Savanna Blazing Star - and How Science Works (sort of)

Unlike other blazing stars here, this one's flower heads are on stalks.

For 35 years the rare Savanna Blazing Star (Liatris scariosa) and its noble animal and plant associates have given us a wild ride at Somme Prairie Grove. We learned along the way. For starters, let’s take a peek into some of the bad-mannered backstory. 

 

Henry Alan Gleason, an Illinois farm boy, briefly became the cutting edge of ecology - before he was banished by academics who couldn’t tolerate new ideas. As compellingly presented by Jack White, Gleason made a great start. He produced “the first quantitative descriptions of vegetation of any kind recorded by any ecologist anywhere in the world.” So far so good. But he also questioned some principles that back then were considered sacrosanct – for example, that “succession” was always good … and fire always bad. 

 

In five articles published from 1917 to 1939, Gleason battled establishment science – and lost. Today most ecologists would largely agree with Gleason, but at the time, as White put it, “His colleagues shunned him. Not a single ecology text quoted and used Gleason’s ideas for thirty years.” Gleason quit ecology, got hired as a taxonomist at the New York Botanic Garden, and put his mind to assembling the monumental New Britton and Brown Illustrated Flora of Northeastern United States and Adjacent Canada (1952).

 

His book deserves credit for the rehabilitation of the savanna blazing star. Just so you know, two warring books long provided authoritative catalogues and descriptions of the plant species of northeastern North America – Gleason’s book and Fernald’s. (See Endnote “What is a species?”) The traditionalist nature of botany is apparent in the very names of these two standard books. The first, by Merritt Lyndon Fernald, was entitled Gray’s Manual of Botany (1848). But Gray didn’t write it. Asa Gray (of Harvard) did write the first version, but it was later revised by others until, more than one hundred years later, the 1950 version which the title page admitted was “largely rewritten and expanded by” Fernald. The second book was initially written (1896) by Nathaniel Lord Britton and Addison Brown of the N.Y. Botanical Garden; its text was “entirely rewritten” by Gleason, but it’s still called The New Britton and Brown Illustrated Flora of Northeastern United States and Adjacent Canada (1952). 

 

I knew none of this, but I’d been inspired to contribute to conservation, and as this mission clearly needed help, I was eager to learn as much as I could. I bought a copy of Gleason’s flora – three-big-volumes – because I wanted the pictures. Gray is 1632 pages of mostly technical language, with no illustrations of most species. Gleason gives us 1732 bigger pages, with plenty of important technical language too, but also lots of pictures. 

 

As I worked to protect Nature Preserves, I learned to distinguish more and more wildflowers, grasses, and trees. I’d been finding one plant that puzzled me; in Gleason’s book it seemed to be New England blazing star (Liatris novae-angliae), a species not in Swink and Wilhelm’s highly-respected Plants of the Chicago Region (1979).

 

Swink and Wilhelm weren’t alone. It turned out that none of the Illinois plant books recognized the species I thought I was seeing. As my work grew in effectiveness and reputation, I increasingly had the ear of those great botanists Floyd Swink and Gerould Wilhelm. Though very busy people, they sometimes helped me with difficult plants. But in the case of what turned out to be Liatris scariosa, both experts encouraged me to forget it. Swink told me that there were already too many species, and we ought not to be burdening the flora with more divisions. He was being what botanists call “a lumper.” Wilhelm said that the Liatris genus was full of messy hybrids, and looking deeper into this one wasn’t worth the time. Neither Swink nor Wilhelm wanted to hear about Gleason. They said they preferred Fernald.  

 

Yet, for some reason, this beautiful plant rankled my peace. I kept finding different ways to ask them about it. After all, Gleason’s book had those pictures. And a breakthrough emerged as part of my efforts to get scientists and conservationists to pay attention to the tallgrass savanna (see endnote Savanna. It was a nearly lost ecosystem – nearly gone on the ground and equally gone from the mind. It was not among the habitats discussed in the 1979 Plants of the Chicago Region. The Illinois Natural Areas Inventory had looked for it but found only two questionable acres of Grade B and no Grade A at all. I’d been claiming that we had the wrong search image, that we could learn to recognize remnants, and we could bring the savanna back. 

 

Wilhelm, bless his heart, finally listened when I told him that I was finding these plants only in savanna remnants – and indeed remnants that were fast being lost to brush and shade. He generously said, “Well, okay. It will be a lot of work. But if you can bring me examples, preserved on herbarium sheets, from a number of populations, with lists of associated species, I’ll study it and make a determination.” 

When the dust cleared, we had a formal scientific paper authored by Marlin Bowles, Wilhelm, and me. Liatris scariosa var. Nieuwlandii was now the “Savanna Blazing Star” and officially placed on the Illinois Endangered and Threatened Species list. (See more details in Endnote: Swink, Fernald, and Gleason.)

 

Okay, recognizing the plant’s existence was a good step. But what about its conservation? The populations I found were in open woods, rapidly becoming darker in the absence of fire. If some people still wanted to argue about the proper naming of this thing, we don’t much care. Any name is fine. What was really needed was to save it and the rest of its vanishing habitats and associated species of plants and animals in tallgrass savannas and open woodlands. 

 

Before long, a few sites where it occurred got better management. Others were lost to brush. We also wanted to learn more about it through restoration, and we put its seeds into the North Branch Restoration Project seed mixes. Somewhere in our records is preserved our first mention of planting its seed. In the wild, blazing stars take at least a few years to get robust enough to flower. We first recorded flowering plants at Somme Prairie Grove in 1992. There were six. We found nine in 1995. In 1998 there were six again, but then numbers started to rise. Fourteen in 2000. Eighty in 2006. One hundred eighty-nine in 2010.

 

Heedless of where we planted them, this species has seeds that blow in the wind, and plants spread on their own. The 2016 map showed sub-populations of five or more plants in eleven places. These sub-pop numbers varied widely from year to year. At one point we realized that the savanna blazing star is truly happy only in the first year following a burn. This is a highly fire-adapted plant. The graph below shows the burn/no-burn alternation for a sub-pop that benefitted from fire preceding the 2004 growing season and every second year thereafter through 2019.

Numbers of flowering plants of savanna blazing star in Sub-pop B from 2004 to 2019

Another discovery came only recently. Although we recorded 624 savanna blazing star plants in 2023 and new populations continued to show up where none had been planted, we wondered why some formerly thriving sub-pops had gradually faded out and were now gone. In some cases the plants emerged where the savanna seed mix had been planted in spots where enough brush had been cut to allow a fully vigorous growth of the taller prairie grasses. Here the savanna blazing star couldn’t compete once the grasses had fully taken hold. It seems only able to compete where semi-shade (or other factors?) limit the vigor of those grasses. But then in some of the shadier areas, woodland sunflower had taken over, and our Liatris friend also couldn’t compete with it. This was a Goldilocks plant, as are so many. They can compete successfully with all other plant species, but only in particular circumstances. 

 

Another surprise came from one area where we provided this plant with temporary artificial care. Back when we had first noticed some sub-pops fading out, apparently because of competition from temporary aggressive species, we wanted to be sure we’d have an ongoing seed source for a while. Thus we chose one of the fastest growing sub-pops and annually scythed down the aggressive plants, starting this approach at least by 2006 and continuing to this day. This experience led to a whole new restoration approach for most highest-quality, now expanded to many areas (to be discussed in a upcoming post on the concept of “lo-pro”). 

 

The area being discussed here, called Sub-pop M, from which we seemed to learn quite a bit, had no trees overhead or to the south but did have trees shading it for part of the day on both the east and west. Warm-season grasses including big bluestem, dropseed, and Indian grass are scattered throughout but mostly sparse and young. We expect them to increase. 

For now it’s sufficient to know that this small area, artificially maintained at first for assured savanna blazing star seed production, settled down into something good we hadn’t seen elsewhere. So we began scything aggressive species in a variety of habitats nearby. And bit by bit, our blazing stars began showing up in those semi-shaded, scythed areas. For many years all savanna blazing stars in this part of the preserve were only in that one little scythed area, 40 feet by 15 feet (or 12 meters by 5 meters) . In 2024 that long-scythed area had 87 blooming savanna blazing stars, and in the surrounding areas, scythed just the last few years and where none had been prior to scything, we counted 23 blooming plants. 

 

But that was just a part of the discovery. In the long-scythed area, many other high-quality plants were joining them. We were on to something. Following scything, and with increasing competition from quality species, the formerly aggressive species seemed to be fading back towards a play-well-with-others status. Little scything was needed in this area in 2024. We’ll publish more posts on this approach when we can. 

 

The photo below is not a classic “beautiful image” with  vast swaths of color or precious close-ups. But it has the beauty of health, diversity, and increasing success.

With aggressive species scythed selectively for nearly two decades, the ground here is dense with savanna blazing star, prairie obedient plant, big bluestem, cream gentian, gray goldenrod, wood betony, sweet black-eyed Susan, tall coreopsis, rattlesnake master, Penn sedge, dropseed grass, wide-leaved panic grass, azure aster, Short’s aster, and so many others. Ten feet away is unscythed dense tall goldenrod and woodland sunflower, with bare ground beneath. 

 

The deepest pleasures of this work come from the actions - some wise and mistaken - that end up producing trial-and-error-with-occasional-success learning and can lead to the recovery of healthy and happy biodiversity. 

Endnotes

 

Endnote: How to distinguish savanna blazing star from rough blazing star, for sure.

Usually in the Chicago area the savanna blazing star is easily separated from all others by the stalks that hold its flower heads away from the main stem. Rough blazing star normally does not have those stalks. 

Savanna blazing star, showing flower heads on long stalks.

But an especially good feature that may be more consistent is the character of the bracts at the base of the flower heads (technically, "phyllaries"). Savanna blazing stars have bracts that look like plain, tiny green leaves. In the rough blazing star (the only one likely to be confused with the savanna) those bracts look like the bases of iceberg lettuce leaves: they're puckered or curled with whitish or translucent parts.   

The green, leafy bracts of the savanna blazing star.

The bracts of the rough blazing star are puckered and crisped. 

Other features that may catch the eye from a distance include flower heads that are typically larger and more deeply purple in the savanna blazing star. This species also starts blooming later in the season.

According to Wilhelm and Rericha, the rough starts to bloom on July 22 while the savanna waits until August 26.

It you might get a kick out of the technical language that distinguishes these two species in Wilhelm and Rericha' key, here it is:

Rough blazing star: Middle phyllaries bullate, glabrous abaxially; distal phyllaries with broad, uneven, irregularly lacerate, eciliate scarious margins; heads all sessile or on peduncles shorter than the involucres.  

Savanna blazing star: Middle phyllaries non-bullate, glabrous, hirsutulous or cinereous abaxially; distal phyllaries uniformly narrow with entire, slightly erose or ciliate scarious margins; heads sessile to very often on peduncles as long as or longer than the involucres.

On the other hand, I can usually tell which is which with a quick glance from many feet or meters away. You need the technical language only to be sure, when learning the gestalt, or in the case of individuals that have quirky mixes of characteristics. In those cases you can, like most botanists, ignore them. 

Endnote: What is a species?  

 

The word “species” had been defined as “a group of living organisms consisting of similar individuals capable of exchanging genes or interbreeding.” But we now know that many species, especially under stressful conditions, breed across species lines quite shamelessly.

 

“Endangered Species” is a status that brings funding and legal protections – determined by legislatures and votes. But what is and is not a species is determined by scientists. At one time, people would turn to Gleason’s or Fernald’s book to determine what was and was not a species. But we learn more, it gets complicated, and our expertise changes. For fuller definitions and problems on “species”, see the discussion in Wikipedia.

 

Biodiversity conservation focuses at three levels: the gene, the species, and the ecosystem. Species are the easiest of the three to study, but the least important. If a species survives but most of its genetic alleles are lost, then most of that species is also lost, although few would know or understand. Most species and most of their genetics are conserved only in living, changing ecosystems

 

As Lynn Margulis, Merlin Sheldrake, and others have pointed out, we can make it harder to understand ecosystems when we retreat too far into specializations. Science needs better ecosystem science that deals with wholes. At one point Sheldrake was asked whether universities should recognize his specialty, fungi, with full scale departments, on a par with botany and zoology. His response was basicly, “God, no! Not more divisions!” It's time to recognize that “studying and saving healthy  ecosystems” and overall “biodiversity conservation” are vastly more important than just “saving endangered species,” as valuable and important as that is. As important and underfunded as specialists are, wholistic generalists are supported even less. Conservation needs both. 

 

When Illinois botanists accepted the savanna blazing star as a species, they also found some populations in the western part of the state. 

At some point Liatris scariosa was removed from the Illinois Endangered and Threatened list, perhaps because of additional populations found, or because conservationists have successfully made its populations more secure; the Endangered Species Board doesn't tell us.  

Endnote: Swink, Fernald, and Gleason 

 

It's a huge amount of work to wrestle with hundreds of species and figure out which ones deserve to be recognized as "good" species and which do not. To his credit, Fernald seems to discuss this plant under L. ligulistylis (his Liatris "no.12") as follows: 

The anomalous X L. Nieuwlandii... considered by Gaiser, from cytological evidence, a hybrid of n. 12 with a second but unidentified species (although occurring almost wholly outside the range of the western inferred parent, i.e. from Mich. and Wisc. so. into OI., Ind., Ill, and Mo., and flowering one to two months later) ... needs further consideration.

In their good wrestling with this "element" in Swink and Wilhelm's 1979 Plants of the Chicago Region, they included the following observation under “Liatris aspera ROUGH BLAZING STAR”:

 “There is a very interesting element which occurs in Cook County, and probably elsewhere in our area, that probably represents a hybrid between L. aspera and some other species; some of these have been named L. X nieuwlandii and L. X sphaeroidea. Some have been referred to L. ligulistylis, although Fernald restricts that species to an area to the north and west of us. Furthermore, Gleason gives the commonest habitat as damp low places for L. ligulistylis, whereas our plants seem invariably to be found in dry clay areas with a former history of cultivation or other disturbance.” 

 

But that's not the only clue found in Gleason. It was his description and drawing of Liatris novae-angliae that had teased me into wondering if that was the plant I was finding in the rare savanna remnants. Swink didn't mention L, novae-angliae, despite the fact that Illinois is within its range according to Gleason, who expressed that range in abbreviations as follows: “s. Me. to n. Mich, s. to Pa, W. Va, Mo. and Ark.” Thus, if you draw a line (for the western boundary of this plant’s range) from northern Michigan to Missouri, Illinois is well within it.   

I was not the first Illinoian to notice. In 1846, S.B. Mead, a country doctor, published a list of plants and habitats mostly from west central Illinois, and he identified Liatris scariosa as a species of the barrens, a name sometimes used back then for the savanna. Mead was a sufficiently accomplished botanist that a number of species now have the specific name "meadii" - in his honor. When I referenced that list and this plant in support of the idea that the savanna was as real as prairies and forests and equally worthy of conservation, some rebutted by questioning whether the plants Mead referred to as scariosa were the same as the ones I was finding. I looked for confirming herbarium specimens but failed to find any. Then by extraordinary luck I was invited to give a talk at the Royal Botanic Garden at Kew near London. There, in one of the world's great herbariums, I found Liatris specimens collected by Mead. On the sheet to which he had fastened his specimen of rough blazing star, Liatris aspera, he wrote the word "prairies". On his specimen of Liatris scariosa, which sure looked like the ones I was was finding, he wrote the word "barrens." A photocopy of that herbarium sheet now is part of the Morton Arboretum herbarium. 

 

Thus, in the 1994 edition of Plants of the Chicago Region Swink and Wilhelm bring us up to date on this species:

 

Liatris scariosa (l.) Willd. var nieuwlandii Lunell SAVANNA BLAZING STAR … According to Bowles, Wilhelm, & Packard (1988), who discuss its taxonomy, distribution, and ecology, this species is prevailingly a plant of savannas on the Tinley and Valparaiso morainic systems in the Chicago region, growing only on the Morely-Markham-Ashkum silt loam soil catena. … Single Cook and Will county sites containing this species have been managed by prescribed burns. All other Illinois populations are very small and appear vulnerable even to minimal disturbances. Quercus macrocarpa is the characteristic overstory tree ... Forty-one herbaceous species have been recorded withing one meter of Savanna Blazing Star, the more frequent and characteristic being … Arenaria lateriflora, … Lathyrus venosus, Polygala senega, … Taenidia intergerrima, Thaspium trifoliatum.” 

 

I wasn't asked to comment on that draft, which I don't at all intend to quibble with. Swink and Wilhelm's noble 921-page book would never have been finished if the authors stopped to check with everyone on everything. But in retrospect I might suggest that “disturbance” was less the threat to the surviving populations than lack of fire. Indeed, much savanna and woodland biodiversity had to date been saved from the deadly dark only by various disturbances; cutting some trees down or some amount of grazing had allowed some species to survive. I might have added that I had also found this plant under white oaks (not just bur). And perhaps it's worth pointing out that there was something special about five of the forty-one ‘prairie’ and ‘woodland’ species that I’d found growing within one meter of the savanna blazing star. Those species are among the ones that seem to be indicators for the black-soil savanna ecosystem, which was only beginning to be recognized by ecologists as a distinctive community type which has some characteristic species.  

Those five, with their common names, are: 

Arenaria (or Moehringia) lateriflora - grove sandwort

Lathyrus venosus - veiny pea

Polygala senega (or Senega officinalis) - Seneca snakeroot

Taenidia intergerrima - yellow pimpernel

Thaspium trifoliatum - meadow parsnip. 

So how well is the species status of Liatris scariosa now established? Internet searches seem to indicate that it's documented well for many states and regions. 

On the national level, the U.S. Forest Service has a good report indicating that this species, which they describe as relatively uncommon everywhere, includes three varieties:

Liatris scariosa var. Nieuwlandii. "Savanna blazing star." Midwest United States.

Liatris scariosa var. novae angliae. "New England blazing star." Northeastern United States. 

Liatris scariosa var. scariosa. "Devil's bite" Southeastern United States.

Though often very helpful, the U.S. Department of Agriculture, Natural Resources Conservation Service's entry for this species in its PLANTS Database leads with a photo that does not seem to have the characteristics by which we identify the savanna blazing star. It maps all three varieties but shows Nieuwlandii - which it calls Nieuwland's blazing star - reaching all the way from Missouri and Illinois into New York and Connecticut. The maps in the PLANTS Database show all three varieties in Pennsylvania and New Jersey. Are all three really there? Interbreeding or not interbreeding? Or is there just not clear information from those states? 

Endnote: What is a Savanna?

 

This is a more difficult question than you might think - especially if the answer is to be interesting enough to read.

 

This blog has posts called What is a Grade A Prairie? and What is a Grade A Woodland? The answers to those questions seemed basically straightforward.

 

But the word savanna has an especially troubled history. It was little used in the Midwest before John Curtis’ seminal book, The Vegetation of Wisconsin, in which he pointed out that the savanna had been the most common vegetation type in the southern half of the state but by then was largely gone. He used the pioneer term Oak Openings for that major savanna type.

 

The Illinois Natural Areas Inventory, which followed Curtis extensively, defined the “Savanna,” and “Sand Savanna,” and "Barrens" as the principal types. In efforts to explain all this to the public, it has been awkward to explain that the term “Savanna” does not include the “Sand Savanna.” I kind of wish it had been called the Black-soil Savanna or the Tallgrass Savanna or such. Trying to clarify, others have renamed this community “Fine Textured Soil Savanna “ and other names perfectly hopeless to become part of common speech.

 

An additional confusion came from the Illinois Natural Areas Inventory's attempt to solve the “missing piece” problem. They understandably decided to use the word savanna for the ecologically wide gradient between the treeless prairie and the dense forest, so their savanna has a tree canopy of up to 80%. That turned out to be too wide, so subsequent work defined an additional community type, the oak woodland. 

 

There was a basic challenge in reaching the definitions and resulting ecological descriptions that are needed to guide conservation action. No high-quality examples of tallgrass savanna survived. Thus there was little point in being as detailed as was done for the prairie and forest, where there were high-quality models to learn from. The only real way now to understand the tallgrass savanna and oak woodland would be to see if good management could restore high quality.

 

An additional challenge arising in the context of this post is that we named the savanna blazing star in the context of the Illinois Natural Areas Inventory definition of the savanna, as it was back then, that is up to 80% canopy. It could be that, as we continue to learn, we’ll find that the savanna blazing star should have been called the woodland blazing star. Or it could be that this species will thrive in both savannas and woodlands.

 

As for what a savanna is, if you’ve been patiently waiting for an attempt:

The tallgrass or black-soil savanna is a community with substantial amounts of both trees and tall grasses, best defined by its dominant and characteristic species – especially those that most depend on it. The principal trees in northeastern Illinois are bur oaks in drier areas and swamp white oaks in wetter areas. In other parts of Illinois, other oak species may predominate. The typical graminoids are the “prairie grasses” plus many sedges, wood reed (Cinna), and rye (especially Elymus virginicus, riparius, and Canadense). 

 

Perhaps more significant for recognizing quality are the less dominant but more characteristic species. Among the plants these include veiny pea, purple milkweed, cream gentian, bearded wheatgrass, and Maryland snakeroot. Among the birds, characteristic species include the ruby-throated hummingbird, black-billed cuckoo, eastern kingbird, orchard oriole, sparrow hawk, and Cooper’s hawk. Many amphibians, reptiles, invertebrates, and fungi also deserve to be on this list. To be sure about all this we need enough large high-quality sites and experts to study them. The animals will likely play a major role in helping the best sites to recover; without them we’re not looking at a real and sustainable ecological community.

"Barrens" and "glades" are related communities which support some of the same species. In early writings, these terms along with prairie and savanna were often used interchangeably. As they are defined today, barrens occur on very poor soils, which limits what plants can grow there. Glades and dolomite prairies occur where shallow soil, often just a few inches of it, lies on bedrock. 

References

The extensive references to how this new Illinois species was verified are summarized in:

Bowles, M.L., G. Wilhelm, and S. Packard. 1988. The Illinois Status of Liatris scariosa (L.) Willd. var nieuwlandii Lunell. A New Threatened Species for Illinois. Erigenia. Illinois Native Plant Society. Number 10. Pages 1- 26.

Acknowledgements

Hundreds of smart, dedicated people did the work on which this report is based.

Eriko Kojima caught an atrocious number of typos and made helpful editorial suggestions.