Why Do Gaps Exist ( Or Seem To Exist ) ?

[DE&i]

I hear more often than not the phrase " new fossil fills the gap " is there a limit for this scenario to keep repeating it self.

Why are there still gaps? And why do many people think that there are even more gaps than there really are?

I find this question intriguing but confusing.

Darren.

[Xiphactinus]

Usually you hear that from people outside the paleo/scientific community, most often in the mass media.

People are too used to seeing video representations of a fish morphing into an amphibian morphing into a lizard...and people expect there to be a fossil representation of every miniscule step in between. We know that there's no way all those intermediate "steps" will be found in fossil form, so there is a "gap". When a new fossil animal is found that has attributes in between known fossils, it is said to "fill a gap". But, when you come right down to it, adding that new fossil just created two new "gaps".

So, no, there is no limit to this scenario repeating itself.

[painshill]

And there are still quite a number of presumptions about the lineages of extant organisms and their antecessors. Sometimes a new fossil helps confirm those presumptions (and sometimes it doesn't).

[Auspex]

The idea that the fossil record is hopelessly incomplete goes directly back to Darwin's original theory (to excuse the lack of intermediate forms of species). To some degree, he was right, but the hypothesis of Punctuated Equilibrium, advanced by Eldredge and Gould, has presented another way of interpreting what appears to be a gappy record by suggesting that evolution happens in bursts, with long intervals of stasis. New fossils are found that fit between those already known, of course, but the whole (deeply ingrained) concept of the 'missing link' is thus fundamentally flawed.

[Missourian]

Significant pauses in sedimentation and erosion of deposits didn't help either.

[DE&i]

I'm still waiting for a breaking news moment where a Tetrapod has been found from the Aalenian , mid jurrasic has this happened yet.

Darren

[RichW9090]

Gaps are wonderful things. Without them, we could never define a fossil species of animal.

Theoretically, if you take a modern species of animal or plant, and think about its most direct direct ancestor - that is, its literal father and mother, how much difference is there between those two generations? They aren't identical, obviously. You have brown eyes, your mother has green eyes. Is that enough of a difference to draw a line between you and your parents and say that you are species A and your parents are species B? Clearly this is not the case. So let's look at your grandparents - the direct ancestors of your parents. They are different from your parents, right? But enough to say that your parents are species B and your grandparents are species C?

Now carry that backwards for one thousand, or a hundred thousand, or a million years. Then 10 million years (the average longevity of a mammalain species) will there ever be enough difference between any individual and its parents to draw the line between species A and species B? No. If we had an absolutely complete fossil record - that is, a fossil of every individual which ever lived - paleontologists could not draw meaningful lines between species.

But gaps in the fossil record come riding to our rescue. They break what would otherwise be a continuous record of evolutionary process into fortuitous chunks. Enough of the record is missing so that the change between fossil population A and fossil population B is enough to say that they are different species. Those gaps are the only thing that allow us to make comparisons between fossil species (chronospecies) as we do with livng species, which we characterize, for the most part, under the Biological Species Concept.

Edited June 8, 2013 by RichW9090

[DE&i]

Thank you very much for this answer I have lots of questions such as these but can't think for the life of me how to word them in laymans terms you certainly made things a little clearer for me.

p.s. Did you get a chance to watch the Ice Age Giants series in the end.

Darren.

Edited June 8, 2013 by D&E

[FossilDAWG]

Another thing to consider is the conditions under which a new species can arise from a pre-existing one. The biological species concept holds that two species are different if they cannot effectively interbreed. Sometimes you may have a species that has lots of individuals spread over a large area. With a large "effective population size" (number of potentially interbreeding individuals in the population) genetic change (the kind of change needed for speciation) will be very slow, because new mutations or combinations of alleles will just get mixed in with the vastly larger number of genomes in the next generation, pulling everything back to the "average" genotype/phenotype.

However, two things can happen out of this process to result in rapid speciation, or the appearance of rapid speciation. First, our large population will gradually accumulate a lot of genetic variation: lots of slight variations (alleles) for lots of genes. If there is then a change in the environment, such as a climate change or the appearance of a new predator (as only two of many possibilities) our species will have lots of heritable variation for selection to work on. If the selection is strong enough, it may cull most of the genetic variation quickly, leaving only individuals with genotypes/phenotypes best able to cope with the new selection pressure. Remember, natural selection doesn't cause mutations or genetic variability, it selects for favorable genes from amongst the range that existed before the new selection was applied. As one rather artificial example, consider insecticide resistance. Resistance can evolve rapidly, because in any insect population (typically large numbers of individuals) there will be a lot of variation; in any large population there will be a pretty good chance that there will be a few individuals that have genes (alleles of genes, really) that confer resistance to the insecticide, before the insecticide is ever used. When the insecticide starts to be used, it kills all the insects except the few individuals with the resistant genotype, so they are the only ones (or the majority) left alive to reproduce. Natural selection is usually not so dramatic, and so takes longer to shift the average genome. However, in the geological record a change that takes hundreds or rhousands of generations can appear virtually instantaneous.

The second thing that can happen is, starting from a large population of some species, a small group becomes isolated from the rest, so they cannot interbreed with the large population. Think of a population that becomes isolated on an island as sea levels rise, or get cut off from the main population [A] when a river changes course. The small population can change genetically more quickly because there are fewer possible genetic recombinations (possible matings) to swamp out new mutations. Even if there are no environmental differences impacting population B vs population A, genetic drift will ensure that differences between A and B will develop, and B (the smaller population) will diverge more quickly from the ancestoral genotype. If there are environmental differences the divergence will be even more rapid. Also the chances are that the fossil record will reflect A, if population A is large and widespread (more opportunities to get fossilized) and population B is small and confined to a small range (such as an island). If circumstances then change so that A and B are brought back together (e.g. sea level falls, the island is again part of the mainland) then A and B may be different enough that they can coexist (occupy different niches), then the fossil record will show the sudden appearance of B alongside A with no indication of where B came from [no gradual change of A into B]. Or, they may compete. If B is able to outcompete A, the fossil record will show a "jump" from A to B. If the replacement happens very quickly, there may be no recognizable record of the brief period where A and B were both present at the same time.

Don

Edited June 8, 2013 by FossilDAWG

[Boesse]

The issue is primarily one to do with the completeness of the stratigraphic record. If you looked at continental sedimentary rocks, and examined how much time is really "locked up" in those rocks, in most places it's probably only 5% or so of what fossil biostratigraphy or radiometric dates would tell you. Deposition is extremely inconsistent and localized in continental settings. Think about it this way: how many times have you gone out walking in the great outdoors and actually been on an active depositional surface with zero (or minimal) plants? Floodplains, dune fields, etc.? Not often. Every time you're on a floodplain with occasional trees and lots of ground cover, that's the equivalent of a paleosol horizon and several thousand years of non-deposition (the rule of thumb for soil development is that it takes about 10,000 years or so - so every time you see a paleosol horizon in continental sedimentary deposits, that's 10k minimum).

The locus of deposition will change geographically speaking - maybe it will migrate closer to or further from the seashore. Maybe uneven tectonic subsidence will make the depocenter shift in a direction not related to 'base level' (=seashore). Then, during periods of sea level drop, rivers begin to incise more intensely and erode away parts of the record - and during sea level rise, they back up and fill with sediment as base level increases. People have written that the terrestrial record is biased towards periods of sea level rise. On the other hand, sea level rise tends to cause periods of sediment starvation on the continental shelf (that sediment that is deposited in the rivers would normally have made it out on the shelf). When sea level begins to fall again, river transported sediment returns to the shelf and marine sedimentation resumes. The most complete record is in the deep sea, where there is a mostly continuous "rain" of mud-size particles. Deep sea cores from the Deep Sea Drilling Program are studied intensely for their complete record of microfossils, which are cross-correlated with shallow marine rocks on the continent that have radiometric dates, and then correlated further with the Atlantic spreading ridge paleomagnetic "stripes".

One thing that my strat prof in my master's program hammered into our heads is that although erosional time-rich surfaces are present in the rock record, every erosional surface correlates to a body of sediment somewhere else. You can trace erosional sequence boundaries into submarine fan deposits, where sediment was being transported and bypassed the entire continent and shelf (typically during periods of super low sea level and fluvial incision - periods such as the Holocene, for example). On the other hand, sequence boundaries in marine rocks can be traced proximally to packages of sediment on the continent. So although a vertical stack of rocks at any one point will have "gaps", examining the rock record in three dimensions shows that there really aren't gaps, in the sense that that period of time is not represented by rocks anywhere.

As popular as punctuated equilibrium was and as dogmatic as some of its supporters were/are, there are plenty of fossil records of lineages showing somewhat gradual non-branching evolution through time (the Carcharocles and Carcharodon lineages, and my favorite example, the northern fur seal Callorhinus gilmorei-sp.-ursinus lineage). Certainly the rate of evolution does change through time, and there probably are periods of stasis and rapid speciation, but the Gould and Eldredge idea is too dogged by the pattern being explained so well as a stratigraphic artifact for me to take 100% seriously.

[Auspex]

I think Gould realized that PE could have been disqualified with just one proven exception, so he couched everything in ways that did not make it a mutually exclusive, either/or, proposition. Some of his most ardent acolytes missed that part...

[RichW9090]

And the idea that evolution takes place at different rates was not new with Gould and Eldredge. It was clearly stated by Simpson years earlier in his book Tempo and Mode in Evolution, where he coined the term bradyteley for slow evolution, and tachytely for the fast spurts. But Gould, in particular, was a magnificent writer, and his explication of the process garnered more supporters than did Simpsons, more stiffly written earlier version.

[Auspex]

"Paleontology; A Philosophical Introduction", by Derek Turner, (ISBN 9780521133326), is a brilliant dissection of PE (and other modern paleontological theories). I recommend it to anyone who enjoys light being cast into the shadowed corners of scientific thinking. I am in my second reading of it, because my lack of education in philosophical argument impedes a quick understanding, and I am actually looking forward to the third (it's that good).

[DE&i]

Hi Auspex,

Could you recommend the book for someone like myself.

Darren.

[AgrilusHunter]

I second the recommendation by Auspex and add that the book's section on underdetermination should be required reading for all burgeoning scientists/philosophers.