Dating fossils isn’t easy. They seldom pay for dinner. (Sorry, I couldn’t resist the bad joke.) To understand the techniques used to estimate the ages of fossils, it will be helpful first to consider how fossils form. The remains of a dead organism are sometimes buried in sediment, like mud or sand, before fully decaying. Under the right conditions, a process called fossilization preserves the remains, or at least their shape, in the following way. Actually, there are numerous different fossilization processes, which produce different kinds of fossils in different conditions. The process I’m about to describe is one of the most common. As water seeps through the sediment, minerals dissolved in the water begin to crystalize, cementing the particles of sediment together to form sedimentary rock. Minerals also crystalize inside the tissues of the organism’s buried remains, and these mineral deposits fill the empty spaces as the biological material dissolves away, leaving a solid mineral cast which is shaped like the original remains. Usually, only hard structures—like seashells, wood, and bones—are preserved; but in some rare cases, the fossilization process happens so quickly that even soft tissues are preserved in exquisite detail.
Sedimentary rocks normally form underwater and beneath other layers of sediment. Over long periods of time, however, geological processes may shift these rocks to higher elevations, where erosion can expose the fossils and we can discover them. Fortunately for paleontologists, the fossil record is abundant. Fossils can be found almost everywhere—in every country in the world, and in every state of the US. (Want to find some yourself? See here or here for a list of over 15,000 sites where fossils have been discovered in the US and Canada, organized by state/province and by county.) Finding fossils is the easy part. Determining when the organisms lived and where they fit in the history of life, on the other hand, is not so straightforward.
Recall from chapter 5 that the ages of some ancient objects can be determined by radiometric dating methods, which involve comparing the abundance of a radioactive isotope to the abundance of its decay products. For example, as explained on this page, it is sometimes possible to estimate the ages of old bones and other organic remains by measuring the proportion of carbon-14 to carbon-12 that they contain. Unfortunately, that particular technique—called radiocarbon dating—has a serious limitation: it doesn’t work on most fossils! The half-life of carbon-14 is about 5,700 years, but fossils typically are much older than that. Fossils usually contain little or no carbon-14: practically all of it has decayed into nitrogen-14 by now. Moreover, most fossils contain little carbon of any isotope. As explained above, the organic (carbon-based) molecules typically dissolve away and are replaced by minerals during the fossilization process. For these reasons, radiocarbon dating is limited to relatively recent remains that have not undergone fossilization.
However, other radiometric dating techniques can be used to measure the ages of certain kinds of rocks and minerals. One of the best techniques is the uranium-lead method. When magma cools into rock, various kinds of mineral crystals form. The mineral most often used in uranium-lead dating is zircon, which consists of crystals of zirconium silicate (ZrSiO4). As these crystals form in cooling magma, uranium atoms from the magma are incorporated into the crystal structure; but atoms of the element lead remain in the magma (and are eventually incorporated into other minerals as the magma continues to cool). If you find a zircon crystal with some lead atoms in it, you can conclude that these atoms came from the decay of uranium: no lead was in the crystal when it formed. Since the half-lives of uranium and other isotopes on its decay chain are known, it is possible to calculate how long ago the zircon crystal formed, just by measuring the ratio of lead to uranium in the crystal.
The uranium-lead method is more reliable if you measure the specific isotopes of lead and uranium. (This can be accomplished using a mass spectrometer—a device that ionizes the atoms and then shoots them through an electric or magnetic field to separate them by mass.) The decay chain of uranium-238 culminates with lead-206, while the decay chain of uranium-235 ends with lead-207. So, you can compare the ratio of lead-206 to uranium-238 to calculate the age of the crystal, and you can also calculate the crystal’s age by comparing the ratio of lead-207 to uranium-235. If you get the same answer from both calculations, you can be pretty confident that you’ve determined the true age of the crystal.
The half-life of uranium-238 is about 4.5 billion years, so the uranium-lead method can be used to estimate the ages of some truly ancient rocks. It yields an age of roughly 4 billion years for the oldest rocks found on Earth’s surfaceThe oldest known rock formation on Earth is the Acasta Gneiss formation in the Northwest Territories, Canada. Zircon crystals in those rocks formed between 4 and 4.03 billion years ago, according to this uranium-lead measurement. and even older ages for some zircon crystals, eroded off of rocks that have long since disappeared. Extremely old meteorites have also been dated using the uranium-lead method and other similar techniques.
Unfortunately, like the radiocarbon method, the uranium-lead method doesn’t work on fossils. The zircon crystals used in the uranium-lead method only form in igneous rock—that is, rock formed from cooling magma or lava. Fossils, however, are almost always found in sedimentary rock, which forms via the cementation of sediment like sand or clay. Zircon crystals can be found in sedimentary rocks, but they didn’t form there; they are simply part of the sediment, which came from the erosion of other rocks.
Moreover, almost all other radiometric dating techniques have the same limitation: they only work on igneous rocks, so they cannot be used to measure the age of a fossil—at least, not directly. There is, however, a workaround for this problem. By studying geological processes and the relations between different rock formations in Earth’s crust, we can determine the order in which the rock layers formed. For example, if we find a layer of sandstone in which the sand consists of pulverized igneous rock from the erosion of a nearby mountain, we can conclude that the sandstone in the valley is younger than the igneous rocks on the mountain. Or, suppose we find a layer of sedimentary rock that has been cracked, and the crack is filled with igneous rock. Magma must have filled the crack and formed the igneous rock after the sedimentary layer already existed, so in this case we can infer that the sedimentary layer must be older than the igneous rock intrusion. For yet another example, a sedimentary rock may include a layer of ash from a nearby volcano. In that case, we can conclude that any fossils it contains must be from organisms that lived around the same time as the volcano’s eruption, and the date of that event can be determined by dating igneous rocks that formed from lava during the eruption. These are just a few of the many ways in which geologists can estimate the ages of the sedimentary rock layers in which fossils are found.
However, the age of a fossil cannot always be determined by radiometric dating. Sometimes, there just aren’t enough clues from nearby igneous rocks to make even an educated guess about the sedimentary rock in which a fossil is found. In such cases, paleontologists must rely on other kinds of clues. One frequently-employed strategy is to look for index fossils—fossils of organisms known to have lived during a certain geological period—in the same rock layer. If you find a dinosaur bone in a rock layer with fossilized ferns that grew between 150 and 200 million years ago, for example, you can infer that the dinosaur probably lived during that same period. Of course, you’d have to know (based on fossil discoveries elsewhere) when that species of fern existed, and you’d have to be reasonably sure that it grew only during that period. If all other rocks in which the fern appears can be dated to the same period, this might be a reasonable assumption. Unfortunately, interpreting the fossil record isn’t always so easy. We’ll consider some other difficulties in what follows.