Some scientific hypotheses, however likely or unlikely they are to be true, refuse to die – mostly because they cannot be “killed” by a refutation within the grasp of human scientists. SETI is an example of this – the hypothesis “Aliens exist in our Galaxy” could be falsified if we could search every planet in the Milky Way, but it is simply not possible. This paper attempts a clean kill of another hypothesis which tests our astronomical skills to the limit – the wonderfully named Nemesis hypothesis.
The primary proponents of this hypothesis are palaeontologists, not astronomers. Analysis of the fossil record indicates there may be a periodicity in it – i.e. the extinction rate increases and decreases rhythmically over time, just as day and night ebb and flow, or the moon’s phase changes. Admittedly, the rhythm is a lot slower than a day or a few weeks – the palaeontologists suggest the period is 26 million years. Well, not all of them – this detected periodicity is the source of a great deal of debate in the field, with no clear consensus.
Let’s leave that argument alone for a moment, and assume that the periodicity exists. What is the source of the periodicity? The Nemesis hypothesis solves this problem by proposing that the Sun is not the only star in our Solar System, but in fact has a binary companion. The reason we haven’t seen it is because it isn’t very bright – perhaps it is a brown dwarf, a failed star which cannot “burn” hydrogen and shine like our Sun, or simply a very small, cool star like an M dwarf (see artist’s impresion below) that we would find difficult to spot. If the eponymous Nemesis did exist, then its orbit would take it within close distance of Earth every time it orbited the Sun, its gravitational field and radiation kneading and grilling our planet, causing the peak in extinction rate every 26 million years.
A very nice argument, but almost impossible to falsify. A non-detection of Nemesis with our current telescopes can’t rule it out. It could be simply too faint to see, plus we’re not entirely sure where on the sky we should look. The next generation of telescope surveys like Pan-STARRS and LSST would have a better shot – one of their functions is to detect faint objects like asteroids in our local neighbourhood. But, even if the telescopes of the 22nd Century (or the 23rd, for that matter) didn’t detect Nemesis, the hypothesis would probably still linger on, as the consequences of the Sun having a binary companion are too staggering to ignore.
What this latest paper does instead is go back to the fossil record, and compare the results of two independent data sets. By cross-correlating the data to look for common trends, they show that a statistically significant periodicity does come out in both data sets, with a value of around 27 million years. Not only that, but they show the periodicity exists for a much longer stretch of history than previously expected, around half a billion years.
However, their conclusion is that the Nemesis hypothesis is false. This is because the periodicity is too regular – in fact, it has hardly changed at all. If Nemesis was to blame, then the period will almost certainly change with time. Why, you ask?
Imagine you have two stars, one orbiting the other, in a Galaxy full of other stars and interstellar bric-a-brac. If something with a large enough gravitational pull comes close to the binary system, then the orbit will be changed – if the gravitational pull is strong enough, the orbit could even be broken, spitting Nemesis out into the Galaxy. Over billions of years, it is highly unlikely that the Sun-Nemesis system could escape close encounters with another star. The orbit of Nemesis must have changed over time, and the periodicity must have changed too.
We must therefore conclude that because the periodicity has not changed, then Nemesis cannot be responsible. If Nemesis isn’t making a dent on our fossil record, then it probably doesn’t exist.
Of course, the paper still hasn’t achieved a clean kill – Nemesis only probably doesn’t exist – but it’s the best that Science can do today. We will have to wait for better data and instruments to put this theory to bed. And that I guess is the deep truth about scientists, and the scientific endeavour – we’ll always have to try again next time, when the next generation of instruments and data comes along. And we’ll relish the challenge, as we always have done.