Annals of the Former World

by John McPhee

  From the Serra Geral to the present island, the Tristan da Cunha hot-spot track is so well defined and dated that, as Morgan says, “it really ties down Africa.” Not to mention South America.

  An automatic inference from the theory is that hot spots perforating the same plates at the same times must make parallel tracks. On the floor of the Pacific, the tracks of the Line Islands, the Tuamotu Archipelago, the Marshalls, and the Gilberts parallel the track of Hawaii and the Emperor Seamounts. In the Atlantic, the Canary Islands have traced a curve parallel to Madeira’s. Both are hot spots, and have left tracks that conform to Great Meteor. The Cape Verde Islands are a hot spot. A hundred and seventy million years ago, it was under New Hampshire, on a track nearly coincident with the later track of Great Meteor. The most voluminous intrusions of granite in the White Mountains are dated around a hundred and seventy million years. Cape Verde was where Charles Darwin first got off the Beagle with Charles Lyell’s Principles of Geology in his hand, and quickly developed such admiration for Lyell’s presentation of the science. Had Lyell told him that the Cape Verde Islands had also been on a voyage—that in a deep geophysical sense they had come from New England—Darwin might have thrown the book overboard.

  Even on the best-defined tracks, not everything falls patly into place. There is some granite in New Hampshire that is two hundred million years old—still too young to be part of the Appalachian orogenic story but too old to be explained in terms of the two passing hot spots that left other granites. Possibly the two-hundred-million-year-old rock has something to do with magmas that came up at that time as the crust tore apart to admit the Atlantic. When Great Meteor arrived at the edge of the Canadian Shield, under the present site of Montreal, it presumably made the Monteregian hills, for one of which the city is named. The Monteregian hills are volcanic, but their potassium-argon age disagrees by twenty million years with the date when, by all other calculations, Montreal was over the hot spot—an exception that probes the theory. Morgan attributes the inconsistency to “random things you can’t explain” and mentions the possibility of faulty dating. He also says, quite equably, “If the Monteregian hills really don’t fit the model, you have to come up with another model.”

  The hot-spot hypothesis was put forward in the early nineteen-sixties by J. Tuzo Wilson, of the University of Toronto, as a consequence of a stopover in Hawaii and one look at the islands. The situation seemed obvious. James Hutton, on whose eighteenthcentury Theory of the Earth the science of geology has been built, understood in a general way that great heat from deep sources stirs the actions of the earth (“There has been exerted an extreme degree of heat below the strata formed at the bottom of the sea”), but no one to this day knows exactly how it works. Heat rising from hot spots apparently lubricates the asthenosphere—the layer on which the plates slide. According to theory, the plates would stop moving if the hot spots were not there. Why the hot spots are there in the first place is a question that seeks its own Hutton. For the moment, all Jason Morgan can offer is another shrug and smile. “I don’t know,” he says. “It must have something to do with the way heat gets out of the lower mantle.”

  From very deep in the mantle (and perhaps all the way from the core) the heat is thought to rise in a concentrated column, and for this reason is alternatively called a plume. Its surface features are not proof in themselves that they are the product of some plantstem phenomenon that is (or was) standing in the mantle far below. The chemistry of hot-spot lavas suggests that the rock is coming from below the asthenosphere, but there is no direct evidence of fixed hot spots in the mantle. They exist on inference alone. There is no way to sample the mantle. It can only be sensed—with vibrational waves, with viscosity computations, with thermodynamic calculations of what minerals do at different temperatures and pressures. Sound waves move slowly in soft rock, and some modes of the sound can be stopped completely where the rock is molten. The speed and patterns of seismic waves tell the story of the rock. Seismology is not quite sophisticated enough to look through the earth and count hot spots, but it approaches that capability, and when it gets there hot spots should appear on the screen like downspouts in a summer storm. If they don’t, that may be the end of the second-greatest story in the youthful explorations of geological geophysics.

  Hot spots seem to be active for roughly a hundred million years. Some of their effects on overriding plates last, of course, longer than they do. If they begin under continents, their initial manifestations at the surface are likely to be flood basalts. Hot-spot tracks have gone forth not only from the flood basalts of the Columbia River and the Serra Geral but in India from the flood basalts of the Deccan Plateau, in South Africa from the flood basalts of the Great Karroo, in East Africa from the flood basalts of the Ethiopian Plateau, in Russia from the flood basalts of the Siberian platform. Flood basalts are what the term implies—geologically fast, and voluminous in their declaration of the presence of a hot spot. In Oregon and Washington, in the middle Miocene, two hundred and fifty thousand cubic kilometres flowed out within three million years. Having achieved the surface in this form, the plume begins to make its track as the plate above slides by, just as Yellowstone, starting off from the flood basalts of Oregon and Washington, stretched out the pathway that has become the Snake River Plain.

  An event of the brevity and magnitude of a great basalt flood is an obvious shock to the surface world. “We don’t know what flood basalts do to the atmosphere,” Morgan remarked one day in 1985, showing me a chronology he had been making of the great flood basalts that not only filled every valley “like water” and killed every creature in areas as large as a million square kilometres but also may have spread around the world lethal effects through the sky. Morgan’s time chart of flood basalts matched almost exactly the cycles of death that are currently prominent in the dialogue of massextinction theorists, including the flood basalts of the Deccan Plateau, which are contemporaneous with the death of the dinosaurs—the event that is known as the Cretaceous Extinction.

  The perforations made by hot spots may be analogous to the perforations in sheets of postage stamps. Plume tracks might weaken the plates through which they pass, so that tens of millions of years later the plates would break apart along those lines. Madeira, for example, first drew the line where Greenland broke away from Canada. The Kerguelen Hot Spot, in the Indian Ocean, may have helped India break away from Antarctica. The Crozet Hot Spot, also in the Indian Ocean, seems to have helped Madagascar get away from Africa. In the interior of the southern supercontinent of three hundred million years ago, a hot spot punched out the line that is now the north coast of Brazil. The same line is the Gold Coast and Ivory Coast of Africa. The hot spot now stands in the Atlantic as the island St. Helena.

  The oldest rocks in Iceland are at the eastern and western extremes of the island, because Iceland is a hot spot whose track comes down from the northwest and at present intersects the Mid-Atlantic Ridge where Europe and America diverge. Iceland, for the time being, is spreading with the Atlantic. A hundred million years ago, the Mt. Etna Hot Spot was under the Ukraine, and seems to have cleaned off the Ukrainian Shield. A hot spot has made Ascension Island, on the South American Plate beside the Mid-Atlantic Ridge, fourteen hundred miles east of Brazil. It spent a hundred and ten million years under Africa after starting off from the Bahamas in early Jurassic time, when the transatlantic crossing was instantaneous, because there was no Atlantic. The high-standing Bahamas—eighteen thousand feet above the Hatteras Abyssal Plain—are defined as a carbonate platform, its wide shallow seas underlain by limestones and corals. Morgan says, “I would hope that if you drilled through them you would end up with basalt.” The Labrador Hot Spot is thought to be “blind”—a hot spot that has not found a way to drive a plume to the surface but has nonetheless raised the terrain. This would account for the otherwise unaccountable altitudes of Labrador, not to mention more cleansing of the Canadian Shield. The Guiana Shield is also thought to lie above a blind ho
t spot, which has lifted the country and produced, among other things, the world’s highest falls—a plume of water twenty times the height of Niagara.

  Bermuda is the last edifice of a faint but evident hot spot, which underlies the ocean crust east of the present islands. The domal swell of the seafloor is classic—like Hawaii’s, a thousand kilometres wide. (Under continents, upwelled masses analogous to the Bermudian and Hawaiian swells can be shown by satellite measurements of gravity anomalies.) Bermuda has not been active for thirty million years, but its track can be extrapolated westward in conformity with the track of Great Meteor and the well-established motions of the North American Plate. Seen in its former contexts, Bermuda proves to be a good bit less interesting for where it is now than for where it has been. If you could somehow look into the side of the American continent from Georgia to Virginia, you would see a great suite of Cretaceous strata dipping north and south, descending like a rooftop from an apex at Cape Fear. Something lifted up that arch, and, as one can readily discern from the stratigraphy and structure, whatever did the lifting did it in Paleocene time. Since the Paleocene, the North American Plate has moved the exact distance from Bermuda to Cape Fear.

  Bermuda came through there like a train coming out of a tunnel. Or so it would appear. In the Campanian age of late Cretaceous time, when Great Meteor was in mid-Atlantic, Bermuda was under the Great Smoky Mountains. The Appalachian system consists of parallel bands of kindred geology sinuously winding from Newfoundland to Alabama, where they disappear under the sediments of the Gulf Coastal Plain. Why this long ropy package would stand high in two places and sink low in others is not explained by plate tectonics. It can be explained by hot spots. Great Meteor and Cape Verde seem to have lifted New England’s high mountains, Bermuda the Smokies. Uplift accelerates erosion. The rock of the Permian period—the last chapter in the Appalachian mountain-building story—has been removed everywhere in eastern America except in West Virginia and nearby parts of Ohio and Pennsylvania, halfway between the hot-spot tracks, halfway between New Hampshire and North Carolina. Because plate motions have shifted over time, the tracks of all hot spots, ancient and modern, form a plexus on the face of the earth. Untouched areas between lines often prove to be continental basins—the Michigan Basin, the Illinois Basin, the Mississippi Embayment, the Williston Basin—while the rims of the basins are structural arches lined up on the tracks of the hot spots. Morgan thinks the large continental basins may have been created when hot spots elevated the edges. The Great Meteor track runs between the Hudson Bay Basin and the Michigan Basin. A Paleozoic hot spot seems to have made the Kankakee Arch, which separates the Michigan and Illinois basins. The Bermuda track runs between the Illinois Basin and the Mississippi Embayment. “Every basin gets missed,” comments Morgan, with his hand on a map. “I don’t think that’s a coincidence.”

  Bermuda made the Nashville Dome. It lifted the Ozark Plateau, in middle Cretaceous time. “How much erodes off the top when a hot spot lifts something up depends on the durability of what’s there,” Morgan goes on. “If it’s coastal mush, or Mississippi River mush, it goes quickly and in great volume. If it’s quartzite, it resists. The resistant stuff stands up higher.” Much of a hot spot’s energy is expended in thinning the plate above it. Where the plate is already thin, most of the energy will appear at the surface in outpourings such as lava flows. When a plume has to come up through thick old craton, it makes kimberlites, carbonatites, gas-rich blowouts. The plume is expressing itself as a diatreme, the extremely focussed volcanic event that brings diamonds out of the mantle and explodes them into the air at Mach 2. The conduit is called a pipe because it is so narrow. The rock left inside it after the explosion is kimberlite. When Bermuda was under Kansas, it sent up the Riley County kimberlites. For many years, these diamond pipes were described as cryptovolcanic structures, meaning that nobody knew what they were. Later, they were thought to be meteorite strikes. In 1975, in Riley County, a hole was drilled with a tungsten-carbide bit that could smoothly cut its way through anything but diamonds. It went down sixty feet, where all penetration ceased. The bit was pulled. It was grooved and scarred. There are “meteor impacts” along the Bermuda track in Tennessee, southern Kentucky, and Missouri. Morgan thinks they are diatremes, or, as he puts it, “hot-spot blasts.” They lie in a matrix of Paleozoic rock. If in fact they are meteor impacts, the hot spot would have lifted up the country and caused the erosion that exposed them to view. In Morgan’s summation, “the thing works for me either way.”

  When Bermuda was under Wyoming, in Neocomian time, the Rockies did not exist, but the magmas of the Idaho batholith had recently come in, a short distance up the track. When Bermuda was under the State of Washington, the State of Washington was blue ocean. If the track is followed back to two hundred million years, Bermuda seems to have been under Yakutat, Alaska. Hearing most of this for the first time at a colloquium in Princeton, a graduate student said, “This is like playing chess without the rules.”

  During the past twenty million years, the region that we like to call the Old West is thought to have been passing over not one but two hot spots, which have done much to affect the appearance of the whole terrain. The other one is less intense than Yellowstone, and is at present centered under Raton, New Mexico. Volcanoes are at the surface there. The Raton plume has lifted the Texas panhandle, the southern Colorado high plains. Its easternmost lava flow is in western Oklahoma. Its track, parallel to Yellowstone’s, includes the Jemez Caldera, above Los Alamos, and may have begun in the Pacific. To the question “What lifted the Colorado Plateau, the Great Plains, and the Rocky Mountain platform?” the answer given by this theory is “The plumes of Raton and Yellowstone.” As Utah and Nevada crossed the hot spots, the plumes are thought to have initiated the extensional faulting that has separated the sites of Reno and Salt Lake City by sixty miles in eight million years, breaking the earth into fault blocks and creating the physiographic province of the Basin and Range. Work done in the rock-dating laboratory of Richard Armstrong, a geochemist and geochronologist at the University of British Columbia, showed that Basin and Range faulting began at the western extreme of the region and moved eastward at a general rate of twenty-eight miles per million years—a frame commensurate in time and space with the continent’s progress over the hot spots now positioned under Yellowstone and Raton. The Tetons began to rise eight million years ago and are clearly not products of the Laramide Orogeny. They are a result of extensional faulting, and conform to hot-spot theory as the easternmost expression of the Basin and Range. The Colorado Plateau lies between the two hot-spot tracks, and Morgan believes that their combined influence is what lifted it, setting up the hydraulic energy that has etched out the canyonlands. How the plateau avoided the rifting and extension that went on all around it—why it, too, did not break into blocks—is a question that leaves him baffled. That the two hot spots, at any rate, are progressively lifting the country is a point reinforced by a remarkable observation: a line drawn between them is the Continental Divide.

  Inevitably, it has been suggested that someday North America may split apart along the Yellowstone perforations of the Snake River Plain. “That gives me a caution,” says David Love. “I think there are some problems there. I have a feeling that the hot-spot ideas have been somewhat enlarged beyond the facts. The term itself probably means different things to different people. To me, a hot spot is an area of abnormally high temperature gradients, so high that it can be interpreted as having an igneous mush down below. In the Snake River Plain, the volcanics do get older east to west—in a broad sense, yes. But when you get down to details you get down to discrepancies. We don’t know all the ages we should, on the various sets of volcanics. We need to learn them, and plot them up in geographic and time perspective. We will—but to my satisfaction we have not, yet. I would like to see a lot more regional information. In northwest Wyoming, volcanism began in the early Eocene, fifty-two million years ago. You got the Absaroka volcanic centers
. Volcanic debris from them was spread by water and wind across the Wind River Basin, the Green River Basin. Then what happened? Everything went blah. The Yellowstone-Absaroka hot spot abruptly terminated at the end of Eocene time. Where the hell did that hot spot go? Twenty-five to thirty million years later, it was reactivated in the same place. What was that plume doing for all those millions of years? How do you reactivate a plume? We need answers to this sort of thing, and we don’t have them. If the plume theory is correct, you’ve got to answer those questions.”

  The hail over the interstate turned to snow, and we passed a Consolidated Freightways tandem trailer lying off the shoulder with twenty-six wheels in the air—apparently overturned (a day or two before) by the wind. Abruptly, the weather changed, and we climbed the Rock Springs Uplift under blue-and-white marble skies. As we moved on to Green River and Evanston—across lake deposits and badlands, and up the western overthrust—the sun was with us to the end of Wyoming. On the state line was a flock of seagulls, in the slow lane, unperturbed, emblematically announcing Utah—these birds that saved the Mormons. Mormon traffic, heading home, did not seem intent on returning the favor.