In Memoriam – Gayle H. Nelson

It was a very productive first day in the Black Hills of South Dakota, with several rather significant finds. However, I’m going to forego an update on these and instead dedicate this post to the memory of my friend and colleague, Dr. Gayle H. Nelson, who passed away on this day three years ago. Gayle was not only one of North America’s premier experts on Buprestidae but was also an outstanding teacher of human anatomy. With a career spanning nearly five decades and generating some 70+ beetle publications, Gayle had the opportunity to interact with many of the world’s most important coleopterists. Despite this, he was one of the most humble and accessible persons I’ve had the honor to meet. I think about Gayle often, especially while on collecting trips – remembering the places we visited and the lessons he taught me. On this 3rd anniversary of his passing, I reproduce here a remembrance that I wrote for a memorial issue of the The Pan-Pacific Entomologist, published September 24, 2006 on the 1st anniversary of his death.

“I had the privilege of calling Gayle Nelson both a mentor and a friend. I first corresponded with Gayle in 1984 as a young collector with a budding interest in beetles, and my first communication with him probably mirrors that of many others – me asking him for help identifying specimens. And, as he likely did for those many others, he graciously agreed. For the next several years, I would send him my “catch” at the end of each season and anxiously await the return shipment. Opening a box of specimens after he’d looked at them was as exciting as Christmas morning, not only to see how well I had fared in my tentative identifications, but also in anticipation of the “gifts” Gayle more often than not included for my collection. On one occasion, I had included examples of a strange looking Agrilus from south Texas that I had found during one of my earliest collecting trips outside of Missouri. They turned out to be A. toxotes, known previously only from Mexico, and a species not represented in Gayle’s collection (a true rarity by that point in his life). In his return letter, Gayle’s excitement about this find was obvious as he politely asked permission to retain a male/female pair. I agreed readily, and when the box of beetles was returned, I found added to its contents several dozen especially colorful examples of western U.S. Acmaeodera. To this young Midwesterner, those beetles were as “exotic” as if they had come from Brazil or Africa. During those early years, Gayle’s letters were rich with advice on collecting and suggestions for localities I should explore, and his kindnesses did much to solidify my passion for buprestids and eventually led to the first of our several coauthored publications.

“It was not until 1991, however, that I finally met Gayle in person while he was still residing in California. I had moved from St. Louis to Sacramento and was eager to explore the “buprestid heaven” that is southern California. Gayle had extended an open invitation to collect with him, so in early June I traveled to his home in Rancho Cucamunga, where he and his wife Jean graciously hosted me for the first two days of a weeklong collecting trip. That first evening I marveled at what was undoubtedly the most impressive private insect collection I had ever seen. Not only was it larger than any collection I had seen, but the exacting and careful manner in which the specimens had been curated and organized was enough to impress even the most retentive among us. We talked about the collecting localities he planned to show me and what species we might find there. To this still relatively “green” buprestophile whose collecting experience was limited primarily to the Missouri Ozarks, the prospect of collecting species of such “exotic” genera as Acmaeoderoides, Anambodera, Prasinalia, and Lepismadora – in one trip – almost seemed too good to be true. But true it was! Our first day in the field I met his longtime friend George Walters, and the three of us visited several of their favorite collecting localities in the San Bernardino Mountains near Wrightwood and in Lone Pine Canyon. I collected around 15 species of buprestids that day – more than I had ever collected on any previous field trip. The next day he took me to the beautiful Santa Rosa Mountains and its fabled Pinyon Flats, Whitewater Canyon and Palm Desert localities, where I added another dozen or so species to my catch – all different from the previous day. During those two days, I was not only astounded by Gayle’s endurance – he was well into his 60s by then – but also impressed with his dogged persistence in searching for his quarry. It didn’t take me long to figure out that this was one of the secrets to his great success as a collector. By the end of the second day, I was so exhausted that I slept during most of the long drive back to his home. I spent the rest of the following week visiting many of the other southern California localities Gayle had recommended, looking on the plants he had suggested, and ended up with a whopping trip total of ~60 buprestid species. During the years that followed, I had the good fortune to accompany Gayle on field trips as far away as southern Mexico and close to home in Missouri and Kansas. Each time he taught me something new and re-energized my passion for collecting buprestids. I knew I was “learning from the Master!” Gayle Nelson was large in stature and in life. He was a scientist, a teacher, a dedicated family man and a friend to us all. He will be missed by all who knew him.”

Below is a photograph of Gayle and several other buprestophiles (including a much younger me!), taken July, 1992 in Tehuantepec, Oaxaca, México during a world buprestid workers gathering (this photo was also published in the Pan-Pac memorial issue). The day this photograph was taken, Gayle and I had collected the first specimens of what we would later describe as Oaxacanthaxia nigroaenea – the second species in an odd little genus with Old World affinities that had been described just one year earlier by Chuck Bellamy from specimens collected in the very same area.

L-R: Svata Bílý (Czech Republic), Chuck Bellamy (USA), Hans Mühle (Germany), Gayle Nelson (USA), Dave Verity (USA), me, Mark Volkovitsh (Russia). Photo by Rick Westcott (USA).

Goin’ fishin’

For the past two days I’ve been in Sioux County, Nebraska – just east of Wyoming and just south of South Dakota. As I traveled up through the western panhandle to arrive at this spot, I was pleasantly surprised by the varied terrain – not at all the monotonously flat landscape that I expected. The landscape in this so-called Pine Ridge area is even more surprising – an impressive escarpment drops 1,400′ from the high shortgrass prairie down to an eery badlands below. The escarpment itself is forested with Ponderosa pine and is studded with numerous impressive buttes. The photos shown here were taken in Sowbelly Canyon – typical of the landscape along the escarpment – and in the badlands below Monroe Canyon a little further west.

Enough about pines and buttes – my business here is tiger beetles. I met up yesterday with tiger beetle aficionado Matt Brust, who recently took a position here at Chadron State College after finishing his Ph.D. in Lincoln. I’ve been corresponding with Matt for a bit now, and when I told him of my interest in doing a tiger beetle trip through western Nebraska, he was more than willing to show me around and hopefully help me find some of the more unusual species I was looking for. Of course, tops on the priority list was Cicindela nebraskana (prairie long-lipped tiger beetle). This beetle isn’t common anywhere within its range and just sneaks into the northwest corner of Nebraska, where the type locality is located. Until recently, the species was known from very few specimens in Nebraska. Matt did some intensive sampling a few years ago and located a few limited populations in the vicinity of the type locality. Yesterday, he took me to two of these localities, and we succeeded in finding one individual at the first and several at the second. It was at the type locality where I succeeded in getting this field photo. While admittedly harshly-sunlit, it is as far as I know the only field photograph of the species – all others that I’ve seen have been taken in terraria. I’ll fix it up a bit with Photoshop and re-post once I get back home. This species looks similar to the black morph of Cicindela purpurea audubonii, which co-occurs with C. nebraskana in Nebraska, but it lacks the bright white labrum and elytral markings of the former. Also, as I would learn during these past two days, it can be instantly recognized in the field by its shinier appearance and “stubbier” legs. A few days before my arrival, Matt succeeded in finding the species in the next county to the east, an eastern range extension of about 60 miles, and today I located the species at another new locality between the two. It is gratifying to have played a small role in increasing our knowledge about this unusual species.

My success with C. nebraskana has come despite uncooperative weather. A series of frontal systems has moved through the area since my arrival, resulting in several rain events and lots of cool, cloudy weather. Tiger beetles are sun-loving insects, and when it gets too cold or wet the adults dig in and don’t come out until the sun shines through or temps warm enough to trigger them to dig out. Matt had taken me to another locality – a sand embankment – where we might find the beautiful Cicindela lengi (blowout tiger beetle), but it rained prior to our arrival and we saw no activity. I tried finding adult burrows to dig them out, but the rain had obliterated any trace of the diggings, making their burrows impossible to find. We did, however, note an abundance of larval burrows. I went back to the spot today hoping to see some activity, but thick clouds and cool temps made that unlikely. This is when I decided to “go fishing.” Tiger beetle larval burrows are easily recognized by their perfect circular shape and clean “beveling” around the entrance (1st photo). Burrows of 3rd instars (the last larval instar in tiger beetles) are distinctly larger than those of 2nd instars (2nd photo), while those of 1st instars are smaller still (not shown). Larvae sit at the burrow entrance and ambush any suitable prey that comes too close. During cool, cloudy weather, however, they drop to the bottom of their burrow – up to a foot or more deep. A technique useful for extracting inactive larvae from their burrows is called fishing and involves inserting a thin grass stem down to the bottom of the burrow in an attempt to coax the larva into “taking the bait” and biting the end of the grass stem (3rd photo). The grass stem is then pulled up rapidly – much like setting the hook when fishing – in an attempt to pull the larva out of its burrow before it has a chance to let go of the stem. It can take a few tries, but with practice one can more often than not succeed at removing the grotesquely odd, yet beautiful larva (4th photo). Note the huge, heavily sclerotized head with upward facing jaws. The hump in the middle of the back is armed with forward-curved spines that helps the larva avoid being pulled out of the burrow by struggling prey (but they’re not so effective against obsessive cicindelophiles!). As I managed to “fish” larvae I placed them in a plastic container with their native soil. In the 5th photo, four larvae have already begun digging new burrows, and one more 3rd instar (L) along with a 2nd instar (R) has just been placed in the container. I’ll bring this container back with me and continue to feed the larvae live insects in the hopes of rearing them to adulthood. I cannot say with certainty that the larvae I collected represent C. lengi – other species that could potentially occur at this site include C. scutellaris (festive tiger beetle) and C. limbata (sandy tiger beetle). However, the locality is known for the abundant occurrence there of C. lengi, so I’m hopeful that that is what I’ve collected – we’ll know in a few weeks. In the meantime, I’ll have additional opportunities to look for this species, along with C. limbata, as I pass through the Sand Hills region later this week.

Tomorrow morning I head to the Black Hills in South Dakota, where I hope to find not only Cicindela longilabris laurentii (Boreal long-lipped tiger beetle) in the high pine forests, but also intergrades between this species and the closely related C. nebraskana in the more open habitats of the middle latitudes. Look for an update in a couple days or so.

Digging tiger beetles

After driving almost 600 miles on Friday, I was ready for some tiger beetle action. A quick 2-hour drive this morning got me to the first set of localities in extreme southwestern Nebraska and just across the border into Colorado (“B” thru “D” in the previously posted map). Compared to the other localities I have lined up for the trip, these were the most vague – based on published records from more than a century ago with no recent ‘ground intelligence.’ I wasn’t too concerned about having success at these spots, I just wanted a potentially interesting area to explore after a day of driving before completing my voyage to the first real target areas up in the northwestern corner of Nebraska. I could’ve taken a more direct line along I-80 and explored areas along that route, but I’ll be exploring that route on my return trip.

My first stop at Benkelman was a bust – I was looking for some sand blows in the hopes of finding Cicindela limbata, but a half hour of so of driving around the area turned up nothing. I stopped at some exposed clay to see if I might encounter C. denverensis but only saw a couple of the ubiquitous C. punctulata. Haigler – a half hour to the west – was more productive. I found a spot with access down along the Republican River (at only 2 yards in width, it is beyond me how this qualifies as a “river”) and found several typical sand associated species – C. scutellaris (pictured) and C. formosa, both nominotypical forms which look very different from the populations we have in Missouri, and several C. tranquebarica individuals that show the bold markings characteristic of the northern Great Plains subspecies kirbyi (the first time I’ve seen this subspecies). My 104 year-old literature reference suggested C. fulgida occurred around salt marshes in the area, but no such habitats were found – probably degraded long ago, or even worse obliterated in the interest of “improving” the land for agriculture. Another half hour drive got me to Wray, just across the border in Colorado, where C. limbata again had been recorded from sand hills above the Republican Trickle. These sand hills were easily located, but the completely barren blows that the species requires, again, were not found – probably allowed/encouraged to grow over. As it was now the heat of the day (a very pleasant low 80’s), the common sand species C. formosa and C. scutellaris were now digging in to avoid the heat.

Although it is well known that adult tiger beetles dig burrows to spend the night, I haven’t really seen much in the literature that talks about their daytime burrowing habits, how to recognize the burrows, and how to dig them out (okay, so now my double entendre of a title makes sense, cuz you see I really ‘dig’ tiger beetles… and I’m digging them from their burrows… oh well, it seemed funny when I thought of it). Many other insects also burrow in the sand, and some of them – especially those of the many ‘digger’ wasps and ground nesting bees – can look very similar to (and sometimes outnumber) those of tiger beetles. In my experience, adult tiger tiger beetle burrows have a more ‘flattened’ aspect to them (1st photo), while those of wasps and bees are more rounded. Fresh burrows from both groups will have moist diggings thrown to one side (in old burrows – often uninhabited – the diggings will be dry), but those of tiger beetles appear more ‘fanned’ while those of hymenopterans are more ‘piled.’ I use a knife to excavate the sand away from the entrance in thin vertical slices (2nd photo), slightly undercutting the burrow to prevent sand from falling into it (and making it impossible to follow). Wasp and bee burrows look more round during excavation and often make a hard turn downward, while those of tiger beetles continue looking very flattened and usually stay rather shallow. If somebody is home, they will be encountered without too much digging – these photos show a C. scutellaris adult peaking out head first and a C. formosa who was still in the act of excavating (done by kicking the sand backwards while backing towards the entrance). I stumbled onto this technique last year during my first encounter with C. limbata – had I not done so I probably would’ve walked away with only a single specimen.

Tomorrow, I hope to have encountered C. nebraskana and C. lengi. Cicindela decemnotata is a long shot, but hope springs eternal…

A hunting we will go!

Maps have been prepared. Relevant emails from my esteemed colleagues to the northwest have been read and re-read. Summary sheets on the distribution, biology, and biogeography of the many different species I hope to encounter are in hand. Google Earth images of each locality I plan to visit – annotated with potential species occurrences and pinpointing precise locations of their likely habitats – have been assembled into a Powerpoint presentation, and detailed driving directions from Point “A” to Point “B”… all the way to Point “X” (home!) have been determined. All of this has been printed out and organized into a 3-ring binder. Why the extraordinary attention to detail? Because…

It’s time for the annual fall tiger beetle trip!

View Larger Map

The annual fall tiger beetle trip started several years ago when I, along with my friend and colleague Chris, began studying Missouri’s tiger beetle fauna. At first it was a diversion – buprestids and cerambycids are pretty well played out by fall, but tiger beetles across much of the U.S. exhibit a unique spring/fall fauna that is quite distinct from the summer fauna. Chris and I would go to different parts of Missouri, documenting the species encountered to fill in distributional data gaps. It was on these trips that I discovered how much I truly love early fall collecting – the cool air, the crisp smells, the long sharp shadows, and a landscape of foliage ever so lightly tinged with shades of red and yellow while grasses morph into fields of gold. In recent years, I’ve begun adventuring beyond Missouri’s borders on these fall trips, allured by the diversity of species found in the Great Plains – species alien to Missouri in an equally alien landscape. First, it was Barber County, Kansas, with its red gypsum hills inhabited by the aptly named Cicindela pulchra (beautiful tiger beetle) – deep wine-red and iridescent purple flashing across the barren red clay. Then last year I got my first taste of the Sand Hills of Nebraska at their farthest eastern extent. I watched in amazement as Cicindela limbata (sandy tiger beetle) – vivid white and metallic green – danced across the surface of sand blows, undaunted by scouring 30 mph winds. It was on that trip that I decided a long weekend wasn’t cutting it – I needed to take a whole week and get myself into the heart of the Great Plains. The annual fall tiger beetle weekend has just become the annual fall tiger beetle week.

As the map above indicates, I’ve got a rather ambitious itinerary of locations that I’d like to visit – 22 in all. I leave tomorrow, and if I have planned properly (and have a little luck) I might be able to visit all of them in the 9 days I have set aside for the trip. My “trip bible” will be my constant companion, along with my already worn copy of the newly issued Tiger Beetles of South Dakota & Nebraska (Spomer et al. 2008), as I explore deep into the Sand Hills and experience for the first time ever the Black Hills of South Dakota. I’ll even sneak over into Colorado and Wyoming for a spot or two. Unfortunately, my faithful colleague isn’t able to join me. I tried to seduce him with visions of Cicindela limbata and C. lengi (blowout tiger beetle) in the numerous sand blows, C. fulgida (crimson saltflat tiger beetle) around countless alkaline lakes, C. longilabris (Boreal long-lipped tiger beetle) in the high pine forests, and C. nebraskana (prairie long-lipped tiger beetle) and (if we’re really really lucky) C. decemnotata (Badlands tiger beetle) just sneaking into the shortgrass prairies of the extreme northwestern corner of Nebraska. I reminded him of my (wanting) photographic skills and the images we would have to settle for if his talent and equipment didn’t accompany me. I almost had him, but in the end he muttered some lame excuse about his 15-month old baby and wife needing him (just kidding, Chris!).

The map above should be fully interactive, so give it a click and follow me along on this adventure. If you happen to be at any of the spots marked by a balloon and see a khaki-clad fellow – insect net in one hand, camera in the other – how’s about joining me for a bit of tiger beetle hunting.

“All the better to see you with, my dear!”

Click me!

Cicindela formosa (the big sand tiger beetle) is a not uncommon species that occurs across much of North America east of the Rocky Mountains in deep, dry, open sand habitats. It is absent in Appalachia and much of the Interior Highlands, understandable given the rarity of deep sand habitats on these elevated landforms; however, its absence across much of the southeastern coastal plain as well as south and west Texas, despite the widespread presence of apparently suitable habitat, is not easily explained. In Missouri, dry sand habitats are rather limited, occurring primarily along the Missouri and Mississippi Rivers, a few of the larger Ozark rivers, and along Crowley’s Ridge and the Blodgett Terrace in the Mississippi River Alluvial Plain. The individual in these photos was seen last weekend at Sand Prairie Conservation Area (on the Blodgett Terrace), where I also recorded it earlier this year. Despite its relative commoness, I always get a little excited whenever I find this species – it’s a big, chunky thing with bold markings and sufficient habitat specificity to keep it from being too pedestrian (unlike Cicindela repanda and C. punctulata, which usually evoke only a groan – okay, maybe western forms of the latter, with their gorgeous suffusion of green and blue excite me a little bit). Cicindela formosa populations to the west are even more brilliantly colored and localized – it’s a handsome species, indeed! Adults are powerful fliers that terminate their long escape flights with a comical tumble or two across the sand before ending up on their feet. Normally a difficult species to get close to, cool temps and overcast skies on this morning resulted in a cooperative subject and excellent lighting for this series of photos.

Missouri populations are assignable to the eastern subspecies generosa – mostly, that is. There is a population known from the Ozarks, along the beautifully pristine Current River, that exhibits tendencies towards the bright coppery-red dorsal and metallic purple ventral coloration of the nominate subspecies found further west. I’ve also located another population in the northeastern Ozarks on “sand” flats – not true sand, but expansive dumpings of pulverized limestone tailings from former lead mining operations – that shows a similar intergrading with nominotypical characters. The occurrence of these populations near typical generosa populations and disjunct from nominotypical populations several hundred kilometers to the west, coupled with the existence of a broad intergrade zone between the two forms along both sides of the Missouri River through Nebraska, Iowa, and the Dakotas, raises interesting questions about the validity of a subspecific distinction for generosa. Additional subspecies have been described from eastern Texas and adjacent areas of Arkansas and Louisiana (pigmentosignata), southeastern New Mexico (rutilovirens), and southwestern Saskatchewan (gibsoni). Each of these populations is at the edge of the nominate subspecies’ range of distribution and exhibits consistent differentiation in multiple characters – primarily color and maculation – from nominotypical populations. As a result, the case for according subspecific status to these populations is more convincing despite the occurrence of intergrades along narrow zones of contact with nominotypical populations. A truly allopatric population center occurs in northwestern Colorado and southeastern Utah – separated from nominotypical populations to the east by a distance of 230 km. The Colorado population strongly resembles and has thus been assigned to subspecies gibsoni. However, it is hard to imagine a mechanism by which the Colorado and Saskatchewan populations – with over 1,000 km separating them – derived from a common ancestry. A more likely scenario is independent adaptation to similar conditions in their respective habitats. Differences in coloration of the larval head capsule between these two populations lend support to this idea, which if true should qualify the two populations for consideration as distinct subspecies despite the similarity in their appearance. Interestingly, the Utah population resembles nominotypical forms further east, although intergrades with the adjacent Colorado population do occur along a narrow zone of contact.

The subspecies concept has been hotly debated for many decades now. E. O. Wilson and W. L. Brown (1953), in their seminal paper, The subspecies concept and its taxonomic application, questioned the validity of many subspecies on the basis that they failed to exhibit concordance across multiple characters and argued that subspecies that interbreed were not “real taxa” because the flow of genes and characters between them prevented divergence. This restrictive concept essentially limited subspecies to populations that showed significant divergence from their relatives but relied upon external mechanisms (i.e., allopatry) rather than internal (i.e., genetic) for reproductive isolation. Many of North America’s described tiger beetle subspecies would not meet these criteria, since there often exist zones of contact where intergrades (a result of gene flow within hybrid zones) do occur. Ernst Mayr took a more pragmatic approach in Animal Species and Evolution (1963), defining subspecies as “an aggregate of local populations of a species inhabiting a geographic subdivision of the range of the species, and differing taxonomically from other populations of the species” – in other words, subspecies are taxonomic units and not evolutionary units. Viewing subspecies as strictly taxonomic units is more convenient, since the presence of hybrid zones does not invalidate a subspecies as long as it retains its taxonomic distinctiveness. I acknowledge that taxonomic subspecies units are useful – named subspecies provide a convenient shorthand for discussing geographical variation within species and stimulate interest in their study and characterization. Also, as emphasized by cicindelid experts D. L. Pearson et al. (2006), the application of formalized subspecies names for distinctive, local populations makes conservation policy decisions more palatable to policians and legislators, thus enhancing the potential for protection. However, I also agree with O’Neill (1982) that the subspecies concept must be connected to an evolutionary unit to be truly meaningful, and the recent application of molecular techniques is now providing a genetic basis for assessing subspecies validity. Interestingly, some such studies have shown near complete blockage of gene flow across hybrid zones, even when hybridization is frequent, providing genetic evidence of “real taxa” that nevertheless interbreed at their boundaries (Mallet 2007). It would be interesting to apply molecular techniques to populations of generosa, nominotypical formosa, and the Missouri intergrades to understand their degree of genetic divergence, the presence of which could convince me that their status as distinct subspecies should be maintained.

The Loess Hills in Missouri

The term Mountains in Miniature is the most expressive one to describe these bluffs. They have all the irregularity in shape, and in valleys that mountains have, they have no rocks and rarely timber. – Thaddeus Culbertson, missionary, 1852


One of the things I enjoy most about the natural history of Missouri is its diversity. Lying in the middle of the North American continent, it is here where the eastern deciduous forest yields to the western grasslands. Coinciding with this transition between two great biomes is a complex intersection of landforms – the northern plains, recently scoured by glaciers; the southeastern lowlands, where the great Mississippi River embayment reaches its northern extent; the Ozark Highlands, whose craggy old rocks comprise the only major landform elevation between the Appalachian and Rocky Mountains; and the eastern realm of the vast Great Plains. This nexus of east and west, of north and south, of lowlands and highlands, has given rise to a rich diversity of natural communities – 85 in all according to Paul Nelson (2005, Terrestrial Natural Communities of Missouri). Despite the overwhelming changes wrought upon Missouri’s landscape during the past 200 years, passable examples of most of these communities still exist in many parts of the state and provide a glimpse of Missouri’s rich natural heritage.

Last month I talked about the critically imperiled sand prairie community in extreme southeast Missouri. This month, we travel 500 miles to the distant northwestern corner of the state to visit another critically imperiled community – the dry loess prairie. These communities are confined to thin slivers of bluff top along the Missouri River in Atchison and Holt Counties. The bluffs on which they lie are themselves part of a unique landform called the Loess Hills. Like the sand prairies of the southeastern lowlands, this angular landscape owes its birth to the glacial advances of the Pleistocene epoch (2.5 million to 10,000 years ago), when streams of meltwater – swollen and heavily laden with finely ground sediments (i.e., glacial “flour”) – filled river valleys throughout the Midwest during Pleistocene summers. Brutal cold during winter reduced these flows to a trickle, allowing the prevailing westerly winds to pick up the sediments, left high and dry, and drop them on leeward upland surfaces across Iowa and northern Missouri. The thickest deposits occurred along the abrupt eastern border of the Missouri River valley – at least 60 feet deep, and in places up to 200 feet. Loess (pronounced “luss”) is a homogeneous, fine-grained, quartz silt – undisturbed it is highly cohesive and able to stand in near vertical bluffs. It is also extremely prone to erosion, and as a result for 10,000 years now the forces of water have reshaped the Loess Hills into the landform we see today. Loess itself is not rare – thick deposits can be found in many parts of the world and over thousands of square miles across the Midwest. It is here, however, along the western edge of Iowa and northern Missouri – and nowhere else in North America – where loess deposits are deep enough and extensive enough to obliterate any influence by the underlying bedrock and dictate the form of the landscape.

It is this form that makes the Loess Hills so unique. The depth of the soil, its cohesiveness, its natural tendency to slump on steep slopes and sheer in vertical planes, and the action of water over the past several millenia have created a landscape of narrow undulating ridges flanked by steep slopes and numerous side spurs, intricate drainages with sharply cut gullies, and long, narrow terraces called “catsteps” cutting across the steep upper hillsides. It’s a sharp, angular, corrugated landscape, stretching 200 miles north and south in a narrow band of varying width from north of Souix City, Iowa, to its southern terminus in northwestern Missouri. Its western boundary is sharply delimited by the Missouri River valley, where lateral erosion (now halted by channelization of the river) and vertical sheering have created precipitous bluff faces. The eastern boundary is harder to delimit and is dependent upon the thickness of the loess. Deposits that fall below 60 feet in depth are unable to mask and reshape the rolling terrain of the eroded glacial till lying beneath. In general, this happens at distances of only 3 to 10 miles from the western edge of the landform.

Its southern terminus in Missouri, however, is the most arbitrary boundary. Discontinuous patches of deep loess terrain do occur as far south as Kansas City, but the dry hilltop prairies, common in the north, are gradually replaced by woodland in the south and disappear completely just north of St. Joseph. It is this interdigitation of two great biomes – the great deciduous forest to the east, and the expansive grasslands stretching far to the west – that give the Loess Hills such a fascinating natural history. This is due as much to the physical character of the Loess Hills themselves as to their ecotonal position at the center of the continent. Rapid drainage of rainwater off the steep slopes combines with direct sun and prevailing southwesterly summer winds to create very dry conditions on hilltops and south and west facing slopes, especially on the steeper slopes along the landform’s western edge. Such xeric conditions favor the growth of more drought-tolerant species derived from the western grasslands. North and east facing slopes and valley floors, protected from direct sun and drying winds, are able to retain more moisture, favoring the growth of woody plant species more common in the eastern forests. Seasonal moisture also shows a north-south gradient, with southern latitudes receiving higher annual rainfall totals that also favors the growth of woody plants, while the lower rainfall totals further north result in larger, more expansive grassland habitats. The steep slopes and rapid drainage create much more xeric conditions than those found further south in the flat to rolling terrain of the unglaciated Osage Plain, resulting in a more drought-tolerant mixed-grass prairie rather than the tallgrass prairie of western and southwestern Missouri. The distribution patterns of prairie versus woodland are dynamic and ever-changing, influenced by both natural and anthropogenic processes. Climatic conditions over much of the Loess Hills are capable of supporting either community type, both of which repeatedly expand and shrink as the balance tips in favor of one versus the other. In the past, the major influence was shifting periods of greater or lesser rainfall. During drier periods, grasslands expanded and woodlands shrank, finding refuge in only the moistest streamside habitats. Wetter periods allowed woody plants to migrate out of the valleys and up the slopes, especially those facing north and east. One particular very dry “hypsithermal” began about 9,000 years ago and lasted for several thousand years. Tallgrass prairies expanded as far east as present day Ohio, and todays tallgrass praires in the eastern Great Plains were invaded by even more drought-tolerant species from the shortgrass prairies further west. Eventually the hypsithermal abated, moisture levels increased, and the grasslands retreated in the face of the advanding forest. Not all of the drought-tolerant species were driven back, however, and scattered populations of these “hypsithermal relicts” still remain on locally dry sites far to the east of their normal range of distribution. Conspicuous examples of such in Missouri’s Loess Hills are soapweed yucca (Yucca glauca var. glauca) and the leafless-appearing skeletonweed (Lygodesmia juncea) (plant above, flower right). Both of these plants are normally found further west in the mixed grass prairies of the western Great Plains but are considered endangered in Missouri due to the great rarity of the dry loess prairies on which their survival depends. (Incidentally, note the crab spider legs extending from behind the petals of the skeletonweed flower). In total, more than a dozen plant species occurring in Missouri’s dry loess prairies are listed as species of conservation concern, along with one reptile (Great Plains skink) and one mammal (Plains pocket mouse).

As is typical, the insect fauna of the Loess Hills has been far less studied than its plants, but many of the species that have been documented in its prairies also show affinity to the Great Plains fauna. Both soapweed and skeletonweed have insect associates that rely exclusively on these hosts for reproduction, and as a result they are also highly restricted in Missouri. Evidence of one of these – a tiny cynipid wasp (Anistrophus pisum) that forms small spherical galls on the stems of skeletonweed – can be seen in the photo above. However, my purpose for visiting the Loess Hills this summer was to look for the rare and possibly endangered tiger beetle, Cicindela celeripes (see this post). Cicindela celeripes has not yet been recorded from Missouri but is known to occur in the Loess Hills of southwestern Iowa, and while I have not succeeded in finding it (yet!) I did observe several adults of this unusual May beetle species, Phyllophaga lanceolata. This May beetle occurs throughout the Great Plains in shortgrass prairie communities. Larvae feed in the soil on roots of grasses and other plants, while adults feed above ground on flowers and foliage. The heavy-bodied adults are unusual in the genus due to their conspicuous covering of scales (most species of Phyllophaga are glabrous or with sparsely scattered and indistinct setae) and by being active during the day. They are also relatively poorer fliers and are thus usually observed moving about on foot – as seen with this individual who was found on bare soil below a vertical cut. This snakeweed grasshopper (Hesperotettix viridis, ID by Eric R. Eaton) is another species more typically seen in the western United States, although populations have been found from across the continent. Preferred host plants include a variety of asteraceous shrubs, but as suggested by the common name snakeweeds (Xanthocephalum spp.) are highly preferred and account for its greater abundance in the west. Populations in northern and eastern portions of its range, which would include northern Missouri, are considered subspecies pratensis, while the more southern and western populations are considered the nominotypical subspecies. Interestingly (and unlike many grasshoppers), this species is considered beneficial by ranchers, since the plants on which it prefers to feed are either poisonous to livestock or offer little nutritional value while competing with more desirable forage plants for soil moisture. While exploring the upper slopes, I encountered sporadic plants of two of Missouri’s more interesting species of milkweed – whorled milkweed (Asclepias verticillata) and green milkweed (Asclepias viridiflora), raising my hopes that I might encounter one of the many Great Plains species of milkweed beetles (genus Tetraopes). However, the only species I observed was the common milkweed beetle, Tetraopes tetrophthalmus, which occurs broadly across eastern North America on the equally broadly distributed common milkweed (Asclepias syriaca).

It is a familiar refrain, but Missouri’s dry loess hill prairie communities are critically endangered. Historically, these communities were probably never as well developed as those further north, and only a few small remnants remain today due to significant woody encroachment following decades of fire suppression. Much of this encroachment has occurred in the past 50 years – Heinman (Woody Plant Invasion of the Loess Hill Bluff Prairies. M. A. Thesis, University of Nebraska at Omaha, 1982) used aerial photographs to show a 66 percent encroachment of shrubs and trees into the loess hill mixed-grass prairies between 1940 and 1981. Additional threats include overgrazing, erosion, invasion by exotic plant species and homesite development. Fewer than 50 acres of native dry loess hill prairie remain in Missouri – only half of which are now in conservation ownership. The majority of these can be found at Star School Hill Praire and Brickyard Hill Conservation Areas in Atchison County and at McCormack Conservation Area just to the south in Holt County. Controlled burning and selective cutting are being used at these sites to control woody plant invasions, but even these management techniques present challenges. Spring burns have been shown to promote the growth of big bluestem (Andropogon gerardii), which could allow it to encroach drier areas where mid-grasses such as little bluestem (Schizachyrium scoparium) and sideoats grama (Bouteloua curtipendula) typically dominate (Rushin 2005). Increases in tall grasses could shade out and eliminate some of the rarer low-growing forbs such as downy painted cup (Castilleja sessiliflora), locoweed (Oxytropis lambertii) and low milkvetch (Astragalus lotiflorus). Fall or winter burns may be more beneficial to forbs because the plants are allowed to complete flowering and seed set, but the steep slopes on which these communities occur make erosion a potential concern. Clearly, all factors must be considered when designing management plans for this rare and significant slice of Missouri’s natural heritage.


In addition to the links and references provided above, I highly recommend Fragile Giants: A Natural History of the Loess Hills, by Cornelia F. Mutel (1989). All of the above photographs were taken at Star School Hill Prairie Conservation Area on July 12, 2008. Additional photographs of Loess Hill habitats in extreme southwestern Iowa appeared in my earlier post, The hunt for Cicindela celeripes. The plants shown in photographs 5-7 are purple praire clover (Dalea purpurea), white prairie clover (D. candida), and lead plant (Amorpha canescens), respectively. Lastly, I would like to apologize for the length of this post – a consequence of my inability to temper my utter fascination with the natural world and desire to understand the depths its connectedness.

Trichinorhipis knulli

Just a little diddy on one of the more interesting species I’ve encountered over the years while I finish up a longer piece on the Loess Hills of Missouri. The specimen shown here is a male Trichinorhipis knulli. This quirky little species belongs to the equally quirky little tribe Xenorhipidini (family Buprestidae). Members of this tribe are among the few groups of Buprestidae in which evolution of the male antenna has diverged dramatically from the typical condition (i.e., serrate). In the Xenorhipidini, this condition may be considered very extended flabellate or even lamellate. As I mentioned, only males exhibit this antennal modification – females possess typical serrate antennae. The functional significance of this almost certainly involves detection of female sex pheromones. The surfaces of the flabellae in these species are covered with numerous presumably olfactory sensillae that are lacking on female antennae, and males of a related species (Xenorhipis brendeli) have been observed attracted in large numbers to caged live females. This antennal condition appears to have arisen independently in three other groups of Buprestidae as well, but Xenorhipidini is the only non-monotypic tribe in which males of all member species possess the condition.

Trichinorhipis knulli is restricted to southern California and has been encountered most often in the vicinity of Mountain Springs in Imperial County (just north of the Mexican border), where it breeds in dead branches of jojoba, Simmondsia chinensis. Very few individuals have actually been observed in the field – most existing specimens have been reared from caged, infested branches (as is the case with this specimen, which emerged August 1994 from a dead branch I collected in October 1992 – patience prevails!). At only 3.6 mm in length, it is one of the smallest members of the family, but I think you’ll agree that it is just as impressive under the microscope as any of the larger members of the family. The genus is monotypic (although I hear rumor of an undescribed species from west Texas) and has been placed in its own subtribe (Trichinorphidina) within the Xenorhipidini due to unique characters that distinguish it from the other included genera (Hesperorhipis and Xenorhipis). These include its entire (not abbreviated) elytra and broadly rounded pronotum lacking lateral margins. In Hesperorhipis and Xenorhipis the elytra are abbreviated, and the pronotum is quadrate with distinct lateral margins. The organization of the antennal sensillae also differs between Trichinorhipis and these other genera.

The tribe Xenorhipidini is currently being revised by my colleague and friend, Dr. Charles Bellamy, California Department of Food and Agriculture, Sacramento.

The hunt for Cicindela celeripes

One of the more enigmatic tiger beetle species in North America is Cicindela celeripes LeConte (swift tiger beetle). This small (6-8 mm), flightless species has been recorded from a restricted area of the eastern and southern Great Plains – from eastern Nebraska and westernmost Iowa south through Kansas to western Oklahoma and the Texas panhandle (Hoback and Riggins 2001, Pearson et al. 2006). Unfortunately, populations of this species appear to have suffered severe declines. It apparently is holding strong in the Flint Hills region of Kansas, but many of the records from outside of that area date back more than a century. Reportedly once common on the bluff prairies along the Missouri River, it has not been seen in Nebraska since 1915 and may have been extirpated from that state (Brust et al. 2005). The reasons for this decline undoubtedly involve loss of preferred habitat – upland prairies and grasslands with clay or loess soils and sparse or patchy vegetation. Areas supporting these native habitats have been drastically reduced since European settlement of the region, and suppression of fire – so vital to prairie ecosystems – has led to extensive woody encroachment on the few prairie relicts that do remain. Unlike many other tiger beetle species that have been able to adapt to these anthropogenic changes, this species apparently cannot survive in such altered habitats.

Chris Brown and I have been interested in this species ever since we began surveying the tiger beetles of Missouri. It has not yet been recorded from the state, but we have long suspected that it might occur in extreme northwest Missouri. It is here where the Loess Hill prairies along the Missouri River reach their southern terminus. (Incidentally, the Loess Hills are themselves a globally significant geological landform, possessing natural features rarely found elsewhere on earth. They will be the subject of a future post). We have searched several of what we consider to be the most promising potential sites for this species in Missouri, though without success. Nevertheless, we remain optimistic that the species might eventually be found in Missouri and has simply been overlooked due to the limited temporal occurrence, small size, rapid running capabilities, and tendency of adults to dart rapidly to the bases of grass clumps where they hide (Pearson et al. 2006). Furthermore, even though the species has not been seen recently in adjacent areas of Nebraska where it has been recorded in the past, it has been seen recently in a few Loess Hill prairie remnants just to the north in Iowa.

A few weeks ago, I was fortunate to receive specific locality information for one of the recently located Iowa populations. Armed with site descriptions, Google maps, photographs, and whatever book learnin’ I had gained about this species, my colleague and I made the long drive to southwestern Iowa in hopes of locating the population for ourselves, seeing adults in their native habitat, and using the learnings we would gain about their habitat preferences and field behavior to augment our efforts to eventually locate the species in northwestern Missouri. At mid-July, we were nearing the end of the adult activity period, but adults had been observed at the site the weekend prior, so we felt reasonably confident that adults might still be found. Additionally, fresh off of our recent success at locating the related Cicindela cursitans in Missouri (another small, flightless, fast-running species), we were hopeful that we now possessed the proper “search image” to recognize C. celeripes in the field should we have the good fortune to encounter it.

Walking into the area, I was impressed at the extensiveness of the prairie habitat – much larger than any of Missouri’s Loess Hill prairies. The presence of large, charred red-cedar cadavers on the lower slopes revealed active management for prairie restoration. We later learned from the area manager that the restoration area had been acquired from a neighboring landowner who had used the land for grazing and sold it when it became unproductive. I can only imagine the second thoughts that landowner must have had when subsequent burn regimes and woody growth removal prompted a return to the beautifully lush sea of prairie vegetation that now covered the hills. As we approached the area where we decided the beetles must have been seen, we started searching slowly and deliberately – looking carefully for any movement between the clumps of grass. It didn’t appear to be prime habitat for C. celeripes – the vegetation was just so thick, with only small openings among the plants. We continued to scour the area closely but saw nothing, and my optimism began to wane. Wrong spot? – I don’t think so. Bad search image? – hard to imagine, considering its similarity to C. cursitans. Too late? – could be.

After it became obvious we were searching the same gaps in the vegetation repeatedly, I started walking towards a small cut further down the hillside that I had noticed earlier (just visible in the previous photo). I had thought, “That’s tiger beetle land down there!” My optimism increased when I reached the cut, seeing the remains of an old, overgrown 2-track leading through the cut and on down the hillside. Vegetation was much sparser within and below the cut – it looked perfect. Chris had become distracted taking photographs of something, so I began searching. I’d been in the cut a few minutes when I thought I saw something flash across a bare patch out of the corner of my eye – was that it? It had to be. I carefully inspected around the base of every clump of vegetation at my feet but found nothing. It must have been wishful thinking – just another spider. I continued on down the cut, and within a few more minutes I saw the flash again – this time there was no doubt as to what it was, and I had a lock on it. I started slapping the ground frantically as the little guy darted erratically under, around, and over my hands. In the few seconds while this was happening, I was simultaneously exuberant at having succeeded in finding it, utterly astounded by its speed and evasiveness, and desperately afraid that it was getting away – swift tiger beetle, indeed! Persistence paid off, however, and eventually I had it firmly in my grasp.

We would see a total of seven individuals that day. Most of them were within or immediately below the cut, while another individual was seen much further down the 2-track. Mindful of the population declines this species has experienced, we decided to capture just three individuals (even though by this point in the season mating and oviposition would have been largely complete) in hopes that at least one would survive the trip back to the lab for photographs. Our primary goal – to see the species in its native habitat – had been accomplished. We now turned our attention to attempting in situ field photographs. This would prove to be too difficult a task – each beetle we located immediately ran for cover, and flushing it out only caused it to dart to another clump of vegetation. This scenario repeated with each beetle until eventually it simply vanished. We would have to settle for photographs of our captured specimens in a confined arena – a few of which are shown here. The beetles were photographed on a chunk of native loess taken from the site, and no chilling or other “calming” techniques were used. Spomer et al. (2008, Field Guide to the Tiger Beetles of Nebraska and South Dakota) state that C. celeripes is a delicate species that does not do well in captivity. It has never been reared, and the larva is unknown. Nevertheless, I placed the chunk of native loess in a plastic tupperware container and transplanted into one corner a small clump of bluegrass from my yard. The soil around the grass clump is kept moist, and every few days I have placed various small insects in the container. Of the three individuals that we brought back, two died within two days. The third individual (these photographs), however, has now survived for four weeks! Moreover, it is a female, and during the past two weeks six larval burrows have appeared in the soil (and another egg was seen on the soil surface just yesterday). Indeed, an egg can be seen in the upper right of the first photo. It remains to be seen whether I will be successful at rearing them to adulthood; however, I’m hopeful this can be accomplished using methods described for C. cursitans (Brust et al. 2005).

Do I still think C. celeripes occurs in Missouri? I don’t know – on one hand, the mixed grass Loess Hill prairie habitats in which the beetle lives in Iowa do extend south into Missouri, and the beetle could be inhabiting them but be easily overlooked for the reasons I’ve already mentioned. However, Missouri’s Loess Hill prairie relicts are small, both in number and in size, and highly disjunct. Such features increase the likelihood of localized extinctions and hamper recolonization through dispersal, especially in flightless species that must traverse unsuitable habitat. With its adult activity period winding down, renewed efforts to locate this species in Missouri will have to wait until next season. Hopefully, the knowledge we gained this season will help this become a reality. For now, the hunt continues…

(closing photo by C. Brown)