Expert Review of Exhibits 36 and 165b

[Perizia genetica sui reperti 36 e 165b, pp. 66-93; translated by katydid]

The first-level Corte di Assise held it was able to find the two current appellants guilty of the crimes with which they are charged based on a series of circumstantial elements, in which a preeminent position is undoubtedly occupied by the investigations performed by the Scientific Police: on the knife seized from Sollecito’s home (Exhibit 36), and on the hook of the bra worn by the victim, discovered in the room where the murder was committed (Exhibit 165b).

And indeed – according to the Scientific Police – Amanda Knox’s DNA profile was present on the handle of the knife and that of Meredith Kercher on the blade, while Raffaele Sollecito’s DNA profile was present on the hook of the bra.

Hence – despite the difficulty in explaining how a knife from Sollecito’s kitchen ended up in the house on Via Della Pergola – the conviction that the knife was the murder weapon or, more specifically, one of the murder weapons; and that culpability had to be attributed to Amanda Knox, who had obviously held it, so that she left her own DNA on the handle; and equally to Raffaele Sollecito, who, at the moment of the violence which culminated in the murder, must necessarily have been in Meredith Kercher’s room, intent on trying to tear off the bra and so leaving his own DNA on the hook.

We are certainly dealing here with circumstantial elements, but their central position in relation to other (also circumstantial) elements is clear: if the DNA found on the hook really belongs to Sollecito, this circumstantial evidence – while remaining just that – is particularly important; and the same may be said for the DNA found on the handle and blade of the knife seized from Raffaele Sollecito’s home, provided that it is certain to be one of the weapons used by the attackers.

During the first-level trial, attorneys for the accused had, through their own consultants, already criticized many aspects of the procedure followed by the Scientific Police regarding the correctness of the evidence collection methods, the genetic tests, and the reliability of the conclusions reached. However, their request that the Court order an expert review to resolve the conflict on this point went in vain: whence the repeat of the request, subject to partial reopening of trial proceedings [rinnovazione parziale della istruttoria dibattimentale].

This Court explained the underlying need for this measure in an order arranging for an expert review on 12.18.2010: the identification of DNA on several items [reperti] and its attribution to the defendants is, in truth, particularly complex due to the objective difficulty for people without scientific knowledge of making choices and forming assessments on particularly technical matters, without the aid of a court-appointed expert.

Furthermore, that this is a particularly complex question is also apparent from the observations made on this point by the first stage Corte di Assise. Indeed, in explaining in the ruling why it rejected the request for a court-appointed expert panel under C.P.P. Article 507, the Court writes the following: “…At this point it should also be observed that, with regard to the different interpretations offered by each side, this Court could, as indeed was requested by the defence, have ordered the appointment of experts and entrusted them with a suitable assignment. However, ultimately we might have been faced with another interpretation which would have fully or partially confirmed one of the interpretations already offered, and the problem as to which was the most appropriate interpretation would have remained. Therefore, ordering an expert review on the matter under C.P.P. Article 507  was not considered to be necessary”.

Essentially, it is as if the issue – already complicated due to the conflicting assessments (Scientific Police on the one hand, defence consultants on the other) – would have become even more complicated with the hypothetical formation of a third assessment, that of the expert possibly appointed by the Court, who would certainly have confirmed in whole or in part one of the different positions. Therefore the Court may as well directly resolve the problem.

The first-level Corte di Assise, then, saw fit to resolve a scientific controversy, recognized as particularly complex, based on scientific assessments formed directly by the Court itself. In contrast, this second-level Corte di Assise did not consider that the personal knowledge of the judges, professional and popular, was such as to allow us to resolve an essentially scientific controversy – to resolve it, therefore, on the basis of scientific criteria – without the aid of trustworthy experts appointed by the Court, and who could carry out the assignment entrusted to them with full consultation of the parties [nel pieno contraddittorio delle parti].

And indeed, while assessing the importance of the evidence (once its material existence has been established) is an appropriate task and matter for the Judge, a problem s/he can resolve with the tools of legal reasoning, ascertaining the material existence of the evidence [in the first place] – especially when it requires particularly technical investigative procedures and complex scientific knowledge – while not formally outside the jurisdiction [potere dovere, lit. “power-duty”] of the Judge, cannot truly be addressed and resolved without the aid of people expert in that field.

Nor is the power and duty [potere dovere] to order an expert review to resolve problems too complex for the collective knowledge of the judges, popular and professional, diminished purely because the tests [accertamenti] carried out by the Scientific Police in the course of the investigations used unrepeatable testing methods, with no special evidentiary hearing [incidente probatorio] having been requested: firstly, because the inability to repeat [the test] does not stem from the methods used, but from that test truly being unrepeatable; and secondly because, in any case, the methods used do not serve to plug the gaps [in knowledge] of the trial judge, who does not become less ignorant solely because the test was performed using particular methods. After all, it is precisely for this reason that C.P.P. Article 224 allows the Judge to order an official expert review.

Furthermore, in the case in question, the defence consultants during the first stage trial criticized the evidence collection [refertazione] and testing activities performed by the Scientific Police both in terms of method and results, based on arguments worthy of particular regard due to their depth and their inception in professors and technicians of unquestioned respect: as a result, the completion of a court-ordered expert review in order to determine the existence [or otherwise] of the aforementioned circumstantial elements appeared, to this Court, to be indispensable.

The expert review was entrusted to a panel composed of university professors at the faculty of forensic medicine at one of the most prestigious Italian universities (La Sapienza): Prof. Vecchiotti, a specialist in forensic genetics (as is shown from the titles she possesses, the material she teaches and her experience in previous cases); and Prof. Conti, a forensic scientist and also an expert in his field, and necessary in view of the complexity of the investigation, its dimensions being forensic as well as genetic. Both deserve the full faith of the Court on a professional and human level.

Equally, the facilities used for the testing, along with the staff, are fully trustworthy due to their academic level, and can be considered wholly adequate with regard to the issues [quesiti] considered and the nature of the investigations.

Well, in response to the assignment [quesito] posed to them, following a careful investigation carried out in full consultation with the parties [nella pienezza del contraddittorio delle parti], they came to the following conclusions:

Based on the considerations explained above, we are able to respond as follows to the inquiries posed at the assignment hearing:

Having examined the record and conducted such technical investigations as shall be necessary, the Expert Panel shall ascertain:

1. whether it is possible, by means of a new technical analysis, to identify the DNA present on items 165b (bra clasp) and 36 (knife), and to determine the reliability of any such identification

– The tests that we conducted to determine the presence of blood on item 36 (knife) and item 165B (bra clasps) yielded a negative result.

– The cytomorphological tests on the items did not reveal the presence of cellular material. Some samples of item 36 (knife), in particular sample “H”, present granules with a circular/hexagonal characteristic morphology with a central radial structure. A more detailed microscopic study, together with the consultation of data in the literature, allowed us to ascertain that the structures in question are attributable to granules of starch, thus matter of a vegetable nature.

– The quantification of the extracts obtained from the samples obtained from item 36 (knife) and item 165B (bra clasps), conducted via Real Time PCR, did not reveal the presence of DNA.

– In view of the absence of DNA in the extracts that we obtained, with the agreement of the consultants for the parties, we did not proceed to the subsequent amplification step.

2. “if it is not possible to carry out a new technical analysis, shall evaluate, on the basis of the record, the degree of reliability of the genetic analysis performed by the Scientific Police on the aforementioned items, including with respect to possible contamination.

Having examined the record and the relevant documents, we are able to report the following conclusions regarding the laboratory analyses performed on Item 36 (knife) and Item 165B (bra clasps):


Relative to the genetic analysis performed on trace A (handle of the knife), we agree with the conclusion reached by the Technical Consultant regarding the attribution of the genetic profile obtained from these samples to Amanda Marie Knox.

Relative to trace B (blade of the knife) we find that the technical analyses performed are not reliable for the following reasons:

1. There does not exist evidence which scientifically confirms that trace B (blade of knife) is the product of blood.

2. The electrophoretic profiles exhibited reveal that the sample indicated by the letter B (blade of knife) was a Low Copy Number (LCN) sample, and, as such, all of the precautions indicated by the international scientific community should have been applied.

3. Taking into account that none of the recommendations of the international scientific community relative to the treatment of Low Copy Number (LCN) samples were followed, we do not accept the conclusions regarding the certain attribution of the profile found on trace B (blade of knife) to the victim Meredith Susanna Cara Kercher, since the genetic profile, as obtained, appears unreliable insofar as it is not supported by scientifically validated analysis;

4. International protocols of inspection, collection, and sampling were not followed;

5. It cannot be ruled out that the result obtained from sample B (blade of knife) derives from contamination in some phase of the collection and/or handling and/or analyses performed.


Relative to Item 165B (bra clasps), we find that the technical analysis is not reliable for the following reasons:

1. There does not exist evidence which scientifically confirms the presence of supposed flaking cells on the item;

2. There was an erroneous interpretation of the electrophoretic profile of the autosomic STRs;

3. There was an erroneous interpretation of the electrophoretic profile relative to the Y chromosome;

4. The international protocols for inspection, collection, and sampling of the item were not followed;

5. It cannot be ruled out that the results obtained derive from environmental contamination and/or contamination in some phase of the collection and/or handling of the item.

This Corte di Assise, in explaining why it is able to accept the conclusions reached by the Expert Panel, observes firstly that the reasons on which this agreement is based can certainly not be of an essentially scientific nature. Were this to be the case, [the Court] would contradict itself in that, after appointing an expert panel to compensate for understandable gaps in knowledge in a particularly complex field, both scientifically and technically, it would then instead – on what basis, one can only guess – appoint itself arbiter of a scientific and technical dispute, perhaps identifying technical-scientific assessment criteria to determine the superiority of one theory over another. No, clearly this is not acceptable, both due to the need for congruity and for a properly non-presumptuous approach [per una esigenza di non contraddizione e per un doveroso atteggiamento di non presunzione].

The reasons for [the Court’s] acceptance of the conclusions formulated by the Expert Panel are, instead, grounded in judicial reasoning [di natura logico giuridica].

The issue, in truth, is one of explaining why the judge holds these conclusions, and obviously the arguments developed by the Expert Panel to illustrate them, to be persuasive and useful for the purposes of the decision, which consists not in the resolution of a purely technical and scientific controversy, but in the recognition or otherwise of criminal responsibility.

Well, as regards Exhibit 36 (knife), the Expert Panel – following sampling and DNA extraction procedures certainly correct in themselves, not having been criticized by anyone – held that the DNA extracted in the course of the expert investigations and quantified according to the method currently in use was not suitable for carrying out the subsequent stages of amplification and electrophoresis, due to its very scarce quantity. Therefore, they continued to the second part of the assignment: assessing the degree of reliability of the genetic tests performed by the Scientific Police on the item, including with reference to possible contamination.

The Expert Panel laid out the established principles on the subject within the Scientific Community, recalling the [various] stages of the technical-scientific procedure for the identification of a subject’s genetic profile in a trace present on an item (sampling, DNA extraction, quantification, amplification, electrophoresis and interpretation of the resulting graph) and the problems, with scientific knowledge as it currently stands, relating to establishing the nature of the sample, storage methods, greater or smaller amounts of extracted DNA, and the precautions to adopt in order to ensure that the result obtained is not caused by possible contamination.

On this last point, it is worth reproducing here some passages from the report:

…Budowle B. et al. (2009) call for caution, and suggest the use of LCN exclusively for identifying missing persons (including the victims of mass disasters) and for research purposes. On the other hand, the aforementioned authors advise against the use of current LCN methods in criminal proceedings, since current methods, technologies and recommendations  still do not allow the problems which characterize LCN typing to be overcome.

In particular, since by definition LCN typing cannot provide reproducible results, and therefore the same result may not be expected if the same sample is analyzed twice, the method cannot be considered robust according to conventional standards (Budowle B. et al., 2009). […]

3. Problems associated with low template quantities

The problems which arise from the analysis of sub-optimal quantities of template DNA in a PCR are numerous, and these problems become even more apparent when the template quantity is reduced. Moreover, the interpretation of mixtures has yet to be well-defined.

The fundamental issues of the low template quantity problem are the following: stochastic effects, the detection threshold, profile interpretation, allele drop-out and heterozygous peak imbalance, stutter, contamination, replicate analysis, appropriate controls, and limitations of use.

3.1 Stochastic effects: Due to the kinetics of the PCR process, a low quantity of starting template will be subject to stochastic effects. In fact, during the first cycles, the primer may not bind in the same way for each allele at a given locus, and therefore a significant imbalance between allelic products may be noted or, in some cases, the total loss of one or both alleles. In other words, an LCN DNA template in a PCR may show stochastic amplification phenomena, visible both as a substantial imbalance of two alleles at a given heterozygous locus, and as allelic drop-out or an increase in stutter (Gill P. et al., 2000; Whitaker J.P. et al., 2001; Kloosterman A.D. et al., 2003; Smith P.J., Ballantyne J., 2007; Forster L. et al., 2008; Budowle B. et al., 2009).

3.2 Detection threshold: Usually, it is recommended that quantities of DNA able to reduce stochastic effects to manageable levels are used for the PCR. However, since differences in the quantification of the template and possible inaccuracies in the pipetted volume may influence the amount of DNA actually present in the PCR, a stochastic interpretation threshold must be used for STR typing results (Minimum Interpretation Threshold “MIT”, Budowle B. et al., 2009).

Given that the Public Minister criticized the fact that the Expert Panel listed principles and problems faced by scholars of the subject within the report, it should be pointed out that this formed an essential part in the completion of the assignment, since the various components of the Court, popular and professional, needed to be provided with the minimum conceptual tools necessary to at least understand the complexity of the issue.

The Expert Panel then examined whether the actions performed by the Scientific Police complied with the standards and methods recognized by the scientific community.

And, while it endorsed the identification of Amanda Knox’s genetic profile on the handle of the knife, since the quantity of extract allowed a reliable profile to be obtained at the end of the amplification and electrophoresis stages, in contrast it rejected the reliability of the result produced by the Scientific Police regarding the alleged presence of Meredith Kercher’s DNA on the blade of the knife.

The Expert Panel noted first that the quantification stage is notably absent in the investigations carried out by the Scientific Police: this stage is, in fact, indispensible in [obtaining] a reliable result, given that “…determining the amount of DNA”  — so one reads on page 45 of the expert report — “present in a sample is important for the majority of PCR-based analyses, because an excessive quantity can cause the appearance of extra peaks, or peaks which are outside the limits of the measurement technique, while too scarce a template quantity can cause allele drop-out in which the PCR reaction is affected by stochastic phenomena. PCR amplification can also fail due to the presence of inhibitors extracted along with the DNA from the sample, DNA degradation, an insufficient quantity of DNA, or a combination of all these factors.…”

Furthermore, since best results are obtained when the template DNA is added in an amount within the range suggested by the Kit, the need to [quantify] the extract in advance and not to determine it in hindsight is understandable, it being a completely pointless action at that stage.

On the other hand, Dr. Stefanoni (Scientific Police), apparently due to an understandable memory lapse, stated before the GUP that she quantified the extract using Real Time PCR (a system which allows the extract to be precisely quantified before it is amplified), and then clarified that she had not used Real Time for this extract, but the Quibit Fluorimeter, which gave the result “not interpretable”. However, Dr. Stefanoni nonetheless considered it suitable for continuing to the subsequent stages, at the end of which she identified Meredith Kercher’s DNA in the graph.

Now, according to the Expert Panel, the most likely outcome if the sample which was amplified had been correctly quantified using the Real Time PCR method, is that it would have been classified LCN/Low Copy Number (due to the fact that the electrophoresis graph shows peaks below the 50 RFU threshold and allele imbalance (Hb=φab >0.60) indicative of a Low Copy Number sample). That is to say: an amount not able to give reliable results if treated in the same way as an extract of greater quantity, and, therefore, only able to be processed further if particular measures [accorgimenti] are taken, as advised by the authors who have dealt with the problems associated with LCN samples. However, these measures do not appear to have been employed, or correctly employed, by the Scientific Police.

These measures concern all phases of the procedure, from evidence collection to quantification, amplification and interpretation of the graph resulting from the electrophoresis. They are intended to reduce the contamination risk in each phase to the minimum possible (a risk which understandably increases as the amount of DNA to be examined decreases) as well as [reducing] the occurrence of stochastic phenomena and, ultimately, errors in the interpretation of the profile.

Amongst these measures, one which is particularly important in ensuring a reliable result is replicate analysis:

…The most commonly used approach for the designation of an allele in an LCN sample requires the subdivision of the sample into two or more aliquots, and recording only those alleles which are common to at least two replicates (Gill P. et al., 2000; Gill P., 2001; Gill P. et al., 2007). The advantage of this approach is that if drop-in occurs randomly and infrequently, observing an allele several times increases the probability that it actually derives from the sample being examined  (assuming that no contamination happened during the sampling phase).

Most scientists who work with LCN stress the need to perform 2-3 repeated [runs] and state that an allele must be observed at least twice to be denominated as such (Taberlet P. et al., 1996, even invoke up to 7 repetitions to increase the reliability of allele denomination): allele redundancy in replicates is therefore the cornerstone of reliable LCN typing.

However, the exact number of repetitions, the number of times an allele is observed, and the degree of reliability (quantitatively or qualitatively) need to be better defined.

In practice, it needs to be considered that performing more than 2-3 repetitions is often not possible; therefore most interpretation guidelines and degree of reliability assertions must be applied and declared based on the analysis of 2-3 repetitions.

Common sense would suggest that dividing a sample into multiple aliquots increases the limitations of LCN typing (Budowle B. et al., 2009) and that every possible effort should be made to concentrate as much template as possible into a single reaction: however, to date, allele redundancy is the only accepted approach.

Studies on dilutions and redundancy have, up to now, been based on laboratory control samples: [these are] completely different to [samples] originating from items from real cases, possessing undetermined (and often degraded) quantities of DNA, and which may contain PCR inhibitors also able to affect allele drop-out.

Now, that the extract obtained by the Scientific Police was an LCN [sample] is not in doubt: both because the limits which define this category are objective reference parameters for the scientific community as a whole, as also stated by the defence consultants (who, aside from their role in the present proceedings, are all scholars and professionals of established reputation, who certainly cannot be induced by the role they have taken on to affirm the existence of scientifically erroneous principles and concepts); and because this was also substantially confirmed by the consultants of the civil parties and Dr. Stefanoni herself, as well as Prof. Novelli, the Public Minister’s consultant (again, certainly none of these could be induced by their role [in the present proceedings] to make statements contradicting established principles within the scientific community). But it is equally certain that at least one of the most important measures (aside from the problems associated with the evidence collection and the <chrome_find class=”find_in_page findysel”>lack of traceability of all stages of the analytical procedure) was not implemented: the sample was not divided into at least two parts [aliquote], subjecting each to the procedure, in order to then record the presence or otherwise of the same alleles in the two replicated samples [replicati].

Dr. Stefanoni explained that, since the quantity of the trace was small, she decided that she should not divide it in order to perform specific tests of its biological nature. Instead, she chose to concentrate it in a single sample to try and obtain a result, being convinced that dividing the sample into several parts would not have allowed one to be obtained.

To compensate for this failure to divide the sample into several parts, she performed a second electrophoretic run of the same amplified sample [amplificato] and compared the two graphs, that of the first run and that of the second: but this is a palliative, since – as Sollecito’s defence consultant, Prof. Tagliabracci, stated at the hearing on 9.6.2011 – repeating the run on the same amplified sample is certainly not equivalent to processing two different samples in order to reproduce the result and, hence, assure its quality, given that getting the same result from two different amplified samples signifies a scientifically reliable result.

Moreover, when Dr. Stefanoni herself was questioned before this Court by Mr. Dalla Vedova at the hearing on 9.6.2011, she was asked whether or not she agreed with a prominent author like Butler on the need to repeat the amplification when dealing with an LCN sample, to [ensure] the reliability of the result. She acknowledged that yes, she agreed with this scholar, but only for cases where she had been able to classify the extract as LCN beforehand, while in this case, since she had used the fluorimeter rather than Real Time, it had been impossible to classify the sample as LCN.

Now, here the intent is not to criticize Dr. Stefanoni’s actions: that is, the aim is not to establish whether she was right or wrong to continue to the subsequent stages without having first accurately quantified and classified the sample, and therefore without having divided it into several parts. The issue is purely to assess the reliability of the result, even if it cannot be reproduced, once it is established that it was LCN

But it was Prof. Tagliabracci who explained that it is not, as said, a question of criticizing the actions of the Scientific Police, but only of assessing the reliability of the result [dato] from a scientific perspective and, therefore, its usefulness for probative purposes. The following is from the transcript of the hearing record of 9.6.2011: “…I think that major problems probably also arise for the jurors, those in this room who are not expert in forensic genetics, regarding the validity of this technique when faced with a rather heated conflict among the experts [i.e. Conti and Vecchiotti], the consultants of the Public Minister, and now yours truly. A conflict which, however, stems essentially from a different approach, I would say two philosophies, two schools in genetic testing. Two approaches and two different philosophies: the philosophy of those who want a sure, reliable result, a solid and robust result, which can be used without problems, even faced with proceedings of a certain complexity, of a certain duration; and on the other hand those who take pains just to get a result. We must bring home a result, these are the two different philosophies of…this isn’t just an issue in Italy, there’s an international debate on these issues which began with that case in New York which Prof. Vecchiotti described: following the trial in which very low amounts of DNA were used for the first time, and used in the trial, a debate began which continued for several issues of the international journal of forensic genetics, and which still has not ended. But this debate basically concerns the question of what is, what are the limits, how reliable is an examination performed in such critical conditions regarding the quantity of DNA, critical in terms of quantity and probably also the quality, which could be altered or degraded DNA, which complicates things further.

“Well, as I will say shortly and as others have already said, the International Society for Forensic Genetics  has taken an overall position which was expressed in one of the recent articles of this journal, when it stated that a trace below these DNA levels could also be considered, provided that it is repeated several times so that the result is a consensus of repeat amplifications.

“I subscribe, let’s say, to classical forensic genetics, and I think we cannot go beyond a certain limit. But anyway, if we do go below a certain limit, important precautions have to be taken, measures and procedures adopted which avoid the risk of getting false positive results, results which could wrongly incriminate [incastrare] someone who left biological material in large quantities even at different times. Therefore, we need to proceed with caution to get a result which can be considered reliable even when dealing with low quantities of DNA.

“Certainly, in any case, the correct methodology needs to be followed in all stages of the investigations, independently of the amount of DNA which we ultimately find to analyse, and in this case, as Prof. Vecchiotti and Dr. Conti recalled in their report – which I obviously agree with, not least because I had expressed many of the concepts put forward in that report during the first stage trial – well, it is necessary anyway to follow a series of steps, to use a correct methodology. And if we examine the work carried out by the Scientific Police service, we can’t say that there was an appropriate, correct procedure [in place], such as not to create problems with the result then obtained. The items to consider here are basically the bra and trace 36B on the knife…”

In essence, this Court holds that the risk of obtaining a result which is not particularly reliable due to the correct methodology not having been followed, and in particular due to the failure to perform two amplifications despite the quantity of extract being very low (LCN), may be acceptable simply for the purposes of orienting the overall investigation in its initial stages, but cannot be accepted when the genetic tests form the basis for evidence of guilt beyond any reasonable doubt.

From Dr. Stefanoni’s statements, it also appears that where the quantity of extract is low, lower than the level recommended by the Kit to obtain a good result, the sensitivity threshold of the machine also has to be lowered: however, this increases the occurrence of stochastic phenomena, which only a comparison between the charts from several amplified samples could detect. As a result, the possibility cannot be excluded that a particular profile, while hypothetically actually belonging to a subject, originates from contamination occurring in one of the stages during the activities of collection and analysis.

Now, Prof. Novelli and then the Public Minister himself claimed that it is not enough to say that the result stems from contamination: the burden is on those claiming contamination to prove its origin.

However, this argument cannot be accepted, because it would ultimately mean treating the possibility of contamination on a legal level as an objection of a civil nature.

That is, one cannot say: I have proven that the genetic profile is yours; now you prove that the DNA was left on the item not by direct contact, but by contamination. No, it cannot work like this.

In the context of criminal proceedings – as is well known – it is incumbent upon the Public Minister who sustains an accusation in court [sostiene l’accusa in giudizio] (the terminology is used in article 125 of the Regulations for Implementation of the Code of Criminal Procedure) to prove the existence of all the elements upon which the accusation is based. Therefore, when one of the elements has a scientific component, the outcome of an analytical procedure, the burden is also that of proving the result was obtained using a procedure which guarantees the integrity of the item from the moment of collection to the moment of analysis.

Above all, in identifying a genetic profile on an item, it is important that the whole procedure is carried out in full observance of the rules set out by the Scientific Community. These are certainly not legal rules (we are not talking here of a State law, as Dr. Stefanoni pointed out) but instead guarantee the reliability of the result; and given that the essential precautions to prevent possible contamination are amongst these rules, it is understandable that respect for them cannot be assumed, but must be proven by those using that result to sustain their accusation.

Therefore, once there is no proof that the precautions which guarantee the result is not due to contamination were respected, it is by no means necessary to also prove the specific source of the contamination.

From this it follows that, for our purposes, the result obtained by the Scientific Police cannot be considered reliable, due to being the outcome of a procedure which did not implement the measures recommended by the International Scientific Community – or in any case its reliability is seriously weakened, making it necessary to seek confirmation in other elements outside the scientific process.

This also explains why the Expert Panel did not proceed further in analysing the sample it collected from the blade of the knife: the quantity was determined to be wholly insufficient (again, LCN) for performing two amplifications, so that if they had continued further, the court-appointed experts would have committed the same error found in the investigations [accertamenti] performed by the Scientific Police. Furthermore, it is clear from the concepts outlined above that the need to divide the sample into several parts applies to each individual trace, the intent being to guarantee the reliability of the result of the analysis of that trace. Hence, analysing two individual traces, both LCN, without subjecting each to that procedure which guarantees the result, does not compensate for the failure to repeat the procedure on each individual trace: the sum of two unreliable results, neither having been subjected to a scientifically correct procedure, cannot give a reliable result, regardless of the possible similarities.

In fact, Prof. Novelli argued that systems currently exist able to analyse such low quantities, albeit still on the cutting edge [in uno stato di avanguardia]. This Court holds, however, that it is precisely the fact they are still cutting-edge, practically in an experimental phase, which precludes us from basing a belief in guilt on the results obtained with the application of such systems: the Judge can do no else but base his or her opinions on the technical systems and established scientific knowledge from a particular time period – the period in which s/he is called to judge – and not on others still in an experimental phase. This, once again, to reach a decision of guilty beyond any reasonable doubt.

It is, therefore, precisely this greater guarantee of a reliable result, due to the rules developed for this purpose by the Scientific Community having been respected, which leads this Court to accept the conclusions of the Expert Panel, consistent with the aforementioned rules.

It should again be observed that, since the unreliability of the result primarily renders the circumstantial evidence which should be represented by that result materially non-existent, rather than equivocal in its significance (in fact, it is difficult to understand how one can be scientific by halves: the result is scientifically accurate or it is not, in a true sense, a result) any attempt to clarify its significance by placing it in the context of other results from the proceedings should be resisted, once its unreliability has been determined. Nonetheless, if the aforementioned result is placed in the context of other results from the present proceedings, its unreliability remains definitively proven.

Firstly, the cytomorphological tests carried out by the Expert Panel on the blade of the knife did not detect the presence of cellular material: in particular, there was no trace of blood. Furthermore, the confirmed presence of starch granules on the blade, particularly in the area where the blade is inserted [into the handle], discovered after a microscopic study and revealed by their structure to be vegetable material, shows that the knife was not washed. Therefore, the absence of blood cannot be attributed to washing.

With a brief experiment, documented in a report whose veracity went unchallenged both by the Public Minister’s consultants and by those of the civil parties, Knox’s defence consultant Prof. Torre explained and demonstrated that the granules of starch, present in many vegetables intended for consumption, are easily removable simply by washing with water. This shows, as said, that the knife had not been washed when it was seized. Furthermore, given that these starch granules are very absorbent when in contact with liquid, they would probably have absorbed blood if the knife had been used to injure and kill. To the contrary, the granules showed no trace of blood.

In the course of the debate, the Public Minister proposed that the starch granules came from the type of gloves used by the Police (powdered with vegetable starch) as an explanation for the presence of starch granules on the blade. But this last minute claim finds no confirmation in objective case elements, and furthermore the location of the starch granules (mainly corresponding to the insertion of the blade) renders this explanation not at all persuasive.

Therefore, it remains a firm point that the knife was not washed; or that although having been badly washed, so much so that it failed to remove the alleged DNA traces attributable to Meredith Kercher, it was used again for cooking after the perpetration of the murder: which really seems a bit of a stretch [un po’ troppo].

Secondly, Amanda Knox’s DNA on the handle of the knife – explained by the fact that she spent time at Raffaele Sollecito’s house in that period, sometimes also staying there to eat and presumably helping him on those occasions – confirms that the knife was not washed; or that, hypothetically, it was used again for cooking after the perpetration of the murder, given that the location where the analysed trace was found, again on the handle, is not at all incompatible with a purely domestic use (contrary to the Public Minister’s claim), since even for domestic use a knife can be gripped in various ways according to various needs.

Thirdly, it should be remembered that reasons have already been set forth above to show that, if the genetic traces are excluded, no objectively significant and truly substantial elements exist to prove that the knife seized from Sollecito’s home is the weapon, or at least one of the weapons, used to commit the murder.

And so, considering it was clearly the fact that the Corte di Assise of first level held the Scientific Police investigations to be reliable which led it to identify the aforesaid knife as the murder weapon – since other elements, taken by themselves, are of ambiguous significance (in particular, compatibility with the injuries), while the proposed explanation for the knife’s presence in the house on Via della Pergola at the moment of the murder is rather peculiar – the failure of this reliability for the reasons outlined, or at least its very serious weakening, cannot be replaced by those elements which were only given value [in the first place] by this reliability.

To conclude, the circumstantial evidence represented by Meredith Kercher’s DNA on the blade of the seized knife is held to be without foundation.


Turning to Raffaele Sollecito’s genetic profile, which the Scientific Police indicated was present on the hook of the bra worn by the victim, it should be noted that the Expert Panel were not able to obtain any suitable DNA from [the hook] to test (and nor from the second hook, since there were in fact two). This is most likely to be a result of the way in which the item was stored: the hooks were presented to the experts covered in a red-brown crusty material, probably due to oxidation of the salts in the extraction solution used by the Scientific Police, and from rusted elements of the metal.

The Expert Panel then went on to assess the procedure performed by the Scientific Police, noting errors in the interpretation of the graph and, once again, an absence of the precautions needed to avoid possible contamination.

The general arguments set forth above about the need to comply with the rules and standards established by the Scientific Community in order to obtain a reliable result – for probative, and not merely investigative purposes – should be recalled here. But it should also be added that the need to respect those rules and standards is, if possible, even more necessary in this case, since it concerns what is certainly a mixed trace, a mixture of DNA, left by several unidentified contributors. Indeed – as both Prof. Tagliabracci and the Expert Panel explained – while the trace is substantial [in quantity], it is nonetheless subject to problems in its interpretation, not excluding contamination, when it is split between several contributors.

And even those not particularly familiar with the subject can easily understand why.

Identifying a genetic profile is very different from and more complicated than reconstructing a photograph. We are not talking about bringing out the contrast in a faded picture or of reassembling the pieces of a torn photograph in order to identify an individual with a very distinct profile, discernible simply by sight. Rather, it involves using complex procedures and technologies to convert the DNA components into a graph, characterized by peaks of differing heights (alleles) and located in different positions, and then of “pairing” peaks with a specific height and placement, so obtaining a profile which can be compared with another DNA profile obtained with the same system, but which definitely belongs to a particular subject.

Therefore, dealing as we are with a graph produced by a complex procedure and technology and sensitive to a multitude of factors, it is easy to understand that when a mixture is present (a sample with several contributors) the problem of tracing a specific profile – that is, of identifying the peaks to pair in the graph, distinguishing them from those which are irrelevant and from other pairings – is particularly complex, since it is very often not possible to exclude alternative but equally plausible pairings.

And this is why Prof. Vecchiotti was able to say that it was fundamentally possible to identify anyone’s profile in the graph obtained from the Scientific Police, even her own.

Hence, it is certainly true to say that alongside the victim’s profile (major contributor), there is also a profile attributable to Sollecito in the graph: but there is no guarantee that this profile is really accurate, since if one considers other peaks also present in the graph but not taken into account by the Scientific Police, different conclusions might be drawn.

Nor does it seem that the biostatistical calculation carried out by Prof. Novelli to confirm the attribution of this profile to Raffaele Sollecito is able to overcome this difficulty. While this calculation is understandably able to exclude the possibility that another person with a profile very close to Raffaele Sollecito’s deposited his own DNA in the house on Via della Pergola (an error due to chance compatibility, described by Prof. Novelli on page 13 of his comments), it does not appear able to exclude the configuration of Raffaele Sollecito’s profile based on the pairing of different peaks to those proposed.

The mixed nature of the trace should have entailed a different calibration of the machine, to make sure that peaks which might have been important were detected: but – observed the P.G. in his closing speech – how could Dr. Stefanoni have known beforehand that she was dealing with a mixed trace?

And yet, here too the issue is not to establish whether Dr. Stefanoni was right or wrong to set the machine beforehand with a high threshold, but to establish the reliability of the result. It is not possible, albeit in hindsight, to ignore the fact that while the [size of the] trace taken as a whole was considerable, the proportion of it belonging to the minor contributors was not.

But the reliability of the result obtained by the Scientific Police is still more undermined in this case by the evidence collection methods, so that the integrity of the item cannot be guaranteed: that is, we cannot exclude the possibility that the DNA, hypothetically actually belonging to Raffaele Sollecito, ended up on the hook not because it was left there by direct contact from Raffaele Sollecito during the alleged attack on Meredith Kercher, but because it was transferred there accidentally by other people attending the crime scene.

Indeed, it is opportune to recall here that the hook, along with the piece of material from the bra to which it was attached and which had been cut from the rest of the garment, was discovered under the victim’s body during the inspection on 2 November, and also photographed. However (presumably due to having been forgotten or because it was not considered important, the entire bra being present – according to Dr. Stefanoni’s statements before the GUP) it was not collected and tested.

The hook appears, in fact, to have only been collected and then tested a month and a half later, on 18 December, when it was found in the same room but in a different position, about a metre from where it was seen during the inspection on 2 November.

On 2 November, then, the piece of material with the attached hooks was found under the victim’s body, while on 18 December, when it was finally collected to be analysed, it was found in another area of the room, near the desk and under a mat, and about a metre or a metre and a half from where it was seen during the first visit.

When Dr. Stefanoni was questioned on the issue, she said she did not know how or why it was moved, nor could she specify how many people had entered the house at Via Della Pergola 7 between the first inspection and the one on 18 December, nor the accesses which took place.

Well, aside from the fact that Prof. Conti, displaying a [still] photograph taken from a film of the activities and shot by the Police themselves,  pointed out that when the hook was collected during the inspection on 18 December, the gloves worn by the Scientific Police operatives already showed traces of dirt at the finger tips [of the gloves] with which it was being held; aside from this, due to the difficulty of fully assessing the nature of the marks through the photograph, interpretable as possible signs of previous soiling; and aside from the fact that, as we see in the film, the hook appears to have been collected and then replaced on the ground again to be photographed, more or less in the same spot; aside from all this, then, it is certain that between the Scientific Police inspection immediately after the discovery of the murder and the second inspection by the Scientific Police on 18 December, the house on Via Della Pergola was subject to several searches with the aim of finding other elements possibly useful to the investigations, in the course of which the house was turned upside down, as also documented in the photographs shown by the defence of the accused but produced by the police themselves. And these searches were understandably carried out without the precautions which accompanied the Scientific Police investigations, in the belief that by this time the items to be subject to scientific investigations would [already] have been acquired.

In this context it is likely that the DNA, hypothetically Raffaele Sollecito’s, was transferred into the room and directly onto the hook by others, by means of contact with hands or even by contact between objects and clothing on which it was present (for instance, inside a washing machine, as Dr. Gino mentioned), such a method of depositing DNA not being particularly unusual. The fact that it is not an unusual occurrence is proven by the studies cited by the Expert Panel as well as by the defence consultants of the accused, faced with which neither the Public Minister’s consultants, nor those of the civil parties made any objections regarding their scientific value, but only about the concrete occurrence of contamination in that context.

Furthermore, Dr. Stefanoni (pages 221, 228) recognized that it is not possible to date the moment at which DNA was deposited, nor to establish the chronological order in which several traces were left, even one on top of another.

Therefore, leaving aside the ambiguous interpretation of the graphs, which has already been discussed, this Court holds it can accept the theory of a probable contamination: because when the item was collected, none of the necessary precautions were taken to guarantee its integrity; because it seems very improbable that Raffaele Sollecito’s DNA was left only on the hook, and not on the parts of the bra material which were easier to grasp, and [indeed] necessary to grasp when trying to cut or tear it from the young woman’s body; because it is not likely that Raffaele Sollecito (as well as Amanda Knox), hypothetically protagonists on a par with Rudy Guede, took part in the attack inside a room which was certainly not large, without also leaving DNA and their own prints on other parts of the body or on objects and clothing, where in contrast Rudy Guede, who is certainly guilty, left DNA and prints inside the room on several parts of the body of the attacked woman (in particular in her vagina) and on her clothing (sweatshirt, material from the bra) and on objects present there.

Nor can it reasonably be maintained that the failure to find DNA and prints from the current defendants (at least in the room where the murder happened) was the result of prompt cleaning activity by the two: it is impossible to think that in the process of cleaning – certainly quickly, due to the circumstances in which it happened – the two were able to distinguish their own prints from Rudy Guede’s in such a way as to remove theirs and leave his, perhaps in an attempt to place all the blame on Rudy Guede. And after all, the surroundings did not appear to have been cleaned, since dust, hair and traces of dirt were present…

It should also be observed that since the band of the bra as well as the shoulder straps were cut – probably by Rudy Guede, who left his own DNA on the material – there was no reason to grasp the hook of the bra at that point.

In addition, the statistical study, this time “inspected” by experts in probability calculus – as Prof. Novelli writes in his comments on the report (page 27) – is not useful for excluding the significant probability of contamination. It refers only to cross-contamination between items analysed inside the laboratory and therefore does not seem to take into account all the practical circumstances of the case, and in particular the peculiar methods and times of the item’s acquisition.

Therefore, it appears to this Court that contamination did not happen in the later stages, when the item was processed at the Scientific Police laboratory, but rather before it was acquired by the Scientific Police.

It is worth noting that Prof. Novelli, questioned at the hearing by the Public Minister about the likelihood of contamination during the collection stage, did not in fact give detailed responses, being mainly focused on excluding contamination in the laboratory – so much so that the President let him know at that point that his response was not relevant, because the question concerned the earlier collection stage. And it was only at the end that Prof. Novelli, in order to exclude the probability (not the possibility) of contamination during collection, said that if it had occurred the presence of other subjects would have been noted (but the DNA of other unidentified subjects was indeed present on the hook) or of Sollecito on other objects (but Sollecito’s DNA was certainly present in the rest of the house, so that it was found, for example, on a cigarette end; nor can it be excluded that it was on other objects which were not collected).

In any case, the arguments set forth above, which refute the idea that the burden to prove the source of contamination rests on the defendant making that claim, should be recalled here: it is, on the contrary, those using that result [as evidence] to support an accusation who have to prove that the procedure and, prior to that, the collection stage happened in accordance with the methods and precautions necessary to avoid contamination. As noted above, this did not happen here.

Therefore, the possibility of using the presence of Raffaele Sollecito’s genetic profile on the hook of the bra as a reliable piece of circumstantial evidence ceases to exist.

Next: Print on the Mat