IN t named Mike Vooring around onsome grassy farmland in eastern Nebraska, not far from ttle toed a curious glint in to  ly preserved skull ofa young r by recent heavy rains.

    A fe turned out,  extraordinary fossil beds everdiscovered in Norter  ooturtles. All erious cataclysm just under time knoogeology as tood on a vast,  plain very like ti of Africa today. to tenfeet deep. t  t, and never had been, any volcanoes inNebraska.

    today, te of Voorate Park, and it ylisors’ center and museum, ful displays on tory of ter incorporates a lab ors can cologists cleaning bones. orking alone in t ary in wured.

    t get a ors to Asate Park—it’s sligo sook me to t atop a ty-foot ravine where he had made his find.

    “It o look for bones,”  I  looking for bones. Iern Nebraska at time, and really just kindof poking around. If I  gone up t just  skull,I’d ed a roofedenclosure nearby, e. Some ther in a jumble.

    I asked   o  for bones. “ell, if you’re looking forbones, you really need exposed rock. t’s ology is done in , dry places.

    It’s not t t’s just t you ting them.

    In a setting like ture across t and unvarying prairie—“you  kno stuff out tto so start looking.”

    At first t tated as mucional Geograpicle in 1981. “ticle called te a ‘Pompeii of preoricanimals,’ ” old me, “e because just after t died suddenly at all. tropeodystrop you  if you  of abrasive as  of it because tt and crumbled itinto my   sligty. “Nasty stuff to o breat on,“because it’s very fine but also quite so tering urned ter into an undrinkablegray sludge. It couldn’t  all.”

    tary ed t tence of so muc, Nebraska’s s  for a long time. Foralmost a century to make  andAjax. But curiously no one  to wonder whe ash came from.

    “I’m a little embarrassed to tell you,” Voor t I tabout it  tional Geograpo confess t I didn’t know. Nobody knew.”

    Voor samples to colleagues all over tern United States asking if t it t ter a geologist named BillBonnic in toucold  tc from a place called Bruneau-Jarbidge in sout Ida tkilled t big enougo leave an asen feet deep almost a tern Nebraska. It turned out t under tern United States t spot, aclysmically every600,000 years or so. t sucion  over 600,000 years ago. t spot isstill t Yelloone National Park.

    e knotle about  is fairly remarkable tot Ford  t tinents moveabout on tion.

    “Strange as it may seem,” e Ricand tribution of matterin terior of tter tand terior of th.”

    tance from to ter is 3,959 miles, w so very far.

    It ed t if you sunk a o ter and dropped a brick into it, it ake only forty-five minutes for it to  ttom (t t point it  rat).

    Our otempts to penetrate to indeed. One or to a dept most mines on Eart a quarter of a mile beneat  yet even come close.

    Until sligury ago, -informed scientific minds kneEarterior  muc a coal miner kne you could digdoance and t rock and t  it. t named R. D. Oldemala, noticed t certain srated to a point deep an angle, as if tered some kind ofbarrier. From t ter a Croatianseismologist named Andrija Moudying grapiced a similar odd deflection, but at a s and tely belole; tinuity, or Mo.

    e o get a vague idea of terior—t really  until 1936 did a Danisist named Inge Leudyingseismograp t o be solid and an outer one (t Oldected) tis t to be liquid and t of magnetism.

    At just about time t Leanding of terior by studying ts at Calteco make comparisons bet. ter and Beno Gutenberg, t o do  at once as Ric o do ereit fello al“tude Scale.”)ter scale ood by nonscientists, ttle less so nos early days o Ricen asked to seeed scale, t rary measure of tremblings based on surfacemeasurements. It rises exponentially, so t a 7.3 quake is fifty times more poimes more pohquake.

    At least tically, t for an earto t, a lo. t says not damage. A magnitude 7quake le—say, four  cause no surfacedamage at all, ation. Mucoo, depends on ture of tion, ty of aftersting of ted area. All t t fearsome quakes are not necessarily tforceful, ts for a lot.

    t eartion ered on Prince illiam Sound in Alaska in Marcer scale, or one in t of Cially logged at 8.6 magnitude but later revised upies(including ted States Geological Survey) to a truly grand-scale 9.5. As you  al science, particularly ing readings from remote locations. At all events, bot only caused al Sout also set offa giant tsunami t rolled six to more victims as far ahe Philippines.

    For pure, focused, devastation,  intense eartory  struck—and essentially so pieces—Lisbon, Portugal, on All SaintsDay (November 1), 1755. Just before ten in ty ed at magnitude 9.0 and ses.

    t t ter rus of ty’s urnedin a y feet o truction.  last tion ceased, survivorsenjoyed just tes of calm before a second sly less severe t. A t t all, sixty tually every building for miles reduced to rubble. timated 7.8 on ter scale andlasted less ty seconds.

    Eartude 2.0 or greater—t’s enougo give anyone nearby a pretty good jolt.

    Altend to cluster in certain places—notably around t anyates, only Florida, eastern texas, and t seem—so far—to be almost entirely immune. Neude 6.0 or greater in t tensive local damage and (I can attest) knocking pictures from walls and childrenfrom beds as far away as New hampshire.

    t common types of eartes meet, as in Californiaalong t. As tes pus eacilone or terval beter t-up pressure and ter t. ticular okyo,  University College London, describesas “ty ing to die” (not a motto you ourism leaflets). tokyo standson tectonic plates in a country already s seismicinstability. In 1995, as you y of Kobe, to t,ruck by a magnitude 7.2 quake, edat $99 billion. But t ively little—compared may a tokyo.

    tokyo  devastating eartimes. OnSeptember 1, 1923, just before noon, ty  Kantoquake—an event more ten times more poime, tokyo , so trainbeneaty years. Eventually it is bound to snap. In 1923,tokyo ion of about today it is approacy million. Nobodycares to guess  die, but tential economic cost  asrillion.

    Even more unnerving, because tood and capable of occurringanyime, are type of sraplate quakes. te boundaries, er deptend to propagate over muc notorious suco  ted States er of 1811–12. ture started just after midnig by tiveness of animals before quakes is not an old ale, but is in fact ablis at all understood) and ty rupturing noise from deep o t  deep. A strong smell of sulfur filled ted for four minutes ating effects to property. Among tnesses ist Joo be in ted out it knocked do least one account, “s in East Coast harbors and .

    . . even collapsed scaffolding erected around tol Building in ason, D.C.” OnJanuary 23 and February 4 furtude follo not surprisingly, since suconing. tone could be under Co guess. And raplate rupturings? Somet know.

    By tists ly frustrated by tle tood ofterior t to try to do somet it. Specifically, t to drill tinental crust oo to tinuity and to extract a piece of tle for examination at leisure. t if tand ture of tbegin to understand eracted, and to predict earts.

    t became kno inevitably, as t ty rous. to lo of Pacific Ocean er off t of Mexico and drill some 17,000 feet tively tal rock. Drilling froma sers is, in trying to drill a op tate Building using a strand of spagti.”

    Every attempt ended in failure. t trated  600 feet. ted s and noresults, Congress killed t.

    Four years later, Soviet scientists decided to try t onRussia’s Kola Peninsula, near t to oa depteen kilometers. ted, but ts ent.  last teen years later, to adepters, or about 7.6 miles. Bearing in mind t t of ts only about 0.3 percent of t’s volume and t t cuteven one-t, we can o erior.

    Interestingly, even t, nearly everyt it was surprising.

    Seismic udies ists to predict, and pretty confidently, t ter sedimentary rock to a depters, folloe for t 2,300meters and basalt from t, tary layer deeper ted and tic layer  all. Moreover, ted, emperature at 10,000 meters of 180degrees centigrade, nearly ted level. Most surprising of all  t t depturated er—somet  been t possible.

    Because  see into to use otecly involvereading ravel terior. e also knotle bit about tlefrom e pipes,  fires, in effect, a cannonball of magma to t supersonic speeds. It is a totally random event. A kimberlite pipe could explode inyour backyard as you read to 120 milesdoe pipes bring up all kinds of t normally found on or near tite, crystals of olivine, and—just occasionally, in about one pipein a s of carbon comes up e ejecta, but most isvaporized or turns to grape. Only occasionally does a  s up at just tspeed and cool doness to become a diamond. It ive diamond mining city in t t  kno. Geologists kno someern Indiana t may be trulycolossal. Diamonds up to ty carats or more  scattered sites tt no one es, it may be buriedunder glacially deposited soil, like ter in Io Lakes.

    So  tle. Scientists aregenerally agreed t ter crust, amantle of , viscous rock, a liquid outer core, and a solid inner core.

    1e kno ted by silicates, for t’s overall density. t be uff inside. e kno togenerate our magnetic field some be a concentrated belt ofmetallic elements in a liquid state. t muc everyt—eract, o bet any time in ture—is a matter of at least some uncertainty, and generallyquite a lot of uncertainty.

    Even t of it , is a matter of some fairly strident debate.

    Nearly all geology texts tell you t continental crust is to six miles t ty-five miles tinents, and forty to sixty miles tain c ties ions. t beneatains, for instance, is only aboutnineteen to ty-five miles to quicksand. (Some people tailed picture of terior, o 40 km (25 mi) is t. From 40 to 400 km (25 to 250 mi) is tle. From 400 to 650 km (250 to 400 mi) is a transition zone betle.

    From 650 to 2,700 km (400 to 1,700 mi) is tle. From 2,700 to 2,890 km (1,700 to 1,900 mi) is t;Dquot; layer. From 2,890 to 5,150 km (1,900 to 3,200 mi) is ter core, and from 5,150 to 6,378 km (3,200 to3,967 mi) is the inner core.

    its crust are questions t divide geologists into t ly early in tory and t er. Strengtters. Ricrong of Yale proposed an early-burst t t of ing t agree  slybefore  at ics in a polemic in an Australian eart cuating myto a report inEarth magazine in 1998.

    “ter man,” reported a colleague.

    t and part of ter mantle togetone”), s on top of a layer of softer rock called t strengt sucerms are neverentirely satisfactory. to say t ts on top of ts adegree of easy buoyancy t isn’t quite rig is misleading to terials flo only in t glass is. It may not look it, but all tless drag of gravity. Remove a pane of really old glassfrom t iceably t ttom t top. t is t of “floalking about. t ten times faster tle.

    ts occur not just laterally as tes move across t upand doion.

    Convection as a process  deduced by tric Count von Rumford at teentury. Sixty years later an Englislysuggested t terior migents to move about,but t idea took a very long time to gain support.

    In about 1970,  urmoil  came as a considerable s it in t ists  decades figuring out tmospropospratosp about wind.”

    ion process goes ter of controversy ever since. Somesay it begins four refil  “ts of data, from t disciplines,t cannot be reconciled.” Geocs say t certain elements on Eartle, but must h.

    terials in tle must at least occasionally mix.

    Seismologists insist t to support suchesis.

    So all t can be said is t at some sligerminate point as er of Earto pure mantle. Considering t itaccounts for 82 percent of t of its mass, tle doesn’tattract a great deal of attention, largely because t interest Eartists andgeneral readers alike ism) or nearer t to a dept a le consistspredominantly of a type of rock knoite, but ain. According to a Nature report, it seems not to be peridotite. More t know.

    Beneatle are ter one. Needless tosay, our understanding of ture of t, but scientists can make somereasonable assumptions. t t ter of tly imes t to turn anyrock tory (among ot t retaining its . Alt is little more t is t t inover four billion years temperature at tly  t estimates range from someto 13,000°F—about as  as the Sun.

    ter core is in many ood, tt it is fluid and t it is t of magnetism. t forward by E. C.

    Bullard of Cambridge University in 1949 t t of t makes it, in effect, an electrical motor, creating tic field. tion is t ting fluids in t somes in wires.

    Exactly  it is felt pretty certain t it is connected s being liquid. Bodies t don’t ance—don’t ism.

    e kno Eartic field cime to time: during t o times as strong as no it reverses itselfevery 500,000 years or so on average, t average ability. t reversal  750,000 years ago. Sometimes it stays put formillions of years—37 million years appears to be t stretc otimes it er as little as 20,000 years. Altoget 100 million years it self about t  est unansion in the geological sciences.”

    e may be going tic field  in t century alone. Any diminution in magnetism is likelyto be bad neism, apart from es to refrigerators and keeping ourcompasses pointing t al role in keeping us alive. Space is full ofdangerous cosmic rays t in tic protection ear tatters. ic field is o ts. teract icles in tmospo createtc knohe auroras.

    A big part of terestingly enoug traditionally ttle effort to coordinate ’s going oninside. According to Ss and geops rarely go to tings or collaborate on the same problems.”

    Perter demonstrates our inadequate grasp of terior t out s up, and it ary reminder of tations of our understanding tion ofMount St. on in 1980.

    At t time, ty-eiged States  seen a volcanic eruption for oversixty-five years. t volcanologists called in to monitor and forecast St.

    ion, and t turnedout,  t all.

    St. arted its ominous rumblings on Marc ingmagma, albeit in modest amounts, up to a imes a day, and being constantly sed to o be a safe distance of eigain’s rumblings gre. ourist attraction for the world.

    Nes on t places to get a vieelevision creedlyfleo t, and people ain. Onone day, more ty copters and lig circled t. But as to develop into anytic, people greless, andt t going to bloer all.

    On April 19 tain began to bulge conspicuously. Remarkably,no one in a position of responsibility sa trongly signaled a lateral blast. ts resolutely based t blo side t somet  a community college in tacoma. ed out t St.  , as o be released dramatically and probablycatastrop part of team and ionsattracted little notice.

    e all kno 8:32 A.M. on a Sunday morning, May 18, t and rock rusain slope at 150 miles an   landslide in ory andcarried enougerial to bury ttan to a dept. Aminute later, its flank severely . omic bombs, sing out a murderous  cloud at up to 650 miles anoo fast, clearly, for anyone nearby to outrace. Many people obe in safe areas, often far out of sigaken. Fifty-seven people y-toll  it een miles away.

    t person on t day e student named ion post 5.7 miles from tain, but interviee tion. aken by David Joon. Joon  to report ts later emporary.

    Eleven years later y-tists and journalists fatally caugpouring of supered asen rock—icflo Mount Unzen in Japan astrophically misread.

    Volcanologists may or may not be t scientists in t making predictions,but t question t in t realizing ions are.

    Less ter tastropcanley illiams of ty of Arizona, descended into tive volcanocalled Galeras in Colombia. Despite t years, only teenmembers of illiams’s party y s or otective gear. ted, killing six of tists, along ourists whers, including illiams himself.

    In an extraordinarily unself-critical book called Surviving Galeras, illiams said er ed t ant seismic signalsand be is to snipe after t, to apply to ts of 1993,” e. y of notiming  to do. I  I ake responsibility. But I do not feel guilty about t. tion.”

    But to return to ason. Mount St.  teen  of peak, and 230square miles of forest ated. Enougrees to build 150,000 s)  $2.7 billion. A giant column ofsmoke and aso a  of sixty t in less ten minutes. An airlinersome ty miles aed being pelted h rocks.

    Ninety  minutes  after  t, aso rain doy of fifty t eig, turned day to nig into everytors, generators, and electricalsc, crians, blocking filtration systems, and generally bringingto a . t s doy were closed.

    All te, just do  Yakima y’semergency broadcast system, o action during a crisis, did notgo on taff did not knoe t.”

    For t off from ts airport closed, itsapproacogety received just five-eiger tion of Mount St.  in mind, please, as one blast would do.