containing bone four and the bag containing the blood tubes but not the other bags containing bones. As to the hairbrush and wands I consider it likely that he unzipped the cloth bag and looked at the brush and wands but did not remove them from the cloth bag.
765 Dr Atchison was assisted by a technician Ms Benton. A mitochondrial DNA test was carried out in respect of bone 4 and Christine Strachan's blood. No other items were tested.
766 The test results were indistinct and Dr Atchison reported:
"As with my previous report on the DNA tests in this case on 20/2/92 the results of the tests were not of sufficient clarity to make definite conclusions of identity."
767 Dr Atchison gave the following evidence in cross-examination:
Q. Your past experience had been that you had been able to extract DNA from bones that was reliable and reportable?
A. Yes.
Q. And I think you informed Sgt Dayment that the inabilities for you to obtain results with the nuclear DNA tests indicated that the DNA was extremely degraded perhaps consistent with your expectation of such small bones, is that right?
A. That is what I would expect yes.
Q. The mitochondrial DNA is that more susceptible to contamination than nuclear?
A. The tests procedures are yes.
Q. If these bone that you had been presented with and the ones that you examined had been completely denuded of any DNA, it is possible if they came into contact with other DNA that DNA may be incorporated into the substance of the bones?
A. Yes it is possible.
Q If the bones were degraded completely of DNA, an occurrence had occurred or contamination in that case had occurred by DNA from another source of bone would give the DNA reading of the contaminant?
A. Yes.
Q. If there was some DNA contained in the bones and there had been a contaminant which effected the bones, you could get a mixture of DNA?
A. You could. It depends on the amount of contaminant, how much DNA was in the bone.
Q. Are you saying by that if a contaminant was present it could overwhelm and disguise any other residual DNA that was in the bones?
A. I believe there is a limit of detection of mixtures.
768 He also agreed that contamination can occur in laboratories as a result of human error.
769 In support of the contention that the bones may have been relevantly contaminated whilst at the Forensic Science Centre Mr Hanley examined in some detail the tests carried out and the apparent results obtained by Dr Van Daal and Ms Harrington, at that time a forensic scientist assisting Dr Van Daal.
770 After the first attempt to extract DNA failed a number of further attempts were made. Dr Van Daal gave evidence in chief:
Q. Now at some stage, was a contaminant or a contamination apparent during some of the processes?
A. Yes. Towards the end of the things that I tried, I tried something that was a way of dealing with very, very small amounts of material, but as a result of that, it is prone to contamination and I believe I have results that reflect that.
Q. When you say the 'results that reflect that', do they reflect what, that you were aware?
A. From the same sample I have three different results, which is not a problem, but by definition they are a result of contamination.
Q. And how did you attempt to push the envelope as it were?
A. When you do a PCR reaction you put the DNA through cycles of heat, and a different heat goes which allows the DNA to be copied. That's normally done somewhere between 25 and 35 times cycles, and that's an exponential process. What I did is took some of that already amplified material and put it through that same process again. So, in essence, it was going through that process 60 odd times so it's, as you said, 'pushing the envelope'.
Q. And were you there dealing with very minute amounts of material?
A. Yes.
Q. Did you form a view as to whether the contamination was due to the bones themselves being contaminated?
A. I did. I believed it is not as a result of the bones being contaminated, because if the bones, themselves were contaminated, I probably would have seen results, without going to the extremes that I did.
771 Ms Harrington, whose transcript from an earlier trial was tendered, had established by use of a Satellite 3 Probe that DNA was extracted from JK4. The relevant result sheet was dated 5 January 1993. Dr Van Daal did point out that the Probe is extremely sensitive.
772 Subsequently a DQ.A1 Typing: Cetus Amplitype Kit (Ex 14) produced typing results.
773 Further, a DQ.A1 Typing: Cetus Amplitype Kit bearing date 29/1/93 (Ex 32) showed positive responses on the relevant "strips".
774 Dr Van Daal conceded that the strips showed alleles from up to 5 people. She considered the fact that these results were not repeated showed that they did not come from DNA from the bone.
775 Mr Hanley drew attention to certain result sheets missing from the file. Dr Van Daal said that she would assume that the tests were done. Mr Hanley submitted that this was supported by the fact that the blood analysis was then carried out suggesting that there was something to compare it with. Dr Van Daal said that it might have been done to add to their database, however, missing records seems more likely. It was said that the Centre would hold the original of the file, however, nothing further was tendered,
776 Looking at the Adelaide operation as a whole I think it unlikely that anything emerged in further testing to support the contention that the bones were relevantly contaminated or that test results excluded Baan or Christine Strachan. I should note that some of the results rejected by Dr Van Daal did, with one possible exception, appear to exclude them.
777 Mr Hanley cross-examined Dr Van Daal as to apparently unfavourable audits of some of her work, however, I do not draw any conclusion as to the quality of her work in this case based upon that cross-examination.
778 I should note also that there was no scientific evidence that put a different interpretation upon the records appearing from the Centre's file.
779 An examination of the events at Adelaide and Melbourne does not lead me to a different conclusion to that drawn as a result of the testing in America.