Wisconsin Pine Barrens

Historical Accounts of the Northwest Sands

Herbert Wagner



Detailed early accounts of the Wisconsin's Northwest Sands (or "pine barrens" as the area was long known) are rare. Most early visitors traveled by Indian canoe and therefore stuck close to the Bois Brulé, St. Croix, and Namekagon water routes that cross the region. Except for a few cases they didn't see or record much about the upland areas away from these streams. While sparse, the information that exists is important in understanding the ecological history of this landscape.

Buffalo Range?

With a long history of wildfire due to sandy drought-prone soils, broad upland areas of the Northwest Sands were prairie-like or semi-open pine/oak savanna in the pre-contact period. This grassland and brush prairie habitat brings up the intriguing possibility whether the Northwest Sands was once habitat and home to the wild American bison.

In his classic 1937 study "The Range of Bison in Wisconsin," A.W. Schorger places traditional bison habitat south of the pine barrens of Washburn, Douglas, Bayfield, and Burnett counties. He then contradicts this when interpreting Radisson and Groseilliers' account of their Third Voyage of Exploration in the region during 1658-1660. In that account Radisson states that "Buffs" sometimes reached the shore of the "upper lake" (Lake Superior) by chance and he also provided details about a winter buffalo hunt. In his analysis, however, Schorger states that after "careful reading of [Radisson's] voyage leads to the conclusion that his first experience with the buffalo was obtained on the upper Mississippi, probably near the Bois Brule-St. Croix waterway." Clearly, this would place bison range in the 1600s north of Schorger's line across the Northwest Sands and towards the Bayfield Peninsula and Lake Superior. Indeed, it is thought that Radisson and Groseilliers wintered on Chequamegon Bay near the site of modern Ashland in 1659-1660 and explored the country to the south and west, possibly as far as Mille Lacs in Minnesota.

Radisson's account also agrees with Ojibwe Indian tradition recorded in the 1850s by William Warren who stated: "The buffalo, also, are said in those days to have ranged within half a day's march from the lake [Superior] shore, on the barrens stretching towards the headwaters of the St. Croix River."

According to Warren's chronology, wild bison on the Northwest Sands would have dated to the 1790s. This is backed up in a published letter or report written at La Pointe in 1843 by Alfred Brunson. When describing the portion of Wisconsin Territory tributary to Lake Superior, Brunson wrote: "tradition states that within a century past the prairies extend[ed] far past their present limits, and even to the Lake shore over which the Buffalo roamed in large droves."

In 1820, a member of the Gov. Cass expedition, while at La Pointe on Lake Superior, recorded that the Ojibwe Indians there still performed "the buffalo dance." This tradition suggests that in the not too distant past previous to 1820 they had hunted bison in the region and had strong memories of it. This evidence also agrees with an 1874 account obtained from an old Ojibwe woman living at Wood Lake in Burnett County. She was said to be 110 years old at that time and could remember when the Ojibwe people drove the Dakotahs out of the region and when her people wore blankets "made from the skins of the moose, elk and buffalo."

Although such evidence is sketchy and ancedotal, it seems that Schorger should have drawn his bison line with a tongue-like extension northeasterly corresponding to where the brush prairie and pine barrens of the Northwest Sands traditionally extended in Wisconsin nearly to the shores of Lake Superior.

Due to intense hunting pressure, however, these large grazing animals were increasingly scarce here after 1800 and were soon extirpated from northwestern Wisconsin entirely. In 1820 Henry Rowe Schoolcraft reported that the eastern fringe of bison range had shifted into Minnesota. On his 1821 map Schoolcraft marked the area west of the St. Croix River and north of modern St. Paul as "Buffalo prairie." In 1834 he defined the northern point of "buffalo prairie" as near Sandy Lake in Minnesota. By that time, it would seem, bison were no longer regular inhabitants of the Northwest Sands. Backing this up is Governor Dodge's statement to Racine's Dr. Hoy when he said that the last time a buffalo was killed on the Wisconsin side of the St. Croix River was around 1833. However, as late as 1836 when Wisconsin Territory was established, one report advised that, "In the vicinity of Lake Superior there is a fine section of the country for hunters -- buffaloes, elks, bears and deers are common, and afford fine sport, as well as great profit to trappers and hunters." By 1850, elk were still a presence on the pine barrens around the St. Croix River, but by that date bison had been exterminated towards the west into Minnesota until one had to go to the head of the Red River near the Dakota border in order to hunt them.

The only evidence remaining today that the Northwest Sands was once American bison habitat is the Ojibwe word for buffalo -- Pijiki -- on a Douglas County lake and river as seen on a few early maps. Sadly, the name Pijiki was later mistranslated as "Ox Lake" and "Ox River" and remain that way on maps today. There is also a Great Spirit Rock in this same area known to have sacred meaning for Native peoples that some have interpreted as resembling in shape the head of a great bison.

(Detail from Nicollet's 1843 map shows Ojibwe word "pijiki" for buffalo along with location of Kabbamappa's village and the trail across the barrens to Lake Superior. The name of this ancient byway appears to be "Women Portage" and is probably another old Indian reference.)

19th Century Accounts

The journals from Schoolcraft's 1831 expedition give the first good description of the Northwest Sands. Note that even as the first Euro-Americans entered the region fire was observed as being a dominant feature of the landscape. By this time the buffalo had been exterminated and whitetail deer may have been taking their place. In the following passage Schoolcraft describes an area adjacent to the Namekagon River in Washburn County: "The country as we descend assumes more the appearance of upland prairie from the repeated burning of the forest. The effect is that nearly all of the small trees have been consumed and grass has taken their place. One result of this is that the deer are drawn up from the more open parts of the Mississippi, to follow the advance of the prairie and open lands towards Lake Superior."

Edmund Ely was a missionary in the region during mid-century. He made several trips between Fort Snelling (St. Paul) and LaPointe on Lake Superior (Madeline Island/Bayfield). The early Indian trail between these two points (later a stage route) roughly followed the St. Croix River north and then veered northeast across the barrens in Burnett, Washburn, Douglas, and Bayfield counties. During one trip in 1834 Ely gives this account: "This morning came on to Kaboniob's village before breakfast; no Indians at the village; we followed up the valley of the St. Croix traveling on the edge of the high land until noon then struck the plains; encamped at night on the north shore of a beautiful lake."

Notice that Ely refers to an area of "plains" clearly indicating an absence of thick forest. No large herbivores are mentioned. The location of Kaboniob's (Kabbamappa) village was on the St. Croix River a few miles west of modern Gordon. The next day Ely traveled three hours on the plains before he entered "the wood country."

Writing in 1843, Alfred Brunson gave this brief description in a work titled Northern Wiskonsan: "A little west of LaPointe, and ten to twelve miles south of the Lake [Superior] shore, the prairie country commences, which extends to and beyond the St. Croix and Mississippi, and offers great inducements to agriculturists..."

In 1846, a traveler who signed himself "Rambler" left his impressions of going on foot from La Pointe to the Upper St. Croix River near modern Gordon. In part he wrote: "There is one singular spot we passed. It is what is called a prairie or barren, being about four miles across, covered with a stunted growth of grass, and a few stunted shrubs. It reminds one of the moors in Scotland...To help the resemblance, whirr! goes a grouse, and then another! They are marked down--the repeater, which has three charges left in it, is unslung, and presently, bang! bang! we have two grouse for supper, and these, with a pheasant shot by Le Gros, make a beautiful meal, and we feast at night."

In 1848 another traveler (an Englishman) described the change in landscape between LaPointe and the St. Croix River: "Towards three o'clock we came upon an elevated table-land, more free of timber than any we had yet seen, and for several hours made good progress. On this plain, which was covered with a coarse description of grass, with occasional clumps of trees, we saw many pigeons and some wild turkeys, and in the distance deer, but these disappeared on our approach."

In 1852 an Austrian traveler noted a distinct change in landscape as he ascended the upper Bois Brulé River and entered the Northwest Sands region: "At points, where the dense forest grew thinner, the land flatter, and the river broader...the region assumed a more refined and park-like appearance...and gradually changed from the mountainous landscape of Lake Superior into the flat prairie lands of the west."

This choice of terms that the forest grew "thinner" and the region more "park-like" leads one to suspect that pine savanna rather than open prairie existed along the Upper Bois Brulé.

In 1858 a government surveyor made the following observations on a trip across the Northwest Sands to Bayfield. Note that he distinguishes "Big Woods" from "barrens" in this passage:

April. 6: Camped 8 miles from Nimekagon.

April 8: Camped at Ox lake.

April 9: Camped in Big Woods. No water.

April 10: Met Larush & Steve with mail.

April 11: Crossed State Road Line & camped on Barrens.

April 12: Got to Pikes at sundown.

A Fire Blessed Landscape

The impact of large burn events and the fire-dependent regeneration of pine (especially jack pine) on the Northwest Sands is clearly evident in an 1863 account recording a stage journey from St. Paul to Bayfield. Here the country and events above the confluence of the Namekagon and St. Croix rivers are described: "After driving two miles [we] became aware of a proximity to burning woods. Trees and grass in flames seemed to surround us. As we drove on the fire extended to the right and left...After passing the burning district we came to a country where we saw numberless evergreens, occasionally many acres being overgrown with young pines...Then, again, appeared a large district covered with half-burnt trees -- charred trees still standing, others lying on the ground in wild confusion -- no signs of vegetation to be seen. We passed numerous small lakes, many of them very beautiful...But the conductor advised going as far as Antoine Gordon's, the usual stopping-place."

Scientists were late in coming to the region, but when they finally arrived they left some pretty good accounts. This one from the 1870s describes the Northwest Sands around the St. Croix headwaters: "The soil is sandy and barren, supporting only a stinted growth of 'jack' pines and 'scrub' oaks. Fire has killed the timber over wide areas, on which grass was growing exhibiting before our eyes natures's simple method of converting woodland to prairie. The reverse process is just as simple. When prairies are no longer swept by fire, timber springs up reconverting prairie into woodland. Grass, with fire as an ally, can beat timber. Timber can beat grass when it has no fire to fight."

Expanding on the this theme of prairie returning to forest another wrote: "Analogous to the 'barrens' are the areas known as 'brush prairies' and simply 'prairie.' These are covered with a scattered growth of shrubs that are usually associated with the more open timber [pine savannah] of the region, but fully developed trees are absent. On some of these 'prairies,' however, young trees are springing up, and bid fair, if undisturbed, to attain the usual size. These have been appealed to as examples of prairies returning to forest, since annual fires are no longer permitted to ravage the region...which, on account of the character of the soil, was especially subject to dryness, and thus to the destructive action of the annual fires; while moister adjoining areas escaped."

In 1888 one writer described the pine barrens as: "sandy stretches of undulating, though sometimes...level lands, sparsely covered with a growth of young pines, generally of the Black Prince variety. In some places, where the trees are crowded thickly together, they are not unlike immense cane-brakes. The trees, from their proximity, have grown very tall and slender. The lateral branches, crowded together and deprived of sunshine, have perished early and the growth of the young trees is chiefly vertical. The lower tree limbs remaining attached to the trunks give the young forest a peculiarly ragged and tangled appearance. There is abundant evidence to prove the existence of ancient pine forest where these pine barrens are now the only growth. In fact some of the larger trees are still standing, and the charred trunks and decaying remnants of others. The graduations from the younger to the older growth may be very plainly seen. Fire is undoubtedly the efficient cause of the stunted and irregular growth of the pine barrens. The matured forest are destroyed by fire and are succeeded by the young pines which are further reduced and injured by annual fires....In most of them there is a dense undergrowth of blueberry bushes producing....small but luscious fruit."

From these descriptions we see a complex and ever-changing pattern of fire and vegetation across the historical landscape of the Northwest Sands. Fires frequently swept the unobstructed and more level parts of the region resulting in a mosaic of grasslands, brush prairie, and youthful regenerated jack pine along with areas of older pine woods. In other places where widely spaced or groves of large bur oak or red pines had become established and that were impervious to groundfires, big tree savannah was encountered. In some areas topographic disruptions such as lakes, rivers, and swamps acted as natural fire-breaks. These more protected areas, often with somewhat moister soil, resulted in tracts of more closed canopy white and red pine forest. Surprisingly diverse plant communities were described in these areas by naturalists. One wrote: "The timber occupying these tracts is peculiar...some portions are covered with scrub pine to the exclusion of all else save underbrush. Most nearly similar to these are patches of Norway pine. Other areas are covered with burr, black, and even white oak bushes, with occasional trees of these species. With these are associated the common white poplar, or trembling aspen...the great toothed aspen [also] associated with it....also areas where white pine occurs associated with both poplars and the three species of oaks...with the soft maple, and with scrub and red pines, forming a very strange association of plants that usually seek quite diverse conditions of soil and moisture."

Modern Era

By the 1850s, the Native American practice of burning the grass prairies was ending. One observer noted: "As the Indians get driven back, the prairies in consequence escape the annual burning, and trees spring up." This lessening of annual burns may reflect the changing native presence on the Northwest Sands from the more prairie-oriented Dakota peoples to the more forest and water-oriented Ojibwe during the 18th and 19th centuries. Perhaps the Dakota were more prone to use fire to alter the landscape than the Ojibwe. In any case it seems evident that the open areas were already descreasing in size by the mid to late 19th century as young forest began to grow in the absence of annual low intensity fires. More fuel in the form of young pine trees, however, would lead to hotter and more severe burns in decades to come.

Loggers first entered the upper St. Croix basin in the late 1850s. For the next 50 years they carried out a rapacious cutting of the big white and red pine timber. Actual homesteaders, however, wouldn't enter the Northwest Sands until the late 1890s. To what extent the area as a whole was altered during the logging decades has not been well documented, but changes must have been profound as new young forest, logging debris, and more human activity resulted in larger and more intense forest fires.

One 1909 glimpse of a logging camp near the edge of the Northwest Sands records the course of several wildfires over a span of several days:

26 April: Big fire from south.

3 May: Big fire in SE.

5 May: Big fire in SW.

6 May: Fire to SE & North. Solid wall of flame. Went out on East Trail to see if fire could reach camp, burning in big timber 1-1/2 miles to NE, 2 m. to E. & 2 M. to South. Stayed up all night and watched.

7 May: Fire coming from South.

8 May: Fire burning on Dingle [Creek]. 8 pm: Call hands, back firing...backfired all night.

9 May: Fire coming now from NE. Backfired near wood piles until 2 pm. -- Rain!

After 1900 the region around Gordon was described in this manner: "Land south and east of Gordon are sand and of uncertain value...Most of the area has been burned over and covered with ferns and sunflowers, grasses, and weeds. South in Washburn County such land has been plowed up and made into beautiful farms."

But most of those "beautiful farms" were also drought-prone and wore out after a few years of cultivation. Well into the 20th century wildfires still dominanted the landscape. An original "jackpine farmer" named Heinrich Steinhilpert who lived into the 1990s to age 102, once told me: "In 1910 there was no snow and that spring a big fire came through. It started west of here where Hoosier was burning brush on a homestead. The fire went south but the next day the wind switched and it came back here right up to our doorstep. The only fire protection them days was a shovel and lots of energy."

No snow, logging slash, poor soils, and settlers clearing land resulted in fires becoming a spring ritual. And the Northwest Sands could not sustain this early small-scale farming industry. Steinhilpert put it this way: "This country dried up after World War One. Hanson tried to raise sheep but they grazed the grass so short they ate sand and died."

In the 1930s wildfire prevention and suppression became the norm. Since that time the former open and fire-swept lands have been planted or have grown up into pine, oak, and aspen forest. The only areas that resemble the pine barrens of old are fire-managed DNR lands at Moquah, the Douglas County Bird Sanctuary, Namekagon Barrens, and Crex Meadows. These open prairie-like sweeps of countryside are but small remnants of the much greater original barrens, but still portray how the region may have looked historically back when fire ruled the landscape. Certain plant communities such as big tree red pine savannah have vanished almost entirely. The bison are gone too, forever it seems, and an increasing human population will restrict most fires to open DNR lands. Except for that rare large wildfire that erupts on occasion, defying Smokey Bear's dictat and brushing aside all human endeavor to stop it as it runs amok through the piney woods.

Click to return to: Herb's Woods!

Click to return to Herb's "At the Creation" hompage!

Contents copyright 2006-2008 Herbert Wagner. All rights reserved.

Created on a Apple Laptop Computer Using Carbon Neutral Electricity Generated by the Sun!

Updated January 30, 2008