Good Luck Finding Fibres By Chance

Fibres Deep Dive Pt 4. How likely are you to find fibres if you start looking for them?

Introduction

If I was to pick a single surface at random, at any given moment, recover all the fibres present upon it and compare them to fibres from a single garment chosen at random, how likely would I be to find any at all? This is the question that forms the basis for target fibre studies. You pick a garment, or a small number of garments, and you look for fibres from them on hundreds of other garments or surfaces.

Once again, historically we have the FSS to thank for being the original source of these studies, the first one being published in 1986 by Cook and Wilson¹. Since then over the last 40 years, there have been many more published since, by independent research groups, from Germany, USA and Australia amongst others. The results have all generally led to the same conclusion, that finding fibres by chance is an unlikely event. But, its important to understand the findings and in particular the nuance that is contained within them.

In 2004, the late Ken Wiggins (RIP) published a target fibre study² and within it he helpfully summarised every target fibre that had been published before. We’re going to have a look at some of these studies in more depth.

Technology

As we know the FSS was really ahead of its time in terms of its ambition to perform research that underpinned types of forensic evidence. The world should be grateful for that now, because the circumstances in which forensic science laboratories and universities operated within in the 1980s to the noughties, allowed and encouraged research like that to take place. Research for research sake, no bells and whistles required to ensure that the outcome of a research project included a new widget for immediate sale on Amazon or the birth of a new tech bro’ founded spin out.

However, the downside to older research is always the technological limitations the scientists were working within at the time the research was conducted. In the 1980s, micro-spectrophotometry (MSP) was in its infancy, and even if instruments were available, they weren’t very good by today’s standards. On that basis the discriminating power of the processes used by scientists in the 80s was likely to have been less powerful than today, making it possible that the finding of any coincidental matches may have been overstated. Nonetheless, the levels of exclusions the scientists in the 80s were able to make based largely on microscopy, MSP and thin-layer chromatography was still very powerful and in my view, they generally still hold up to the test of time.

By the 1990s the technology for MSP had improved greatly, whilst innovations in the noughties focussed more on speed of acquisition and sensitivity improvements.

Choice of Target Fibre

It is of no surprise that as far as I know, there has not been a target fibre study conducted where the target fibre was black or grey cotton. The reason for this is a practical one - there’s thousands of them on any one garment and the scale of the challenge to look for a target fibre like that on hundreds of garments is too much for any one research team to do. The lack of such a study remains a gap in the collective knowledge, however we know from colour block studies just how well we can discriminate between certain types of black cotton. On that basis we can apply the collective knowledge gained from other target fibre studies to help us.

The choice of blue lambswool, dark blue wool and blue cotton as target fibres in the studies is remarkable because natural fibres can be very challenging to compare under the microscope. In contrast to synthetic fibres, natural fibres from the same source can be so variable in terms of their physical appearance that their comparison under the microscope is immensely challenging. Nonetheless, the results of the studies were ground breaking.

Snapshot in Time

A major weakness in the target fibre studies done to date is that none of them have been able to capture the fibre populations on the surfaces of garments at precisely the same time. The design of such a study would require several hundred garments worn by volunteers for a period of at least 8 hours on the same day and to have all the fibres recovered from them examined.

Whilst target fibre studies are so important as a guide and help demonstrate the power of forensic fibre evidence, finding small numbers of fibres on a few hundred garments that are possible as a coincidence, we really need to explore what that means.

What did all these studies find?

Below is the table of findings from Wiggins et al:

This is what Wiggins et al concluded

Ok. That’s a pretty good assessment, but its important to stress that if we think of these findings in a case context one thing we are not doing in casework, is examining hundreds of garments picked at random in the world all at once. In casework, fate selects items of clothing from a victim and police officers indirectly select the target fibres by their choice of suspect and the seizure of their clothing.

The science says finding any target fibres by coincidence in a small number of items of clothing is so unlikely that it remains a wonder that fibre examiners find any matching fibres at all. If I was to pick two people at random, who have had no physical connection or shared environment, right now at this moment, recover all the fibres from their clothing and search for them on each other, I would most likely not find any fibre connection between them whatsoever. The most important message from target fibre studies remains not “We searched the surfaces of hundreds of garments and we found matching fibres” it is in fact “look at how many garments we had to search to find any at all”. So when fibre links are found, they have meaning.

The target studies also show that the more target fibre types you find on a garment the lower the chances become of a co-incidental finding. So, for example, if a suspect is wearing a blue and red striped top constructed from blue polyester fibres and red polyester fibres and both are found on the victim’s top, that’s a big deal.

When you also consider a two-way transfer where fibres from the victim’s clothing are found on the suspect’s clothing and vice-versa, that’s an even bigger deal, because a coincidental finding would require the victim to have made contact with something else made of the same fibres as the clothing of the suspect and vice-versa, all within a short time frame.

What target fibre studies have given us fibre experts is a logical framework on which we can assess the strength of a forensic textile fibre link between two surfaces. Generally speaking the more links that are found, the stronger the link becomes. It follows therefore that to get the best out of fibre evidence you should always be looking for as many fibre links that you can.

This is where the private forensic market in England and Wales controlled by police procurement teams steps in and says “hold my beer”. Imagine setting up a system where police decide on whether they should pay for a one way transfer or a two way transfer, that would be completely mad right? A system engineered to generate poor outcomes (fewer positive cases) with crappier evidence (weaker evidential strength). More on that in future posts.

We do need to explore the number of target fibres found more deeply. It does appear to be the fact that in the unlikely event that a co-incidental finding occurs, the fibres are found in low numbers. But does that mean that when you find fibres in low numbers they are likely to be the result of a co-incidental finding and that any finding of fibres in low numbers should be dismissed? We know from a previous post³ the effect persistence has on the numbers of fibres found, but to consider the findings from the target studies we also need to learn about how fibres are transferred. That’s for another post, but for now, its good to explore a couple of target fibre studies in more detail to show what precisely is going on “under the hood” as our American cousins might say, because what’s under the hood really matters and tells the story of coincidental findings far better than merely how few fibres are found on garments at random. I’ve picked two really good ones IMO.

A Study on The Random Distribution of a Red Acrylic Target Fibre

This study⁴ referenced by Wiggins et al previously, is important because of the extent of the collaboration across Europe. 39 laboratories in 19 different countries all looking for precisely the same fibre on random surfaces. Wow.

Furthermore, there was so much information available about the target fibre, that no previous team conducting a target fibre study had the luxury of knowing.

Finally this was real caseworkers searching real items of clothing submitted to forensic laboratories, so confidence in the search and analysis could be ranked very highly indeed.

The target fibre was a red acrylic fibre found in a red “Cashmink” scarf which had annual sales of 5000-10,000 scarves per year, across Europe and the US. Information from the manufacturer showed that a fabric was dyed in batches which would be used to manufacture 1000 scarves⁵. Although there were limited samples available to test from different batches, forensic scientists could not discriminate between them at least to batches made since 1990, 7 years prior to the research being conducted. On that basis, the scientists concluded that batch to batch variation amongst the thousands sold would be minimal and unlikely to be detectable.

38 working forensic laboratories across 19 countries examining 435 items of clothing for these fibres. There was one constraint placed upon the searchers, that was to search tapings used to recover fibres only from the collar and shoulder areas of upper clothing - where a scarf, if worn, would be likely to transfer fibres. This is an understandable constraint, but a weak point in the design, because it fails to consider fibres picked up by secondary transfer mechanisms across a wide range of clothing types, but more on transfer later. A further weak point in the design is the small number of garments examined by every laboratory, but as a body of work it retains immense value.

Across the 435 items of clothing two target fibres were found, one fibre on a garment submitted as part of a case, one on a jacket belonging to staff member from a laboratory.

The authors concluded that:

The point of view that one matching found fibre can be ignored, is one to which the authors cannot subscribe. When only one matching fibre is found, it must be explained that it is impossible to prove that it is not an airborne contaminant⁶. The precautions taken against this happening should be stressed. If the one fibre occurs in combination with collectives of other matching fibre types or together with other types of evidence, it is reasonable to conclude that it strengthens the overall picture.

The investigation into the production of the scarves described in this study, proves, as already shown elsewhere [9], that even textiles that are described as having widespread mass production (like scarves that can be found in department stores) have, in fact, relatively small production numbers. The probability that a collective of fibres of a particular generic type/colour/morphology combination will appear as a foreign fibre on any piece of clothing which one may care to select can be said to be extremely remote.

The Prevalence of Two ‘Commonly’ Encountered Synthetic Target Fibres within a Large Urban Environment

I really like this study⁷ because it builds on the Wiggins paper and places the findings in the real world context of activity on surfaces. Yes, the headlines are that a lot of surfaces were searched for two target fibres and none were found, but there’s more to this study than that.

Tapings from seats relating to cinemas, public buses and public houses within Newcastle upon Tyne were taken to recover fibres. The authors made the point that the surfaces were chosen owing to their likelihood that they would have contact with a huge number of different sources of fibres from multiple individuals over time. On that basis, it would not be unreasonable to expect a greater chance of finding adventitious matches with the target fibres, compared to surfaces such as garments and car seats.

Data from the cinemas where fibres were recovered showed that:

…approximately 18000 tickets were sold in cinema 1 and 85000 tickets were sold in cinema 2. This translates to approximately 900 potential contacts per seat for cinema 1 and 300 potential contacts for each seat in cinema 2. Whilst these figures would of course assume that all of the seat contacts were evenly distributed and that all tickets sold were actually used, it nevertheless serves to illustrate that the surfaces sampled are likely to have been subjected to a large number of random contacts with a large variety of garments in the population.

In relation to the bus:

The bus service from which the seats were sampled, ran daily between Newcastle city centre to Easington in county Durham (with 10 stops in between). The service is run approximately 70 times a day, with an average of 1500 people using the service. As with the cinema seats, it can be seen that there is a large number of potential contacts between clothing of the populace and the recipient surfaces (it may however be argued that the range of bus seat contacts is less random than that of cinema seats).

The surfaces were searched for target fibres from a pair of gloves (black acrylic) and a fleece (blue polyester):

both target fibres types are commonly encountered (in the generic sense) in casework and have been shown to be the most frequently encountered synthetic fibres in fibre population studies. Despite this, no fibres matching the either of the two target fibres were found on any of the seats sampled.

What I really like about this study is that it shows that the dynamic nature of the fibre exchange has to be taken into account when considering the “all fibres are mass produced, therefore the same, therefore found everywhere, all the time, at any given moment, therefore of little evidential value” argument we hear all the time from those who are uneducated in the field.

Fibres are constantly being transferred and lost at the same time, so when there is high traffic from multiple sources on a surface (like public transport or a cinema or a pub) whilst the likelihood that the seat may enounter a particular textile is increased, the likelihood that the fibres will be lost quickly after transfer is also increased, therefore reducing the likelihood that they will be found at all.

There are so many complex mechanisms at play, that unless you are an expert in forensic fibre examination you will not be in a position to understand, never mind make a decision over a strategy to employ in a criminal investigation.

Wearing clothes, buying clothes, seeing other people wear clothes, is insufficient knowledge, it tells you nothing about the complex behaviours that single textile fibres experience in the microscopic world in which they exist.

Next up transfer of fibres and the role of numbers.

1. Cook R and Wilson C . The significance of finding extraneous fibres in contact cases. Forensic Science lnternational 1986; 32: 267-273.

2. A study in relation to the random distribution of four fibre types on clothing (incorporating a review of previous target fibre studies) K. Wiggins, P. Drummond , T. Hicks Champod. Science & Justice Volume 44, Issue 3, July 2004, Pages 141-148

3. Fibres are Lost Quickly, What Does That Mean?

4. Bruschweiler and Grieve, Science & Justice 1997; 37(2): 85-89

5. Although that number could and did rise to 5000 occasionally

6. The authors discuss contamination at length in their research owing to issues with laboratories having handled the cashmink scarves in the past, prior to this research being conducted. I haven’t gone into that here, but it is a good discussion on laboratory contamination if the reader is interested.

7. Sheridan et al, Science & Justice 63 (2023) 509–516