Stories in Stone

by David B. Williams

  In describing their momentous day, Playfair had written,“The palpable evidence presented to us, of one of the most extraordinary and important facts in the natural history of the earth, gave a reality and substance to those theoretical speculations, which, however probable, had never till now been directly authenticated by the testimony of the senses . . . What clearer evidence could we have had of the different formation of these rocks, and of the long interval which separated their formation, had we actually seen them emerging from the bosom of the deep? . . . The mind seemed to grow giddy by looking so far into the abyss of time.”8

  Geologists now know that Hutton, Playfair, and Hall were looking back 425 million years in time, to the age of deposition of the graywacke. The unconformity they saw represented a gap of 80 million years, which from a geologic point of view is a blink, but it was enough time for four-legged amphibians to have moved onto land, for forests to have spread widely across the globe, and for sharks to diversify and dominate the seas.

  Hutton fleshed out the ideas the three men had discussed at Siccar Point in his book Theory of the Earth; or an Investigation into the Laws Observable in the Composition, Dissolution, and Restoration of Land upon the Globe. Playfair followed several years later in 1802 with Illustrations of the Huttonian Theory of the Earth. He had been forced to write his book, which covers the same material, because Hutton’s dense and obtuse prose had prevented most readers from understanding his revolutionary geologic ideas.

  As he noted in his title, Hutton based his ideas on what he could see out in the field. This in itself was a new beginning for most scientists, but he also observed that the planet was in a constant state of change and ever recycling itself. Dead organisms accumulated in the sea and made limestone. Mountains eroded to a “confused mass of stones, gravel, and sand,” which consolidated into puddingstone. Rivers carried sands that became sandstones. Hutton’s geologic reasoning has been summed up with a very Hillel-like statement, “The present is the key to the past.”

  One man who did penetrate Hutton’s dense language was Charles Lyell, a twenty-seven-year-old lawyer and amateur geologist. In the year following Keating’s visit to the Minnesota River valley, Lyell visited Sic-car Point with James Hall. Lyell recognized the significance of the great unconformity between the schistus and the sandstone. In 1830 he wrote Principles of Geology, Being an Attempt to Explain the Former Changes of the Earth’s Surface, by Reference to Causes Now in Operation.

  Principles took Hutton’s great statement and backed it up with numerous examples. Lyell also made two additional points: that natural laws did not change over time and that change occurred slowly and gradually and not catastrophically. Together these three principles form the core of uniformitarianism, one of the central tenets of geology. In addition, Lyell’s book further pushed geologists to start pondering an Earth much older than the one described in the Bible. With Principles geologists now had their own scripture.

  Although Principles pointed the way, acceptance of an ancient Earth came slowly. When Charles Darwin wrote in the first edition of On the Origin of Species (1859) that the valley of the Weald in southern England took 306,662,400 years to erode, critics lambasted him and he pulled the number in the third edition. Yet in 1862 one of Darwin’s detractors, William Thomson—later Lord Kelvin—proposed an age for Earth as great as 400 million years.

  Kelvin did not base his calculation on uniformitarianism but on the second law of thermodynamics, that a hot mass will cool over time. Since Earth had started out hot, as indicated by rocks such as the gneiss of the Minnesota River valley, it must be cooling. Few could dispute Kelvin’s observation. As he gathered more data, however, he dropped his number to 100 million years and finally to 24 million years.

  Not all geologists agreed with the physicist Kelvin. What about the slow, steady rates of geological and biological processes, which required more time than Kelvin proposed? By looking at sedimentary processes, such as deposition and erosion, geologists derived numbers between 3 million and 15 billion with the most popularly accepted age of Earth being around 100 million years.9

  Compared to what we now know the age of the earth to be—4.5 billion years—a range of 24 to 100 million years seems to be puny. But consider the relative change between six thousand years and 100 million years. To make that leap requires four orders of magnitude compared with just one order of magnitude between 100 million and 4.5 billion. Hutton and Lyell had helped to push geologists across their biggest relative chasm of time. To get across the bigger, absolute chasm would require another radical shift in our understanding of the planet.

  Geologists had long known that the Morton Gneiss was very old but not until 1956 when four University of Minnesota researchers published a crystallization date of 2.4 billion years did they learn how old.10 The quartet based their age on a fundamental property of many elements, radioactivity, which simply means that an element is naturally unstable and constantly decaying. The best-known element for radiometric dating, as this technique is called, is carbon-14, which works for ages less than sixty thousand years. For ages of millions or billions of years, geologists turn to potassium, argon, lead, and, the most useful for the Morton, uranium.

  When an element decays, or breaks down, it changes from an unstable isotope, called the parent, to a stable isotope, called the daughter, sort of like a 1960s flower child giving birth to an accountant. Isotopes refer to different forms of an element that have the same number of protons in their nucleus but a different number of neutrons. For uranium, which has several isotopes, the most important for geologists studying Morton Gneiss are uranium-238 (written 238U), which decays to form the lead isotope 206Pb, and 235U, which decays to 207Pb.11

  What makes radioactive decay of uranium, and other elements in radiometric dating, useful to geologists is that the change from parent to daughter occurs at a measurable rate, called the half-life. The half-life of 238U is 4.47 billion years, which means that 4.47 billion years after crystallization, half of the parent 238U will have become the daughter, 206Pb. (Carbon-14 has a half-life of 5,700 years.) To figure out the age of formation of the uranium, and hence the age of formation of its surrounding rock, all geologists have to do is count the number of parent uranium isotopes and number of daughter lead isotopes.

  Chemist Ernest Rutherford was the first to calculate the amount of the parent and daughter elements in a mineral. In 1904 he obtained an age of 497 million years. Three years later, an American chemist, Bertram Bolt-wood, showed that a uranium-bearing mineral (uraninite-UO2) from Canada had formed 2.2 billion years ago.

  Rutherford and Boltwood weren’t geologists, so radioactivity for them was more of a technical challenge than a way toward understanding geologic time. Geologists quickly recognized the insights to be gained with radiometric dating, and by the 1930s, it had become the accepted form of ascertaining the age of rocks. Researchers have continued to perfect their techniques both through improved technology and improved understanding of radioactivity. They have continued to refine the geological timescale, precisely dating eons, eras, periods, epochs, and ages. They have also continued to seek out and identify older and older rocks, including a beautiful building stone from Minnesota.

  To analyze uranium in the Morton Gneiss, geologists studied zircon,12 an extremely resistant mineral because of its high melting temperature and extreme hardness. In addition to zircon grains’ long life, they have an unusual crystalline structure, which facilitates the incorporation of uranium from a magma, usually in concentrations of a few hundred parts per million.13 (Zircon isn’t radioactive because it contains so little uranium.)

  Zircon’s heat resistance also aids geologists in age dating. During deep burial or when intruded by magma, many minerals cannot withstand the heat these processes generate and end up melting. In contrast a crystal of zircon resists melting and instead attracts any uncrystallized zircon, and uranium, that might be in the introduced liquid. The new zircon forms a layer or rind on the original.
Each subsequent high-temperature geologic event produces additional layers and each rind creates a time stamp as the uranium begins to decay, allowing geologists to date when that event occurred.

  Adding new layers, like a tree adding rings, creates a technological problem. Although zircons occasionally grow to five-eighths of an inch long, the ones used to date the Morton Gneiss, and most other rocks, are about as wide as a two human hairs. The rinds are more miniscule, on the order of spider silk. When geologists first started to study the Morton they did not have the technology to date separately the rinds and the core. Instead they analyzed the entire crystal and came up with what Pat Bickford, professor emeritus of petrology at Syracuse University, called “precise but inaccurate” ages for zircons.14

  As their dating tools improved, geologists pushed the date of the Morton back. By 1963 it was 3.2 billion years old,15 and in 1974 two researchers reported an age for the Morton of 3.8 billion years old, the oldest rock on Earth.16 As one can imagine, there was much rejoicing. Fame was fleeting though. By 1980 the commonly accepted age of the Morton had dropped to 3.5 billion years. At that number, it was still the oldest, most commonly used building stone in the world, but the title of Earth’s oldest rock belongs to gneiss found near Slave Lake in the Northwest Territories of Canada. Its age is 4.03 billion years.

  In 2006 Pat Bickford led a team of researchers who obtained the Morton’s most up-to-date age of 3,524 million years ago.17 His date will probably be the one that sticks, mostly because he was able to take advantage of technology developed in the 1990s to analyze individually the core and rinds of a zircon crystal.

  “If you want to know the answer badly enough, you will do tedious and persnickety work,” said Bickford. To obtain a radiometric date, he starts by breaking rocks into plum-sized pieces, grinds them into powder with a jaw crusher, and separates out denser material (“like panning for gold”). He further isolates the zircons by floating the residue in a heavy liquid, which suspends the lighter particles, and by running the material through a magnetic separator. Once he obtains a good supply of zircons, he places them under a microscope and uses tweezers to handpick pure, crack-free crystals, each of which is half the size of a grain of salt. Finally, he mounts the zircons in epoxy and polishes them to half their original thickness. A typical puck of epoxy contains as many as a thousand zircons. Bickford does all this work at his lab in Syracuse before shipping the mounts out to Stanford.

  There he uses the Sensitive High-Resolution Ion Microprobe, or SHRIMP, one of the tools that has enabled geologists to tease out the rind’s geologic information. To date a zircon, the SHRIMP bombards the crystal’s surface with a tightly focused beam of oxygen ions, which excavates micrometer-sized pits from the zircon crystal and strips uranium and lead atoms of some of their electrons. This gives the uranium and lead atoms, now called ions, an electrical charge.

  Once ionized, the uranium and lead are accelerated through a strong magnetic field, which separates the ions on the basis of their mass. By changing the pull of the magnets, Bickford can precisely focus and measure specific uranium and lead isotopes. He usually counts 238U and 206Pb, as well as 235U and its daughter 207Pb, which have a half-life of 747 million years. The second set of isotopes provides an independent, corroborating age date.

  “The SHRIMP has completely revolutionized geochronology. It means we can pinpoint specific regions of a zircon and get specific dates for specific events,” said Bickford. With the SHRIMP uranium-lead method, Bickford can narrow down his dates to the point where the range of uncertainty is small enough that the number can be counted as accurate. For example, one zircon crystal from Morton Gneiss recorded three dates, which reflect initial formation and two subsequent geologic events. The average margin of error for each date is .005 of a percent.

  Time is one of the hallmarks and central challenges of geology. How does one relate to billions and millions, or even tens of thousands, numbers not typically bandied about in daily conversation? To do so requires an openness and respect for the possibilities of time. For example, I can’t see back in time but I can look at a sandstone and easily grasp that it was once sand that washed out of an eroding mountain chain, settled in a dune, disappeared under more sand, and finally got converted to rock.

  I also trust the numbers used by geologists. The numbers are based on the laws of science and in the realms of observation, experimentation, and hypothesis. Scientists have tested and retested these numbers, not just on the same rocks or in the same laboratory or under the same conditions, but on different rocks, in different labs, and under different conditions around the world. The numbers have withstood the intellectual challenges of other scientists, scientists who had no interest, either financially, intellectually, or emotionally, in the truth of the numbers.

  John McPhee coined the term “deep time” to describe the great abyss that made John Playfair so giddy. Deep time is what makes possible the almost imperceptible spreading of two plates to become the Atlantic Ocean. It is what allows a single finch on a small group of volcanic islands in the Pacific Ocean to evolve into thirteen species of finches, each adapted to a specific niche. Deep time is what helps with understanding the billions of years necessary for the formation of a rock unit in the Minnesota River valley that looks like a mixture of bubble gum and fudge.

  According to our Morton amanuenses, Pat Bickford’s zircons, the story of the Morton Gneiss began 3,524,000,000 years ago. Bickford found that the oldest zircons originated in the gray layers of the Morton. Because the gray layers have the chemistry of a type of rock known as tonalite, Bickford and many other geologists who have studied the Morton believe that the original source for the gneiss must have been a tonalite. Tonalites (named for rocks found near Tonale, Italy) are similar in composition to granite but a bit darker, thus in the beginning, the Morton was probably a drab, gray igneous rock with lighter speckles of biotite and chocolate chips of hornblende.

  Bickford does not know where the Morton originated. He can, however, theorize about how it formed by looking at more modern tonalites. For example, 40 million years ago, during the collision of Africa with Europe, oceanic crust subducted continental rock and generated a magma that later solidified into the Italian tonalite. Colliding plates probably generated the Morton’s tonalite, but geologists face a problem with making an absolute statement about the collision. Planet Earth 3.5 billion years ago did not look anything like the green and blue planet we now inhabit.

  First, the color green probably didn’t exist, or at least green life did not exist. The oldest fossil evidence for life are microbes found in 3.45-billion-year-old rock in western Australia, which means that when you see the Morton, you are seeing a rock that first cooled and solidified on a planet devoid of life. Second, what we think of as continents may not have existed either or they were a much less significant feature of our budding planet.

  I use the word “may” because one of the great questions in geology is “When did plate tectonics begin?” At its most basic, plate tectonics describes the interaction of the dozen large and several smaller plates, consisting of continental or oceanic crust, which constitute Earth’s outer layer. The word “tectonic” comes from the Latin tectonicus, pertaining to building. New crust forms at spreading centers in the oceans and moves away from its point of origin. Plates disappear when they dive beneath other plates at subduction zones but at the same time subduction generates continental crust. Although plate tectonics is one of the best known, most widely accepted, and easily explainable founding concepts of any field of science, a controversy centers on when plates began to form, interact, and disappear, or die, at least in the mode seen on modern Earth.

  “All geologists were suckled on plate tectonics and they never accept that something other than plate tectonics existed,” says Robert Stern of the University of Texas at Dallas.18 He is one of the few who believes plate tectonics began as recently as only 1 billion years ago. Stern holds that early Earth was much hotter than it i
s today and crustal plates may have been too buoyant and too thick to subduct each other. If subduction was occurring on a young Earth, it should have generated three types of rock: ophiolites, blueschists, and ultrahigh-pressure (UHP) metamorphic terranes. Stern sees little evidence for these rocks prior to a billion years ago and without them, no plate tectonics.

  Furthermore, Stern argues that a shift to an Earth dominated by plate tectonics must have had a profound affect on global environments. An increase in subduction-generated volcanism would have shot gases and fine particles into the atmosphere, cooling the planet and leading to a climatic condition that geologists call “snowball Earth,” one of which froze the entire planet around 710 million years ago. “All in all, the arguments for early plate tectonics are fairly unconvincing,” Stern concludes.

  Kent Condie of New Mexico Tech counters that plate tectonics has operated since at least 2.7 billion years ago and possibly as far back as 4.4 billion years, the age of the oldest known mineral, a zircon from Australia. He challenges Stern’s concerns about lack of subduction-related rocks by observing that a hotter young Earth altered how plates sub-ducted, and that resulted in fewer ophiolites, blueschists, and UHPs, although all three rock types do exist as far back as perhaps 3.8 billion years ago. Condie also cites rocks such as the Morton’s original tonalite, which are widespread in the Archean era (3.8 to 2.5 billion years ago) and require oceanic crust and subduction for formation. “It may be that plate tectonics did not begin globally all at once,” says Condie. “It may have been episodic but by 2.1 billion years it was continuous.”19

  Condie makes an additional point. He observes that no other planets in our solar system have higher life forms, mostly because plate movement fostered our oxygen-bearing atmosphere by creating continents. Continents allowed for the evolution of photosynthetic plants, the source of oxygen. Furthermore, the shallow marine shelves that develop on the edges of continents are an ideal place for carbon dioxide to be deposited, in the form of calcite. If such deposition did not occur, carbon dioxide would accumulate to dangerously high levels in our atmosphere. “Without plate tectonics, humans wouldn’t exist,” says Condie.