Glades of Jefferson County

We stood a moment to contemplate the sublime and beautiful scene before us, which was such an assembly of rocks and water—of hill and valley—of verdant woods and naked peaks—of native fertility and barren magnificence… – Henry Rowe Schoolcraft, 1818-1819

In the Ozark Border south of St. Louis, a series of natural openings punctuate the dry, rocky forests of Jefferson County. Commonly called “glades” or “cedar glades,” these islands of prairie in a sea of forest are home to plants and animals more commonly associated with the Great Plains region further to the west. Extending in a narrow arc from central Jefferson County east and south into northern Ste. Genevieve County, these glades occur most commonly on south and southwest-facing slopes below forested ridges and are characterized by thin soils and exposed dolomite bedrock of Ordovician age. Glades are, in fact, a common natural feature throughout much of the Ozark Highlands, an extraordinary plateau where the great eastern deciduous forest begins to yield to the western grasslands. A much more extensive system of dolomite glades occurs in the White River Hills of southwest Missouri, where they often extend up steep slopes and over the tops of knobs to form what Schoolcraft called “naked peaks” and are now called “balds” (and spawning the “Baldknobbers” of Branson fame). Additional glade complexes occur throughout the Ozark Highlands on different rock substrates – igneous glades abound in the St. Francois Mountains, sandstone glades dot the Lamotte landscape in Ste. Genevieve County and the northern and western Ozarks, limestone glades can be found in the northern Ozarks near Danville and Lake of the Ozarks, and chert glades occur in extreme southwest Missouri. These different glade systems share a common feature – shallow soils where tree establishment is limited due to summer moisture stress. They differ vegetationally, however, due to differences in hydrology and soil chemistry as a result of their different substrates. Floristically, dolomite glades exhibit a high degree of diversity relative to other glade types.

The term “glade” is derived from the Old English “glad,” meaning a shining place – perhaps the early settlers found their open landscapes a welcome respite after emerging from the confining vastness of the eastern deciduous forest. Whatever the meaning, the glades of Jefferson County hold a special place in my heart, for I “grew up,” entomologically speaking, in those glades. As a young entomologist, fresh out of school, I spent many a day scrambling through the glades and surrounding woodlands. It was here where my interest in beetles, especially woodboring beetles, was born and later grew into a passion. For eight years I visited these glades often – attracted by the extraordinary diversity of insects living within the glades and congregating around its edges. My earliest buprestid and cerambycid papers contain numerous records from “Victoria Glades” and “Valley View Glades” – the two best-preserved examples of the glades that once occurred extensively throughout the area (more on this later). My visits to these glades ended in 1990 when I moved to California, and although I moved back to the St. Louis area in 1995, the focus of my beetle research has more often taken me to places outside of Missouri. It had, in fact, been some 10 years since my last visit to these glades until last week, when I was able to once again spend some time in them.

Ozark glades differ from the true cedar glades of the southeastern U.S. in that they are not a climax habitat – they depend upon periodic fires to prevent succession to forest. Some recent authors have suggested the term “xeric dolomite/limestone prairie” be used to distinguish the fire-dependent glades of the Ozarks from the edaphic climax cedar glades of the southeast (Baskin & Baskin 2000, Baskin et al. 2007). Fires have been largely suppressed throughout Missouri since European settlement, leading to encroachment upon the glades by eastern red-cedar (Juniperus virginiana). Pure stands of red-cedar have developed on many former glades, crowding out the herbaceous plants that depend upon full sun and leading to soil formation that supports further encroachment by additional woody plant species such as post oak (Quercus stellata), blackjack oak (Q. marilandica), flowering dogwood (Cornus florida), and fragrant sumac (Rhus aromatica) from the surrounding woodlands. Fire has returned to many of the Ozark glades situated on lands owned or managed by state and federal agencies such as the Missouri Department of Conservation, Missouri Department of Natural Resources, and U.S. Forest Service, as well as private conservation-minded organizations such as The Nature Conservancy. These agencies have begun adopting cedar removal and fire management techniques to bring back the pre-settlement look and diversity of the Ozark Glades. This is particularly true at Victoria Glades and Valley View Glades, the two largest and most pristine examples of the Jefferson County dolomite glade complex. Fires have been used to kill small red-cedars in the glades, as well as rejuvenate their herbaceous plant communities. Larger red-cedar trees are not killed outright by fire and must be removed by chainsaws. This above distant view of the TNC parcel at Victoria Glades shows many such burned red-cedars. The glades themselves are not the only habitat to benefit from this aggressive management – when I was doing my fieldwork here in the 1980’s the surrounding woodlands were a closed post oak forest bordered by fragrant sumac and with little or no understory in the interior. The photo at right now shows an open savanna with a rich understory of not only sumac and other shrubs, but also many herbaceous plants as well such as black-eyed susan (Rudbeckia hirta) and American feverfew (Parthenium integrifolium). Such open woodland more closely resembles what Schoolcraft saw across much of the Ozarks during his journey almost two centuries ago.

Victoria and Valley View Glades are dominated by little bluestem (Schizachyrium scoparium), Indian grass (Sorghastrum nutans), big bluestem (Andropogon gerardii) and prairie dropseed (Sporobolus heterolepis). A smaller but highly charismatic non-grass flora is also found on the glades – species such as Missouri evening primrose (Oenethera macrocarpa) (left), pale purple coneflower (Echinacea simulata) (pictured above and below), and prairie dock (Silphium terebinthinaceum) not only add beautiful color but also support both vertebrate and invertebrate wildlife. The Fremont’s leather flower (Clematis fremontii) is a true endemic, occurring only in this part of Missouri and entirely dependent upon these glades for its survival. Less well studied is the vast insect fauna associated with the glades. It is here where I first discovered the occurrence of Acmaeodera neglecta in Missouri. This small jewel beetle is similar to the broadly occurring A. tubulus but at the time was known only from Texas and surrounding states. In collecting what I thought were adults of A. tubulus on various flowers in the glades, I noticed that some of them were less shining, more strongly punctate, and exhibited elytral patterning that was often coalesced into longitudinal “C-shaped” markings rather than the scattered small spots typical of A. tubulus. These proved to be A. neglecta, which I have since found on many glade habitats throughout the Ozark Highlands. Both species can be seen in this photo feeding on a flower of hairy wild petunia (Ruellia humilis) – the lower individual is A. neglecta, while the upper individual and two inside the flower are A. tubulus. Another interesting insect-plant association I discovered at these glades was the strikingly beautiful Dicerca pugionata – another species of jewel beetle – and its host plant ninebark (Physocarpus opulifolius). Only a single Missouri occurrence had been reported for D. pugionata, despite the common occurrence of its host plant along rocky streams and rivers throughout the Ozark Highlands. This plant also grows at Victoria and Valley View Glades along the intermittent streams that drain the glades and in the moist toeslopes along the lower edges of the glades where water that has percolated through the rocks and down the slopes is forced to the surface by an impermeable layer of bedrock. Unlike the tall, robust, lush plants that can be found in more optimal streamside habitats with good moisture availability, the ninebark plants of Victoria and Valley View Glades are small and scraggly, usually with some dieback that results from suboptimal growing conditions. I surmise these plants have reduced capabilities for fending off attacks by insects, including D. pugionata, and as a result a healthy population of the insect thrives at these glades. Some might be inclined to call this beetle a pest, threatening the health of one of the glade’s plants. In reality, the insect finds refuge in these glades – unable to effectively colonize the vast reserves of healthy plants that grow along streams throughout the rest of the Ozarks, it strikes a tenuous balance with plants that are themselves on the edge of survival.

Despite the success in moving Victoria and Valley View Glades closer to their pre-settlement character, the integrity of these areas continues to be challenged. Poachers take anything of real or perceived value, and ATV enthusiasts view the open spaces as nothing more than tarmac. Pale purple coneflower occurs abundantly on these Jefferson County glades (but sparingly in other habitats – primarily rocky roadsides), where they provide a stunning floral display during June and sustain innumerable insect pollinators. Plants in the genus Echinacea also have perceived medicinal value, as herbalists believe their roots contain an effective blood purifier and antibiotic. There are no conclusive human clinical trials to date that fully substantiate this purported immune stimulating effect (McKeown 1999). Nevertheless, demand for herbal use has skyrocketed in recent decades, prompting widespread illegal harvesting of several coneflower species throughout their collective range across the Great Plains and Ozark Highlands. I witnessed massive removals of this plant from both Victoria and Valley View Glades during the 1980’s, but the pictures I took this year suggest that such illegal harvests have been suppressed and that the populations at both sites are recovering nicely.

The same cannot be said for the practice of rock flipping. This was a problem I witnessed back in the 1980’s, and I saw fresh evidence of its continued occurrence at both sites. The thin soils and sloping terrain leave successive layers of dolomite bedrock exposed, the edges of which shatter from repeated freeze-thaw cycles to create rows of loose, flat rocks along the bedrock strata. Lizards, snakes, tarantulas, and scorpions find refuge under these loose rocks, only to be ripped from their homes by flippers and transferred to a dark, cold terrarium to endure a slow, lingering death. As if poaching the glade’s fauna and watching them slowly die isn’t bad enough, the flippers add insult to injury by not even bothering to replace the rock in its original position after stealing its inhabitant, amounting to habitat destruction three times greater than the area of the rock itself. Firstly, the habitat under the rock is destroyed by sudden exposure of the diverse and formerly sheltered microfauna to deadly sunlight. Next, the habitat onto which the rock is flipped is also destroyed, as the plants growing there begin a slow, smothering death. Lastly, the upper surface of the rock, sometimes colonized by mosses and lichens that might have required decades or longer to grow, usually ends up against the ground – its white, sterile underside becoming the new upper surface. Rock flipper scars take years to heal, and nearly all of the flat, loose rocks seen in the more accessible areas of the glades exhibit scars of varying ages next to them. If a scar is fresh (first photo), I generally return to the rock to its original position – the former inhabitants cannot be brought back, but at least the original habitats are saved and can recover quickly. However, if a scar is too old (2nd photo) it is best to leave the rock in its new position – replacing it only prolongs the time required for recovery.

Even more damaging is ATV use. Herbaceous plants and thin soils are no match for the aggressive tread of ATV tires, and it doesn’t take too many passes over an area before the delicate plants are killed and loose soils ripped apart. I witnessed this become a big problem particularly on Victoria Glades during the 1980’s – actually finding myself once in a face-to-face confrontation with an ATV’er. Fortunately, he turned tail and ran, and it appears (for now) that such abuses have stopped, as I saw no evidence of more recent tracks during this visit. But the scars of those tracks laid down more than two decades ago still remain painfully visible. I expect several more decades will pass before they are healed completely.

My return to Victoria and Valley View Glades was a homecoming of sorts, and I was genuinely pleased to see the progress that has been made in managing these areas while revisiting the sites where my love affair with beetles was first kindled. Sadly, however, the larger glade complex of Jefferson County continues to deteriorate. Restoration acreage aside, red-cedar encroachment continues unabated on many of the remaining glade parcels – large and small – that dot the south and southwest facing slopes in this area. It has been conservatively estimated that as much as 70% of the original high quality glades in Missouri are now covered in red-cedar. Many of these are privately held – their owners either do not recognize their ecological significance or are loathe to set fire to them. An example can be seen in the picture here – this small parcel is part of the Victoria Glades complex but lies on private land in red-cedar choked contrast to the Nature Conservancy parcel immediately to the south. Small numbers of herbaceous plants persist here, but without intervention by fire or chainsaw their numbers will continue to dwindle and the glade will die. Aside from the loss of these glades, the continuing reduction of glade habitat complicates management options for preserved glades as well. Many glade associated invertebrates are “fire-sensitive” – i.e., they overwinter in the duff and leaf litter above the soil and are thus vulnerable to spring or fall fires. While these fires are profoundly useful for invigorating the herbaceous flora, they can lead to local extirpation of fire-sensitive invertebrate species within the burn area. Recolonization normally occurs quickly from unburned glades in proximity to the burned areas but can be hampered if source habitat exists as small, highly-fragmented remnants separated by extensive tracts of hostile environment. Grazing also continues to threaten existing remnants in the Jefferson County complex. Grazing rates are higher now than ever before, with greater negative impact due to the use of fencing that prevents grazers from moving to “greener pastures”. Over-grazing eliminates native vegetation through constant depletion of nutrient reserves and disturbance of the delicate soil structure, leading to invasion and establishment of undesirable plant species. Eventually, the glade becomes unproductive for pasture and is abandoned – coupled with fire suppression this leads to rapid woody encroachment. It is truly depressing to drive through Jefferson County and recognize these cedar-choked glades for what they were, able to do nothing but watch in dismay as yet another aspect of Missouri’s natural heritage gradually disappears. The continued loss of these remnant glades makes careful use of fire management on Victoria and Valley View Glades all the more critical – ensuring that a patchwork of unburned, lightly burned, and more heavily burned areas exists at a given time will be critical for preventing invertebrate extirpations within these managed areas.

I close by sharing with you a few more of the many photographs I took during this visit – stiff tickseed (Coreopsis palmata), three-toed box turtle (Terrapene carolina triunguis), climbing milkweed (Matelea decipiens – see the excellent post about this plant on Ozark Highlands of Missouri), downy phlox (Phlox pilosa), green milkweed (Asclepias viridiflora), and a “deerly” departed native browser.

Ozark Trail – lower Courtois Section

The Courtois Section is the northern terminus of the Ozark Trail (OT). Despite its proximity to the St. Louis metro area, it feels just as remote and wild as the more southern sections. Rich and I played hooky from work on Friday and made our first visit to this stretch of the Ozark Trail. At 40 miles in length, we’ll need to break it up into at least three parts, so for our first attempt we hiked the lower portion from Hazel Creek (where the Trace Creek section begins) north to the Hwy 8 trailhead. Apparently this portion of the OT is very popular with mountain bikers and equestrians; however, we didn’t encounter a single person all day.

I expected the terrain to be rather mild at this northern end of the OT, but the first few miles were quite up and down. There was still some snow on the ground from a big storm a few days earlier – mild temps and sunny skies since then had caused a lot of melt. As a result, south facing slopes were completely devoid of snow cover, while north facing slopes still had and inch or two of snow, creating “split” scenes such as this:

Right away we noticed a lot of fresh woodpecker damage on oak trees. This is likely the result of infestations by the red oak borer (Enaphalodes rufulus), a cerambycid beetle that preferentially attacks red and black oaks suffering from drought or other environmentally-induced stress. The larvae of these beetles mine beneath the bark on the trunks of these trees before tunneling into the sapwood to pass the winter. Overwintering larvae are tasty morsels for woodpeckers, who hammer into the trunks with their beaks and extract the larvae with their barbed tongues. Interestingly, conventional wisdom has it that the tongue “stabs” the larva, and the barbs aid in pulling the larva out of its gallery. However, recent experiments with a West Indian species suggest this is not the case. Rather, the larva “sticks” to saliva on the tongue, and the barbs help to grab the larva as the tongue is wrapped around it. This picture shows a small black oak (Quercus velutinus) tree with fresh damage, probably from a pileated woodpecker (Dryocopus pileatus) judging by the size, going after one of these larvae.

A few miles into the trail, we came upon some curious “pits” covering one hillside. We speculated what they might be – sinks was an early thought, but I didn’t think that was so because the ground was mounded around the edge like they had been intentionally dug. Rich then remembered reading something about miners digging such pits in past years looking for minerals – we decided that must be what they were, and this was later confirmed in our Ozark Trail guidebook. Certain hillsides were literally covered with these pits, spaced ~10-15 feet apart.

After passing through Snapps Branch (where we noticed a small calcareous wet meadow, or fen – thankfully fenced), the trail leveled out for awhile before descending down to Boiling Springs Hollow where we stopped for lunch. Many of the larger valleys along the OT show some evidence of prior habitation – either by remains of old structures or by the stage of succession exhibited by the bottomland forest. Right at Boiling Springs, I noticed this large, old oak tree along with several large sugar maples (Acer saccharum) surrounded by younger forest – I suspect these “founder trees” were planted at some point when people lived near the spring (or at least spared from “the saw”) and remain as the only evidence of the people who lived here in the past.

I love bones and pick them up whenever I get the chance. After leaving Boiling Springs I noticed this half mandible of a white tailed deer (Odocoileus virginianus) laying on the trail, still partially embedded in the snow. It was remarkably clean and complete, containing all of its dentition and with no remaining tissue except for a small piece attached to the nerve fossa. It’s completeness begged the question – where was the other half? We looked around and couldn’t find it. We then wondered if it had been dragged there by a scavenger, although we thought that if that was the case it should show signs of gnawing or at least have lost some of its dentition. At any rate, I have a white tailed deer cranium in my collection but not a mandible, so this will be a welcome addition.

Eventually we entered Machell Hollow, where we followed a beautiful stretch through the upper reaches of the valley. In this area we noticed a large number of dead white oaks (Quercus alba) that were all about the same size (~4-8″ dbh) and in about the same stage of decay, as if they had all died about the same time (maybe 4-5 years ago). There were still plenty of larger living trees, and I began to suspect that a fire had moved through this area and began looking for the evidence. Soon we found several larger trees showing some blackening around the base of the trunk that seemed to confirm this thought. We had a lot of fun “pushing over” some of these trees, with one in particular probably representing our champion pushover to this point. I didn’t think it was gonna go, but Rich chipped in, and against our formidable combined weight the tree gave way and came down with a crash. I noticed evidence of tunneling by wood boring beetles (probably a species of Buprestidae) inside the trunk of this tree where it cracked upon falling and lamented that I could not take a piece with me for rearing. All of the dead white oaks had this one type of shelf fungus growing from their trunks, which were particularly numerous on this already fallen tree:

Climbing up (briefly) out of Machell Hollow, we saw this cut shortleaf pine (Pinus echinata) laying by the side of the trail. Interestingly, the accumulated ice on the cut end of the trunk was not the result of water running off the trunk, but through the trunk, apparently through insect galleries and perhaps even the vascular bundles of the wood itself. The slow melt and freeze resulted in these interesting little ice columns joining the trunk to the moss-covered ground below.

Back down into the lower reaches of Machell Hollow, evidence of prior settlement was obvious, as the bottomland forest in this area was replaced by young successional forest comprised primarily of chokecherry (Prunus virginiana), honey locust (Gleditsia triacanthos), and brambles (Rubus sp.). We saw this lone little fruticose lichen growing on a small honey locust. Apparently, of the three main groups of lichens, fruticose lichens are the most sensitive to environmental disturbance. Perhaps the existence of this one colony suggests that the health of this bottomland forest is returning as succession proceeds along the path to maturity.

Here’s a picture of Rich taking his own picture of the lichen. I don’t know why he didn’t just wait and steal mine once it got posted 😉

Much more abundant on the honey locust trees were these foliose lichens. Lichens in this group are probably the most commonly noticed lichens in the Missouri Ozarks (although the less conspicuous crustose lichens may actually be more diverse). If you click on the photo to see the full-sized version, you can see long, black “hairs” around the margin of each “leaf” – if anyone knows the identity of this or any of the other lichens pictured on this site please let me know.

While ascending out of Machell Hollow, we noticed this small canyon about a hundred yards off to the left and decided to go investigate. Along the way we noticed the small creek coming from it was actually a ‘losing creek’ – which means that the water flows into the ground at certain points and is ‘lost.’ This is another feature of the limestone/dolomite-based Karst geology so common here in southern Missouri that results in its abundance of caves and springs. When we got to the canyon we saw it was comprised of a layer of sandstone. This must be a rare western exposure of the LaMotte sandstones that are more common just to the east in Ste. Genevieve County (see earlier posts on Hawn State Park and Pickle Springs Natural Area). This sandstone layer overlying dolomite has created an interesting geological feature, where a losing creek originates from a box canyon. Ice stalactites were dripping from the north facing slope of the canyon walls.

Back down into another hollow leading to Lost Creek we saw more dead white oaks with shelf fungi growing from the trunks. This one was interesting in that the shelf fungi were themselves supporting the growth of algae on their surface – an exquisite example of the interconnectedness of life.

We had seen a flock of wild turkeys (Meleagris gallopavo) moving through the forest earlier in our hike. We were too clumsily noisy to get close enough for more than a cursory look at them as they trotted off on high alert, but evidence of their activity was obvious as we saw their fresh “scratchings” over a wide swath through the forest as they searched for acorns to eat. Tracks were abundant in the snow around the area also, but I couldn’t get a good picture of them. Later, as we neared Lost Creek, I saw more tracks in the mud, so I was able to get a good picture of one. It looked fairly fresh (well defined, with nail holes evident):

Lost Creek represented the end of our hike, but it proved to be a more than insigificant final hurdle, as the water level was quite high due to all the recent snow melt. There was no choice, we would have to get wet. Rich is smarter than I and had thought to bring along some flip flops, so he took off his boots and socks, rolled up his pants, and forded the creek. I let him go first to see how deep the water was – it reached above his knees and got is rolled up pants wet. I decided to get my boots wet – I didn’t want to walk on those rocks barefoot, which would slow me down far more than I wanted in that cold water. I could handle wet boots for the final quarter mile in exchange for the comfort and speed they would provide on the rocks. Rich may be smarter, but I took a better line and didn’t even get my pants wet, so for me it was only a matter of changing into my comfy shoes back at the car, with no need for a change of clothes (which I also wasn’t smart enough to bring, either). We completed the hike in 7 hours – yes, we’re lollygaggers, constantly distracted by little things that most people either don’t see or don’t care about. It was a wonderful hike on another beautiful day, and we ended it with another traditional post-hike visit to the nearest pizza parlor before the short drive back to St. Louis.