Seeds and cuttings I collect around North Carolina, what they turned out to be, and how to grow them for a NC Piedmont native-plant nursery.

← All field guides
🔭 HuntingcuttingPiedmont: Excellent

Coral honeysuckle

Lonicera sempervirens · Caprifoliaceae

The top leaf pairs fuse into a disc with the flower stalk rising through the middle. Nothing else on a Piedmont fencerow does that.

Reference photos

via iNaturalist

Community photos of Lonicera sempervirens to study before I head out: leaf, flower or bract, ripe fruit, bark, whole plant.

🔭 Where & when to hunt

Regions
Roxboro, NC · Person County · north-central Piedmont. Nearest documented wild site is Caswell Game Land at Country Line Creek, 36.399561 / -79.287462, about 27 km west at the WRC gate on the west end of Cobb Rd. Person's own three records: Hurdle Mills, NW Person at 36.4641 / -79.1331, and Moores Mill Rd at Rougemont. Zero-record but high-habitat county targets: Mayo Lake Park, Hyco and Mayo game lands, South Hyco Creek Slopes, Hyco/Castle Floodplain Forest. Central Piedmont backup ranked by wild record count: Eno River corridor 99, Uwharrie NF 22, Umstead 17, Duke Forest 17, Lower Haw SNA 9, Penny's Bend 7, Occoneechee 4, and the Haw near Saxapahaw at 8, mostly roadside and the weakest of the named targets
Season
Late Mar – Apr is peak bloom and the only time the vine is genuinely easy to locate: NC logs 1,316 April iNaturalist records against 76 in August. Ripe fruit late Jul – Sep, tailing into Oct, with sporadic rebloom Jun – Nov. Winter works too, since it holds leaves and the dried fused discs and old whorled peduncles persist at the shoot tips
Habitat
Sunny woodland edges, powerline cuts, woods roads, old fence rows, road and cut banks, upper ravine slopes and bluff rims, grown-up old fields, thinned pine plantations, rock outcrop margins. Dry to mesic and tolerant of damp. Piedmont clay is fine, but it's over-represented on base-rich and rocky ground: diabase, granite outcrop edges, river bluffs. Light is the controlling variable, since deep-shade stems stay sterile and unnoticeable. It twines up 3–5 m saplings or drapes over shrubs, fences and brush piles. Associates: loblolly and Virginia pine, eastern redcedar, sweetgum, oak-hickory edge, greenbrier and blackberry scrub

Native and secure statewide, G5 / NC S5, not tracked and not listed. Weakley calls it frequent in the Coastal Plain and Piedmont, infrequent in the Mountains. Person County's three iNaturalist records and one undatabased NCU sheet reflect survey effort rather than scarcity: Orange 250, Durham 263, Alamance 56, Chatham 132, statewide 2,889. Within 25 km of Roxboro there are 14 wild records, mostly around Rougemont and Bahama and several at residential addresses; within 50 km, 328. In practice it's one vine at a time on a sunlit edge, far less common than Japanese honeysuckle. The online record is also contaminated by garden plants, with 15% of Durham records and 11% of Orange records user-flagged cultivated

✅ Field ID checklist

  • The top one or two leaf pairs are connate-perfoliate, fused into a single elliptic to round disc with the flowering or fruiting shoot rising through the middle. Lower leaves are free, so a mid-stem sample shows nothing. Check shoot tips
  • Leaves opposite, simple, entire, 3–8 cm long, thick and leathery. No teeth, no lobes, no stipules
  • Flip a leaf. The underside reads pale blue-white glaucous rather than plain green. This is the one vegetative check that separates it from Japanese honeysuckle
  • Corolla 30–50 mm long and 10–15 mm at the mouth, a narrow tube, scarlet-orange outside and yellow-orange in the throat, with five short nearly equal rounded lobes 4–8 mm. Barely flared, and no scent
  • Inflorescence a terminal peduncle 2–7 cm carrying one to four spaced whorls of 2–6 sessile flowers above the fused disc. Stacked rings on a bare stalk
  • Berry red to orange-red, translucent, 6–10 mm, in the terminal whorl above the disc. Never in stalked pairs out of the leaf axils
  • Climbs by stem twining only. No tendrils, no aerial rootlets, no thorns. Twigs terete, smooth, glaucous at the tips, and vines twist around one another in ropey pairs
  • Older basal stems yellowish to orange-brown, shredding in thin papery strips. Japanese honeysuckle peels too, so this confirms nothing on its own
  • Winter, Nov–Mar: partial leaf retention, dried fused discs and old whorled peduncles at the shoot tips, crescent leaf scars with three bundle scars

🌱 What to collect

Window
Late Jul – Sep, tail into Oct
Material
Fully red translucent berries 6–10 mm, picked by hand from the terminal whorl above the fused leaf disc. Macerate in water the same day, float off pulp and empty seed, keep the sinkers at roughly 2–5 per berry and about 4 mm. Revisit a vine at two-week intervals, since fruit ripens over a long window and birds strip it fast. Collect from a site with several separate vines rather than one long clone

Best bet: Cuttings, not seed. June softwood, 3–6 in tips off non-flowering shoots, cut below a leaf pair, strip the lowest one or two pairs, 1,000 ppm IBA talc, straight perlite, an inch deep, vented dome in bright shade. Roots in 2–4 weeks. Second wave mid-Jul to mid-Aug on semi-hardwood. For seed, 90 days cold-moist at 34–40°F, cover 3–6 mm, and split the lot with an unstratified control, since no germination trial exists for this species

🏡 Piedmont grow-out

Site fit
Excellent

Person and Alamance sit at 7b–8a, inside every published hardiness range. Heavy red clay is a listed tolerance rather than a caution, though not clay in a low spot, since waterlogging rots it. Give it six hours of sun or more: shade costs bloom, costs pollinator visits and therefore fruit, and raises mildew. Powdery mildew is the main local problem, so water at the base, cut nitrogen, and prune after flowering for airflow. Bloom sits on the previous year's wood, so late-winter pruning removes the spring display. It twines, so it needs wire or lattice and won't cling to a flat wall. Spacing 36–60 in. Semi-evergreen here with occasional hard-freeze dieback. Rated high flammability, so keep it off structures

Look-alikes & how to tell them apart

reference photos via iNaturalist

Each look-alike with the one character that settles it. Tap a photo to zoom.

Japanese honeysuckle (Lonicera japonica) — same twining woody honeysuckle in the same fencerows, opposite simple entire leaves, and it carries red berries in late Aug on the way to black

All leaf pairs separate and free, with flowers and fruit in stalked pairs out of the leaf axils along the stem; coral honeysuckle fuses its top one or two pairs into a perfoliate disc and bears flowers and fruit in whorls at the shoot tip. Backup: flip the leaf, japonica is green beneath and often pubescent where coral honeysuckle is whitish-glaucous. In flower, japonica is white aging cream-yellow, deeply two-lipped with lobes about as long as the tube, and strongly fragrant; coral honeysuckle is scentless with lobes far shorter than the tube. Ripe fruit blue-black against red. 7,945 NC records against 2,889, so this is the one I'll actually meet.

Crossvine (Bignonia capreolata) — reddish-orange two-inch trumpets, yellow inside, Apr–May, on a woody vine in Piedmont bottomland, often up the same tree

Leaves compound, two leaflets with a branched tendril between them, climbing by disc-tipped tendrils; coral honeysuckle leaves are simple with the top pairs fused, and it twines. That works year-round. Backup: cut the stem and crossvine shows a dark Greek cross in cross-section. Third: crossvine fruit is a flat bean-like capsule 4–7 in of stacked winged seeds rather than a berry, which settles it in the collection window. Already in the repo as field guide 03.

Trumpet creeper (Campsis radicans) — orange-red tubular flowers on a woody vine in sun-exposed fencerows and powerline cuts, and coral honeysuckle reblooms straight into its window

Pinnately compound leaves with 7–11 coarsely serrate leaflets, against simple entire leaves. Backup: Campsis climbs by aerial rootlets clinging flat to bark where coral honeysuckle twines on bare stems. Third: the Campsis corolla runs to 3 in and flares widely in loose terminal cymes, and its fruit is a 3–5 in capsule of winged wind-borne seed. Its sap causes contact dermatitis. Already in the repo as field guide 09.

Amur honeysuckle (Lonicera maackii) and the other invasive bush honeysuckles — smothered in bright red paired berries in Sep–Oct, and an arching shrub with a honeysuckle vine tangled through it is an easy mis-collection

A freestanding multi-stemmed shrub to 15 ft, not a vine, with hollow brown pith in stems more than about two years old. Backup: berries paired and nearly stalkless in the leaf axils along the twigs, against a terminal cluster above a fused disc. Third: no connate leaves anywhere on the bush honeysuckles, and maackii leaves taper to a long acuminate point. Berries are mildly toxic; keep them out of a seed lot. Durham 286, Orange 98, Person 0.

Carolina jessamine (Gelsemium sempervirens) — a wiry twining evergreen vine with opposite entire glossy leaves on the same fencerows; the confusion happens out of season, on a leafy non-flowering stem in December

Leaves lanceolate and pointed, glossy green beneath, and never fused — no connate disc anywhere on the plant. Coral honeysuckle leaves are oval to oblong, whitish-glaucous beneath, top pairs fused. Backup: jessamine fruit is a thin flattened dehiscent capsule, brown-green Sep–Nov, not a red berry, and its flowers are golden-yellow and fragrant Feb–May. Highly toxic in all parts, which is reason enough never to taste a honeysuckle flower or nectar to confirm an ID.

Goldflame honeysuckle (Lonicera heckrottii) — a common landscape hybrid that escapes near houses and trailhead beds, and it has connate terminal leaves just like the native, so the fused disc doesn't save me

Deep purplish-pink in bud, opening rose-pink with a creamy-orange throat, and strongly fragrant; wild L. sempervirens is scarlet-orange with a yellow throat and scentless. Fragrance is the check. Backup for the related 'Dropmore Scarlet' (L. × brownii), which is scarlet with connate leaves: it carries hairy leaf undersides and ciliate margins where typical wild sempervirens is glabrous. Var. hirsutula is a real pubescent native variety, so weigh hairs together with how close the vine sits to a planting.

Limber honeysuckle (Lonicera dioica) — the one other native climbing honeysuckle with a connate perfoliate disc, glaucous underleaves, red berries and a terminal cluster; structurally the closest thing in the genus

Corolla strongly two-lipped and gaping, only 15–30 mm, dull yellow-green to purple-bronze, swollen at the base; coral honeysuckle is a narrow nearly regular tube 30–50 mm, scarlet outside and yellow inside. Backup: dioica is a low clambering plant 3–10 ft blooming Jun–Aug, and in NC it's a Mountains-only S2 plant of bogs and rock outcrops with 44 state records and zero in Person. Piedmont pins for it are most plausibly misidentified coral honeysuckle.

Red morning glory (Ipomoea coccinea) and its relatives — small bright red tubular flowers with a yellow throat on a twining vine in a hedgerow in September, right across the fruiting window

Alternate simple heart-shaped or dissected leaves on a herbaceous annual, with a dry capsule for fruit; coral honeysuckle is a woody perennial with opposite leaves, the top pair fused, and a fleshy red berry. Alternate against opposite ends it in one second. Backup: Ipomoea flowers are flat-faced and last one morning where coral honeysuckle tubes persist for days. Its seeds carry ergoline alkaloids and are not a collection target.

How to find it

Light is the controlling variable, so I hunt edges: woods roads, powerline cuts, old fence lines, road and cut banks, upper ravine slopes and bluff rims above rivers, grown-up old fields, thinned pine plantations. Deep-shade stems stay sterile and I'd never notice them. It twines up a 3–5 m sapling or drapes over a shrub, fence or brush pile. It is not a canopy climber like trumpet creeper. Base-rich and rocky ground is over-represented in the record: diabase sills, granite outcrop margins, river bluffs. Penny's Bend and a 1968 Durham sheet both sit on diabase.

The scan target this week is the leaf, not the flower. Look at the tops of saplings and edge shrubs for a single elliptic or round disc with a bare stalk rising through the middle of it. That's the top one or two leaf pairs fused into one, and nothing else on a Piedmont fencerow does it. Then flip a leaf: the underside should read pale blue-white, not plain green. Those two checks together clear Japanese honeysuckle, which is the vine I'll actually be looking at 99 times out of 100.

Timing is honest but awkward. This week sits inside the fruiting window, so a located vine can be carrying ripe translucent red berries in the terminal whorl, full green foliage, and a few late flowers. Records from these exact weeks confirm it: ripe berries at Old North Durham on Aug 27, a tight cluster of about eight in Forsyth on Aug 15, ripe fruit on a vine in a pine in NW Person County on Jul 28, and a fresh whorl of coral tubes photographed in hand within 50 km of Roxboro four days ago. The problem is finding the vine at all. NC has 1,316 April records against 76 in August, and the April number rides entirely on a flower display that carries 20–30 m. That display is gone now.

Field ID checklist

Look-alikes & how to tell them apart

What to collect & when

Best single collection bet is early to mid September, when fruit is ripe and the rebloom flowers are still up to act as a locator. Pick fully red translucent berries by hand from the terminal whorl, macerate the same day, float off pulp and empty seed, keep the sinkers. Nobody publishes a seeds-per-berry count for this species — Go Botany says two or more, the ovary is 2–3 locular, Illinois claims one — so count the first batch and record it. Fruit set is sparse and erratic and the species is at least partly self-incompatible, so an isolated vine may be carrying nothing at all. Revisit at two-week intervals and collect across several separate vines rather than one long clone.

Seed is orthodox. Dried and sealed at 1–3°C it held viability 15 years in the Woody Plant Seed Manual data, though L. maackii seed dropped off fast after dispersal, so test-sow year-one stored seed before committing a batch. Stratification standard is 60–90 days cold-moist at 4°C. There is no germination trial for coral honeysuckle anywhere; it's absent from the WPSM tables, and genus germination runs 37–100%. The real complication is that several Lonicera species carry underdeveloped embryos at dispersal, only 20–40% elongated, which is morphophysiological dormancy — the embryo has to grow before it germinates. That fits Cullina's observation that fall-sown seed may sit until midsummer after a winter outdoors. Cover the seed 3–6 mm; the "surface sow, needs light" advice circulating on garden sites is unsupported and WPSM says light is not necessary. Plan quantity against roughly 15% sown-seed to usable seedling for the genus, not against lab germination percentages, and note that cut tests don't correlate with germination in Lonicera. Piedmont schedule: clean in September, dry-store sealed at 34–38°F, start stratification about January 1, out about April 1 into flats. Run a fall-sown flat alongside it the first year, since Piedmont chill gets interrupted by 60°F breaks.

The route I'd actually bet on is cuttings. Hayden found preformed root initials at the nodes, which is why it roots so readily. Take 3–6 in tips from vigorous non-flowering shoots after the spring flush, cut just below a leaf pair, strip the lowest one or two pairs, insert about an inch in perlite or sand, vented cover, bright shade. Roots come in a few weeks with no hormone, mist or bottom heat needed, though 1,000 ppm IBA talc is cheap insurance on a batch. Cullina's semi-hardwood version runs mid-July to mid-August. No published rooting percentage exists for this species, so I'll log my own. It self-layers at the nodes, so pinning a stem in spring and lifting it the following spring is nearly automatic. Keep six to ten distinct wild genotypes as stock or the block won't set seed.

Provenance screen matters more here than on most targets, because this is one of the most widely sold native ornamental vines. Stay away from houses, old homesites, cemeteries and trailhead landscape beds. Reject anything fragrant, rose-pink or cream. Yellow flowers don't mean L. flava — 'John Clayton' and 'Sulphurea' are pure-yellow selections of this same species, and flava is a western NC granitic-dome plant with zero records here. A vine covered in flowers with no fruit hints at a heavy-blooming selection like 'Major Wheeler'.

One thing I'm deliberately not asserting: pith. The technical descriptions say the twig is hollow with no pith, one look-alike source insists it's solid white and builds a field triage on it, and the "native honeysuckles have solid pith" rule is a shrub-honeysuckle rule that doesn't transfer. Don't use pith either way.

Hazards are mild. The berries are inedible and may cause nausea, but there's no dermatitis or sap hazard and no gloves needed for collection or maceration. The real risks on the trail are the neighbors: Carolina jessamine on the same fencerows is lethally toxic, trumpet creeper sap causes contact dermatitis, and poison ivy climbs the same posts I'll be reaching into.

Where to look near you

Best-documented publicly accessible wild site in range is Caswell Game Land at Country Line Creek, at the WRC gate on the west end of Cobb Rd, 36.399561 / -79.287462. NCU sheet collected 2023-04-16, habitat noted as "vine in small tree," about 27 km due west of Roxboro.

Person County's own three are Hurdle Mills (flowering, 2020-09-21), NW Person at 36.4641 / -79.1331 (ripe fruit on a vine in a pine, 2021-07-28), and Moores Mill Rd at Rougemont (2023-04-11). A bounding box over the whole northern half of the county, Hyco and Mayo lakes included, returns exactly one wild record. Mayo Lake Park, the Hyco and Mayo game lands, South Hyco Creek Slopes and the Hyco/Castle floodplain forest all sit at zero. Those are data holes rather than confirmed absences, and any find there is new county data worth documenting properly.

For a higher-probability day, the Eno River corridor from Hillsborough to Durham carries 99 wild records, and Penny's Bend has 7 on diabase. Uwharrie NF 22, Umstead 17, Duke Forest 17, Lower Haw River SNA 9, Occoneechee 4. The Haw near Saxapahaw returns only 8, all roadside around Graham, Swepsonville and Snow Camp, nearly all April, and not obviously wild — the weakest of the named targets despite being the trail I know best.

Then go back late March through April and map vines properly while they're in bloom, for next year's cutting and seed trips.

References