Following some initial analyses of the imported Horsfield's tortoises an interesting pattern has emerged.  In an earlier post I reported some preliminary results: nitrogen signature (d15N) varies significantly between individual tortoises and the carbon signature (d13C) of growth of a medium age is significantly higher than that of new growth.  When I took the samples I also noted whether I thought the specimens had lumpy or smooth shells: a lumpy, or 'pyramided' shell is often a sign of a captive tortoise. 
Figure 1. Changes in carbon and nitrogen signature for individual tortoises with lumpy (blue), smooth (red) and unclassified (yellow) carapaces.

When pooled, this data sorts the tortoises into two distinct groups, suggesting that they come from different populations (figure 1).  My guess would be that this reflects a split between wild-caught and ranched individuals; though this is speculation it does seem to warrant further investigation... 
Being unable to work at any speed in the lab and being liable to dropping things due to my dodgy shoulder, I have taken the opportunity to catch up on some website pages.  The species profiles are finally undergoing some work - the page for Hermann's is now up (link) and I'm getting to work on Horsfields.

I ran a plate of claw samples this week and so far it would seem that claw is more enriched in both carbon and nitrogen than the shell (by about 1.5 permil for carbon, and by about 2 or 3 permil for nitrogen).  I can't say much more than that at the moment as I used whole claw and do not know what time period it reflects.  The enrichment could be due to different trophic enrichment factors (TEFs) (tissues formed using the same foodsource may end up with different isotope ratios to each other due to different metabolic pathways), may reflect a change in location or diet between when the shell was grown and when the claws were grown, or be a combination of these factors.  
Well, it's been a while since my last post, but much has happened since then.  Unfortunately I have now come to a bit of a halt with lab work due to a fractured shoulder blade following an unintentional gallop and fall off a horse. 

I have weighed and run nearly all the carapace and plastron samples for carbon and nitrogen analysis and the results seem promising.  These samples are of unknown provenance - they were re-exports from the Ukraine but that is all I know.  I am therefore using them purely to test variation within individual tortoises.  As I had hoped, my preliminary results suggest that there is no significant difference between scutes (the hexagonal sections that make up the shell), and no significant difference between the top (carapace) and underside (plastron) of the shell.  This means that I will be able to pool material from multiple scutes to get a better sample weight.   
So: no significant variation between the scutes.  There is however a fair amount of variation between the different ages of growth, particularly for nitrogen: the tissue the tortoise was born with (the juvenile scute) is more enriched in Nitrogen than other growth layers (figure 1).  Nitrogen becomes enriched in the heavier isotope as trophic level increases; a predator will be more enriched in 15N than it's prey.  My results are therefore not surprising as Horsfield's tortoises are capital breeders, which means they use stored resources for reproduction.  The tissues of the offspring are produced from metabolised fat and are thus effectively a trophic level higher than the tissues of the mother.  Had I not seen a difference between the juvenile and newer growth I would have been concerned about my sampling. 
Figure 1: Isotope signatures accross the scute. Left: nitrogen; right: carbon

Interestingly, there is also a significant change in the carbon signature across the scute (figure 1), which could be due to a change in diet or location, but I don't have enough information about the tortoises to know what the cause is.  Again, although I don't know the reason, there are also significant differences between individual tortoises in both carbon and nitrogen signatures (figure 2).  This might be because some are ranched and others wild-caught - the higher nitrogen signatures do belong to the smoother shelled tortoises (ranched tortoises tend to have lumpier shells than wild tortoises) but this is sheer speculation.
Figure 2: Nitrogen signatures of individual Uzbekistan re-exported Horsfield's tortoises
Before I came up to York I used a freeze dryer (another machine new to me) to dry the soft tissue samples so that I wouldn’t have the headache of transporting frozen samples halfway across the country without them defrosting.  Samples have to be dried out before analysis so this would have had to be done at some point anyway.  I’m a little worried about the smell: I think a couple of the specimens were not put in the freezer soon enough…or defrosted on their journey from Heathrow to Falmouth.   I took notes on the condition of the specimens during dissection though, so I will be able to remove the less well preserved ones from analysis.

Having been in York for almost a week (during which time I have been through the whole induction process, got lost multiple times, and found a house), I am now sitting on a train back to Falmouth – I feel the Christmas lecture, Christmas meal and a couple of meetings almost justify this.  Before I left I put my bulk material and some of the keratin samples in the oven (finally a machine I don’t need instruction for) so they can dry whilst I am away and will be ready for weighing when I get back. 

So I'm finally going up to York tomorrow to use the machines at the Fera Laboratories for analysing my samples.  I thought I would be going earlier in the project but with one thing and another it just didn't happen till now!  

I'm quite excited about it, if slightly worried about the fact that I have yet to find somewhere to live - for I don't know how long... I'm sure this will sort itself out soon enough though!

Things are finally moving :D
Sampling the carapace took a while – but at least it was a relatively clean job.  A week ago every time I closed my eyes I saw an image of a tortoise carapace – quite pretty but after a while somewhat annoying.  Now I close my eyes to see the internal anatomy of a tortoise, and am followed around by the accompanying smell.  Needless to say I preferred the former…

And yet I always did like dissections however morbid they may be. 

Tortoises present a problem – how do you get into them?  Rather foolishly I initially tried using a scalpel – I have no idea why I thought this was in any way logical but I’m going to use the excuse of being tired.  A good night’s sleep and three broken blades later I realised my mistake and went out to buy a hacksaw.  This was far more productive, though drew some strange looks from others in the lab.  Mind you it still took a while to actually get the carapace off – these are incredibly tough little beasts!  The experience has definitely renewed my respect for lions, having watched them crack open tortoises with their teeth when I was in Africa several years ago…

Anyway – why the dissections?  Well, different tissues turnover at different rates: the keratin of the carapace once formed does not change, so it reflects the diet at the time of growth.  Blood, liver and muscle have varying rates of turnover, with blood reflecting the shortest time period.  As far as I know turnover rates have not been calculated for tortoises, but those for the pondslider, Trachemys scripta, are surprisingly slow with a Nitrogen turnover rate of 155 days for whole blood (Seminoff et al, 2007).  Although all the evidence points to the pattern of scute growth shown in my previous post, there has been little work on how tortoise scutes grow.  It is therefore worth checking that the isotopic values of soft tissues correspond best with the values for what I believe to be the most recent growth.  

diagram of the supposed growth of laminae of the scute.
I have now photographed, measured and weighed the dead imported tortoises from Heathrow and have taken samples from the top (carapace) and underside  (plastron) of the shell.  It seems that the best tool to use is a dental scaler that has been bluntened slightly This gives more control over how much material is taken, which is important for two reasons: 1) If the technique is to be used on live tortoises it must be possible to do without fear of getting too near the nerve layer under the keratin, and 2) because I want to be able to get relatively good temporal resolution  (see image).

areas sampled on plastron and carapace
As these are the first samples I’ve taken, I’ll be using them to check whether isotopic values are consistent within individuals.  I therefore took samples from 4 different scutes on each tortoise (see image).  I chose these scutes as I assumed they would be least susceptible to wear, though I’m not entirely sure I’ve got this right…

I also want to see if I can consistently sample a particular period of growth and have therefore sampled three areas of growth on each scute (following the growth rings).  I am crossing my fingers on this one as it would be incredibly useful if I could sample multiple areas of growth to track changes in diet or location.  

Next up: dissections!

(Image of scute cross section adapted from Seltzer and Berry (2005))

scute from an old long dead Hermann's tortoise
Following my intent to post a bit more regularly, this is just a quick note of where I have now got to.

A while back a vet kindly gave me some scutes from a long-dead tortoise (see picture on left) , so I have now ground these down into a fine powder to use as my bulk material (see explanation below).  This involved using a cool machine I've not used before   - a high velocity ball mill (image below).

When analysing samples for stable isotope analysis multiple samples are put in a run, many of which will be the samples that have an unknown isotopic value (the one's we're interested in).  Others are samples of known isotopic value.  These are called standards and are used to calibrate the data.  It is also good to include a sample of bulk material that is similar to the unknown samples.  Bulk material can be included in every run to check consistency within and between labs.  

a dental scaler (ie a small blade on a stick)
It has been an incredibly long time since my last post and I must admit I feel rather guilty about this.  I’ve been trying to work out exactly how I want to use this space and in doing so have failed to use it for anything. To remedy this I have now given myself a target of writing a post at least every other week, either an update of the project or a commentary on other work in the field.  I think one of the things I can take from the first few months of my masters is the importance of keeping things moving.  Letting things slide is all too easy!  

Since my last post I have spent much of my time going through the literature on tortoises and stable isotope analysis, contacting vets to talk through sampling procedures, and contacting breeders and zoos to discuss the project and source samples.  I’ve also looked at import and export data and paid visits to the Animal Health and Veterinary Laboratories Agency (AHVLA) and a major importer of tortoises,so I can better understand the current state of the tortoise trade in the UK.  Most recently I have set up a survey to find out what tortoises in the UK are fed – and how much variation there is in diet (the technique I am using depends on what tortoises eat and where their food is from in the world).  Results from the survey should give me some idea of how representative my samples are -  If you own or have recently owned a Hermann’s, spur-thighed, or Russian tortoise I would be very grateful if you could take a few moments to complete it here!   

Perhaps the biggest recent development has been that I now have twenty imported tortoises that died in transit to the UK sitting in a freezer in the lab.  I will be taking scrapings from the upper and lower surfaces of the shell (the carapace and the plastron) using a dental scaler and will be taking these up to the laboratories in York  

So in a few weeks I will be so much closer to having some data to analyse (I hope)!

A problem to solve
pet Hermann's tortoise (Testudo hermanni)
When I first read about the MRes that this blog is set to follow, I must admit that one of my first thoughts was-- ‘Illegally imported tortoises? Really? How much of an issue can this possibly be?’--Yet it turns out that alongside rhino horn and elephant ivory, tortoises are one of the big concerns in wildlife trade regulation.

The tortoise trade
Tortoises are a popular pet in many countries including the UK, though perhaps not one of the first animals that spring to mind on hearing the word ‘pet’.  But then when I think about it I do vaguely remember wanting a tortoise when I was younger - maybe Blue Peter and the wonderful George had something to do with it.  My parents had a pet tortoise.  Someone down the road had one that ran away.  And my Grandad wrote a story about one.

In fact thousands are imported every year.  There are breeders in the UK and in mainland Europe, but there is no way they can currently satisfy the market.  The numbers for some species are quite shocking and I’m left wondering how wild populations can possibly sustain such trade. According to CITES data over 200,000 Horsfield’s tortoises (Testudo horsfieldii) have been exported from Uzbekistan alone in the last five years, with over 30,000 of them coming to the UK; all of them wild-caught or ranched.  Horsfield’s tortoise is currently listed as vulnerable on the IUCN red list - it ‘is facing a high risk of extinction in the wild in the medium-term future’.  Although much of this trade is legal, beyond expert opinion there seems to be precious little that can be done to validify claims of origin.

Mediterranean spur-thighed tortoises (Testudo graeca) and Hermann’s tortoises (Testudo hermanni) are also traded in large numbers even though all trade of wild caught and ranched individuals of these species is illegal under CITES regulations.  Accordingly, neither species can legally be traded without Article 10 certificates, on which origin is recorded.  Unfortunately these certificates are not specimen specific until the tortoise reaches a size at which it can be safely microchipped - so it is possible to pass a wild caught tortoise off as captive bred through the use of fraudulently obtained Article 10s.  Worryingly, in a study by the RSPCA it was reported that Testudo graeca could be bought for as little as £1 on market stalls in Marrakech - and I have heard similar reports about other markets.  People who buy these animals may or may not know that they are breaking the law by doing so, but such animals can make a very tidy profit, going for as much as £200 if sold as captive bred. 

Where did it really come from?
With less commonly bred species such as the popular Indian star tortoise (Geochelene elegans), it is possible to check claims of captive breeding using DNA tests to check parentage.  But with high volume species like spur-thighed and Hermann’s the cost and logistics of this become prohibitive.  This is where my study steps in:  Stable Isotope Analysis (SIA) is relatively cheap and looks promising as a tool to confirm claims of origin. 

As this tool has not been used for tortoises before, a major aim of my study is to provide proof of concept.  I will therefore be starting small, developing a sampling protocol and checking that SIA can be used to distinguish between UK and imported tortoises. 

SIA has huge potential in the field of wildlife trade regulation but is largely untapped.  If it can be used successfully on tortoises (and there are multiple reasons why it might not) the possibilities seem almost endless.