The flaws of hair analysis: Longleat Leopard
A while ago I took part in a tracking session in Longleat forest with a group of cryptozoology researchers. We were looking for signs of a big cat being in the area after several eyewitnesses reported seeing a cat in the area of the forest.
The one thing found in the forest that was of interest was a single dark coloured hair that was found on a wire fence in the forest by three of the cryptozoology researchers who were joined by my younger brother, Charlie.
This hair was bagged up as “potential evidence” and taken away and after returning home I thought nothing more of it. A few weeks after the discovery of the hair the cryptozoology researchers that had been with us on the day were attending a conference called “The Weird Weekend” hosted by the CFZ (Centre for Fortean Zoology).
While at the conference they got Lars Thomas, a well-known cryptozoologist, to identify the hair on a piece of equipment he had at the conference that was designed to examine hairs and similar.
He did so and declared that it showed signs of being a leopard hair. the cryptozoology researchers got excited and saw this as evidence that there may be a big cat in Longleat Forest. However, I wasn’t as keen as them to make that assumption.
I didn’t think that one hair equals proof of a cat being present, also, I didn’t know enough about the testing process to gather a clear view of how it had been identified.
I wrote a piece detailing my exact position on the case. This received a lot of criticism from the cryptozoology researchers who accused me of being untrusting and having an agenda, a lot of people couldn’t understand why I didn’t agree with the hair being a leopard hair.
I am taking this chance to finally lay down the reasons I don’t trust the testing and the results and don’t count them as proof.
Problems with the initial hair testing
This is the picture I was provided of the testing of the hair.
This is a photo of the hair being tested.
Lars Thomas used an RGB colour camera which immediately reduces the working resolution to 1/3 of what he probably thinks it is. Apparently, using a black and white camera with a colour slider to create false-colour images would have preserved the resolution of the image much better.
A colour camera uses 3 pixels (a red, blue and a green) to describe a single colour pixel, so you reduce resolution drastically. He has, I note, also zoomed in and taken a photo of the monitor which is why you can see individual pixels in the first picture.
The lighting conditions are not great for microscopic analysis of a sample either. A bright room = bad for microscopy. According to a biologist I am in contact with, Lars should really have used polarised light which it really doesn’t look like he hasn’t done.
Also having your lunch around the microscope is generally not considered good scientific practice…
Not only were there issues with the testing conditions, but a single hair is just not enough. All biological samples display variation and the FBI recommend using at least 25 hairs to build up an accurate morphological profile.
Plus, although the hair in the picture does appear to display the segmented medulla (middle of the hair) that felines have (although due to the resolution of the image, it is hard to say to what degree they are segmented), it does not appear to display the characteristic overlapping cuticle (outer edge) of the feline hair (see picture C & D below) You can also see the segmented medulla in the dog hair photo (See picture E)
Without a forensic analysis of this, the whole testing is inconclusive. There is no DNA evidence, a single sample, poorly imaged…
Ultimately what it boils down to is this: lacking any other evidence of a leopard (no kills, no bones with tooth impressions, no prints) this hair is more or less meaningless.
This is exaggerated by not a) providing images of known leopard hair and b) no side by side comparison with other hair from other animals likely to be in that environment.
Without a DNA profile you can’t say for sure anyway. The one thing taxonomists have come to realise over the last 100 years is that morphology alone is not a good system for defining a species. Let alone the morphology of one hair!
When an animal hair is found, it is identified to a particular type of animal and microscopically compared with a known hair sample from either an animal hair reference collection or a specific animal.
If the questioned hair exhibits the same microscopic characteristics as the known hairs, it is concluded that the hair is consistent with originating from that animal. It is noted, however, that animal hairs do not possess enough individual microscopic characteristics to be associated with a particular animal to the exclusion of other similar animals.
Now, the biologist that I have been speaking to believes this is a dog hair. To see if his was a common view he posted the picture of the hair on a scientific forum that he is a member of. He posted just the photo for analysis – no back story, and the common view from other members is that it looks very much like a dog hair.
Obviously there is no way to be 100% sure, but something I am sure of is this – concluding that this one hair is proof of a leopard being in Longleat forest is clutching at straws.
That is why I held (and still hold) a skeptical and doubtful position on the case. I am not the one using flimsy evidence to support my beliefs and this will be the last I write of this case until more substantial evidence is provided by the cryptozoologists who are making claims about a leopard in Longleat Forest.
photos provided by 'A.D', the biologist who I consulted with over this case and the testing of the hair. A.D wishes to remain anonymous, but can be contacted if you don't believe he is real. Just ask.
Posted on October 1, 2010, in cats, cryptozoology, irrational, misattribution, rational, the team, Wiltshire and tagged big, cat, forest, fortean, hayley, leopard, longleat, stevens, wiltshire. Bookmark the permalink. 4 Comments.