Readers Nuggets:

Ever wanted to know just how hard it is to clone mammoths? PSD explains the pros and cons in this notoriously tricky area of genetic research.

Okay, so what was a mammoth anyway?

Most people think of mammoths as hairy elephants, and they aren't far wrong. Mammoths and elephants are both classified as being in the order Proboscidea, which basically translates as 'trunked animals'. Recent DNA evidence suggests that mammoths were more closely related to Indian elephants than the African, which makes sense geographically. And if youíve ever wondered what the difference between mammoths and mastodons was ñ itís all in the teeth.


Meet Bob. He's extinct. And he's bloody angry about it.

In terms of size the woolly mammoth, Mammuthus primigenius, was about the same height as an Indian elephant, except weighing anything up to three tonnes more, and some of the North American mammoths were even taller, although not massively larger in weight. The weight difference between mammoths and elephants may well have been an adaptation to climate.

So why and when did they become extinct ñ and would clones only become extinct again?

The 'whení is easier than the 'why'. Most estimates place the mammoths as being extinct by 8500 years before the present, with a remnant population surviving a few thousand years later on one remote island. As for why they became extinct, there are three main theories.


Hunting has long been suspected in the case of the North American mammoths as their extinction appears to be correlated with the arrival of the Clovis people. At the same time a large number of other large animals (the ëmegafaunaí) also seem to have become extinct. However direct evidence - such as butchery marks on bones - is so far lacking. Further evidence against the hunting theory is that the European megafauna disappeared at a similar time, despite humans in Europe being more technologically advanced than the Clovis. If hunting was responsible, it would no longer apply as a threat to mammoths that had been recreated - unless someone claims 'ancestral tradition' and demands the right to hunt. (As an aside, has anybody noticed how ëancestral traditionsí of hunting donít seem to include using ëancestralí weapons and clothing?)


It has been suggested that a highly contagious disease may have wiped out the mammoths. Again, this fails to explain why mammoths should have been wiped out across the world at a similar time, rather than in a wave spreading outwards from it's source. This theory does have some serious support, however, and should not be discarded lightly. Resurrected mammoths might face disease still, depending on what had done for them. Those frozen carcases may still hold viruses, for one thing...

Climate Change

Perhaps the most immediately appealing theory is that the change in climate at the end of the last Ice Age killed off the mammoths. However, this may well be too simplistic a view. Mammoths had already survived several cycles of glaciations followed by warmer periods. Tantalisingly, however, summer conditions during the last ice age may have actually been warmer than during the modern inter-glacial - especially during an episode known as the Younger Dryas, a sharp 500 year spell of colder conditions about the same time as the mammoths disappeared from most of the world. Pollen from frozen mammoth corpses has shown that they fed heavily on tundra flowers ñ even a few summers without this source of food might have been enough to tip the mammoths into oblivion.

If climate killed off the mammoths, it wouldnít have been directly. Mammoths were highly unlikely to have been stupid enough to stay somewhere too cold for them ñ they would just have moved south. However a colder summer climate could have killed off their food (flowers not being renowned for rapid movement), thus affecting the mammoths indirectly. It should be perfectly possible to keep mammoths alive in the modern climate.

Most likely, however, is that there were a range of factors involved. The problem with science is that when there are competing theories everyone tends to try to find only one cause, when the likelihood is that several causes acted together. Imagine, for example, that the climate change reduced the population of mammoths by reducing their summer food. The number of herds decreased, and also the numbers in each herd, leaving them more open to disease, inbreeding and hunting. Increased hunting and disease reduces the numbers still further, and a vicious cycle forms driving the population downwards, until some critical threshold is reached and the mammoth is doomed. Populations can be dragged to extinction from surprisingly high thresholds, such as the Passenger Pigeon - whose 'minimum' popuation turns out to have been in the tens of millions. This is known as an Allee effect, when the minimum population size is higher than genetics would suggest - for example mammoths may have needed to form huge herds to migrate, and without the weight of numbers migration to summer pastures became much harder, even if the gene pool was still healthy. Donít confuse this with the Ale effect, where reproduction is hampered by brewers' droop.

So, what's with all these deep frozen mammoths we hear about?

Every so often (well, quite rarely, really) a frozen mammoth is found in the Siberian permafrost. These unfortunate beasts appear to have fallen into deep pits that had opened up in the ice, and become frozen quickly enough to stop major decomposition. You can dissect the stomachs to discover the last meal, if you like. The flesh, apparently, has been eaten, and didn't poison anybody. Smart money has to go on it tasting like chickenÖ

So can we use the deep frozen mammoths to produce new ones and resuurect the species?

There are two issues at stake here. The first is ethical - is it right to bring them back? Iím not going to dive too deep into the ethics, but one question is where the hell do you put them? Put quite simply, the landscape they were a part of no longer exists. Wide open tundra grassland rich in wild flowers is common enough - but they need somewhere to over winter at lower latitudes ñ which is precisely where most of the worldís crops are currently grown. Certainly Britain couldnít support mammoth herds ñ itís simply too cultivated. This means that any mammoths cloned would have to be kept in captivity, raising even more questions about ethics.

The second issue is how to do it. There are two methods that have been suggested: nuclear cloning, and cross-breeding with modern elephants.

Nuclear Cloning

This is the technique used to produce Dolly the sheep. Take an egg, knock out the nucleus (where all the genetic material and suchlike is held) and insert the nucleus from the adult youíd like to clone. In this case the egg would have to come from an Indian elephant (the closest living relative) and the nucleus would come from one of the frozen mammoths. The egg would then be carried by a surrogate mother - an Indian elephant.


This was suggested by a Japanese team a few years back, who were busy looking for a frozen male mammoth. The idea was that Indian elephants may be closely related enough to mammoths to be able to cross-breed with them. Having used frozen sperm and IVF to get a couple of new mammoths, you can then fertilise females with more of the frozen mammoth sperm to increase the percentage of mammoth DNA. After a few generations you could have something close to 100% mammoth.

Incidentally, I have no idea why they were so keen on the mammoth being male, unless it was because a male mammoth will have more sperm than a female will have eggs. Given that they failed to find any frozen mammoths, I donít think they can afford to be choosyÖ

So whereís the catch?

There are loads of catches, to be honest, which is why nobody has so far done it. Here are some of the key problems:

Getting hold of intact DNA

One of the major problems is the nature of DNA. DNA is a remarkably fragile molecule, and cells invest an awful amount of energy in repairing it. As the cell dies the DNA starts to degrade. The hope of those wishing to clone mammoths is that the frozen specimens may have cells that have intact DNA. This is highly unlikely. A mammoth is a huge animal, and while it may have frozen quickly enough to prevent the bacteria in the gut from eating it from within, the chances of it freezing quickly enough to prevent large ice crystals forming are next to nil.

The problem with ice crystals forming is that as they grow they literally rip the DNA apart. This leaves the ëDollyí technique stranded, as it needs intact DNA. The Japanese technique, using frozen sperm, is even more likely to be scuppered by this: the entire sperm cell is liable to have burst, let alone DNA remaining intact.

The other problem that rarely gets mentioned is getting mitochondrial DNA from a mammoth. Mitochondria are best thought of as the batteries that fuel the cell. They have separate DNA to the rest of the cell, and there is evidence that this DNA plays a part in the organisation of the developing embryo and also within each cell. Basically, if it doesn't have mammoth mitochondria, it isnít actually a mammoth. Similarly, the internal structure of mammal eggs appears to be vital for development, and an elephant egg might not have the right internal structure to ever produce a mammoth.

People paying attention may have noticed that up above I said that DNA evidence showed mammoths to be more closely related to Indian elephants than African. Surely if we can get DNA for that we can get DNA for cloning? Unfortunately, it doesnít work like that. DNA fingerprinting only requires tiny quantities of DNA, and it relies upon breaking up the DNA into little chunks ñ and then comparing the chunks form one species against that of another. Cloning requires every single piece of the genome ñ in the right order.

It has also been suggested that you could custom build the correct DNA for a mammoth. People who suggest this point to the way that strands of DNA can now be made in the lab (at much shorter lengths than an entire genome) and that the time required and cost is falling all the time. This is fine - except you need to know what youíre copying to copy it. The same restrictions apply, and if you write the code for an approximation of a mammoth then you get an approximation of a mammoth. It still won't be a mammoth. The latest evidence is that the so-called ëjunkí DNA actually has an important function ñ but how do we work out what the correct ëjunkí would be?

Suitable surrogate mothers

It's all very well saying that mammoths and elephants were related, but that doesn't imply that elephants could actually carry a mammoth calf to full term. For instance, we know nothing about the size at birth of a mammoth. If the adults were half as heavy again as an Indian elephant, how can we be sure that a mammoth calf wouldn't be? There may also be immunological problems that may cause elephant surrogates to abort mammoth embryos.

With the idea of cross-breeding there is another problem with using elephants: they have different numbers of chromosomes. Elephants have 56, mammoths had 58. This means any offspring would be sterile - just like in mules ( across between horses and donkeys, which have different numbers of chromosomes). This would wreck any attempt to breed increasingly mammoth-like animals using frozen sperm - or any other sperm.


One major issue with cloning mammoths is that the smaller a population you breed from, the more likely you are to have a problem with inbreeding. If we can only get DNA from one mammoth, then weíre starting with a population of one. This isnít a good thing.

An inbred population, with little genetic diversity, is susceptible to disease for two reasons. Firstly, genetic diversity means that there will be a wide range of immunity genes in the population as a whole. A disease that can beat a lot of immune systems is pretty sad for the individuals as a whole, but as a population there are likely to be some animals that survive by virtue of having different genes. In an inbred population, there are less possible combinations of genes. This means a disease that can kill an individual has a good chance of killing the rest of the population too.

The second problem is the issue of genetic disease. On average, everyone reading this will have two genes somewhere in their genome that would kill them if they didnít have another version of it on another chromosome. This sounds scary at first, until you realise that the chances of meeting someone with the same malfunctioning gene and having children are pretty remote. If you were unlucky, however, then a quarter of your children might suffer a genetic disease through no other cause than blind chance. As I said, this is unlikely to happen to us, yet it does on a regular basis. Now, with our mammoth clones, weíre starting from a population of one (the dead animal weíre cloning from). The chances of genetic diseases occurring are much higher.

This isnít a hypothetical problem, either ñ itís one that concerns every captive breeding programme in the world. Populations of captive animals in zoos worldwide show less diversity than wild populations, despite our best efforts to prevent it. And those populations started with far more diversity than a population of mammoth clones might. Even managing to clone a herd of mammoths wouldnít be enough ñ the rule of thumb for most species is that inbreeding becomes serious with a population below 500.

There is one possible hope, however. Many species have gone through ëgenetic bottlenecksí, where the population crashed to a very low level. Cheetahs are one such species, and anyone cloning dodos will be pleased to learn that as an island species a bottleneck is virtually guaranteed. The result of a bottleneck is that if a species got through it, then there probably arenít that many harmful genes remaining. If there were, the population would have gone extinct. However, thereís no reason to believe that mammoths went through a bottleneck. Living in large herds, and roaming over wide areas, they simply donít fit the template for one.


Money talks, and in this case it says ìnoî. The Dolly technique required 300 attempts to get just one clone - and that was starting with an intact source of cells to start with. Yes, a mammoth clone might be worth a load of money from people wanting to see it, but enough to pay for all the work needed to get us to the stage where it might be possible? Not at the moment, and Iím willing to bet not in the future either.


Taking elephants as a model, elephant herds have matriarchs and many elephant behaviours are learnt from their own herd. If you don't have a mammoth herd for your mammoth clone to copy, is it behaving as a mammoth - and if not, is it still a mammoth? It's a philosophical question, but pretty important from a scientific point of view.

So what you're saying is that there's no chance of a mammoth being reconstructed?

Outside of books? No.

It might just be possible to make a mammoth/elephant clone, but thatís not a mammoth. Using it to breed a series of increasingly mammoth-like animals still doesn't make them mammoths either. The chances of using cloning techniques is there, theoretically, but youíd need a mammoth egg to get a pure mammoth - and thatís pretty unlikely - and it still might not act like a mammoth. Thatís not to say that we couldnít learn from making such a half-clone if it were possible, but in the end the idea we could recreate a mammoth is about the same as me writing a bestseller.

It's a nice idea, but it ain't going to happen.

Ben Tymens, AKA PSD