seedHigh conf.Piedmont: Good
American witch hazel
Hamamelis virginiana Β· Hamamelidaceae
Understory shrub with spidery yellow flowers in late fall and woody capsules that fling their black seed.
Hamamelis virginiana L., Hamamelidaceae. Also called common or southern witch hazel.
Confidence high: several independent characters agree, including a host-specific gall.
Evidence from the photos
- Leaves (IMG_4969, 4971): broadly oval to obovate, up to ~6 in, with a strongly
oblique base where one side attaches lower than the other, and wavy scalloped
(undulate-crenate) margins rather than sharp saw-teeth. Straight pinnate veins run to
the margin. That base-plus-margin combination is the witch-hazel signature.
- Fruit (IMG_4969, 4970, 4993, 4994): clustered fuzzy (tomentose) two-beaked woody
capsules with a persistent calyx cup. Each splits along two valves and ejects 2 shiny
black seeds. Green now, ripening to light brown.
- Spiny gall (IMG_4970, 4995): the spiny witch-hazel gall aphid, Hamamelistes
spinosus. It galls witch hazel specifically, so it works as a biological fingerprint
for Hamamelis.
- Bark (IMG_4972, 4976): thin, smooth, gray-brown, lichen-mottled. Consistent with a
witch-hazel understory stem. The narrow evergreen leaves in the background are
rhododendron or mountain laurel, part of the surrounding escarpment understory.
Look-alikes considered and ruled out
- Hazel alder (Alnus serrulata), the most likely streamside confusion. Alder fruit is
a hard woody cone (strobile), not a fuzzy capsule, and alder leaves are more elliptic
with finely serrate margins and a symmetric base. No cones or catkins in any photo,
and the bases here are strongly oblique.
- American hazelnut (Corylus americana): nuts sit in leafy bracts, and leaves are
doubly-serrate and pointed. Fruit and margins don't match.
- Witch-alder (Fothergilla spp.): same family, similar leaf, but no fuzzy bivalved
capsules like these, and it isn't the host of the spiny gall.
- American beech (Fagus grandifolia): the spiny gall could read as a beechnut husk at
a glance, but beech leaves are elliptic and sharply toothed.
Habitat check
Matches. Witch hazel favors stream banks, moist coves, and woodland margins on acidic,
humus-rich soil, and occurs throughout western NC. The low-lying streamside valley in
Saluda is exactly that.
What it looks like through the year (so no season surprises you)
The same shrub looks completely different depending on when I see it, which is why an
image search can look wrong next to what I found.
- Spring and summer (July, what I photographed): green leaves, small green developing
capsules. No flowers. Witch hazel doesn't bloom in summer.
- Late summer to fall (AugβOct): capsules ripen green, then tan, then brown, then split
and eject black seed. This is the collection window.
- Late fall to early winter (OctβDec): spidery, crinkled, ribbon-like yellow flowers
with dark-red centers on nearly bare branches. This is the look in most photos and
the definitive field confirmation. Go back to this shrub then to verify and to
collect ripe seed.
- Winter: bare gray, smooth, lichen-mottled stems, often with dried empty capsules
still hanging on.
Most showy witch-hazel photos online are the ornamental Asian and hybrid types
(Hamamelis Γ intermedia, H. mollis), bred for dense flower clusters. Wild
H. virginiana flowers more sparsely: same flower form, less pom-pom density.
The big timing issue: mid-July seed is immature
Witch hazel runs an unusual ~14-month cycle. It flowers in late fall (OctβDec),
fertilization is delayed to the following spring, and the capsules then develop all
summer and ripen AugβOct, exploding their seed by late October. The green capsules I
collected in mid-July are this season's immature fruit, so the seeds inside are very
unlikely to be mature or viable.
Mark this shrub and go back in late AugβSept, collecting capsules just as they turn
from green to tan or light brown but before they split. That one change matters more
than any stratification trick.
Seed propagation (when you have ripe seed)
- Ripen and catch the seed. Put nearly-ripe capsules in a closed paper bag in a warm,
dry room. Over days to weeks they dehisce and ballistically eject the black seeds,
and the bag catches them. Separate seed from capsule debris.
- Double dormancy, slow on purpose. Standard protocol is warm, moist stratification
60β90 days, then cold, moist stratification ~90 days (40 Β°F, moist sand or
perlite). Simpler home route: sow fresh seed in an outdoor protected bed in fall and
let nature cycle it.
- Expect patience. A few may sprout the first spring, but most germinate the second
spring or later. Don't discard flats early.
Vegetative propagation
- Cuttings are hard and mid-July is late. Softwood cuttings root best from spring to
early summer, and research on H. virginiana found rooted cuttings routinely failed
to survive their first winter. If I try: firm softwood, rooting hormone, bottom heat
~68 Β°F, mist or humidity tent, 8β10 weeks. Do not repot until the following
spring. Overwinter the rooted cutting undisturbed; repotting the same season is
what kills them.
- Layering is the reliable home method. Peg a low branch to the ground in spring, wound
it lightly, and let it root over a season before severing.
Grow-out in the NC Piedmont
Well suited. H. virginiana is native to the Piedmont, hardy roughly USDA 5β8. Give it
moist, acidic, humus-rich, well-drained soil and part shade (understory), or sun with
adequate moisture. Water through Piedmont summer dry spells and mulch to keep roots
cool. Large shrub or small tree, fragrant yellow ribbon flowers in late fall, good
yellow fall color. Watch heavier Piedmont clay for drainage.
Bottom line / next step
Good native shrub, wrong month. The capsules I collected in July are immature. Flag
this plant's location and go back in late August or September for ripe capsules, then
dry-in-a-bag, warm-then-cold stratify (or fall-sow outdoors), and be ready to wait into
a second spring. For a faster copy of this exact plant, ground layer next spring.
Sources