The Talk.Origins Archive Post of the Month: March 2000

Subject:    Re: Fossils of Tiny Primates Found
Newsgroups: talk.origins
Date:       March 17, 2000
Message-ID: 8atuuq$4rm$1@darwin.ediacara.org

In article <38D2660C.815669C6@pioneer-net.com> Dan Smith <dan@pioneer-net.com> writes:
|Andrew MacRae wrote:
|> In article <38D20737.4594BB52@pioneer-net.com> Dan Smith
|> <dan@pioneer-net.com> writes:
|> |Larry Handlin wrote:
|> |> On 16 Mar 2000 11:21:20 -0500, Dan Smith <dan@pioneer-net.com> wrote:
|> |> >So, how do they know that these primates are 45 million years old?
|> |> >Well, let me quote from the Scholarly Paper itself (available at
|> |> >http://www.niu.edu/pubaffairs/RELEASES/2000/MAR/primate/):
|> |> >
|> |> >"The vertebrate faunas from these fissures have been estimated to
|> |> >date to approximately 45 Ma on purely biostratigraphic evidence"
|> |> >
|> |> >"purely biostratigraphic evidence" means they used the other fossils
|> |> >found in this area to date the strata these primates were found in.
|> |>
|> |> Dan, do you care to actually discuss the process? You have been shown
|> |> that this isn't circular so what's your point anyway?
|> |
|> |I've been "shown" it isn't circular - you and others said it wasn't -
|>
|> I recommended the following web page:
|>
|> http://www.talkorigins.org/faqs/dating.html
|>
|> Which addresses the "circularity" issue a bit, and cites some
|> papers that address it more specifically.
|
|I read the faq and probably need to read it again, but my main problem
|with the entire index fossil dating method is that it leaves little room
|for fossils to be found outside their assigned "age".

In what way? If a particular fossil is thought to occur, say, only in the Late Cretaceous, and it is subsequently found to extend even lower, into intervals replete with many other fossils of Early Cretaceous age, or simply lower in the stratigraphy than previously sampled, then why wouldn't its stratigraphic range be revised accordingly? Do you need specific examples?

Fair enough, but this is only a beginning, and most of these accounts will deal with individual examples only superficially. That is true even for the FAQ I wrote, which only scratches the surface. That is why I recommended a bunch of other references, although I recognize that these things take considerable time to evaluate and you have limited time.

|I'm not claiming to be an expert by any means.
|There's quite a diversity of opinion out there and I'm not convinced
|that anyone can accurately date fossils.

There are always questions with regards to how certain an age interpretation can be made from a particular suite of fossils. Always. However, it is a question of resolution and the degree to which the distribution of particular fossils has been tested, both by the distribution of other fossils, and by completely independent means. If you want to know whether something comes from the Micula mura nanofossil zone, it can be much trickier to correlate reliably than determining whether a rock correlates to anytime in the Cretaceous Period, or the entire Mesozoic. A paleontologist may be uncertain whether a fossil is from the Lutetian or Bartonian (two subdivisions of the mid-Eocene), but they would probably be more certain it is from the mid-Eocene, from the Eocene at all, and even more certain that it is from the Paleogene, the early part of the Cenozoic. Usually the precision of the age estimate reflects the uncertainty, but in popular accounts, unfortunately, most of that gets glossed over. This problem is not peculiar to paleontology or science in general in terms of how the details of an issue are presented by the popular media.

Like any interpretation, there can be mistakes, contradictory evidence, or simple imprecision, but for the most part, paleontologists have been studying the distribution of fossils for over 200 years, and there are some aspects of the pattern to their succession that are very well established. It is highly unlikely that the fossil occurrence we are talking about actually comes from, say, the Cretaceous, although stranger things have happened, and fissure fills are much more difficult to date than many other geological settings (I can go into the details of why, if you like).

|> |It's still "Rocks date fossils, fossils date rocks".
|>
|> Could you be more specific about how that applies here, or
|> exactly what is wrong with this principle? In this case, rocks in
|> Europe, found intercalated with ones higher and lower in the
|> stratigraphy, have a characteristic succession of fossils. When the
|> rocks and their fossil content were subdivided into discrete intervals,
|> this particular interval was assigned to the epoch called the Eocene.
..[some of the fossils at the new locality are the same]
|> What, exactly, is wrong with this procedure and the hypothesis
|> that (when applied to biostratigraphy) these rocks correlate to rocks
|> of similar age elsewhere, based upon the fossil content?
|
|It's the "characteristic succession of fossils" - which assumes
|evolution - that I object to.

No, it does not at all. The fossils are just there, observed in a succession of rocks in Europe, in the interval that came to be called the Eocene. If what you say were correct, then people would not have been able to do biostratigraphy before evolution was proposed, would they? And yet, they were doing it for many decades prior, all the way back into the late 1700s.

Evolution explains why biostratigraphy works so well. It is not an assumption necessary in order to do it (see below for the Coca-Cola cans example).

|It is assumed - since fossils were found
|in Europe in a particular order and then assigned dates, that that order
|will be consistent throughout the world.

No, that was not "assumed", it was a property of rock and fossil successions that was discovered. In fact, there are plenty of examples where such an assumption would simply be wrong for certain fossils anyway, so it would be an unreliable assumption. The hypothesis that some fossils are found in consistent successions world-wide is a testable one, by simply looking at the succession of fossils seen in different areas. Are they truly consistent? Sometimes, individual fossil species are not. They may not occur in another region at all, or they may occur somewhat earlier or later with respect to the majority of other fossils in the succession. Most importantly, there are sometimes special event beds or indications of other events that have unique characteristics, and which can be found world-wide in a similar succession. This demonstrates that no matter what imperfections exist, the gross pattern of fossil succession can not be some kind of circularly-derived artifact.

Even many "young Earth global flood" creationists accept fossil succession, which makes sense, given that the broad pattern was worked out by creationist geologists in the late 18th and early 19th centuries. For a more modern example, didn't you recently cite Bernard Northrup's "The Genesis of Geology" article? His claims depend greatly upon the reality of fossil succession.

It is a vanishingly tiny minority that would dispute such a basic and well-tested interpretation.

|Those who accept it cite corroborating evidences,
|those who don't cite anomalies.

Like what? Most of the "anomalies" I have ever seen such people cite are inadequately documented or have clear evidence for the reason for the occurrence of fossils in a seemingly inconsistent position. For example, people who cite fossils that occur in association with "thrust faults" as an anomaly are ignoring the need for fossil succession to be determined in undisturbed rock successions, or for any structural discontinuities to be recognized first. They then trivialize evidence for the existence of thrust faults by claiming that the presence of the thrust faults is only proposed to account for the fossils, which is ridiculous, because thrust faults can be recognized even without any fossils at all.

There are innumerable other examples, but most neglect the fact that paleontologists expect certain kinds of "anomalies" to show up. For example, it is expected that, just like any other kind of rock, fossils can be eroded out of older strata and redeposited in younger sediments, in a process known as reworking. You can see such a process yourself if you visit a modern fossil locality and see old, eroded-out fossils rolling on the sediments of a modern beach or river where recent sediments are accumulating too. Is this a problem for conventional geologists to detect, though? Not usually, because the fossils have all sorts of evidence for the reworking process (e.g., they are often abraded, or are infilled with sedimentary rock different from the surrounding, younger sedimens, or they might be mineralized differently). This evidence, and the ability to detect examples of reworking, is commonly neglected by people claiming these occurrences are genuine "anomalies".

|> There often are
|> other ways to establish correlations anyway, and therefore
|> independently test the biostratigraphic correlation hypothesis.
|
|What other methods? And how often are they employed?

I mention one in the FAQ -- the detection of a geochemically distinctive iridium concentration at the boundary between the Cretaceous Period and Tertiary, and the occurrence of impact spherules, shocked quartz, and at a few localities, tsunami deposits. This sort of "event bed" is very distinctive, and is found in some form world-wide at many localities close to the point where, in terms of the fossils, dinosaurs and ammonites and a host of other animals and plants become extinct. Coincidence? Maybe, but if it were, there would be an awful lot of coincidences in terms of the succession of biological and physical events to account for. There are also numerous locations where geochemically distinctive ash beds occur. Each ash bed has a geochemical "fingerprint" that is unique to a particular volcanic eruption. This allows them to be be traced laterally over large distances in order to test hypotheses about how the rock units, and the fauna or flora contained within them, correlate in detail. For example, take a look at this paper:

Mitchell, C.E., Goldman, D., Delano, J.W., Samson, S.D., and Bergstrom, S.M., 1994. Temporal and spatial distribution of biozones and facies relative to geochemically correlated K-bentonites in the Middle Ordovician Taconic foredeep. Geology, v.22, p.715-718.

There are also other techniques, such as paleomagnetic reversals, astronomical cycles, and, of course, radiometric dating, that can provide independent tests of a biostratigraphic correlation. How often are they used? Probably not as often as fossils, but often enough that there can be little doubt that fossil succession is not some kind of made-up artifact without independent constraints. Here are a few papers that employ both biostratigraphic and non-biostratigraphic correlation techniques:

Renne, P.R., WoldeGabriel, G., Hart, W.K., Heiken, G. and White, T.D., 1999. Chronostratigraphy of the Miocene-Pliocene Sagantole Formation, Middle Awash Valley, Afar rift, Ethiopia. Geological Society of America Bulletin, v.111, no.6, p.869-885.

Hilgen, F.J., Krijgsman, W., Langereis, C.G. and Lourens, L.J., 1997. Breakthrough made in dating of the geological record. EOS, Transactions, American Geophysical Union, v.78, no.28, p.285-.

Elder, W.P., 1988. Geometry of Upper Cretaceous bentonite beds: implications about volcanic source areas and paleowind patterns, western interior, United States. Geology, v.16, p.835-838.

Garces, M., Krijgsman, W. and Agustf, J., 1998. Chronology of the late Turolian deposits of the Fortuna basin (SE Spain): implications for the Messinian evolution of the eastern Betics. Earth and Planetary Science Letters, v.163, p.69-81.

Calanchi, N.; Cattaneo, A.; Dinelli, E.; Gasparotto, G. and Lucchini, F., 1998. Tephra layers in Late Quaternary sediments of the central Adriatic Sea. Marine Geology, v.191-209.

No, it does not. The situation is the opposite -- the theory of evolution was derived, in part, from fossil succession. Evolution only explains why it turns out each species origination and extinction is unique and successive. It is not a requirement in order to do biostratigraphy, and even if the theory of evolution were negated tomorrow, the succession of fossils would still be sitting there, waiting to be explained by some other theory.

|It is assumed that evolving species only existed in a certain "window" -
|not before - not after.

|So once a date is determined that "fits", I
|doubt if much further serious inquiry goes into it.

You would be incorrect. The problem is, after 200 years of sampling, most of the changes that happen are at finer resolution than anyone but specialists pay attention to -- at the scale of individual zones or stages that finely subdivide the more commonly-known geological periods. But even the boundaries of those periods and the distribution of fossils with respect to them do get refined by new data. Sometimes it is found that ranges are not what paleontologists thought. For example, it was long thought that inoceramid clams became extinct right at the end of the Cretaceous Period. It turns out this was probably due to sampling limitations. Based upon much more intensive sampling of the final stages of the Cretaceous Period, it is now thought that they became extinct somewhat before the end, to a high statistical confidence.

That last bit about statistics is something I have not mentioned, but it deserves some elaboration -- all these stratigraphic ranges can have confidence intervals placed upon them that reflect the degree to which they have been sampled. This correspondingly influences the certainty of age assignments from an assemblage of fossils, although, again, this is something that specialists deal with most of the time.

|> |Or in this case "fossils date fossils".
|>
|> Fossils whose distributions have been extensively sampled and
|> studied around the world date different fossils whose distribution is
|> unknown. Yes. What is wrong with that procedure? They are not the
|> same fossils.
|
|That's exactly what is wrong with that procedure - they are not the same
|fossils. How do we know that this particular group of fossils is not
|from an entirely different era than the European one? The fossils?

The other fossils -- the other ones found in association with what are interpreted to be tiny primates, or in rocks immediately above and below. Those ones probably are the same, and have been previously seen elsewhere.

|> If people conduct an archaeological dig in a garbage dump, they
|> could hypothetically get a fairly good idea of relative age based upon
|> the distribution of the different types of Coca-Cola cans through the
|> stratigraphy.
|
|Big difference. We know for FACT the exact dates when Coca-Cola changed
|their can styles.

Yes. I chose the example so that whether they had changed through time at all was not an issue.

|We don't know whether evolution really produced any of
|these species at a certain time or not.

No. Because you neglect the value of sampling what is preserved there, even without separate knowledge of what actually transpired through time. If you assume that humanity became extinct, and all written records of the history of the Coca-Cola company and their can development were destroyed, would an alien who came to visit the Earth be able to figure out the succession of cans from what is preserved in the garbage dumps of the world? You bet. I doubt it would even be difficult, even if it would be rather messy to investigate :-) It would not be as precise as if there was a written historical account, and many of the same problems as exist for real biostratigraphy would exist in this hypothetical example too, but that would not diminish the ability to, at a useful level of detail, do garbagostratigraphy if enough dumps were sampled and the consistent succession worked out. The change from pull-tab to single-unit cans would be a pretty distinctive zone, for example. The interpreters would also have the ability to test out their hypothesis with respect to other event beds within the stratigraphy, such as the radioactive and geochemically unique layer that might be near the top of all the garbage dumps, representing some kind of global catastrophe that coincidentally matched the cessation of all garbage deposition, and resumption of normal sedimentation. This would be preceded by a much lower level, also seen world-wide, below which certain non-natural isotopes no longer occurred, corresponding to sometime (in our calendar) dating to the 1940s or so when open air nuclear testing began.

You must be joking. Any paleontologist would love to show that the supposed "Eocene" fossils actually dated from the Miocene, and therefore could not have anything directly to do with the origin of primates. The would be even more interested in showing that the fossil succession at the type (original) areas in Europe is inaccurate. For example, I do not know if it has been tried yet, but it would be interesting to take some samples and look for pollen and spores in the fissure infillings, to see if that fossil assemblage was still consistent with an Eocene interpretation. Pollen and spores have some fairly distinctive events in the Eocene (e.g., grass pollen first shows up in that interval).

If it is thought that a fossil locality with a new fossil is especially important, it attracts paleontologists and geochronologists of other disciplines like flies, in order to better constrain the age. People do not make an initial interpretation, and then it sticks forever, without anyone questioning it, especially if other methods can be employed, as is usually the case.

Garbagostratigraphy

Post of the Month: March 2000