Ever since evolutionary thought began, there have been dissenters who have grappled with the fact that evolution is inconsistent with religious accounts of creation. In recent years, dissenters in North America have pushed so called "intelligent design" as a scientific alternative to evolution, and have advocated for intelligent design to be taught alongside evolution in American science classrooms. Let's take a closer look at this controversy.
Intelligent design is based on the notion that some aspects of the natural world are better explained by an intelligent, supernatural designer, rather than by evolutionary processes. Intelligent design rests on two assumptions: 1) that a supernatural being can work outside the laws of science and nature, and can interfere with natural events and that 2) some natural phenomena are "irreducibly complex". That is, they could not have arisen by gradual evolutionary processes because in simpler, more preliminary forms, these structures would have been non-functional and/or would have conveyed no survival advantage.
The scientific community overwhelmingly rejects intelligent design as a viable alternative to evolution for several reasons. First of all, many of the central claims of intelligent design—such as the argument for irreducible complexity—have been repeatedly dispelled. For example, the bacterial flagellum has often been cited as an example of an irreducibly complex structure. The flagellum is a tail-like projection that helps bacteria to move, and is made of several different protein building blocks. Proponents of intelligent design have argued that removing any of these proteins would result in a non-functional flagellum. However, recent evidence suggests that the flagellum is not irreducibly complex at all—in fact, some of the proteins making up the flagellum serve other purposes, helping bacteria to inject toxins into other cells. Thus, although having a "piece" of a flagellum doesn't help the bacterium move, it does serve another important function, and therefore could have been favored by natural selection.
The second reason scientists reject intelligent design as an alternative to evolution is because intelligent design is fundamentally non-scientific. Science as a process requires formation and revision of hypotheses, experimentation, collection of data and observations, and inferences based on observed evidence. Claims made by proponents of intelligent design are neither testable nor falsifiable, because there is no way to empirically test for the presence of a supernatural being.
The debate on whether intelligent design should be taught in schools arises frequently in the news. A relatively recent case—Kitzmiller vs. Dover—went to trial in 2005 in Pennsylvania. Several parents with children in public schools sued the Dover, Pennsylvania school district because the school board had mandated that intelligent design be taught alongside evolution. After hearing testimony from both sides, the judge ruled that teaching intelligent design was akin to teaching creationism, and was unconstitutional. You can read more about the Kitzmiller vs. Dover case here.
This issue will undoubtedly remain controversial in the years to come, but it's important to recognize that this debate is not a culture war between religion and science. The United States Constitution protects everyone's right to religious freedom. Rather, the issue here is 1) whether intelligent design should be taught in science class, since it is neither testable nor falsifiable and 2) whether teaching intelligent design in public schools violates separation of church and state, or unfairly promotes one religion over others.
We saw how geology—specifically, the principle of uniformitarianism—helped to establish that the earth was old enough for evolution to occur. The early observations of Hutton and Lyell have more recently been corroborated by a series of techniques, collectively known as radiometric dating. Radiometric dating is a way of figuring out how old sediments, fossils, or even archaeological artifacts are. It is based on the fact that some radioactive isotopes occur naturally in the environment, and they change (or decay) at fairly predictable rates over time.
Let's back up a second. Remember that an atom of any given element has protons and neutrons in the nucleus, and electrons orbiting around the nucleus. The number of protons defines what kind of element it is…so carbon will ALWAYS have 6 protons, or it's not carbon. But, the number of neutrons can vary, resulting in different isotopes of elements. For example, carbon exists in three isotopes: 12C, 13C, and 14C. Most (>99%) of the carbon on earth is 12C, and most of the remaining 1% is 13C.
Both these isotopes are stable – that is, they are not radioactive and don't decay over time. 14C makes up a teeny tiny fraction of all carbon, and is radioactive. Over time, it decays into 14N, which is stable. Scientists know that, for any given amount of 14C, it takes about 5,730 years for half of it to turn into 14N. Thus, we say 14C has a half-life of 5,730 years. Radiometric dating requires that you know the half-life of the radioactive isotope you're using.
Since we've been talking about carbon so far, let's see how we could use radioactive carbon to figure out the age of something. We know how much 14C is currently in the atmosphere. Another important piece of information is that organisms can only gain 14C as long as they are alive. Once they die, they can't absorb any more. So, at the moment right before death, the amount of 14C would match the atmospheric levels, but it would steadily decrease thereafter—specifically, there would only be half the expected amount after 5,730 years.
By comparing the actual amount in a fossil sample (for example) to the amount in the atmosphere, you can calculate how many years have passed since that organism died. Pretty cool, right? This is called radiocarbon dating. Because the half-life of 14C is relatively short, after about 60,000 years there is so little 14C left that it's impossible to measure accurately. So if you want to know the age of something older than that, you've got to try another form of radiometric dating.
Potassium-argon (K-Ar) dating is another form of radiometric dating. 40K makes up about 1% of all K in the earth's crust, decays into 40Ar, and has a half-life of about 1.3 billion years. That's much longer than carbon's half-life – as a consequence, K-Ar dating doesn't work very well on things younger than 100,000 years old, because so little of the 4K will have decayed in that amount of time. K-Ar dating also isn't performed directly on fossils, as we saw with radiocarbon dating. Instead, usually volcanic rock that lies close to a fossil is dated, and that date is taken to approximate the age of the fossil or other surrounding sediments.
So why is volcanic rock so great for K-Ar dating? Well, at the time magma is hot, all the naturally occurring 40Ar lying around is able to escape from the molten rock, which basically means that, at the time the rock forms, there is absolutely no 40Ar in the rock. Once the rock cools, however, 40Ar can no longer escape, so any 40Ar being produced by decay of 40K gets trapped. By comparing the ratio of 40Ar to 40K in a rock sample and knowing the decay rate, scientists can figure out how much time has elapsed since the rock cooled.
There are many other kinds of radiometric dating, and all are a little different based on the half-life of the radioactive elements, the kinds of samples that can be dated, and the accuracy of the techniques. These techniques collectively confirm the observations of Hutton and Lyell and show with certainty that the earth is old enough for evolution to have occurred.
One of the most compelling pieces of evidence for evolution comes in the form of transitional fossils—that is, fossils that show intermediate stages in the evolution of modern organisms. Whales are awesome, and their fossil record shows a clear progression of forms from land-dwelling ancestors to the marine forms familiar to us today.
Modern whales, or cetaceans, can be subdivided into toothed whales (like dolphins and orcas) and baleen whales (like humpbacks). Whales are mammals, and like all other mammals, they produce milk for their young and breathe air. Most mammals also have hair, and although some cetaceans have hair, some have lost it over evolutionary time. So, how did mammals with adaptations for living on land invade the seas and ultimately evolve into whales?
One of the earliest known fossil whales is called Pakicetus, and comes from the middle Eocene epoch (about 50 million years ago). We're calling Pakicetus a fossil whale, but superficially, it didn't look anything like what you think of as a modern whale. This carnivorous mammal walked on four legs, had a long tail, and was about the size of a wolf. However, when paleontologists started looking more closely at Pakicetus they realized that the inner ear bones of this mammal share some similarities with the inner ears of modern whales, which are adapted to hearing under water. In fact, Pakicetus also had its eyes on top of its head, which is common in species that like to submerge themselves in water and look up. The isotopes in the bones and teeth of Pakicetus tell scientists that this animal was probably living in an aquatic habitat, perhaps wading in shallow water (Thewissen et al., 2009). Although Pakicetus had a lot of traits that suggest it was a transitional form on the way to modern whales, it also has many traits that link it to artiodactyls—even-toed hoofed mammals, including hippos, pigs, cows, camels, and deer. In fact, DNA evidence confirms that artiodactyls are the closest living relatives to cetaceans.
Ambulocetus is another transitional whale fossil, but unlike Pakicetus, shows evidence of greater commitment to an aquatic lifestyle. Ambulocetus was larger than Pakicetus, and had paddle-like feet and a large, muscular tail; it probably used its hind limbs and tail for propulsion in water. Remingtonocetus also had a very large tail, and may have only needed its limbs for steering. Importantly, this early whale probably had an even more advanced form of underwater hearing than either Pakicetus or Ambulocetus.
All the transitional forms we've talked about so far are known primarily from India and Pakistan, and were probably somewhat localized. Our next group of transitional fossils, the ptotocetids, achieved a more widespread distribution because they were able to swim efficiently through the oceans. In terrestrial mammals, the pelvis (hip bones) are connected to the spine. This setup is great for moving on land, but if you want to be a superstar swimmer, it's better to have your pelvis and spine separated, because it allows for better propulsion. Maybe that's how Michael Phelps does it. Protocetids had a separated pelvis and spine, hence their ability to go global. They also had nostrils on the top of their heads, making it easier to get air. But, in spite of looking more and more like modern whales, their ankle bones were still similar to their hoofed relatives, and at least some protocetids could probably still come out on land.
Finally, about 41 million years ago, yet another ancient whale had come into the picture, and these, called basilosaurids, were the first completely aquatic whales. They had blowholes on the top of their heads, flippers, and hind limbs that would have been too small to support them on land. Several more million years of whale evolution would pass before the first real toothed and baleen whales appeared, but by the basilosaurids, the land to water transition was compete.