What Does the Fossil Record Say About How Speciation Happens?

Below are the answers to a series of questions asked of me by a friend from way back in high school. His questions were interesting enough, that I thought I’d post the answers here. Other folks might be interested, too.

These answers come off the top of my head. I did not research them, so I might have a few details wrong. But the overall story should be about right.

—–

Could you give me an example of a lineage with an abundant fossil record that stretches over a very long period?

Probably the best example of a fossil lineage that is abundant and present throughout the Earth’s history is the foraminifera, or forams for short. These are tiny, single-celled organisms that have existed throughout the last 570 million years to the present. Forams secrete little calcium carbonate shells that are easily preserved in the rock record. They live in oceans and lakes, so they’re not good at telling one anything about terrestrial environments. But oceans comprise what, 70% of the Earth’s surface? So you can learn a lot.

Forams are also unicellular organisms, which makes them a little far removed from more familiar animals that we’d be interested in, like mammals. And you need a microscope to study them. In that regard, they might be a little disappointing. But they’re abundant and they have a great fossil record. If you want to study evolutionary change in a single lineage, this might be a good place to start.

 

If you were to make a time-lapse animation using known fossils at distinct time intervals (e.g. can only use one fossil example per 1 million years), how long could the animation last at 5 frames/second?

Well, if we consider the Phanerozoic (the Era of visible life – when we have the best fossil record) is 570 million years long, there’d be 570 frames. Divide by 5, that’s 114 seconds, which is a little under two minutes. Not very long. Earth’s been around 4.6 billion years, so that’d take 15 minutes and 20 seconds.

 

Would it be more realistic to use a different time span per frame? How many gaps would there be in such an animation (i.e. if you decide to use a fossil per 1 million years for 100 million years, would you have 100 fossils available, or only 20)?

Well, evolution can occur pretty quickly, but species do also tend to linger for more than one million years, so one frame per million years would probably work, provided the fossil record is relatively complete.

 

Would we observe gradual changes, or see punctuated equilibrium?

The honest answer is, “I don’t know.” I don’t know that anyone has attempted such a study. One of the problems (as you point out in your next question) is that we don’t know how quickly speciation has to occur in order for it to be ‘punctuated.’ It’s a semantic argument. The other problem is that even the most complete rock record has gaps, periods of slow deposition, and periods of rapid rock deposition. The first two of these might make the changes in species appear more rapid than they really were. The last would make speciation potentially look more gradual. We’d have to somehow account for this.

An additional problem would lie in how we define species. No one person is likely to have time to actually look at all the individual specimens used to define the various species in the rocks, so they’d have to rely on the identifications of other scientists. Some scientists define species based on more subtle characteristics than others. A ‘splitter’ uses tiny differences to name new species, whereas a ‘lumper’ will only name new species when there are big differences. Rates of evolution calculated from species named by a splitter then will seem more rapid than rates based on species named by a lumper.

 

Do we have enough fossils in enough lineages to say how long speciation typically takes for animals? Is there some kind of timeframe cutoff where you would say “punctuated equilibrium” but below that you would say “gradualism?”

I don’t think we do. Maybe with some of the microfossils, like forams, this approaches being possible. But for larger animals, like dinosaurs or mammals, there are very few examples of places with sufficiently abundant specimens to even attempt such a study, and even in the cases where there are abundant specimens, they usually only cover maybe a million years. Certainly not long enough to make grand statements about how evolution works in all cases.

 

Also, what does the fossil record look like for times of extreme speciation (such as mammals after the loss of dinosaurs, or for the Cambrian explosion)?

Rapid speciation events like the Cambrian explosion, often wind up not being as dramatic as we think they are, once we get sufficient time resolution for the fossils. The Cambrian explosion seems rapid, but it also suffers from a lack of record. The Cambrian was a long time ago, and many of the rocks (and the fossils therein) have long since been eroded or otherwise lost. Now we have numerous gaps in the record making things look more rapid than they really were.

After the dinosaurs went extinct, there was also a lag in evolutionary response from the mammals. It took ten million years before most of the modern groups of mammals appeared, and that was mostly after a second global event that disrupted everything at the Paleocene-Eocene boundary. The only familiar group to have evolved within ten million years of the extinction of the dinosaurs was the rodents, but even that took over six million years!

So the ‘rapidity’ of speciation as we understand it has everything to do with the resolution and completeness of the fossil record. Because of this, the debate about gradualism versus punctuated equilibrium continues.

The one event that has the potential to show us something about potential rates of evolution in mammals is the event that occurred at the Paleocene-Eocene boundary, about 55 million years ago. This was a time a rapid global warming probably due to increased carbon dioxide in the atmosphere (and for obvious reasons is of great scientific interest). We have fairly complete records of this interval, which lasted about 200,000 years, from Wyoming, and those rocks are loaded with mammals fossils. What we see is 1) mammal species experienced dwarfism during the warm time, and 2) mammals experienced rapid evolution with the appearance of many of the modern orders of mammals at this time.

 

Do we have intermediates during times like those, or do a bunch of lineages just suddenly show up?

The term “intermediate” is a difficult term to use. When we name species, we only name them because they are distinct from other species. Once you’ve named a species, then how can it be an intermediate? It’s a semantic problem, and we can go in circles about it.

Anyway, such intervals are so rare, that it’s hard to say much about them. As far as the Paleocene-Eocene boundary goes, we think we know which groups were ancestral to which other groups, but I don’t think anyone’s gone out on a limb and said they’d found the intermediate form between, say, condylarths (a primitive group of mammals thought to be ancestral to modern hoofed mammals), and perissodactyls (the modern group of hoofed mammals that includes horses, rhinos, and tapirs, known to have first appeared at the beginning of the Eocene).

In most cases, it just looks like a bunch of new lineages just shows up. How much of that is punctuated equilibrium, and how much is deficiencies of the fossil record is an active focus of study.