When I talk to people around Arkansas about the ANHC’s work they are often surprised to hear that ecologists are still learning new things about our state’s natural history; most people assume that there aren’t new discoveries to be made. However, that isn’t the case, as exemplified by one of the most exciting Arkansas natural history discoveries in recent decades, the description of the Ozark Pocket Gopher (Geomys bursarius ozarkensis).
A subspecies of the Plains Pocket Gopher, the Ozark Pocket Gopher is known only from a small area in southern Izard County, and nowhere else on Earth. This became official with the publication of an article by Douglas Elrod, Earl Zimmerman, Philip Sudman, and Gary Heidt in the August 2000 issue of the Journal of Mammalogy. These unusual burrowing rodents have coats camouflaged to match the color of the soil where they live and spend most of their lives in underground tunnels, emerging briefly to gather plant foods, which they transport back to their burrows in large fur-lined cheek pouches, or pockets. In addition to having a very narrow range, the Ozark Pocket Gopher has the distinction of being the only mammal that is endemic to, or found only in, Arkansas. The occurrence of an endemic pocket gopher in this particular region of the state, however, is somewhat unexpected and presents something of a mystery – one that has really captured my interest over the past few months.
The presence of narrowly endemic species, those restricted to a small geographic area (and usually also to a specific uncommon or rare habitat), is evidence of great antiquity. Simply put, these unique species or subspecies didn’t just show up recently – their habitat must have been there for a long time. And therein lies the mystery. Pocket gophers are grassland species, preferring open prairie or savanna habitat with only scattered trees. For the endemic subspecies to occur in southern Izard County, there must have been some kind of suitable natural grassland habitat there in ancient times, and at least some of it must have been present up until modern times or they would have gone extinct. So, what is (or was) that habitat?
The literature on the Ozark Pocket Gopher gives its habitat as something along the lines of “pastures and crop fields with deep sandy soil,” and this is an accurate description of where they are found today. More specifically, they occur within these habitats on sandy, elevated terraces along a series of streams that drain southern Izard County, flowing southward into the White River. The mystery is that we have no historic records of there being grasslands in this area historically. They weren’t mapped by the General Land Office (GLO) surveyors who completed the original survey of the Louisiana Purchase (one of our best sources of information on the locations of historical grasslands in Arkansas), nor were they described in the accounts of early naturalists, explorers, or settlers (at least so far as we are aware). We knew there were scattered glades in the hills around these valleys, but these are rocky grasslands with bedrock at or near the surface of the ground and not suitable for pocket gophers. So, what were the sandy, valley grasslands that these gophers called home before humans were here to maintain open pastures and crop fields for them? And are there any remnants of these grasslands left that we could study?
Remnants of ancient, natural grasslands are always interesting botanically, and usually have uncommon plants that aren’t found in other habitats. Concentrations of these grassland indicator plants are the hallmarks of such a remnant, and I knew it would be these botanical clues that could help solve the mystery of the missing grasslands. So back in May I began exploring a couple of interested landowners’ properties on these creeks, hunting for grassland indicator plants and trying to get my head around what grasslands in the area might have been like historically. I knew there wouldn’t be large prairies (if those had existed, they would have made it onto the maps), but I had a hypothesis that maybe there were once sandy savannas and open woodlands throughout these valleys, with small sandy prairies here and there. These would have distinctive plant communities made up of sun- and sand-loving plants.
From studying grassland ecosystems in other areas, I knew that the usual pattern of loss is that open prairies got plowed up for crops or converted to non-native pasture grasses early on and that savannas and open woodlands either got cleared and converted to pasture or have since grown up into dense forest. In the latter case botanical clues can often still be found along open roadsides and utility rights-of-way through these forests, so that was my first strategy – run the roads and look for grassland indicator plants. Unfortunately, a lot of these have been sprayed with herbicide in recent years, reducing rich pockets of diverse grassland flora to low-diversity patches of the weediest, most common, and often invasive, species. But before long I found some clues – big stands of little bluestem (Schizachyrium scoparium), one of our most common denizens of dry grasslands, along with some grassland-obligate wildflowers like pale purple coneflower (Echinacea pallida) and old-plainsman (Hymenopappus scabiosaeus). This was encouraging.
The first property I explored, accompanied by Kyle Lapham (ANHC’s Chief of Acquisitions and Stewardship), didn’t have any significant areas of open ground, but it did have a very sandy forested stream terrace with a heavy component of eastern redcedar (Juniperus virginiana), a common invader of formerly open savannas and woodlands. A quick study of the trees in this forest showed the typical pattern of invasion we see in savannas and open woodlands – scattered larger (and older) individuals of oaks and pines with a more-or-less even-aged stand of smaller, younger cedar filling in between them. This cedar-encroachment generally happened in conjunction with open range grazing and widespread fire suppression following Euro-American settlement.
The soil here was very loose and sandy, and even though the site was nearly flat and located right along a stream, the non-cedar trees were species typical of well-drained, even drought-prone soils: shortleaf pine (Pinus echinata), post oak (Quercus stellata), black oak (Quercus velutina), and chinquapin oak (Quercus muehlenbergii). The shrub component in these woods was right out of the sandhill woodlands of southwest Arkansas’s West Gulf Coastal Plain: fringe tree (Chionanthus virginicus), dwarf hawthorn (Crataegus uniflora), Carolina holly (Ilex ambigua), farkleberry (Vaccinium arboreum) … I felt like I had suddenly been transported down to Miller County Sandhills Natural Area on the Texas border. But my eyes were really drawn to the ground, where the real plant diversity is found, and a few small semi-open pockets of woodland had a little bit of what I was after.
There was more little bluestem, Carolina Larkspur (Delphinium carolinianum), and bastard toadflax (Comandra umbellata) – sure signs that this area was once more open than it is today. There was a lone, shade-suppressed clump of eastern gamagrass (Tripsacum dactyloides), another prairie species, which ranchers sometimes call “ice cream grass” because it’s the first species eaten by livestock if it grows in a field. My pulse really started moving when I found a few plants of the rare hairy rockcress (Arabis pycnocarpa var. adpressipilis), a species of state conservation concern previously known in Arkansas from just a handful of sites. In 25 years of hunting plants in the state I have only seen this at one other site!
There was also a nice list of psammophytes (sand-loving plants). These were the ones I was most interested in finding, as they fit in with my hypothesis that there were ancient sand grasslands in these valleys. Again I was transported to the southwest Arkansas sand barrens as I documented species after species of sand-lover: Texas green-eyes (Berlandiera betonicifolia), erect dayflower (Commelina erecta), velvet-leaf tick-trefoil (Desmodium viridiflorum), woolly rosette grass (Dichanthelium villosissimum), cream avens (Geum virginianum), hairy wild lettuce (Lactuca hirsuta), racemed milkwort (Polygala polygama), starry rosinweed (Silphium asperrimum), pencilflower (Stylosanthes biflora), and southern woolly violet (Viola villosa) to name a few. Several of these are rarely encountered in the Ozarks and are always in fairly open habitat on sandy soils when they are found.
This article is Part 1, which will be continued in an upcoming newsletter issue. Stay tuned for more information about pocket gophers, surprising plant finds, and a major botanical rediscovery …
Photos:
Photo 1: Ozark Pocket Gopher (Geomys bursarius ozarkensis). Photo by Matthew B. Connior.
Photo 2: Ozark Pocket Gopher (Geomys bursarius ozarkensis) mounds. Photo by Matthew B. Connior.
Photo 3: Old-plainsman (Hymenopappus scabiosaeus) in an open powerline right-of-way through a cedar-encroached sand woodland in Izard County. Photo by Theo Witsell.
Photo 4: Dwarf hawthorn (Crataegus uniflora) in a cedar-encroached sand woodland in Izard County. This is a shrub characteristic of similar sand woodlands in the West Gulf Coastal Plain. Photo by Theo Witsell.
Photo 5: Hairy rockcress (Arabis pycnocarpa var. adpressipilis), a species of state conservation concern found in a cedar-encroached sand woodland in Izard County. Photo by Theo Witsell.
Photo 6: Racemed milkwort (Polygala polygama), a sand-loving plant that is very rare in the Ozarks. A population found in a cedar-encroached sand woodland in Izard County is apparently the second known site in the Ozark Highlands Ecoregion. Photo by Eric Hunt.
A subspecies of the Plains Pocket Gopher, the Ozark Pocket Gopher is known only from a small area in southern Izard County, and nowhere else on Earth. This became official with the publication of an article by Douglas Elrod, Earl Zimmerman, Philip Sudman, and Gary Heidt in the August 2000 issue of the Journal of Mammalogy. These unusual burrowing rodents have coats camouflaged to match the color of the soil where they live and spend most of their lives in underground tunnels, emerging briefly to gather plant foods, which they transport back to their burrows in large fur-lined cheek pouches, or pockets. In addition to having a very narrow range, the Ozark Pocket Gopher has the distinction of being the only mammal that is endemic to, or found only in, Arkansas. The occurrence of an endemic pocket gopher in this particular region of the state, however, is somewhat unexpected and presents something of a mystery – one that has really captured my interest over the past few months. The presence of narrowly endemic species, those restricted to a small geographic area (and usually also to a specific uncommon or rare habitat), is evidence of great antiquity. Simply put, these unique species or subspecies didn’t just show up recently – their habitat must have been there for a long time. And therein lies the mystery. Pocket gophers are grassland species, preferring open prairie or savanna habitat with only scattered trees. For the endemic subspecies to occur in southern Izard County, there must have been some kind of suitable natural grassland habitat there in ancient times, and at least some of it must have been present up until modern times or they would have gone extinct. So, what is (or was) that habitat?
The literature on the Ozark Pocket Gopher gives its habitat as something along the lines of “pastures and crop fields with deep sandy soil,” and this is an accurate description of where they are found today. More specifically, they occur within these habitats on sandy, elevated terraces along a series of streams that drain southern Izard County, flowing southward into the White River. The mystery is that we have no historic records of there being grasslands in this area historically. They weren’t mapped by the General Land Office (GLO) surveyors who completed the original survey of the Louisiana Purchase (one of our best sources of information on the locations of historical grasslands in Arkansas), nor were they described in the accounts of early naturalists, explorers, or settlers (at least so far as we are aware). We knew there were scattered glades in the hills around these valleys, but these are rocky grasslands with bedrock at or near the surface of the ground and not suitable for pocket gophers. So, what were the sandy, valley grasslands that these gophers called home before humans were here to maintain open pastures and crop fields for them? And are there any remnants of these grasslands left that we could study?Remnants of ancient, natural grasslands are always interesting botanically, and usually have uncommon plants that aren’t found in other habitats. Concentrations of these grassland indicator plants are the hallmarks of such a remnant, and I knew it would be these botanical clues that could help solve the mystery of the missing grasslands. So back in May I began exploring a couple of interested landowners’ properties on these creeks, hunting for grassland indicator plants and trying to get my head around what grasslands in the area might have been like historically. I knew there wouldn’t be large prairies (if those had existed, they would have made it onto the maps), but I had a hypothesis that maybe there were once sandy savannas and open woodlands throughout these valleys, with small sandy prairies here and there. These would have distinctive plant communities made up of sun- and sand-loving plants.
From studying grassland ecosystems in other areas, I knew that the usual pattern of loss is that open prairies got plowed up for crops or converted to non-native pasture grasses early on and that savannas and open woodlands either got cleared and converted to pasture or have since grown up into dense forest. In the latter case botanical clues can often still be found along open roadsides and utility rights-of-way through these forests, so that was my first strategy – run the roads and look for grassland indicator plants. Unfortunately, a lot of these have been sprayed with herbicide in recent years, reducing rich pockets of diverse grassland flora to low-diversity patches of the weediest, most common, and often invasive, species. But before long I found some clues – big stands of little bluestem (Schizachyrium scoparium), one of our most common denizens of dry grasslands, along with some grassland-obligate wildflowers like pale purple coneflower (Echinacea pallida) and old-plainsman (Hymenopappus scabiosaeus). This was encouraging.The first property I explored, accompanied by Kyle Lapham (ANHC’s Chief of Acquisitions and Stewardship), didn’t have any significant areas of open ground, but it did have a very sandy forested stream terrace with a heavy component of eastern redcedar (Juniperus virginiana), a common invader of formerly open savannas and woodlands. A quick study of the trees in this forest showed the typical pattern of invasion we see in savannas and open woodlands – scattered larger (and older) individuals of oaks and pines with a more-or-less even-aged stand of smaller, younger cedar filling in between them. This cedar-encroachment generally happened in conjunction with open range grazing and widespread fire suppression following Euro-American settlement.
The soil here was very loose and sandy, and even though the site was nearly flat and located right along a stream, the non-cedar trees were species typical of well-drained, even drought-prone soils: shortleaf pine (Pinus echinata), post oak (Quercus stellata), black oak (Quercus velutina), and chinquapin oak (Quercus muehlenbergii). The shrub component in these woods was right out of the sandhill woodlands of southwest Arkansas’s West Gulf Coastal Plain: fringe tree (Chionanthus virginicus), dwarf hawthorn (Crataegus uniflora), Carolina holly (Ilex ambigua), farkleberry (Vaccinium arboreum) … I felt like I had suddenly been transported down to Miller County Sandhills Natural Area on the Texas border. But my eyes were really drawn to the ground, where the real plant diversity is found, and a few small semi-open pockets of woodland had a little bit of what I was after. 
There was more little bluestem, Carolina Larkspur (Delphinium carolinianum), and bastard toadflax (Comandra umbellata) – sure signs that this area was once more open than it is today. There was a lone, shade-suppressed clump of eastern gamagrass (Tripsacum dactyloides), another prairie species, which ranchers sometimes call “ice cream grass” because it’s the first species eaten by livestock if it grows in a field. My pulse really started moving when I found a few plants of the rare hairy rockcress (Arabis pycnocarpa var. adpressipilis), a species of state conservation concern previously known in Arkansas from just a handful of sites. In 25 years of hunting plants in the state I have only seen this at one other site!
There was also a nice list of psammophytes (sand-loving plants). These were the ones I was most interested in finding, as they fit in with my hypothesis that there were ancient sand grasslands in these valleys. Again I was transported to the southwest Arkansas sand barrens as I documented species after species of sand-lover: Texas green-eyes (Berlandiera betonicifolia), erect dayflower (Commelina erecta), velvet-leaf tick-trefoil (Desmodium viridiflorum), woolly rosette grass (Dichanthelium villosissimum), cream avens (Geum virginianum), hairy wild lettuce (Lactuca hirsuta), racemed milkwort (Polygala polygama), starry rosinweed (Silphium asperrimum), pencilflower (Stylosanthes biflora), and southern woolly violet (Viola villosa) to name a few. Several of these are rarely encountered in the Ozarks and are always in fairly open habitat on sandy soils when they are found. This article is Part 1, which will be continued in an upcoming newsletter issue. Stay tuned for more information about pocket gophers, surprising plant finds, and a major botanical rediscovery …
Photos:
Photo 1: Ozark Pocket Gopher (Geomys bursarius ozarkensis). Photo by Matthew B. Connior.
Photo 2: Ozark Pocket Gopher (Geomys bursarius ozarkensis) mounds. Photo by Matthew B. Connior.
Photo 3: Old-plainsman (Hymenopappus scabiosaeus) in an open powerline right-of-way through a cedar-encroached sand woodland in Izard County. Photo by Theo Witsell.
Photo 4: Dwarf hawthorn (Crataegus uniflora) in a cedar-encroached sand woodland in Izard County. This is a shrub characteristic of similar sand woodlands in the West Gulf Coastal Plain. Photo by Theo Witsell.
Photo 5: Hairy rockcress (Arabis pycnocarpa var. adpressipilis), a species of state conservation concern found in a cedar-encroached sand woodland in Izard County. Photo by Theo Witsell.
Photo 6: Racemed milkwort (Polygala polygama), a sand-loving plant that is very rare in the Ozarks. A population found in a cedar-encroached sand woodland in Izard County is apparently the second known site in the Ozark Highlands Ecoregion. Photo by Eric Hunt.