HORTSCIENCE 42(3):470–473. 2007. wide and from 0.9 to 1.5 mm deep.

Cytology
determined a chromosome number of 2n =
Clarifying Taxonomy and 4x = 48 (Weaver, Jr., 1969). In contrast, F.
major is found on upland sites in the pied-
mont and mountains of North Carolina, South
Nomenclature of Fothergilla Carolina, Georgia, Alabama, Tennessee, and
Arkansas (Flora of North America Editorial
(Hamamelidaceae) Cultivars Committee, 1993+; Weakley, 2006; Weaver,
Jr., 1969). This species generally is larger in
and Hybrids stature (7–65 dm) than F. gardenii and is
distinguished by larger leaves ranging from
Thomas G. Ranney1,4 and Nathan P. Lynch2 2.5 to 13 cm long and 4.2 to 12.5 cm wide
Department of Horticultural Science, Mountain Horticultural Crops that generally are toothed from below the
middle and conspicuously asymmetric at
Research and Extension Center, North Carolina State University, the base. Stipules are 2.8–7 (10.2) mm
455 Research Drive, Fletcher, NC 28732-9244 long. Stamens generally number (18) 22–32.
The hypanthium at anthesis ranges from
Paul R. Fantz1 2.4 to 3.9 mm wide and from 1.5 to 3 mm
Department of Horticultural Science, North Carolina State University, deep. Cytology determined a chromosome
Raleigh, NC 27695-7609 number of 2n = 6x = 72 (Weaver, Jr., 1969).
Although there is no known diploid species
Paul Cappiello3 of fothergilla, Parrotiopsis (Niedenzu)
Yew Dell Gardens, P.O. Box 1334, 6220 Old LaGrange Road, Crestwood, C. Schneid. is a closely allied genus with
KY 40014-9550 2n = 2x = 24 (Goldblatt and Endress, 1977;
Li and Bogle, 2001; Weaver, Jr., 1969) and
Additional index words. cytology, DNA content, flow cytometry, Fothergilla gardenii, may represent a parallel lineage from an
Fothergilla major, Fothergilla ·intermedia, genome size, hybrid fothergilla, interspecific ancestral diploid.
hybridization, polyploidy, witch-alder Often, the two species of Fothergilla are
confused, but they can be separated by
Abstract. Fothergilla L. spp. are valuable nursery and garden plants. However, clear comparing key characteristics (Clark,
differentiation among F. gardenii Murray, F. major Lodd., and potential hybrids can be 1988). Also, there has been speculation that
difficult based solely on morphological characteristics. The objectives of this work were the two species of Fothergilla hybridize
to verify and describe the existence of interspecific hybrids and to clarify the proper (Dirr, 1998). Hybrids between these species
nomenclature for cultivars of Fothergilla that are commonly grown in the nursery should have a chromosome number of 2n =
industry. A comparison of morphological characteristics was made among diverse clones 5x = 60. Microscopic determination of chro-
representing both species and potential hybrids. A combination of chromosome counts mosome numbers is not a practical approach
and DNA contents was used to clearly differentiate among F. gardenii (2n = 4x = 48), for separating species and hybrids among
F. major (2n = 6x = 72), and hybrids (2n = 5x = 60). It was determined that the majority large numbers of cultivars. However, flow
of cultivars represented in commerce were hybrids. Fothergilla ·intermedia Ranney and cytometry can provide a fast and accurate
Fantz (hybrid fothergilla) is proposed as the name for these hybrids and is validated with determination of nuclear DNA content that is
a Latin diagnosis. Although certain morphological characteristics can be used to related directly to ploidy level (among
differentiate between F. gardenii and F. major, the hybrids tend to be intermediate closely related taxa) and can be used as a
and are particularly difficult to separate from F. major on the basis of appearance. The taxonomic tool (de Laat et al., 1987; Doležel,
correct classification and nomenclature for 17 different taxa are presented. 1991; Doležel et al., 1998; Galbraith et al.,
1983).
The objectives of this research were to
Fothergilla L. spp. (fothergilla or witch- with fall foliage ranging from and including verify the existence of hybrids between
alder; Hamamelidaceae R. Brown) are excep- multicolored combinations of yellow, F. gardenii and F. major and to clarify the
tional garden plants (Clark, 1987; Dirr, 1998; orange, maroon, and scarlet. Fothergilla have proper taxa designations for clones of Fother-
Flint, 1984; Weaver, 1971) that display few pest problems, and they tolerate a broad gilla commonly grown in the nursery industry.
showy, white, fragrant flowers in a terminal range of climates (USDA hardiness zones
spike that resembles a bottlebrush. Summer 4–9), soil types, and shade. As a result, Materials and Methods
foliage color can be dark green to blue-green Fothergilla have become valuable nursery
and garden plants. Plant material and morphology. Collec-
There are two species of Fothergilla: tions of Fothergilla at the North Carolina
F. gardenii Murray and F. major Lodd. Both State University, Mountain Horticultural
Received for publication 15 Nov. 2006. Accepted are native to the Southeastern United States. Crops Research and Extension Center,
for publication 30 Dec. 2006.
This research was funded, in part, by the North
Fothergilla gardenii is found in wet savannas Fletcher, N.C. (NCSU) and Yew Dell Gar-
Carolina Agricultural Research Service, Raleigh, and pocosins in the coastal plains of North dens, Crestwood, Ky. (YDG), were used for
NC 27695-7643, and the North Carolina Associa- Carolina, South Carolina, Georgia, Florida, and this project (Table 1). Morphological mea-
tion of Nurserymen, Raleigh, NC 27607-4904. Alabama (Flora of North America Editorial surements were taken on lamina length,
The authors gratefully acknowledge the excellent Committee, 1993+; Weakley, 2006; Weaver, lamina width, leaf margin dentation location
technical assistance of Tom Eaker and Joel Mowrey Jr., 1969). This species generally is smaller (strictly above the middle, to the middle, or
at the Mountain Horticultural Crops Research and in stature (3–10 dm) than F. major and is extending to below the middle), symmetry of
Extension Center, the staff at the Mountain Horti- distinguished sometimes by smaller leaves leaf base (symmetrical, variable, or asym-
cultural Crops Research Station, and Cassandra ranging from 1.9 to 6 cm long and from 1.3 to metrical), stipule length, stamen number, and
Finger and JoAnne Fischer at Yew Dell Gardens.
1
Professor.
5.2 cm wide that are generally toothed only hypanthium depth and width at anthesis.
2
Research specialist. on the upper half and symmetric at the base. Twelve measurements were taken for each
3
Executive Director. Stipules are 1.5–4 (6.1) mm long. Stamens leaf morphology character, and six measure-
4
To whom reprint requests should be addressed; generally number from 12 to 24. The hypan- ments were taken for each flower morphol-
e-mail [email protected]. thium at anthesis ranges from 1.5 to 2.6 mm ogy character for each clone.

470 HORTSCIENCE VOL. 42(3) JUNE 2007

 length (mm)
Flow cytometry. Holoploid, 2C DNA

6.2–10.9
Stipule

3.9–8.6
contents (i.e., DNA content of the entire non-

4.0–5.0
4.0–5.0

2.5–3.9
3.9–8.8

3.8–8.6

5.0–9.7

5.0–7.0
3.8–6.9
4.6–8.6
4.8–6.9

6.0–7.0
NMy

5.0

10
6
replicated, chromosome complement irre-
spective of ploidy level) were determined via
flow cytometry (de Laat et al., 1987; Doležel,
1991; Galbraith et al., 1983; Greilhuber

Hypanthium
width (mm)
et al., 2005). Nuclei isolation and staining

2.0–2.2
1.7–2.1
1.9–3.1

1.0–1.8
1.1–1.7

1.7–2.6
1.6–2.5
1.8–2.4
1.8–2.5
2.5–3.4
1.7–2.7
1.7–2.5
1.7–2.4
1.7–2.0
1.0–1.6

2.0–3.0
1.4–2.2
followed protocols provided by Partec GmbH
(Münster, Germany). About 12 stamen fila-
ments were chopped with a razor blade in a
petri dish containing 400 mL of extraction
buffer (CyStain ultraviolet Precise P, Partec).
Hypanthium
Depth (mm)

The suspension was filtered through 50-mm

1.8–2.3
1.2–1.7
1.5–2.1

0.8–1.2
0.7–1.7

1.5–2.2
1.5–2.4
1.6–2.3
1.1–2.2
1.5–2.0
2.0–2.6
1.5–2.3
1.5–2.2
1.9–2.4
0.9–1.4

1.7–2.7
1.0–1.6
nylon mesh, and nuclei were stained using
1.6 mL of staining buffer containing 4#,6-
diamidino-2-phenylindole (DAPI) (CyStain
ultraviolet Precise P, Partec). The suspension
Not available

was analyzed using a flow cytometer with

Stamen no.

fluorescence excitation provided by a mer-

16–28

14–22

12–17
14–18

18–23
19–24
17–23
24–30
27–33
14–21
18–26
18–23
22–26
16–19

18–27
16–23
cury arc lamp (PA-I Ploidy Analyzer, Partec).
The mean fluorescence of each sample
was compared with an internal standard of
known genome size (Pisum sativum L.
Asymmetrical

Asymmetrical
Asymmetrical

Asymmetrical
Asymmetrical
‘Ctirad’, 2C = 9.09 pg; Doležel et al.,
Symmetrical

Symmetrical

Symmetrical
Symmetrical
Leaf base

1998). A minimum of 4,500 nuclei were

Variable

Variable
Variable

Variable

Variable

Variable

Variable
Variable
analyzed to calculate the ratio of sample peak
to the internal standard for determining
genome size [2C pg = (mean fluorescence of
sample peak/mean fluorescence of internal
standard peak) · 9.09 pg]. Two to six sub-
Middle to above
Leaf dentation

Above Middle
Above middle

Above middle

Above middle
Above middle

Above middle

Above middle

Above middle
Below middle

Below middle
Below middle

Below middle
Below middle

samples were analyzed for each taxa.

Chromosome counts. Root tips were col-
lected in the morning from newly rooted stem
Middle

Middle

Middle

cuttings of Fothergilla ‘Mt. Airy’ and placed

in 2 mM 8-hydroxyquioline for 3–5 h at
12 C. Roots were then rinsed with cold
(4 C) distilled water and placed in 3:1 solu-
width (cm)

8.0–11.0
Lamina

2.3–3.2
3.0–3.5
3.5–4.0

2.1–2.7
2.1–3.4

4.3–6.8
6.0–6.5
4.8–6.6
5.4–7.7
5.1–9.5
6.5–7.5
5.5–7.8
4.5–6.3
4.6–6.1
5.5–6.0

6.0–8.0

tion of 95% ethanol/propionic acid fixative

for 24 h at room temperature. Samples
were rinsed with cold distilled water, trans-
Table 1. Comparison of selected characteristics among taxa of Fothergilla species and hybrids.

ferred to a 70% ethanol storage solution, and

placed in a refrigerator at 4 C. The following
length (cm)

8.0–11.1
8.0–10.0

9.0–11.5
Lamina

week, samples were removed from storage

3.9–5.4
4.0–5.0
6.5–8.0

3.4–4.4
4.2–5.1

5.6–8.9
7.0–8.5
5.6–8.0
7.6–9.4

7.0–9.5
5.3–8.1
5.3–7.2
7.0–9.0

6.0–8.5

and transferred to 30% aqueous ethanol for

12 min, followed by two 15-min rinses in
distilled water. Roots were then hydrolyzed
for 30 min at room temperature in 1 N HCl
4.32 ± 0.03Z

and then for 15 min at 60 C, followed by

4.20 ± 0.04
4.46 ± 0.04

4.44 ± 0.04
4.36 ± 0.04

5.52 ± 0.02
5.28 ± 0.02
5.30 ± 0.06
5.30 ± 0.02
5.24 ± 0.01
5.22 ± 0.02
5.25 ± 0.12
5.21 ± 0.08
5.32 ± 0.04
5.36 ± 0.10

6.35 ± 0.06
6.18 ± 0.01
size (pg)
Genome

a quick rinse in distilled H2O. Small samples

of root tips were excised and placed on a
glass microscope slide with a drop of 1%
acetocarmine stain, squashed with a cover-
slip, and viewed at 1500·.
YDG-2005–320-A
YDG 2005–317-A

YDG 2005–322-A

YDG 2004–602-A
YDG 2004–535-A
YDG 2005–323-A

YDG 2005–310-A

YDG 2005–318-A
YDG 2005–312-A
Values for genome size are means, n = 2 to 6, ±1 SEM.
NCSU 2004–069

NCSU 2001–047
NCSU 2005–136

NCSU 2005–051

NCSU 2002–128

NCSU 2005–137
NCSU 2001–234
NCSU 2001–235
Accession no.

YDG-2005–1-A

Results and Discussion

Cytological examination of 14 mitotic
cells revealed that Fothergilla ‘Mt. Airy’
was a pentaploid with 2n = 5x = 60 (Fig. 1),
thereby confirming that it is a hybrid between
tetraploid F. gardenii and hexaploid F. major.
‘KLMsixteen’ May Bouquetä
‘KLMfifteen’ Red Monarchä

Flow cytometry was an effective method for

‘KLMtwo’ Beaver CreekÒ

‘KLMG’ Mystic Harborä

NM, stipules not measured.

determining genome size and ploidy levels of

the species and their hybrids (Fig. 2). Fother-
‘Bill’s True Dwarf ’

‘Arkansas Beauty’

gilla ‘Mt. Airy’, a confirmed pentaploid, was

‘Harold Epstein’

‘Eastern Form’
‘Blue Shadow’

used as a reference to compare the approxi-

‘Red Licorice’

‘Windy City’
‘Appalachia’

F. ·intermedia

mate genome sizes (DNA content) for the

‘Sea Spray’
‘Blue Mist’

‘Jane Platt’

‘Mt. Airy’
F. gardenii

different ploidy levels. Mean 2C holoploid

F. major
F. sp.

genome sizes for F. gardenii ranged from

4.2 to 4.5 pg, hybrids ranged from 5.2 to
Taxa

5.5 pg, and F. major ranged from 6.2 to 6.4 pg

y
z

HORTSCIENCE VOL. 42(3) JUNE 2007 471

Fig. 1. Photomicrograph of root tip cell of Fother-
gilla ·intermedia ‘Mt. Airy’ in prophase with
60 somatic chromosomes.

(Tables 1 and 2). Genome sizes within species

and hybrids had a narrow range, providing
clear distinction between the three taxonomic
groups consistent with variations in ploidy
levels. Mean 1Cx monoploid genome size
(i.e., DNA content of one nonreplicated base
set of chromosomes with x = 12) was similar Fig. 2. Flow cytometry histogram of a combined sample containing nuclei from F. gardenii ‘Jane Platt’
at 1.09 pg DNA for F. gardenii, 1.06 pg DNA (2n = 4x = 48), F. ·intermedia ‘Mt. Airy’ (2n = 5x = 60), F. major ‘Arkansas Beauty’ (2n = 6x = 72),
for the hybrids, and 1.04 pg DNA for F. major, and an internal standard, Pisum sativum ‘Ctirad’, with a known 2C holoploid DNA content of 9.09 pg.
indicating that monoploid genome size is The DNA contents of Fothergilla samples were calculated based on mean sample fluorescence relative
to the internal standard.
highly conserved among species and ploidy
level in Fothergilla.
Differentiation between species was often
ambiguous based on foliar and floral charac- Separating hybrids from parental species F. major Lodd. cum characteribus morpho-
teristics (Tables 1 and 2). Ranges for lamina was particularly challenging when based logica intermedius, tamen distinguibili cytolo-
length, stipule length, and hypanthium depth strictly on morphology. Most ranges for gia ambospecies pentaploidis cum
and width tended to overlap between these morphological measurements of hybrids chromosomatum 2n = 60, et genomibus ampli-
two species. Due to considerable variation overlapped with one or the other parent tude 5.2–5.5 pg DNA, et distinguibili latofolius
within species and overlap in ranges between (Table 2). One exception was that the lamina ad F. gardenia et folius dentibus ad super vs.
species in our sample set, leaf margin denta- width of F. gardenii was consistently nar- infra medium ad F. major. Pentaploid hybrid
tion, symmetry of the leaf base, and stamen rower than either F. major or the hybrids. In shrub, 2n = 60 with genome size of 5.2–
number provided little value for separating general, hybrids tended to resemble F. major 5.5 pg DNA. Leaf blade, 5.3–11.1 cm long,
these two species. Lamina width was the only more closely, likely resulting from higher 4.3–7.8 (9.5) cm wide, base asymmetrical
characteristic, with distinct ranges from 2.1 ploidy level and gene dose that was contrib- or variable, margins toothed from above the
to 4.0 cm for F. gardenii and from 6.0 to uted from F. major. middle to below the middle; stipules, 3.8–
11.0 cm for F. major. Although we did not To help clarify the taxonomy and nomen- 10.9 cm long. Fruit and seed typically
compare plant height and emergence of clature of Fothergilla spp., nothospecies lacking. Flowers with hypanthium, 0.9–2.6
flowers relative to foliage, it was reported F. ·intermedia Ranney and Fantz is proposed mm wide and 1.0–3.4 mm deep; stamens, 14–
generally that F. gardenii had a smaller for the hybrid species name in accordance with 30 in number. Holotype: Fothergilla ‘Mt.
mature height and bloomed before leaf emer- Article H.3–5 (Greuter et al., 2000). The new Airy’, plant, 1.5 m tall, NCSU 2006–137,
gence, while F. major had a larger mature hybrid species is described as follows: Notho- Mountain Horticultural Crops Research Sta-
height and bloomed with the emergence of species Fothergilla ·intermedia Ranney and tion, Fletcher N.C., 25 Sept. 2006, Fantz and
new foliage (Clark, 1988; Weaver, Jr., 1969). Fantz hybrida nova a F. gardenii Murray et Ranney 8911 (NCSC). Isotype: NA.

Table 2. Comparison of characteristics of Fothergilla gardenii, F. ·intermedia, and F. major.

Characteristic F. gardenii F. ·intermedia F. major
Chromosomes
Chromosome no.z 2n = 4x = 48 2n = 5x = 60 2n = 6x = 72
Genome size (2C)y 4.2–4.5 rg DNA 5.2–5.5 rg DNA 6.2–6.4 rg DNA
Leaves
Lamina length (cm) 3.4–5.4 (8)x 5.3–11.1 6.0–11.5
Lamina width (cm) 2.1–4.0 4.3–7.8 (9.5) 6.0–11.0
Leaf dentation location Mostly toothed Toothed above, interm., or Toothed from
above the middle below the middle below the middle
Leaf base Symmetrical or variable Asymmetrical or variable Variable
Stipule length 3.9–8.8 3.8–10.9 6.0–10.0
Flowers
Stamen no. 12–28 14–30 16–27
Hypanthium depth (mm) 0.7–2.3 0.9–2.6 1.0–2.7
Hypanthium width (mm) 1.0–2.2 1.0–3.4 1.4–3.0
z
Chromosome numbers for F. gardenii and F. major were determined by Weaver, Jr. (1969).
y
Ranges for cytometry and morphological traits are a compilation of data from Table 1.
x
Numbers in parentheses indicate extreme ranges, but uncommon occurrences.

472 HORTSCIENCE VOL. 42(3) JUNE 2007

 On the basis of this study, we further Dirr, M.A. 1998. Manual of woody landscape Greilhuber, J., J. Doležel, M.A. Lysák, and M.D.
identified the cultivars ‘Appalachia’, ‘Bill’s plants: their identification, ornamental charac- Bennett. 2005. The origin, evolution and pro-
True Dwarf’, ‘Blue Mist’, ‘Harold Epstein’, teristics, culture, propagation and uses. 5th ed. posed stabilization of the terms ‘‘genome size’’
and ‘Jane Platt’ as F. gardenii. Cultivars Stipes Publishing, Champaign, Ill. and ‘‘C-value’’ to describe nuclear DNA con-
Doležel, J. 1991. Flow cytometric analysis of tents. Annal Bot. 95:255–260.
‘Arkansas Beauty’ and ‘KLMG’ Mystic Har- nuclear DNA content in higher plants. Phyto- Greuter, W., J. McNeill, F.R. Barrie, H.M. Burdet,
bor were found to be F. major. The remaining chem. Anal. 2:143–154. V. Demoulin, T.S. Filgueiras, D.H. Nicholson,
cultivars, representing the majority of named Doležel, J., J. Greilhuber, S. Lucretti, A. Meister, P.C. Silva, J.E. Skog, P. Trehane, N.J. Turland,
selections in commerce, including ‘Blue M.A. Lysák, L. Nardi, and R. Obermayer. and D.L. Hawksworth. 2000. International
Shadow’, ‘Eastern Form’, ‘KLMtwo’ Beaver 1998. Plant genome size estimation by flow Code of Botanical Nomenclature (Saint
Creek, one unnamed clone (YDG 2005–323- cytometry: inter-laboratory comparison. Ann. Louis Code). Költz Sci. Books, Königstein,
A), ‘KLMfifteen’ Red Monarch, ‘KLMsixteen’ Bot. (Lond.) 82:17–26. Germany.
May Bouquet, ‘Mt. Airy’, ‘Red Licorice’, Flint, H. 1984. Fothergilla. Horticulture 63(9): Li, J. and A.L. Bogle. 2001. A new suprageneric
‘Sea Spray’, and ‘Windy City’ were hybrids, 12–16. classification system of the Hamamelidoideae
Flora of North America Editorial Committee. 1993+. based on morphology and sequences of nuclear
F. ·intermedia. Flora of North America North of Mexico, and chloroplast DNA. Harv. Pap. Bot. 5(2):
Literature Cited Vol. 3, Magnoliophyta: Magnoliidae and 499–515.
Hamamelidae. Oxford Univ. Press, New York. Weakley, A.S. 2006. Flora of the Carolinas,
Clark, R.C. 1987. Another southern delight: the 25 Aug. 2006. <http://www.efloras.org/ Virginia, and Georgia, and Surrounding Areas
witch alders (Fothergilla spp.). Morton Arbo- florataxon.aspx?flora_id=1&taxon_id=112962> UNC Herbarium, N.C. Botanical
retum Qrtly. 23(3):33–39. Galbraith, D.W., K.R. Harkins, J.M. Maddox, Garden, Univ. of N.C., Chapel Hill. 25 Aug.
Clark, R.C. 1988. A case of mistaken identity: how N.M. Ayres, D.P. Sharma, and E. Firoozabady. 2006. <http://www.herbarium.unc.edu/
to correctly identify Fothergilla. Amer. Nurs- 1983. Rapid flow cytometric analysis of the flora.htm>
eryman 168(4):52–54, 58–59. cell cycle in intact plant tissues. Science Weaver, R.E. 1971. The Fothergillas. Arnoldia
de Laat, A.M.M., W. G öhde, and M.J.D.C. 220:1049–1051. 31(3):89–97.
Vogelzang. 1987. Determination of ploidy of Goldblatt, P. and P.K. Endress. 1977. Cytology and Weaver, Jr., R.E. 1969. Studies in the North
single plants and plant populations by flow evolution in Hamamelidaceae. J. Arnold Arbo- American genus Fothergilla (Hamamelida-
cytometry. Plant Breed. 99:303–307. retum 58:67–71. ceae). J. Arnold Arboretum 50(4):599–619.

HORTSCIENCE VOL. 42(3) JUNE 2007 473