Making The Record

Fossils have been known since ancient times: Aristotle discussed them, and fossils of the beaked dinosaur Protoceratops may have given rise to the mythological griffin of the ancient Greeks. But the real meaning of fossils wasn't appreciated until much later. Even in the nineteenth century, they were simply explained away as products of supernatural forces, organisms buried in Noah's flood, or remains of still living species inhabiting remote and uncharted parts of the globe.

But within these petrified remains lies the history of life. How can we decipher that history? First, of course, you need the fossils—lots of them. Then you have to put them in the proper order, from oldest to youngest. And then you must find out exactly when they were formed. Each of these requirements comes with its own set of challenges.

The formation of fossils is straightforward, but requires a very specific set of circumstances. First, the remains of an animal or plant must find their way into water, sink to the bottom, and get quickly covered by sediment so that they don't decay or get scattered by scavengers. Only rarely do dead plants and land-dwelling creatures find themselves on the bottom of a lake or ocean. This is why most of the fossils we have are of marine organisms, which live on or in the ocean floor, or naturally sink to the floor when they die.

Once buried safely in the sediments, the hard parts of fossils become infiltrated or replaced by dissolved minerals. What remains is a cast of a living creature that becomes compressed into rock by the pressure of sediments piling up on top. Because soft parts of plants and animals aren't easily fossilized, this immediately creates a severe bias in what we can know about ancient species. Bones and teeth are abundant, as are shells and the hard outer skeletons of insects and crustaceans. But worms, jellyfish, bacteria, and fragile creatures like birds are much rarer, as are all terrestrial species compared to aquatic ones. Over the first 80 percent of the history of life, all species were soft-bodied, so we have only a foggy window into the earliest and most interesting developments in evolution, and none at all into the origin of life.

Once a fossil is formed, it has to survive the endless shifting, folding, heating, and crushing of the Earth's crust, processes that completely obliterate most fossils. Then it must be discovered. Buried deeply beneath the Earth's surface, most are inaccessible to us. Only when the sediments are raised and exposed by the erosion of wind or rain can they be attacked with the paleontologist's hammer. And there is only a short window of time before these semi-exposed fossils are themselves effaced by wind, water, and weather.

Taking into account all of these requirements, it's clear that the fossil record must be incomplete. How incomplete? The total number of species that ever lived on Earth has been estimated to range between seventeen million (probably a drastic underestimate given that at least ten million species are alive today) and four billion. Since we have discovered around 250,000 different fossil species, we can estimate that we have fossil evidence of only 0.1 percent to 1 percent of all species—hardly a good sample of the history of life! Many amazing creatures must have existed that are forever lost to us. Nevertheless, we have enough fossils to give us a good idea of how evolution proceeded, and to discern how major groups split off from one another.

Ironically, the fossil record was originally put in order not by evolutionists but by geologists who were also creationists, and who accepted the account of life given in the book of Genesis. These early geologists simply ordered the different layers of rocks that they found (often from canal excavations that accompanied the industrialization of England) using principles based on common sense. Because fossils occur in sedimentary rocks that begin as silt in oceans, rivers, or lakes (or more rarely as sand dunes or glacial deposits), the deeper layers, or "strata," must have been laid down before the shallower ones. Younger rocks lie atop older ones. But not all layers are present at any one place—sometimes they are not formed or are eroded away.

To establish a complete ordering of rock layers, then, you must cross-correlate the strata from different localities around the world. If a layer of the same type of rock, containing the same type of fossils, appears in two different places, it's reasonable to assume that the layer is of the same age in both places. So, for example, if you find four layers of rock in one location (let's label them, from shallowest to deepest, as ABDE), and then you find just two of those same layers in another place, interspersed with yet another layer—BCD—you can infer that the record includes at least five layers of rock, in the order, from youngest to oldest, of ABCDE. This principle of superposition was first devised in the seventeenth century by the Danish polymath Nicolaus Steno, who later became an archbishop and was canonized by Pope Pius XI in 1988—surely the only case of a saint making an important scientific contribution. Using Steno's principle, the geological record was painstakingly ordered in the eighteenth and nineteenth centuries: all the way from the very old Cambrian to the Recent. So far, so good. But this tells you only the relative ages of rocks, not their actual ages.

Since about 1945 we have been able to measure the actual ages of some rocks—using radioactivity. Certain radioactive elements ("radioisotopes") are incorporated into igneous rocks when they crystallize out of molten rock from beneath the Earth's surface. Radioisotopes gradually decay into other elements at a constant rate, usually expressed as the "half-life"—the time required for half of the isotope to disappear. If we know the half-life, how much of the radioisotope was there when the rock formed (something that geologists can accurately determine), and how much remains now, it's relatively simple to estimate the age of the rock. Different isotopes decay at different rates. Old rocks are often dated using uranium-238 (U238), found in the common mineral zircon. U238 has a half-life of around 700 million years. Carbon-14, with a half-life of 5,730 years, is used for much younger rocks, or even human artifacts such as the Dead Sea Scrolls. Several radioisotopes usually occur together, so the dates can be cross-checked, and the ages invariably agree. The rocks that bear fossils, however, are not igneous but sedimentary, and can't be dated directly. But we can obtain the ages of fossils by bracketing the sedimentary layers with the dates of adjacent igneous layers that contain radioisotopes.

Opponents of evolution often attack the reliability of these dates by saying that rates of radioactive decay might have changed over time or with the physical stresses experienced by rocks. This objection is often raised by "young-Earth" creationists who hold the Earth to be 6,000 to 10,000 years old. But it is specious. Since the different radioisotopes in a rock decay in different ways, they wouldn't give consistent dates if decay rates changed. Moreover, the half-lives of isotopes don't change when scientists subject them to extreme temperatures and pressures in the laboratory. And when radiometric dates can be checked against dates from the historical record, as with the carbon-14 method, they invariably agree. It is radiometric dating of meteorites that tells us that the Earth and solar system are 4.6 billion years old. (The oldest Earth rocks are a bit younger—4.3 billion years in samples from northern Canada—because older rocks have been destroyed by movements of the Earth's crust.)

There are yet other ways to check the accuracy of radiometric dating. One of them uses biology, and involved an ingenious study of fossil corals by John Wells of Cornell University. Radioisotope dating showed that these corals lived during the Devonian period, about 380 million years ago. But Wells could also find out when these corals lived simply by looking closely at them. He made use of the fact that the friction produced by tides gradually slows the Earth's rotation over time. Each day—one revolution of the Earth— is a tiny bit longer than the last one. Not that you would notice: to be precise, the length of a day increases by about two seconds every 100,000 years. Since the duration of a year—the time it takes the Earth to circle the Sun—doesn't change over time, this means that the number of days per year must be decreasing over time. From the known rate of slowing, Wells calculated that when his corals were alive— 380 million years ago if the radiometric dating were correct—each year would have contained about 396 days, each 22 hours long. If there were some way that the fossils themselves could tell how long each day was when they were alive, we could check whether that length matched up with the 22 hours predicted from radiometric dating.

But corals can do this, for as they grow they record in their bodies how many days they experience each year. Living corals produce both daily and annual growth rings. In fossil specimens, we can see how many daily rings separate each annual one: that is, how many days were included in each year when that coral was alive. Knowing the rate of tidal slowing, we can cross check the "tidal" age against the "radiometric" age. Counting rings in his Devonian corals, Wells found that they experienced about 400 days per year, which means that each day was 21.9 hours long. That's only a tiny deviation from the predicted 22 hours. This clever biological calibration gives us additional confidence in the accuracy of radiometric dating.

Was this article helpful?

0 0

Post a comment