Robert Mazrim is haunted by a highway that no longer exists. Tantalizing bits of evidence that it served travelers for thousands of years keep disrupting his regular work as a historical archaeologist. Even the interstates he drives between Saint Louis and Peoria appear to follow the same ghostly corridor, a mile or two wide, through which the old road ran.

“This found me,” he says, sounding a bit like a dog with a burr in its paw. “I have no agenda here. I’ve got plenty else to do.” He’s director of the Sangamo Archaeological Center and a historical archaeology consultant for the Illinois Transportation Archaeological Research Program at the University of Illinois. “I’m not really interested in studying old roads or putting up a sign saying ‘3,000-year-old rut.'”

Hardly anything is left of this road, trail, trace, or path. Never paved, it vanished wherever farmers plowed the prairie. The three fragments that appear to have survived did so because they’re in areas that were never plowed. One fragment is southeast of Springfield; two are on private property on Elkhart Hill next to the tiny town of Elkhart, Illinois. What remains isn’t even a rut. It’s just a strip of flat ground roughly a yard wide that’s a foot or so lower than the adjacent land, tramped down and eroded over the centuries. Even this much is hard to spot once the weeds grow up in midsummer.

But the trail won’t go away. About 15 years ago Mazrim began studying how European-Americans settled the Springfield area in the 1810s and ’20s. He found that they frequently used this trail. Over time he managed to uncover documents showing that it had been used by French settlers and missionaries as early as 1711. When he mentioned it to friends who work in prehistoric archaeology, they added to the shadowy story by telling him about prehistoric sites in the corridor that had always puzzled them.

Despite his reluctance to pursue the topic, Mazrim is now drafting a publication that will lay out the circumstantial evidence–none of it conclusive–that Native Americans were using this corridor on a regular basis as early as 1200 BC. In August his archaeological museum in Elkhart, called Under the Prairie, will, in conjunction with the local historical society, offer wagon rides that show off the distinctive natural and cultural history of Elkhart Hill–including the two flat spots that are probably among the last visible traces of the old road.

Others share Mazrim’s interest, if not his reluctance. Last September the Illinois State Historical Society dedicated a marker identifying a likely piece of the trail overlooking Lake Springfield that was discovered by local historian and bookseller David Brady of Divernon. At the historical society’s annual symposium in Springfield in December, four historians–Brady, Patricia Goitein of Peoria, Anna Clark of Kinderhook, and Roger Matile of Oswego–gave talks on “Lost Highways Found.”

Matile, who writes a column for the weekly Oswego Ledger-Sentinel, deplores the way local histories name the early settlers and dwell on their hardships, but say little or nothing about the roads they traveled on: “It’s as if they were too well-known to be discussed.” Goitein edits the newsletter of the Galena Trail Committee of Peoria County. (Galena Trail is a northwest extension of the old trail, leading from Peoria to the old lead-mining district of northwestern Illinois and southwestern Wisconsin.) “The roads we drive over today are more than they appear to be,” she says. “In many cases they’re the last and best renditions of the old trails.”

The trail that’s bugging Mazrim doesn’t match any one current highway. According to historical accounts, it started near Saint Louis and followed the high ground north by northeast. It passed a spring at the head of Paddock Creek, just northeast of the present-day town of Bunker Hill. According to Zimri Enos of Springfield, water there had washed out the prairie sod and made “a long, wide and deep hole that was filled with a never failing supply of good spring water.” Enos’s notes on the trail were published in 1911, after his death, in the Journal of the Illinois State Historical Society. In dry weather, he wrote, this spring was the only watering place for 50 miles. (Since our predecessors didn’t build rest areas, their trails typically had to go past essential resources such as water and hunting grounds.) The trail crossed the Sangamon River near the mouth of Sugar Creek, just east of where Springfield is now, and continued northeast to the conspicuous hill at Elkhart. Around present-day Lincoln, it turned north toward Peoria Lake in the Illinois River. As Enos wrote, for almost 100 miles this trail “crossed no stream of any size and passed through little timber,” following “the watersheds or divides of the streams through the prairie.”

Today most travelers along these lines take Interstate 55 as far as Lincoln and then swing north on Interstate 155. People of a certain age know I-55 as a revised and improved version of Route 66, and I-155 as a revised and improved version of Illinois 121. Those who take an interest in Illinois history may also know that these old highways follow the corridor of the still older Edwards’ Trace. The question that’s bedeviling Mazrim and some colleagues is this: did Edwards’ Trace follow a much older prehistoric trail?

Edwards’ Trace got its name from a raid carried out during the War of 1812. Illinois wasn’t yet a state, just a territory on the western edge of the frontier. The war here consisted of back-and-forth acts of terrorism between European-American settlers and Indians allied with the British, of which the Fort Dearborn massacre was one.

In the fall of 1812 territorial governor Ninian Edwards led a band of 350 rowdy young men from southern Illinois on a punitive expedition into central Illinois. They left the northernmost European-American citadel, Camp Russell, near Edwardsville, and marched north toward Peoria Lake. John Reynolds–then a private without a horse, later governor of Illinois–wrote in his autobiography that they marched “up the north-west side of Cahokia creek nearly to its source, thence across the prairie to Macoupin creek, not far above the present Carlinville, and at the Lake Fork we stopped to noon. At this point some wild boys dug open an Indian grave, and found in it with the Indian, a gun, broaches, and other articles.” Crossing the Sangamon at Sugar Creek and passing east of Elkhart Hill, they burned an abandoned Indian village “where we saw on the bark of the wigwams much painting, generally the Indians scalping the whites.” The following morning they surprised the Black Partridge village on Peoria Lake, killed 24 warriors, drove the rest of the inhabitants into the surrounding swamps, and destroyed “their horses, camp kettles, corn, and everything.”

Fearing a counterattack, the band returned south at once. In Reynolds’s opinion, their expedition “no doubt did much service in preventing the Indians from marauding around the frontiers.”

In fact, the cycle of terror continued up and down the trace for years to come. In 1814 Kickapoo raiders went as far south as Wood River, killing members of the Reagan and Moore families. “Settlers pursued this band up the Indian Trail to the headwaters of Sugar Creek,” writes John Mack Faragher in Sugar Creek: Life on the Illinois Prairie, “where they killed one warrior in the timber and recovered Mrs. Reagan’s scalp from his pouch.”

It took Mazrim a few years of off-and-on study to confirm that Edwards didn’t blaze the trail his soldiers used. As Mazrim explains in his recently published book, “Now Quite Out of Society”: Archaeology and Frontier Illinois, Father Gabriel Marest’s account of his travels from Kaskaskia to Michilimackinac in the spring of 1711 strongly implies that he followed the same corridor.

Like most early chroniclers, Marest took his route for granted and didn’t describe it, but he did offer three clues. He said the trip made his feet hurt, implying that he was walking and not sitting in a canoe. He said his party was going to see “the Peourias,” a tribe of the Illini located on the Illinois River, either at Peoria Lake or at present-day Naples, west of Springfield. Finally, he said that at one point his party “reached” the Illinois River. He wouldn’t have put it that way if they’d been hugging the riverbank all the way, so presumably they took a route that left the river valley and returned to it later.

If you’re looking for such a path, topography makes Edwards’ Trace the logical choice, based on the nature of downstate landscape, which was flat and treeless except where creeks and rivers cut valleys below the level of the plain. As Mazrim explains in his book, Edwards’ Trace “avoided the more dissected uplands further west, and took advantage of timber lines [in the creek valleys] for navigation. Travel further west would have been slower,” as it would have involved going down and up the valley hills, through woods, and over creeks. Travel farther east, in the open prairie, would have been more difficult, as it would have involved a greater risk of getting lost. (Depending on where the Peoria tribe was at the time, Marest may have left the trace and headed west from Springfield or continued on it north to the river at Peoria.)

After the War of 1812 traffic on the trace burgeoned. In 1816 Stephen Long, an engineer in the employ of the U.S. government, made the first attempt at drawing a map of it to scale. The earliest European-American settlers of the “Sangamo Country” (now the Springfield area) followed the trace north from the existing settlements in the American Bottom across from Saint Louis. You might expect pioneers to fan out north and east, stopping and settling at the first fertile, well-timbered, well-watered places they came to. Not so. For his book Mazrim carefully examined early Government Land Office maps and original surveyors’ notebooks, which show that after the War of 1812, “American farmers did not evenly disperse across attractive environs north of the greater American Bottom.”

Instead they followed existing trails, Edwards’ Trace being the main one. And along those trails, their settlements clustered at the far northern edge of the area being surveyed by the government at the time. For reasons that aren’t yet known, they skipped what look like equally desirable places along the way. One of these settlements was on Edwards’ Trace, along Sugar Creek near Springfield. Comparable places farther from known trails, such as Decatur, weren’t settled until later. “People in Kentucky heard about the Sangamo Country,” says Mazrim. “They picked this over comparable places east of here. That has a lot to do with why the state capital is here.”

Faragher tells some of the Springfield-area pioneers’ stories in Sugar Creek. In October 1817 the peg-legged entrepreneur Robert Pulliam followed the trace north with a pioneer party of four men, one woman, and a string of cattle. They built “a rude log shelter” near Sugar Creek (the first known European-American building in the area), pastured the cattle on prairie grass, tapped the maple trees for sap, which they boiled down to sugar, and returned south in the spring. Local lore has sometimes portrayed Pulliam as venturing into an empty wilderness, but in fact he was trespassing, and his party was lucky to be spared by the Kickapoo. Faragher thinks the Kickapoo must have used that grove to make maple sugar just the year before, and the tribe didn’t cede the territory to the U.S. government until almost two years later.

In the spring of 1818, “guided by a ranger of the late war,” another party of four fathers and six sons led their teams of oxen up Edwards’ Trace to the same neighborhood. In August 1819, a month after the Kickapoo cession, German traveler Ferdinand Ernst came up the trail–which he described as a “fine, well-traveled road”–and counted 60 families raising impressive crops of corn along Sugar Creek. In 1821 the first commissioners of Sangamon County declared their portion of the “old Indian trail” a public road and ordered neighbors to work on improving it. The following year Congress declared it a post road for regular mail service. Faragher writes, “In everyday conversation the route from American Bottom to Sugar Creek soon acquired its third name in recorded history–the St. Louis Road.”

Catholic settlers were pleased when a Jesuit made his way up the road to celebrate mass near Springfield in 1829. In the 1830s and ’40s one of the younger established settlers, Philemon Stout Jr., drove his family’s cattle south along the road to market in Saint Louis.

By that time the Illinois frontier was closing. Increasing wealth and improved technology–such as bridges over creeks that made it less necessary to follow the high ground–altered the road. As early as 1833 new users were shifting the path. According to Zimri Enos, between Honey Point (Macoupin County) and Zanesville (Montgomery County), Edwards’ Trace was “very distinct” but was located “east of and considerably further out in the prairie than the wagon road between the same places.”

In 1821 Elijah Iles built and stocked a store two miles west of the trace, in what became Springfield. His stock came from Saint Louis by water to Beardstown, not up the trace. The only store for miles, according to Mazrim, it pulled the trail west.

(Earlier this spring construction at Second and Jefferson in downtown Springfield enabled Mazrim and helpers to make a weekend dig that uncovered a sandstone wall of Iles’s store cellar and a few remains of his stock-in-trade. This is a characteristic example of Mazrim’s historical archaeology work, which involves comparing what historical documents say about how people lived with the stories surviving physical relics tell.)

Notwithstanding these small shifts, central Illinois probably looks the way it does because of Edwards’ Trace. If the trace hadn’t existed, the next big city north of Saint Louis might well have been Peoria, which is strategically located on the Illinois River. If the trace had passed, say, 40 miles east of where it does, Decatur might have enjoyed that distinction. Without a trail on its doorstep, Springfield would have been settled later, and so it might not even have been in the running for state capital when the issue came up in the 1830s.

So where did the road come from? Did the French voyageurs and missionaries blaze Edwards’ Trace? Or was it here before they were?

It may seem reasonable to assume that they found it already here and used it, just as Robert Pulliam used an existing sugar grove. So says the state historical marker in Lake Park: “For millennia prehistoric people utilized the trail for seasonal trading, hunting, and waging war.” But proving this assertion is difficult, because prehistoric traders, hunters, and warriors didn’t leave a trail of Holiday Inns or burned-out tanks as evidence of their passing.

Illinois State Museum curator of anthropology Michael Wiant tackles the question of prehistoric travel systematically. He starts with what he’s pretty sure of–namely, that from their arrival between 12,000 and 14,000 years ago, Illinois’ inhabitants did indeed travel overland, not just on streams and rivers. That’s plain from seeing where they lived. “If I could show you the map of the 50,000 or so known archaeological sites in Illinois, you’d see a distinct clustering on streams and rivers, but some scattered across the landscape as well.” Even that degree of clustering may be an illusion, because more archaeological digs have been made in river valleys than in the uplands.

When they did travel overland, did these Native Americans use the same trails over and over? Wiant thinks they did. In 1730, for instance, when the Fox were at odds with the French and their Indian allies, “they moved into the interior and constructed a rude fort near Arrowsmith,” in eastern McLean County–far from any rivers or streams. “They had to know where to do this in the middle of the prairie” in order to be able to find springs and wild game; therefore the site must have been on a regularly used trail. Similarly, it makes sense that Native Americans would use the same groves of sugar maple–source of an important early spring food–year after year. Wiant isn’t comfortable with another story, that the Indians followed trails created by the buffalo, mainly because Illinois’ tallgrass prairie never supported the large numbers of buffalo the short-grass prairies farther west did.

There are other reasons to believe in prehistoric overland travel. Rivers don’t always offer the best connection between two points–Edwards’ Trace, for example, is shorter than the corresponding water route between Saint Louis and Peoria. Nor are waterways convenient for travel when they’re frozen or, worse, partly frozen.

It’s even possible that prehistoric Illinoisans traveled overland to avoid confronting hostile neighbors. A 1991 paper coauthored by Ken Farnsworth, now senior researcher at the Springfield Research Lab of the Illinois Transportation Archaeological Research Program, makes a case based on artifacts that the lower Illinois River valley was occupied by a group culturally separate from the Mississippians who lived at Cahokia roughly 1,000 years ago. If his article is correct, and if the two groups were sometimes at odds, then Cahokia traders or emissaries might have chosen to avoid the river and follow Edwards’ Trace when they journeyed north.

Yet Wiant’s systematic statement of the case leaves us well short of Edwards’ Trace. Knowing that there must have been regularly used trails somewhere in prehistoric Illinois is a long way from proving where any particular one went. That’s the task in which Mazrim finds himself enmeshed, and Wiant thinks it’s a daunting one.

In earlier decades this task was close to impossible, because archaeologists had trouble telling just how old a given site might be. “At first we used spear points” to date associated artifacts and village sites, says Farnsworth. But spear-point technology was slow to change–the same style might persist for 2,500 years. The development of radiocarbon dating helped, especially when combined with increasingly careful study of pottery styles. Now, says Farnsworth, any find from the past 2,000 years in Illinois can be dated within about 200 years. It’s easier than ever before to tell what’s typical and what’s weird. And telling the two apart is essential to this quest.

In the summer of 2000 the ancient road disturbed Mazrim’s work again. He was excavating what archaeologists call a “pit feature” on the site of James Latham’s cabin atop Elkhart Hill. (It turned out to be a shallow cellar used to keep food cool.) In 1819 Latham built a “dog trot”–a double cabin with a roofed breezeway in between that was commonly built by Illinois settlers from upland-south states like Kentucky and Tennesee. His family thus became the first European settlers in Logan County. The Lathams lived right on Edwards’ Trace. A straight line connects their cabin site with a spring, with the two segments of Edwards’ Trace still visible on Elkhart Hill, and with the two-story Kentucky House, a tavern and wayside inn Latham’s son Richard later built alongside the trail.

At Latham’s cabin site Mazrim found the expected artifacts from the 1820s, plus a fragment of pottery with a distinctive crosshatched fabric imprint. That design told him this pot was no 19th-century creation; it had been in the ground a long time when Latham dug his cellar. Mazrim labeled it, put it in a zip-lock bag, and took it to Ken Farnsworth, who advises him when he encounters more prehistory than he can handle.

Farnsworth looked at the piece of pottery and said, “You didn’t find this around Elkhart. You just misplaced it.”

Mazrim knew perfectly well where he’d found it. As all professionals know–and as all amateurs should–if you dig up an artifact and carry it off without documenting its original location and what was there with it, your find has lost its informational value. By taking it out of context, you’ve reduced it to little more than a trinket.

Farnsworth explained why this kind of pottery shouldn’t be in Elkhart. It had been made around 500 BC, by people known to archaeologists as the Black Sand culture. “We know of 114 sites,” he says, “and all but a dozen of them are in the floodplain of the Illinois River”; the rest are nearby. Elkhart, at mile 115 on I-55, is more than 40 miles from the river.

Mazrim countered that the problem wasn’t just a single fragment, which might be explained away somehow. He’d also found pieces of stone tools used in that culture dating to about the same time.

How did Black Sand relics wind up on a hill in the middle of the prairie? People of that era might hunt there, but no hunter in his right mind would drag around heavy, breakable ceramics while trying to run down a deer. Pottery means that people stayed put for a while. The conclusion seemed inescapable: contrary to all expectations, a Black Sand village had existed on Elkhart Hill, far from any water transportation corridor. Yet if it was on an overland corridor its location would make sense. And we know the historic Edwards’ Trace was right there.

Maybe it wasn’t a coincidence that James Latham built his cabin where the Black Sand people had lived. Maybe they both settled by the side of the road–the same road, 2,300 years apart.

This tantalizing notion had already occurred to Mazrim because the Black Sand village wasn’t the first anomalous site to turn up on the trace. In the panhandle of Montgomery County, 20 miles or so south of Springfield and right in the path of Edwards’ Trace, is a village site from Mississippian times, where archaeologists have found 1,000-year-old pottery like that used in Cahokia and probably imported from the great city. How did the pottery get there? It’s a long way to the nearest canoeable stream–but Edwards’ Trace is handy.

In Logan County near the trace is a site called Yellow Bluffs, with 2,000-year-old pottery characteristic of the Havana culture. It doesn’t belong there either. “I’ve mapped more than 350 such sites on the lower Illinois River,” says Farnsworth. “They go up the tributaries, like Macoupin Creek–but nothing like this far inland.”

In 1994 Farnsworth had a personal encounter with the oldest anomaly of all. The town of Bunker Hill in southern Macoupin County had been digging trenches for new utility lines. A local collector knew that previous excavations had uncovered unusual arrowheads. When they turned up again this time, he called Farnsworth. City officials agreed to let Farnsworth reexcavate a trench where the leaf-shaped blades had been found.

Working carefully, Farnsworth and his helpers found two walls of a shaft tomb, where a prominent member of what we call the Prairie Lake culture had been cremated and buried between 900 and 1200 BC. Such tombs–typically about five feet in diameter and seven feet deep –were located far enough underground that it’s impossible to find them in a surface survey. The tomb contained some 200 finely crafted blades, five to six inches long, made of a glossy translucent gray stone, a kind of chert found only in the Ohio River valley in far southern Indiana. (No professional archaeological account of this find has been written yet. Ollie Schwallenstecker, a local writer, has produced a brief nontechnical account that appears as part of the “digital history” of Bunker Hill at

Farnsworth calls the Bunker Hill find “shocking,” but not because the chert used to make the blades came some distance. These people are known to have been traders; their goods included copper and marine shells from even farther away. He’s shocked because such a tomb implies a village nearby, which, once again, just shouldn’t be there. “Every village of theirs that we’ve ever found,” he says, “has been on a bluff top overlooking the Illinois, Mississippi, or Missouri rivers.” Bunker Hill is miles from the big rivers. But it sits squarely on Edwards’ Trace.

Mazrim and Farnsworth don’t claim to have proved that Edwards’ Trace existed in some form a thousand years before the time of Christ. But they think they’re onto something. They’ve come across several old village sites in addition to the four mentioned above–1200 BC (Bunker Hill), 500 BC (Latham’s cabin site), 1 AD (Yellow Bluffs), and 1000 AD (the Mississippian village in Montgomery County). All of them ought to be on a stream or river, but instead they’re on Edwards’ Trace.

The evidence is suggestive, not definitive, because Illinois hasn’t been dug over–not even close. For all we know, there may be equally anomalous prehistoric sites all over the state, so many and so widely distributed that the apparent lineup along Edwards’ Trace would just disappear in the crowd. To take one obvious example, nobody has done archaeological digs all the way from Bunker Hill east to the Wabash River. If such a project were done and no similar villages were found in between, then the hypothesis that this village depended on Edwards’ Trace would be strengthened. If several such villages were found, the hypothesis would be weakened.

So just how old is the I-55 corridor? “I do believe there’s something going on here of real antiquity,” says Mazrim. “But I don’t want to Rorschach it. It’s so vague. It lends itself to finding exactly what you want to find. That’s why I started looking in 1990 and haven’t written anything yet. And that’s why my report will just be a bunch of tidbits. You string ’em up.” 7

For more on finding the trace see the Visitor’s Guide in this issue.

Art accompanying story in printed newspaper (not available in this archive): illustration/Elizabeth M. Tamny; photos/Suzy Poling, Yvette Marie Dostatni.