YEC Best Evidence 10: DNA in ancient bacteria

We now come to the last of Answers in Genesis’s top ten claims of evidence for a young earth. This one concerns the discovery of some still viable bacteria in 250 million year old salt crystals — a finding reported in Nature in 2000 by Vreeland, Rosenzweig and Powers.

The brief overview by Georgia Purdom gives very little detail about this discovery. However, Answers in Genesis has a more detailed technical article by Ewert van der Heide that gives some details and attempts an analysis to prove that the bacteria can not be as old as Vreeland et al suppose. The salt deposits, known as the Salado formation, come from the Delaware Basin (which, confusingly, is in Texas, not Delaware). Within these deposits, there are inclusions containing pockets of salt water, and it was in this salt water that the bacteria were found.

Naturally, the age of these bacteria has been greeted with a lot of scepticism by the scientific community as well. Responses by Hazen & Roedder (2001), Graur and Pupko (2001), Willerslev and Hebsgaard (2005), and Nickle et al (2001) among others, insist that the bacteria must be significantly younger than the deposits themselves. Purdom, of course, dismisses their response as “rescuing devices”:

Some scientists have dismissed the finding and believe the Lazarus bacteria are contamination from modern bacteria. But the scientists who discovered the bacteria defend the rigorous procedures used to avoid contamination. They claim the old age is valid if the bacteria had longer generation times, different mutation rates, and/or similar selection pressures compared to modern bacteria. Of course these “rescuing devices” are only conjectures to make the data fit their worldview.

Are the crystals the same age as the deposits?

It’s true that Vreeland et al took great care to eliminate contamination. They sterilised the salt crystal in concentrated hydrochloric acid and concentrated sodium hydroxide, and they reported no bacteria while they were drilling into the crystal, but only when they got to the salt inclusion. So at first glance, this one looks pretty convincing.

But the responses from other scientists aren’t claiming modern contamination. They are proposing that the crystals themselves formed at a later date than the rest of the formation. This could have happened, for example, during a glacial maximum during the past 100,000 years, if water had seeped into the formation, causing some of the salt crystals to dissolve and then re-crystallise. Hazen and Roedder (2001) argued that the clarity of the crystal itself indicated that something along those lines was likely the case, and also pointed out that fluid samples from the Delaware basin vary widely in composition — a fact pointing to a mixture of both ancient and modern waters. Furthermore, as Nickle et al (2001) pointed out, Vreeland et al did not provide any tests to eliminate the possibility that the salt crystals might have subsequently re-formed in this manner:

It is not hard to imagine that water seeped into this formation (e.g., during a recent glacial maximum within the last 100,000 years), resulting in the formation of new salt crystals in an otherwise old geological formation. In contrast to their elaborate controls for contamination, they did not present any data to verify the age of the crystal from which they extracted the bacteria. Hazen and Roedder (2001) have argued that the clarity of the crystal from which they extracted strain 2-9-3 is consistent with this crystal being of a more recent origin. Hazen and Roedder also pointed out that the fluids in the Delaware Basin (the geological region from which Vreeland et al. obtained their sample) are extremely heterogeneous with respect to the “absolute concentration as well as the ratios of halogen, alkali and alkaline-earth ions,” suggesting that this region contains a mixture of ancient and modern waters, though Powers et al. (2001) have countered that such heterogeneity does not necessarily imply that these fluids are from different ages.

The fact remains that Vreeland et al have not managed to convince the scientific community that the bacteria were indeed deposited along with the halite 250 million years ago. There are other quite plausible mechanisms by which they could have been deposited at a later date.

The “rescuing devices” fallacy

Throughout these ten claims, we’ve seen a constant drumbeat refrain of perfectly reasonable old-earth explanations being dismissed out of hand as “rescuing devices.” This reflects the YEC belief that long ages are nothing more nor less than an attempt to fudge things to accommodate the “evolutionary worldview.”

But as I’ve repeatedly made it clear, long ages do not come from an “evolutionary worldview”; they come from measuring things. Remember that in order to build a case that the earth is young, you need to provide robust evidence that hundreds of thousands of high precision and extensively cross-checked measurements are consistently in error by factors of up to a million.

It is simply not sufficient to dismiss any explanations of the data that you don’t like as “rescuing devices” or “only conjectures.” You must provide evidence that rigorously and systematically falsifies them. In every single case that we’ve seen, YEC attempts to falsify these so-called “rescuing devices” have been inadequate at best, in some cases nonexistent, and in others even outright dishonest.

Before YECs can claim a slam dunk with this claim of ancient bacteria, they need to provide solid evidence that the crystals had not recently dissolved and re-formed. They must also provide evidence that the mutation rates really could not account for their great age, if that is what it turns out to be. Nickle et al proposed a couple of tests that could be carried out as a starting point to investigate the matter further. For example, they suggested carrying out a controlled study of mutation rates in this particular strain of bacteria under identical conditions, or testing the samples for the presence of carbon-14.

They will also need to provide other studies to corroborate this one. A single, disputed study is nowhere near sufficient to overturn vast swathes of high-precision, extensively cross-checked data.

Reproducibility is important in science.

This is important. The history of science is littered with extraordinary claims that fell by the wayside because they could not be replicated. One particularly well-known example of spurious results was Fleischmann & Pons’s 1988 claim of having discovered cold nuclear fusion. Other teams were unable to replicate the experiment, and it is now regarded as spurious. A more recent example was the 2011 claim of the OPERA experiment to have discovered faster-than-light neutrinos: this turned out to be a hard to track down experimental error.

This being the case, findings, especially extraordinary ones such as this one, need to be replicated by multiple teams before they can be accepted as evidence. For what it’s worth, this is the same as the Biblical principle that “every matter must be established on the evidence of two or three witnesses.” (Deuteronomy 17:6; Deuteronomy 19:15; Matthew 18:16; 2 Corinthians 13:1.) If we could cite a single disputed study as evidence for anything, we would also be opening the doors to stuff and nonsense like homeopathy, and our hospital emergency departments would look like this:

The fact remains that there is simply not enough data to establish that the bacteria concerned really were the same age as the deposits in which they were found. This claim is yet another example of YECs drawing extraordinarily wide-ranging conclusions from extraordinarily slender evidence. It simply doesn’t work.

Featured image credit: pixabay.com

Advertisements

No, the distant starlight problem is not the same as the horizon problem

The featured image in the header of this post shows two colliding galaxies, formally known as NGC 4676, or more colloquially, as “The Mice.” It comes from NASA’s Astronomy Picture of the Day website.

They are about 300 million light years away, more than 100,000 light years across, and colliding with each other at somewhere in the region of 200 miles a second. The “tail” of the right hand galaxy is the result of tidal forces stretching it out across vast reaches of space as it collides with its partner. A quick back-of-the-envelope calculation tells you that that tail must have taken at least one hundred million years to get spread out like that.

Distant starlight is a massive problem for the young-earth timescale. Not only must light have taken billions of years to reach us from distant galaxies, but when it arrives, it shows clear evidence of processes that must have been going on for millions of years already. Astronomy PhD student Casper Hesp has a series of posts on the BioLogos website where he examines the evidence from distant starlight in considerable detail. Another example that he cites is relativistic galactic jets.

Artist’s impression of the Milky Way. The red circle represents a distance of six thousand light years from the sun; everything outside of it is the distant starlight problem.
Image source: Wikimedia Commons

On a related note, the oft heard YEC claim that galactic spiral arms could not persist for billions of years is not true. Spiral arms have been well understood since the 1960s to be waves of high densities of stars within a galaxy: a theory that has been confirmed by computer simulations showing them to be extremely stable. The Wikipedia article on density wave theory has some animations showing clearly how it works.

How can we see distant starlight in a young universe?

If you read the attempts by YEC organisations such as Answers in Genesis, the Institute for Creation Research, Creation Ministries International and others to address the problem of distant starlight, you’ll find that they all claim that standard Big Bang cosmology has exactly the same problem:

It’s interesting to note that big bangers have exactly the same problem. That is, the background radiation temperature is almost uniform, to one part in 100,000, at about 2.725 K, even when we look in the opposite directions of the cosmos. Since the big bang would predict hugely different temperatures, how did they become so even? Only if energy was transferred from hot parts to cold parts. However, there hasn’t been nearly enough time for this to occur even in the assumed time since the alleged big bang—see the instructive article Light-travel time: a problem for the big bang by Ph.D. astrophysicist Jason Lisle.

There are just two problems with this argument.

  1. It doesn’t answer the question.
  2. It isn’t true.

This problem, also known as the horizon problem, is indeed a real one. But it isn’t even remotely similar to the distant starlight problem. The only thing that the two have in common is the problem of light travel time. Beyond that, the differences are so massive that to call them “exactly the same” is absurd.

The first, most obvious difference is scale. The horizon problem concerns distances of billions of light years: the size of the visible universe. The distant starlight problem, on the other hand, concerns distances of just six thousand: a fraction of the size of our galaxy. That is a difference of six orders of magnitude. It is the difference in size between a mountain and a molehill.

The two involve completely different eras of cosmic history, and completely different laws of physics. The horizon problem only concerns the first 0.002% of the age of the universe (300,000 years). The distant starlight problem concerns the entire history of the universe almost right up to the present day. The horizon problem operates at scales where the laws of physics are not fully understood, and that lie at the very limits of what we can explore experimentally and what we can theorise about. By contrast, the distant starlight problem concerns laws of physics that are well established, far more readily accessible to astronomers, well within the capabilities of modern measurement, and mathematically far more straightforward.

Omphalos, oh omphalos

The other big difference between the distant starlight problem and the horizon problem is that the distant starlight problem requires the creation of evidence for a history of events that never happened. The horizon problem does not.

Young-earth astronomers have made several different attempts to solve the distant starlight problem. These include Barry Setterfield’s c-decay; Jason Lisle’s anisotropic synchrony convention; and Russell Humphreys’ white hole cosmology. All of these make predictions that are not observed in nature; some of them descend into absurdity; and none of them can account for features of the cosmos that show evidence of a lengthy history, such as galactic collisions and relativistic jets.

Now the horizon problem does have a possible solution in cosmic inflation, which proposes that in the first 10-32 seconds after the Big Bang, the universe went through a period of dramatic expansion. To be sure, inflation is a bit of a mind-bender, and it does sound a bit whacked out, but it is a valid solution to the Einstein field equations, and many (but not all) cosmologists believe it to be the correct one.

But even if inflation turns out to be wrong, the universe can be explained in terms of the Big Bang being very finely tuned. Many scientists find that a bit of an ad hoc explanation, but there’s nothing theologically objectionable about it, and in fact, it would be compelling evidence for design. But there is no false history involved, the universe remains the same age as it appears to be, and the integrity of the Creator is upheld.

screen-shot-2016-10-25-at-104428-am
A comparison between the distant starlight problem and the horizon problem.
Source: Casper Hesp, BioLogos.

The fact remains that, far from being the same as the horizon problem, the distant starlight problem is in a completely different league altogether. To claim that the two are the same, when they are separated by six orders of magnitude, is patently absurd. It simply doesn’t make sense.

YEC Best Evidence 9: not enough salt in the sea, or not enough precision in the measurements?

One of the things that’s taken me somewhat by surprise in researching for this blog is the precision that modern radiometric dating methods can achieve. To give just one example, a few years ago researchers at Glasgow University pinned down the date of the K/T impact event, which killed off the dinosaurs, to within just eleven thousand years of 66,038,000 years ago. That’s an accuracy of just one part in six thousand — far tighter than I ever expected!

For what it’s worth, tolerances as tight as these completely falsify the oft-heard young-earth claim that radiometric dating is “guessing at best,” or that long ages are merely a presupposition to try and make space for evolution to happen. It is simply not possible to get results that specific out of vague and non-specific starting points, and “evolutionary presuppositions” are about as vague and non-specific as you can get.

Such high precision results are also the exact opposite of what we see in claims of evidence for a young earth. YEC arguments routinely rely on extremely low precision measurements with huge error bars, poorly known quantities, and rates that nobody expects to have been the same in the past as they are today.

The ninth entry on Answers in Genesis’s top ten list is a textbook example of this. It is the claim that there is not enough salt in the sea for an old earth. This argument says that if you tot up what goes in and what goes out, and divide how much is already there by the difference, you get an upper limit, and the earth (or at least, the oceans) can’t be any older than that.

Exactly what limit does this place on the age of the earth?

Everyone who cites this argument seems to have different ideas about what that upper limit actually is. Many rank-and-file YECs think it’s just a few thousand — I’ve had one person quote me 6,000 and another person quote me 100,000. This person quotes Kent Hovind as thinking it is 5,000 years.

For the most part, if you’re quoting figures this low, you probably just saw this argument on your Facebook feed, shared it without clicking through to read it, and blindly assumed that it must have been somewhere in the region of six thousand years or only slightly more. In actual fact, Andrew Snelling, Answers in Genesis’s geologist-in-chief, who wrote the article in the first place, gives a figure of 42 million years, citing the 1990 paper The Sea’s Missing Salt: A Dilemma for Evolutionists by YEC scientists Steve Austin and Russell Humphreys.

As evidence for a young earth, that is a joke. 42 million years may differ from the modern scientific consensus on the age of the earth by a factor of a hundred, but it also differs from the YEC timescale by a factor of seven thousand. If we are to concede that this falsifies the scientific consensus on the age of the earth, we must also insist that it falsifies the young-earth timescale seventy times more forcefully.

But does it falsify the scientific consensus anyway? In order to answer this question, we must address a question that we need to ask of all young-earth claims.

How large are the error bars?

There’s a deep and fundamental problem with trying to use the amount of salt in the sea to estimate the age of the earth. We are dealing with quantities that are extremely difficult to pin down, highly sensitive to changing climatic and environmental conditions, and as such can not be realistically assumed to have been the same in the past as they are today. There are a lot of different inputs and outputs, some of them not fully understood or completely quantified even today, and it’s very easy to overlook some of them. Even measuring the known quantities is a gargantuan task, requiring massive multi-national surveys over long periods of time. Enormous error bars are par for the course. The values involved change constantly as new and more detailed surveys are undertaken.

Nevertheless, the amount of salt in the sea was actually the basis for some of the earliest attempts to estimate the age of the earth in the pre-radiometric era. The first people to try and come up with a figure were Edmund Halley (1715) and John Joly (1899). Joly’s figure was 90 million years (Hay et al, 2006). More recently, Daniel Livingstone (1963) used data by Clarke (1924) to come up with an estimate of a few hundred million years, with a very wide margin of error that could extend as high as 2.5 billion years. The modern scientific consensus, taking all known data into account, now considers that long-term rates of influx and egress are equal within error bars, and consequently the amount of salt in the sea tells us nothing whatsoever about the age of the earth. (Holland, 2006.)

It is error bars, not evolutionary presuppositions, that have caused scientists to abandon the salt chronometer in favour of radiometric techniques. No matter what your worldview, it is outright ridiculous to reject high-precision results, accurate to one part in 6,000, in favour of low-precision methods whose errors can be ±50% or more.

Of course, Snelling describes all this as a “rescuing device,” and says that “even the most generous estimates” give an upper limit of just 62 million years. However, besides not taking everything into account, Austin & Humphreys’ paper was based on outdated data.

A case in point: halite deposits

One of the most important processes by which salt is removed from the oceans is by evaporation. This leaves behind massive deposits of halite, which can be found in numerous places all over the world. One of the largest halite deposits is found underneath the Mediterranean. There is a lot of evidence that from 5.96 to 5.33 million years ago, the Strait of Gibraltar repeatedly closed off, causing the Mediterranean to dry out and depositing vast quantities of ocean salt on the sea floor. This period, the Messinian Salinity Crisis, came to an end 5.33 million years ago when the Strait of Gibraltar was finally breached one last time and the Mediterranean rapidly re-filled in an event called the Zanclean Flood. A speculative future recurrence of this flood was portrayed in the award-winning xkcd webcomic episode, “Time,” a few years back.

xkcd1190-2907
The sea is rising! The protagonists are shown the future shoreline of the rapidly refilling Mediterranean. From xkcd “Time”.

This episode would have removed a substantial fraction of the salt in the oceans. So too would other large halite deposits that can be found all over the world, in places such as the Dead Sea and the Gulf of Mexico. Austin and Humphreys cite a figure of 4.4×1018 kg for the worldwide inventory of halite, citing Holland (1984). They claim that it is “extremely unlikely” (page 8) that this contains a significant error, as “No major quantity of halite in the earth’s crust could have escaped our detection.”

In actual fact, it turns out that major quantities of halite deposits had escaped our detection, and in the past three decades, many more such deposits have been found in the course of oil exploration. Hay et al (2006) give much more up to date estimates of between 19.6×1018 kg and 35.2×1018 kg, or between four and a half and eight times greater than that cited by Austin and Humphreys. Furthermore, they state that these figures are most likely incomplete (they include no data from Antarctica, for example), and that further exploration and surveys in the future may well push the figure up even higher. They also conclude that, far from increasing, the amount of salt in the oceans has actually decreased since Precambrian times.

Austin and Humphreys’ paper also overlooks several other factors. Glenn Morton points out for example that there are no less than sixteen different known mechanisms for sodium removal that they omitted to take into account, such as plankton concentrating sodium in their bodies, which is then removed from the oceans as sediment when they die.

The 2006 paper by Hay et al (full content here) is a good, comprehensive scholarly survey of the amount of salt in the oceans. It is a lengthy read, but it is easy to understand and it presents a lot of useful data. However, it should be abundantly clear that, once modern, up to date figures are used and all known vectors are taken into account, and given the huge uncertainties and variabilities in the values concerned, the amount of salt in the oceans tells us nothing whatsoever about the age of the earth.

Featured image credit: Gail Hampshire (via Wikimedia Commons)

There is hope

Therefore, since we are surrounded by such a great cloud of witnesses, let us throw off everything that hinders and the sin that so easily entangles. And let us run with perseverance the race marked out for us, fixing our eyes on Jesus, the pioneer and perfecter of faith. For the joy set before him he endured the cross, scorning its shame, and sat down at the right hand of the throne of God. Consider him who endured such opposition from sinners, so that you will not grow weary and lose heart.
— Hebrews 12:1-3

It is all too easy, on seeing the excruciating falsehoods and absurdities presented by young-earth “creation science” and the extreme dogmatism with which they are taught, to grow weary and lose heart, or to even allow them to colour your perception of Christianity in general.

Don’t.

If you’re asking the question, “What else are they lying to me about?” about your pastor or church leaders, rest assured that the answer is, not as much as you think. Many rank and file YECs aren’t intentionally out to deceive: they just lack the skills and training necessary to fact-check the claims that they repeat. You’re only lying if you know — or should reasonably be expected to know — that the claims you are making are untrue.

Furthermore, there are well-informed Christian apologists and ministries out there who are much more honest in their approach to matters such as this. Ministries such as Ravi Zacharias International Ministries, William Lane Craig and Reasonable Faith, and Hugh Ross and Reasons to Believe, provide a lot of valuable resources here. Remember too that the world has agendas of its own. There are plenty of dishonest arguments and ideological biases coming from “the other side” as well. Don’t throw out the baby with the bathwater.

It’s also important to realise that while the scientific evidence takes a firm and insistent line on the age of the earth, the Bible does not. I know that some people think that 2 Peter 3:8 and Psalm 90:4 are not relevant here, or that they are not a lot to go on, but at least in terms of long ages, the Bible gives us something. No matter how old the earth is, or who did or didn’t evolve from what, God’s Word still stands. Of that, I am confident.

Don’t forget what it’s all about.

Christmas is a joyful time. The carols that we sing present a joyful message: your Saviour, Christ the Lord, has come. We send each other cards and gifts to celebrate His birth, a little over two thousand years ago.

For that is what it’s all about. As I’ve said before, we are Christians, not Adam-and-his-pet-dinosaurians. We celebrate the completed work of Christ on the Cross, not the incomplete work of Adam and Eve on the back of their pet T-Rex. Don’t buy into any of this “no six days means no Gospel” nonsense. It’s a denial of the Gospel, it’s what Richard Dawkins wants you to believe, and it is factually untrue.

Remember Colossians 2:8:

See to it that no one takes you captive through hollow and deceptive philosophy, which depends on human tradition and the elemental spiritual forces of this world, rather than on Christ.

Note in particular how it ends with, “…rather than on Christ.” Not with “…rather than on a young earth and dinosaurs on the Ark.” Hold fast to that which is good.

May God bless you all richly this Christmas.

Image credit: pixabay.com

YEC Best Evidence 8: short-lived comets

The eighth of AIG’s ten best evidences for a young earth is the existence of short-lived comets. Comets lose mass as they come close to the sun, when ice evaporates and dust gets dislodged by the solar wind.

The author of this article, Danny Faulkner, tells us that this means we can easily calculate an upper limit to their ages. He doesn’t tell us how to do so, nor does he quote any figures for this, but he is correct in saying that comets don’t last all that long once they venture into the inner Solar System. For example, Halley’s Comet is only expected to last another 10,000 years or so.

He tells us that comets can get kicked out of the Solar System by gravitational interactions with the planets. What he doesn’t tell us is that comets can also get captured into the Solar System by exactly the same interactions. In fact, NASA ran some calculations on Comet 67P/Churyumov–Gerasimenko (the one that was visited by Rosetta and the Philae lander) and found that this was exactly what happened to it:

Analysis of the comet’s orbital evolution indicates that until the mid-19th century, the closest it got to the Sun was 4.0 AU (about 373 million miles or 600 million kilometers), which is roughly two-thirds of the way from Mars’ orbit to that of Jupiter. That far from the Sun’s heat, it would not sprout a coma or tails, so it was invisible from Earth.

But scientists calculate that in 1840, a fairly close encounter with Jupiter must have sent the comet flying deeper into the inner solar system, down to about 3.0 AU (about 280 million miles or 450 million kilometers) from the Sun. Churyumov-Gerasimenko’s perihelion (closest approach to the Sun) drifted a bit closer to the Sun over the next century, and then Jupiter gave the comet another gravitational kick in 1959. Since then, the comet’s perihelion has stood at about 1.3 AU, which is about 27 million miles (43 million kilometers) outside Earth’s orbit.

So it’s not uncommon for long-lived comets to come in from the outer Solar System, be captured by one of the planets, and turned into short-lived comets. The Oort Cloud and the Kuiper Belt are believed to be the source of these long-period comets in the first place.

Yet Faulkner dismisses the Oort Cloud as a “rescuing device” concocted by “evolutionary astronomers” (as a reminder: there’s no such thing as an “evolutionary astronomer”) saying that “there is no evidence for the supposed Oort cloud, and there likely never will be.” He also cites the large size of the Kuiper Belt asteroids and their composition as evidence that the Kuiper Belt can not be the source of these comets.

“Pluto’s surface is young!”

In July 2015, the New Horizons space probe flew past Pluto and sent back pictures of a surface that looked much more pristine and smooth than what we see on other rocky planets and moons in the Solar System. The same author, Danny Faulkner, wrote this in response to these pictures:

Compounding this problem for a 4.5-billion-year age for the solar system is the fact that Pluto is located in a particularly crowded part of the solar system. Pluto orbits the sun in a region with many other large objects that are too small to be planets and are also orbiting the sun. Presumably, thus far we have found only the larger members of this second asteroid belt, the first belt being mainly between the orbits of the planets Mars and Jupiter. We would expect that for each of these bodies in this second asteroid belt there would be many more much smaller bodies. Therefore, Pluto ought to be undergoing impacts today at a higher rate than most other objects in other portions of the solar system.

So … on the one hand, we are told that the Kuiper Belt is too sparsely populated to be able to provide a steady supply of short-term comets, yet on the other hand, we are being told that it is so densely populated that it must be pulverising Pluto’s surface to smithereens! Which is it?

Absence of evidence is not evidence of absence.

The reason for the large size of known KBOs is that the smaller ones are too hard to see, not that they don’t exist. Comet-sized objects (about 10-20 km) in particular are right at the resolution limits of our telescopes at that distance, and looking for them is complex and expensive. We also know that objects in both the asteroid belt (Gladman et al, 2009) and the Kuiper Belt (Fraser & Kavelaars, 2008) follow a power law distribution in terms of size, with smaller ones being far, far more common than larger ones. Consequently, Faulkner’s claim that the Kuiper Belt objects are all too large and sparse to account for the origin of comets is completely out of touch with reality.

After New Horizons visited Pluto, NASA conducted a search for Kuiper Belt objects for it to visit as a follow-up. The potential candidates had to meet some fairly stringent criteria in terms of size, distance and position in the sky, and the fact that they managed to find three possible targets is further evidence that there is no shortage of material in the Kuiper Belt.

Ultimately, this argument boils down to “absence of evidence is evidence of absence.” While absence of evidence may be evidence of absence if it is something that we expect to see, such as sequenceable DNA in 6,000 year old dinosaur fossils, it most certainly is not evidence of absence if it is something that we don’t, such as small objects beyond the resolution limits of our telescopes.

The Oort Cloud may not have been directly observed, but it has good explanatory power, and its existence has never been falsified. Besides, which is more plausible an explanation — that the Oort Cloud exists, or that hundreds of thousands of independent, high-precision measurements whose commercial incentives strongly favour correctness over ideological convenience are all consistently in error by up to a factor of a million?

Featured image: Comet Hyakutake, by Bill Ingalls/NASA (source: Wikipedia)