Monday, July 28, 2014

Another Unexpected Discovery - Out of the Ashes

Michaux's Sumac (Rhus michauxii); Durham County (2014)
A series of unexpected and seemingly unrelated events (including encouragement from great volunteers, a last minute field trip venue change, perfect burn weather one fine spring day, a drenching downpour & flooded creek one late spring day, a new fire management partnership, etc...) conspired to produce spectacular results. 

In mid-May we held a field trip for members of the Friends of Plant Conservation ( Guests and leaders alike were impressed with the results of a spring-time prescribed burn. Photo opportunities abounded (see image, left). There is nothing like seeing freshly sprouting plants invigorated after a burn!

While most people were snapping photos, some on their hands & knees, Lesley Starke called me over. She pointed to the small, hairy shrub (pictured above) and said, "is that what I think it is?"  But she already knew; we both instantly knew!  It's not every day one finds a new population of an endangered species, especially one of the rarest shrubs around.

Lesley Starke admiring her unexpected find


What made this such an unexpected surprise? No previous survey work had associated Michaux's Sumac with these particular soils, and we did not believe the species native to the county.  Sure, there was a planted population elsewhere in the county but what did that mean? We had been frankly fixated on the other rare species present, namely Smooth Coneflower (Echinacea laevigata). Finally, the area had been literally "out of site and out of mind". Other than a single prescribed fire we conducted years ago, this area had received no management attention. The last time we burned nearby,  the area was avoided and had been languishing ever since. Fortuitously, early in the year I made a brief visit to the area with Herb & Pat Amyx.  We observed excellent restoration potential evidenced by numerous rosettes of Parthenium, and Helianthus atrorubens; we speculated on what might appear with appropriate restoration. Game on......and the hard work began.

We got an answer to our earlier speculation this May when over 40 stems of Michaux's Sumac were found!

Sumac Site shortly after prescribed fire & clearing

An abandoned railroad track runs through the middle of the site. In preparation for a controlled burn woody vegetation was cleared around the RR line (which also served as the fire line). The dense, dark woods toward the back right remained uncleared but were included in the initial fire; this tree density was typical across the entire length of the tract. 

Same view as above (approx.) several weeks later;
note flowering Sumac in foreground

Fire scars are evident on boles of shortleaf pines from the prescribed burn & orange-brown vegetation (back left) was top-killed by a more recent prescribed burn. The brush pile is one of many created from hand clearing.

During the prescribed fire, dense piles of pine needles and other leaves burned deeply exposing the RR ties underneath (we had stationed a pump unit there to wet the ties before they could ignite). All we really hoped for was flowering of the common species to benefit pollinators.

Michaux's Sumac (Rhus michauxii) flowering stem (foreground) with fire (background)

Michaux's Sumac (Rhus michauxii) was listed as federally endangered in 1989.  Although also endangered in North Carolina, the species is doing reasonably well in the sandhills region, where protected land and prescribed burning are relatively common. Populations in North Carolina's Piedmont have been all but wiped out. Loss of habitat to development and fire suppression are among the main culprits.

NOTE: Herb Amyx produced an excellent article on the Sumac which can be viewed here: (

Sunday, July 27, 2014

Restoring the Larkspur Savanna

Tall Larkspur (Delphinium exaltatum) approaches its southern range limit in North Carolina. Only a handful of sites have been documented and the species is considered endangered.  Of the 2 documented records in the Piedmont region, one is now extirpated (by the creation of a water line), and the second persists in a narrow slice of habitat between a paved road and a frequently disturbed water line.

Several years ago with the financial support of the NC Natural Heritage Trust Fund (now defunct), we were able to purchase a wooded tract adjoining the lone remaining Piedmont population.  The plan was to restore the densely forested tract hoping the Larkspur could re-occupy more natural habitat away from the constant threats associated with living on the roadside.

Dense patch of Eastern Red Cedar killed by
prescribed fire under large shortleaf pine

Since then, we have conducted 3 or 4 prescribed burns across the tract, combatted invasive species, and cleared some of the dense tree growth present.

The forest has scattered remnant trees with growth forms indicative of savanna growth. However, these were shrouded in dense growth of young Eastern Red Cedar (Juniperus virginiana) and little light reached the forest floor. Prescribed fires and selective tree removal have begun to produce desired results.

After several years, the upland edge of the site has begun to appear much more open.  In the following images a shortleaf pine and dogwood can be seen before and after preliminary restoration; more shortleaf pines become apparent deeper in the woods on the latter image.

Maintaining and expanding the developing "Larkspur Savanna" continues to demand resources. 
Restoration in progress at Larkspur Savanna, 2014 (above)
w/small pile of Eastern Red Cedar logs & brush piles
browned ground was dense Japanese Stilt Grass (Microstegium vimineum

Aerial views show the expanding opening.
Site in 2010, before hand-clearing
Site in 2013, after hand-clearing

Over the last several years we have been lucky to have several dedicated individuals carry the load of our restoration efforts at this site.

Earlier this year, Thomas Blaine has taken on thinning and brush-cutting to expand potential habitat. He shocked me, and perhaps himself, with discovery of the first Tall Larkspur plants outside of the original patch!

Not to be outdone, our local Preserve Stewards, Herb & Pat Amyx discovered a few more plants upslope of Thomas' just days later.  Taken together, these finds have expanded the extent of the Piedmont's lone Tall Larkspur population by more than 100 meters.

Friday, July 25, 2014

Prairie Flora in North Carolina: Buffalo Clover

I don't get phone calls about Buffalo Clover (Trifolium reflexum) very often, but when I do I take them seriously! 

Threatened in North Carolina, barely half of the historically reported locations are believed to still be extant. Like many other species associated with open, historically fire maintained habitats, the populations have dwindled or disappeared along with the habitat.

Nearly all North Carolina sites have been documented in the Piedmont, so hearing of a potential site in the North Carolina mountains (Madison County, north of Asheville), was a bit unexpected. In fact, this would be the western-most population in the state. I traveled out to take a look.

Patti Waltz discoverer of the Buffalo Clover in Madison County.
My first surprise was the steepness. The site climbs approximately 300' in elevation (topping out around 1960') in less than 400 linear feet (that's an 80% slope)!

The site is densely bouldery with numerous ledges and crevices. Widely scattered prickly pear cactus (Opuntia humifusa) made resting on these problematic. Other open sun-exposed rocks have large clumps of species associated with high pH rocks; such as Limestone Goldenrod (Solidago sphacelata) and Shale-barren Aster (Symphyotrichum oblongifolium). Spreading Sunflower (Helianthus divaricatus) was apparent in places, while moister, shadier areas had dense patches of River Oats (Chasmanthium latifolium).

Patch of River Oats under Black Walnut above outcrop

Eastern red-cedar, black walnut, Biltmore Ash (Fraxinus biltmoreana), Georgia hackberry (Celtis tenuifolia), and Red bud are common trees along with patches of aromatic sumac (Rhus aromatica), coralberry (Symphyocarpus orbiculatus), and an occasional fringe-tree (Chionanthus virginicus). Where Patti Waltz removed invasive plants, especially Japanese honeysuckle, the rich, black soil is often exposed. In such pockets, the Buffalo Clover has emerged.

There are a few "protected" sites where Buffalo Clover occurs in the Piedmont, but none are found in such unusual ecological settings.  This is one of the reasons our goal is to protect multiple examples of each species across their natural range in North Carolina. 

Wednesday, July 23, 2014

Purple Loosestrife

I spotted a small patch of Purple Loosestrife (Lythrum salicaria) recently along Interstate 40 in Forsyth County, North Carolina.

Although I have heard rumours that beekeepers have spread Loosestrife in some areas, I saw no European honeybees on these plants. However, several species of apparently native bees were swarming over the plants.

Purple Loosestrife is a well known invasive of many northern states and a host of these have passed regulations restricting imports.  It remains poorly established in North Carolina, where it is listed as a "Class B Noxious Weed" (along with any other Lythrum spp. not native to the state).  Class B status is defined as, "Any noxious weed that is not native to the state, is present in fewer than 20 counties statewide, and poses a threat to the state" (Section .1700, 02 NCAC 48A.1701). One impact of this regulatory listing status is that sale or distribution of Purple Loosestrife is prohibited in North Carolina.

It turns out the small population noted above is scheduled for control action by the North Carolina Department of Agriculture, and will hopefully be eradicated before it sets seed.  Please report any other populations found in NC. 

Sunday, July 13, 2014

The Evil Disappears along with Canada Barberry?

Berberis canadensis showing the bristly leaf margins
Canada Barberry (Berberis canadensis) and the non-native Common Barberry (Berberis vulgaris) are very similar in appearance. According to Weakley's flora the most obvious difference between the two is the number of bristles on the leaf margins (the image on the right shows the relatively sparser bristles on B. canadensis).

Distinguishing these two may be especially relevant given that Common Barberry was apparently heavily naturalized in the eastern US by the early 1900's. In a 1920 survey, over 1,600,000 "escaped bushes" were found in the north-central states (1). The status of B. vulgaris in NC is unclear although it has been documented here (8)

Although Common Barberry was introduced for various beneficial uses, it had also long been considered a potential threat to agriculture for well over a century; "One extraordinary fact is that the barberry bush will produce smut, or something very similar to it, in all corn growing within a considerable distance of soon as the barberry has been thoroughly extirpated, the evil disappears" (2)  

This "extraordinary fact" ended up contributing to what has been considered the largest and most intensive non-native plant control effort in US history (3).  By 1918, USDA and numerous states were implementing broad reaching programs to eradicate Barberry in an attempt to protect wheat and other cereal crops from infestation by black wheat rust, the spores of which were determined to be spread especially by Common Barberry (Berberis vulgaris). At least one botanist posed objections to the wholesale slaughter of the Common Barberry, but never mentioned the potential loss of, or confusion with, Canada Barberry (4) .

Perhaps, this lack of concern was initially based on statements by USDA that native barberries...."seem to be of very little or no importance in spreading rust" (5). However, by 1921, it had been determined that "only one species of native barberry (Berberis canadensis) rusts abundantly enough to be dangerous" (1).  It is unclear how aggressively, if at all, Canada Barberry was pursued for eradication, but given the similarity of appearance to Common Barberry and the wide range of citizens involved in control programs, it is likely that Canada Barberry would have been destroyed if encountered.

Control efforts focused on the north-central states (1) and continued through at least 1981 although it was believed that 98% of the infected areas had been eradicated nearly a decade prior to that (6). It has been estimated that over 500 million Berberis were removed by cooperative eradication programs (3).

The current rarity of Berberis canadensis in the northern parts of the known natural range (PA, VA, IN, IL, MO) could well have been facilitated by Wheat Rust-associated eradication programs. However, such programs apparently did not extend south of Virginia (6); wheat rust spores in the south (Missouri and Kentucky southward) lose viability (1) and therefore, eradication programs were less needed. Although at least one southern state (7) still lists eradication programs as a threat, I have found no reliable evidence for barberry eradication in NC or southward, and am not convinced it was a significant factor in creating rarity in NC, SC, GA or other southern states.

Berberis canadensis with dangling inflorescences

Trifurcate thorns are typical of B. canadensis
(but also B. vulgaris)
It is unclear why Canada Barberry is so uncommon in NC and places southward. Possible explanations are its tendency to associate with higher pH soils (rare in and of themselves) and relatively open sites (also rare and becoming rare).  A couple of years ago I found a large population (which Harry LeGrand later estimated to be NC's largest) in the Uwharrie region of NC, along an open roadside with a host of prairie associates; the site has since been destroyed by road construction. I know of one apparently thriving and protected population which is the source of the images here. More commonly, Japanese Barberry can be found naturalizing in NC.  
Japanese Barberry (Berberis thunbergii);
this specimen apparently planted at the Linville Falls trailhead
Today, in North Carolina's "natural areas" we may need to reconsider the message on the sign below.
(1) Annual Reports of the Department of Agriculture , 1920
(2) The Principles of Agriculture, Thaer 1844, 2 volumes, London
(4) American Botanist, 1918 (William Clute essay)

Saturday, July 12, 2014

Parrot Pitchers

Considered by at least one author, "the most bizarre member of it genus" (Cheek 2008), the Parrot Pitcher Plant (Sarracenia psittacina) has always been one of my favorites. On a recent return to north Florida I relished the opportunity to see these special carnivores once again.

The plants shown here were found in a beautiful wet pine savanna not far from Tallahassee.

[Click the map for a larger view]

Sarracenia psittacina                  
range map                  Source: 

The embedded map suggests, this pitcher plant is largely a Gulf Coast endemic ranging from the Florida parishes of Louisiana eastward through the outer coastal plain of Georgia (although some sources include western Louisiana, it is my understanding this species does not naturally occur there).  These little guys are easy to overlook and just so darn cute I thought I'd just post a few images of them! 

There is no doubt, Pitcher plants are best appreciated in their natural habitats

Friday, July 11, 2014

Prairie Flora in North Carolina: The Partheniums

Dense Quinine patch in Durham County;
plants in the foreground are P. auriculatum, taller
stems in background are P. integrifolium
Wild Quinine (Parthenium integrifolium) and Glade Wild Quinine (Parthenium auriculatum) are frequent and sometimes common members of North Carolina Piedmont diabase woodlands. For unknown reasons, they may co-occur (as in the image to the left) or be found singly at a given site.

Both flower profusely in open, prairie-like sites, especially after burning although individuals may persist vegetatively for long periods under somewhat shaded conditions.

Wild Quinine is the wider ranging of the two, being found nearly throughout the eastern US (where often associated with high quality, prairie or savanna-like habitats). Glade Wild Quinine is the southern cousin ranging roughly from West Virginia to Alabama. The former may be found nearly across NC, while the latter is restricted to a handful of Piedmont counties.


Typical P. auriculatum
basal leaf

Typical P. integrifolium
basal leaf


Both species have similar appearance with large, coarse, basal leaves.

P. integrifolium is much less hairy, often has a dark reddish mid-vein apparent on the larger, mostly basal leaves. It also tends to be a taller and larger plant when flowering (up to 3' or so).  P. auriculatum has a more petite, compact form, and longer blooming period. The toothier leaf margins help distinguish Glade Wild Quinine, and many plants show a sheen of white pubescence even from a distance while the hairs on Wild Quinine are much less developed and obvious.

The upper image (right) is Parthenum auriculatum, the lower is P. integrifolium; both in full flower (an insect has chewed into one on the lower plant.


Wild Quinine patch flowering after late spring
prescribed fire in Durham County
on exposed diabase outcrop

Large expanse of Parthenium integrifolium on a clayey, upland flat with deeper soils derived from diabase.
Site burned early Spring

Wednesday, July 9, 2014

Two Lupines

Sundial Lupine is a characteristic species of many sandy habitats across the eastern US.  In former savannas and barrens of the upper Midwest, southern Ontario, extending into New Hampshire and New York this native Lupine is the only known host plant for the larvae of the Karner Blue Butterfly, a federally endangered species. With the loss of open habitats, lack of fire, and decline of host and nectar plants the butterfly remains in trouble with its fate inextricably linked to the Lupine. During June 2014 visits to the "Oak Openings" region of southeastern Michigan and northwestern Ohio I had my eyes open for this keystone species.

Sundial Lupine (Lupinus perennis) in southeastern Michigan
For the most part, I was a bit too late to see extensive flowering; many of the plants I saw either showed no signs of flowering or were already in fruit, such as shown in the image below.

Sundial Lupine with dehiscent pods at Lou Campbell Preserve (Ohio)
The twisting and natural opening of the pods helps disperse the seeds 

Later, as I traveled northward to the Sanilac region of Michigan, where the sandy lakeplains had disappeared, I was surprised to see a couple extensive Lupine patches interspersed within the matrix of agricultural fields (see below). Of course I was hoping they were remnant savannas.

Unfortunately, upon closer examination, these plants had larger and more numerous leaflets and larger flowers than Sundial Lupine, and the leaflet tips were shaped differently. They turned out to be so-called Garden Lupines (Lupinus polyphyllus) which are not native to Michigan, but rather the Pacific Northwest.  I noticed patches in 3 different locations where plants were clearly naturalizing and apparently had been present in non-cultivated margins of agricultural fields for some time (well away from existing homes).
Lupinus polyphyllus in Michigan
Dots indicate previously documented;
stars show the counties where I
observed populations

I consulted the Michigan flora online ( and found documentation for Lupinus polyphyllus only in the Upper Penisula of Michigan, relatively far from the "Thumb" region of my observations (Tuscola & Huron counties) near Lake Huron.

Normally a range extension like this would be positive, but given the origin of these Lupines, not so least they're pretty!