Soil samples taken from the forest where a sex worker's body was found "corresponded" with those discovered in the van of the man accused of murdering her, a court has heard.
Iain Packer, 50, is on trial at the High Court in Glasgow accused of murdering Emma Caldwell, 27, in 2005, and faces 46 charges involving a number of women including rape as well as abduction and assault.
He denies all the charges against him, and has lodged special defences of incrimination, consent, defence of another and self-defence.
Giving evidence on Wednesday, Dr Stefan Uitdehaag, from the Netherlands Forensic Institute in the Hague, said he wrote a report on palynology - the study of soil - after being commissioned by Police Scotland.
He calculated "ecological distances" between six samples from around Limefield Woods in Biggar, South Lanarkshire, where Ms Caldwell's body was found on 8 May 2005, and a sample found in the footwell of Packer's van, examining pollen composition.
The six samples from the site included from a molehill on a route to the body site, two from ditches where Ms Caldwell was found, from under spruce needles, from an area where footprints were found a few metres from the body, and from under moss.
The forensic scientist said the results showed the soil was "much more likely" to have come from the same location as the sample in van, and ranked as "odds 100 times to 10,000 times more likely", or a "99.99%" chance, that they came from the same spot, rather than another random site.
The calculated "ecological distance" considered the sensitivity of the results such as mixing and selective loss of material after transfer from location to the vehicle.
Dr Uitdehaag's report examined "proposition one" (P1) - that the sample from the van came from the same location as other samples; and "proposition two" (P2) - that it came from a random other location, defined as "within Scotland outwith 100 metres".
The report said the results of the examination "fit proposition P1 well", but three samples fitted P1 "very well", it added.
Dr Uitdehaag, who has worked on "best practice guides" for Europe, told advocate depute Richard Goddard KC that the three samples "were more likely to have come from the same location than from a different location".
Mr Goddard said: "There was a slight difference which was the amount you would have expected. You found a wide variety of pollens which corresponded between the van sample and the samples taken from the woods."
He added: "The chances of getting these results are 100 to 10,000 times more likely that P1 applies than P2."
Dr Uitdehaag replied: "The proposition of P1 is 99% to 99.99% more likely than P2. The numbers are based on my knowledge and expertise. It means the results are more likely in P1 or P2, in this case much more likely."
The trial, in front of Lord Beckett, continues.