Content uploaded by Thomas Molnar
Author content
All content in this area was uploaded by Thomas Molnar on Oct 11, 2017
Content may be subject to copyright.
Chapter 2
Corylus
Thomas J. Molnar
2.1 Introduction
The Corylus L. genus contains a wide diversity of
deciduous shrub and tree species that are important
components of many temperate forests across the
Northern Hemisphere, all bearing edible nuts. Its most
widely known and well-studied member, the European
hazelnut (Corylus avellana L.), is also an economi-
cally valuable commercial tree nut crop, ranking fifth
in world production behind cashews (Anacardium
occidentale L.), almonds [Prunus dulcis (Miller)
D.A. Webb], walnuts (Juglans regia L.), and chestnuts
(Castanea spp.) (FAOSTAT 2010). The top hazelnut
producing country in the world is Turkey, which typi-
cally produces more than 70% of the world’s crop,
which was 1,052,001 tons in 2008. Turkey is followed
by Italy, which produces around 15–20% of the total,
and the US, which produces <5%. Other countries
producing significant crops include Azerbaijan,
Spain, Georgia, Iran, France, and China (FAOSTAT
2010). Commercial production is limited to regions
with climates moderated by large bodies of water
that have cool summers and mild winters, such as the
slopes on the Black Sea of northern Turkey or the
Willamette Valley of Oregon, where 99% of the US
crop is produced. The demand for hazelnuts world-
wide is predominated by a desire for round, high-
quality, well-blanching kernels for use in chocolates
and other confectionaries, baked goods, spreads, and
other products. Only 10% or less of the world’s crop
is sold as in-shell nuts. Superior kernels of the Italian
cultivars “Tonda Gentile de Langhe”, “Tonda di Gif-
foni”, and “Tonda Romana” set quality standards of
comparison for the industry.
Hazelnut has a long history of utilization and
production by man, likely predating the Roman era
(Rosengarten 1984; Boccacci and Botta 2009). Despite
this long history, hazelnut breeding is in its infancy
compared to most other domesticated crops. Until only
recently, world production has been based entirely on
traditional selections made from local populations,
whose exact origins have been largely lost with antiq-
uity. Public breeding programs were initiated in Italy
and the US in the 1960s, Spain and France in the 1970s,
and Turkey in the 1980s (Thompson et al. 1996), but
the Corylus genus as a whole remains essentially
untouched by plant breeders. Outside of the US, tradi-
tional cultivars and local selections still represent a
majority of the hazelnuts being established in produc-
tion orchards today (Bozog
˘lu 2005; Tombesi 2005;
Tous 2005; Sarraquigne 2005). Nevertheless, over
the past several decades much has been learned about
hazelnut genetics, biology, and production, and very
effective traditional and molecular genetic improve-
ment techniques have been developed (Mehlenbacher
1994; Thompson et al. 1996; Chen et al. 2005; Molnar
et al. 2005; Mehlenbacher et al. 2006;Go
¨kirmak et al.
2009). While most work has been centered on
cultivated forms of C. avellana, the interspecific
hybridization potential and genetic diversity of the
genus is high, and substantial opportunities exist to
utilize wild species in genetic improvement and
research efforts (Mehlenbacher 1994; Erdogan and
Mehlenbacher 2000a,2001). In this chapter, the his-
tory, current status, and breeding potential of wild
Corylus are discussed, prioritizing the need to con-
serve and better study underutilized wild species.
Few surveys of wild Corylus have been made in recent
T.J. Molnar
Department of Plant Biology and Pathology, Rutgers University,
Foran Hall, 59 Dudley Road, New Brunswick, NJ 08901, USA
e-mail: molnar@aesop.rutgers.edu
C. Kole (ed.), Wild Crop Relatives: Genomic and Breeding Resources, Forest Trees,
DOI 10.1007/978-3-642-21250-5_2, #Springer-Verlag Berlin Heidelberg 2011
15
years, and overdevelopment in many regions has
increased the possibility that some species are exp-
eriencing unchecked genetic erosion. One species,
C. chinensis Franch., is even considered endangered
by the International Union for Conservation of Nature
and Natural Resources (Sun 1998).
2.2 Botany of Corylus
The Corylus genus is widely distributed across tem-
perate regions of the Northern Hemisphere, with spe-
cies found in Japan, Korea, and China, through Tibet,
India, northern Iran, Turkey, the Caucuses, Europe,
and in North America, with none endemic to the
Southern Hemisphere (Kasapligil 1972; Thompson
et al. 1996). Most taxonomists place Corylus L. in
the subfamily Coryloideae of the family Betulaceae,
order Fagales (Chen et al. 1999; Yoo and Wen 2002).
Corylus comprises anywhere from 9 to 25 species,
depending on the authority, with current revisions
based on morphological, molecular, and hybridization
studies suggesting around ten polymorphic species
assigned to four subsections (Thompson et al. 1996;
Erdogan 1999; Erdogan and Mehlenbacher 2000a,b).
Subsection “Corylus” includes three species whose
major similarities include leafy, overlapping invo-
lucres (husks) covering the nuts (Corylus avellana,
C. americana Marshall, and C. heterophylla Fisch.);
subsection “Siphonochlamys” includes three species
with tubular, bristle-covered involucres (C. cornuta
Marshall, C. californica Marshall, and C. sieboldiana
Blume.); subsection “Colurnaea” includes three spe-
cies that grow as single trunk trees (C. colurna L.,
C. chinensis, and C. jacqemontii Decne.); and subsec-
tion “Acanthochlamys” includes only C. ferox Wall,
which has a spiny chestnut-like (Castanea L.) invo-
lucre unlike any other species in the genus (Erdogan
and Mehlenbacher 2000a,b; Whitcher and Wen 2001).
The little-studied paperbark tree species Corylus
fargesii (Franch.) C.K. Schneid. (likely syn. C. papyr-
aceae Hickel) has yet to be officially placed in a
subsection.
Species range in size from small, multi-stemmed
bushes (1 m) to large trees (up to 40 m). All members
are deciduous, with simple, alternate leaves and
monoecious wind pollinated flowers that undergo
anthesis before leaves develop in the spring. Plants are
self-incompatible and diploid; most researchers agree
the chromosome number across the genus is 2n¼2x
¼22 (Thompson et al. 1996). Corylus avellana – the
European hazelnut of commerce and the most widely
studied of the genus – is believed to have a relatively
small genome (0.48 pg/1C nucleus, 413 Mbp) (Bennet
and Smith 1991), with all other members of the genus
unreported.
2.2.1 Subsection Corylus
Corylus avellana: European hazelnut. Plants are
multi-stemmed shrubs 3–10 m tall, with a growth
habit ranging from very erect to drooping. Ornamental
forms also exist that have weeping or contorted
branches. Plants spread by suckers, but the rate and
number of suckers produced from the base of the plant
varies considerably. Shoots are glandular pubescent
and vary in their thickness and branching density.
Leaves range from 5–10 cm in length and are elliptic
to ovate to rounded in shape, slightly cordate at the
base, and have doubly serrate margins. Nuts develop
in clusters of 1–12, each separately enclosed in an
involucre made up of two overlapping, leafy bracts
that vary considerably across the species in terms of
the length, constriction around the nut, indentation and
serration at the apex, and thickness at the base. Nuts of
cultivated forms, which may or may not be released
from the involucre at maturity, are by far the largest of
the genus, although they vary tremendously in size,
shape (from oblate to long and tapered), shell thick-
ness, and percent kernel (ratio of kernel to shell by
weight). Commercial production is found near coastal
areas of Europe, the Caucasus, Asia Minor, and the
Pacific Northwest of the US where the climate is
moderated by large bodies of water. However, the
native range of C. avellana is quite extensive, span-
ning northward to nearly 68N in Norway and 61N
in Finland, eastward through St. Petersburg to 58Nin
the Ural Mountains of Russia, and southward to 32N
in Morocco, bounded in the west by the Atlantic
Ocean. It typically grows as a common understory
shrub and forest edge species in mixed deciduous
forests. Most authorities now include the previously
named species Corylus maxima Mill., C. pontica
Koch., and C. colchica Alb. as members of C. avel-
lana, exemplifying its very diverse and polymorphic
16 T.J. Molnar
nature (Mehlenbacher 1991a; Thompson et al. 1996;
Erdogan and Mehlenbacher 2000a,b). Corylus avel-
lana was one of the first species to colonize Europe
after the last ice age, with pollen records and chloro-
plast DNA variation studies suggesting expansion
from refugia in southwestern France into most of
Europe, except for southern Italy and the Balkans,
where expansion was from local populations (Palme
and Vendramin 2002; Boccacci and Botta 2009).
While it is not certain when the domestication of
hazelnut began (Zohary and Hopf 2004), Boccacci
and Botta (2009,2010) suggest, based on genetic,
historical, and archaeological data, that the species
was independently domesticated in the Mediterranean
(Spain and Italy), Turkey, and Iran. More than 400
cultivars have been described (G€
urcan et al. 2010).
Descriptions were derived from Smolyaninova (1936),
Kasapligil (1972), Deacon (1974), Mehlenbacher
(1991a), and Thompson et al. (1996).
Corylus americana: American hazelnut. Plants are
small multi-stemmed shrubs, 1–3 m tall, that spread
by abundant suckers. Shoots and leaf petioles are
glandular pubescent. Leaves range from 5–16 cm in
length and are generally broadly ovate to round in
shape with the base rounded or slightly cordate. The
leaf apex is acuminate with serrate to doubly serrate
leaf margins. Nuts develop in clusters of 2–8 with each
nut enclosed in an involucre made up of two over-
lapping, enlarged, glandular-pubescent leafy bracts
that are typically two times the length of the nut,
with deep, irregular indentations at the apex. Nuts,
which are retained in the involucre at maturity, are
round to round-compressed in shape and 1.0–1.5 cm in
diameter. Nuts are generally smaller and thicker-
shelled then those of C. avellana but are similar in
flavor and quality with variability observed in produc-
tivity, size, and other attributes across populations and
individual plants. The species is native to a wide part
of eastern North America from Saskatchewan and
Maine in the northeast to Minnesota and southern
Manitoba in the northwest, all the way south to north-
ern Florida and westward to eastern Oklahoma. Plants
grow as a forest edge species and along roadsides,
fence rows, ravines, and streams, as well as in waste
places and tall- and mid-grass prairie habitats. Corylus
americana is considered an important wildlife and
riparian species that is used as a component of shel-
terbelts and as an ornamental due to some plants
expressing attractive bright red and/or pink fall color.
Nuts of C. americana have been harvested and used
locally from wild plants. Several cultivars producing
larger-sized nuts have been selected in the past
(see Sect. 2.5.2). Descriptions were derived from
Weschcke (1954), Drumke (1964), Duke (1989),
Mehlenbacher (1991a), Boufford (1997), and Gleason
and Cronquist (1998).
Corylus heterophylla: Siberian hazel. Plants are
multi-stemmed shrubs 1–3(7) m tall that spread by
abundant suckers. Shoots and leaf petioles are glandu-
lar pubescent. Leaves range from 4–13 cm in length
and are quite variable in shape from elliptic, elliptic-
obovate, broadly ovate, or obovate to suborbicular.
Some plants express two distinctly different leaf
shapes with some having an acuminate apex and
others an abruptly acuminate apex (truncate) and
somewhat bi-lobed leaves with the apex not exceeding
the lateral lobes. Leaves are cordate at the base with
margins irregularly or doubly serrate. Nuts generally
develop in clusters of 1–7 with each enclosed in an
involucre made up of two slightly pubescent to
densely pubescent glandular, bell-shaped leafy bracts
that are normally slightly longer in length than the nut
(although some can range from twice the nut length to
being equal or shorter than the nut, while others are
very short and not well developed). The husk displays
deep, irregular indentations at the apex. Nuts, which
may or may not be retained in the involucre at
maturity, show great variability in size, shape, and
shell thickness, but are generally round to ovoid and
0.7–1.5 cm in diameter. The species is native to a large
area of Korea, Japan, China, eastern Mongolia, and
the Russian Far East where it grows as an understory
shrub in open forests, on forest edges, on deforested
hills, in dry river valleys, and in vast thickets on
mountain slopes. The nuts of C. heterophylla are reg-
ularly harvested from the wild and sold in domestic
markets for food and oil, with some cultivars and
interspecific hybrids selected and grown commercially
in China and Korea. Two botanical varieties are recog-
nized that have a more southerly distribution in China,
which are considered separate species by some autho-
rities. Corylus heterophylla var. sutchuensis (syn.
C. kweichowensis Hu) grows 3–7 m tall and is distri-
buted throughout the Shangxi, Sichang, Hubei, Hunan,
Jiangxi, Zhejiang, and Guizhou provinces. Corylus
heterophylla var. yunnanesis Franch., which can
be found growing in high density in some areas,
grows 1–3(5) m tall primarily in the Yunnan and
2Corylus 17
Sichuan provinces. Descriptions were derived from
Smolyaninova (1936), Kasapligil (1972), Mehlenbacher
(1991a), and eFloras (2009).
2.2.2 Subsection Siphonochlamys
Corylus cornuta: Beaked hazel. Plants are multi-
stemmed shrubs 1–3 m tall, which spread by suckers
and abundant below ground stolons. Shoots and
petioles can be glabrous to pubescent, but non-
glandular. Leaves range from 5–12 cm in length and
are ovate to obovate or narrowly elliptic in shape,
with an acuminate apex. They are slightly cordate
at the base with margins irregularly doubly serrate.
Nuts develop in clusters of 1–6, each tightly enclosed
in a tube-like involucre (beak) that is constricted
beyond the nut, measures 2–4 times its length, and is
densely covered with bristly, irritating hairs. The nuts
are retained in the involucre at maturity and are typi-
cally 1.0–1.5 cm in diameter, very thick shelled, and
ovoid. The species is native to a broad section of North
America, farther north and west than C. americana,
although their ranges overlap significantly. It can be
found in the northeast from Newfoundland and New
Brunswick, Canada, through Maine to South Carolina
and the mountains of Georgia in the southeast, and
west to Alabama. In its northern range, it can be found
west of New Brunswick, through southern Canada,
across the upper Midwest US states, and north into
Manitoba, Saskatchewan, and Alberta. It grows as an
understory plant in open woodlands and clearings, as
well as along moist to dry roadsides, at the edges of
woods, and along streams, often at higher elevations.
It can sometimes form very dense thickets, due to its
stoloniferous habit, and has the ability to re-grow after
forest fires. Corylus cornuta is not widely harvested or
cultivated for its small, thick-shelled nuts, although
several interspecific hybrids have been developed in
attempts to access its extreme cold hardiness. Descrip-
tions were derived from Buckman (1964), Drumke
(1964), Mehlenbacher (1991a,2003), Boufford
(1997), and Gleason and Cronquist (1998).
Corylus californica (syn. C. cornuta var. californica).
Plants are multi-stemmed shrubs 3–4 m tall that spread
by suckers, lacking the stolons of C. cornuta. Young
shoots and petioles are typically glandular pubescent,
although glabrous in maturity. Leaves range from 4 to
7 cm in length and are rounded or obovate to broadly
elliptic, with an obtuse to acute apex. Leaves are
typically more cordate at the base than C. cornuta,
with margins coarsely doubly serrate. Nuts develop in
clusters of 2–4, with each tightly enclosed in a tube-
like bristly involucre that is constricted beyond the nut
and is generally two times its length or shorter. The
nuts are usually larger than those of C. cornuta and
are retained in the involucre at maturity. The species is
native to western North America, from southern Brit-
ish Columbia southward through western Washington,
Oregon, and central California. Plants are generally
found along streams, on damp rocky slopes, and in
cool canyons in the coastal mountain ranges. Corylus
californica is not widely harvested for its nuts.
Descriptions were derived from Drumke (1964), Meh-
lenbacher (1991a), and Boufford (1997).
Corylus sieboldiana (syn. C. mandshurica Maxim.).
Plants are multi-stemmed shrubs, 2–6 m tall that spread
by suckers. Young shoots and petioles are glandular
pubescent. Leaves range from 5–12 cm in length and
are round, broadly ovate, oblong, or oblong-obovate,
with a mucronate-acuminate or caudate apex and cor-
date base. Leaf margins are dentate to irregularly and
coarsely serrate. Nuts develop in clusters of 1–9 with
each tightly enclosed in a tube-like bristly involucre
that is constricted beyond the nut and is generally two
times its length or longer. Nuts are typically small,
thick-shelled, ovoid-globose, and pointed, with some
having thin shells up to 1.5 cm in diameter. They are
retained in the involucre at maturity. The species is
native to Korea, Japan, northern China, and the Rus-
sian Far East (Primorsky and Khabarovsky regions),
where it significantly overlaps the range of C. hetero-
phylla but is much less abundant. It occurs in forest
areas with moist, fertile soil high in organic matter.
Nuts of C. sieboldiana are harvested from the wild,
although harvesting is complicated by the bristly invo-
lucres. Descriptions were derived from Smolyaninova
(1936), Kudasheva (1965), Mehlenbacher (1991a), and
eFloras (2009).
2.2.3 Subsection Colurnaea
Corylus colurna: Turkish tree hazel. Plants are large
single trunk, pyramidal trees, 20–40 m tall, with trunk
diameters ranging from 30–60(120) cm. Young
18 T.J. Molnar
shoots are glandular pubescent, with distinctive corky
and furrowed bark that is light-gray to gray in color.
Petioles are slightly pubescent glandular although
sometimes glabrous. Leaves range from 7–18 cm
in length and are round, oval, ovate, obovate, wide-
elliptical, to slightly lobed in shape with an acute apex
and a cordate to deeply cordate base. Leaf margins are
dentate to doubly serrate. Nuts develop in clusters of
2–10 with each enclosed in a fleshy, glandular-pubes-
cent involucre that is 2–3 times longer than the nut,
open at the apex, and deeply dissected almost to its
base into numerous long-acuminate lobes. Nuts are
ovoid-globose to nearly round to flat-compressed,
sometimes extended-elliptical or angular and
1.0–1.5 cm in diameter, with thick shells that are con-
nected strongly to the involucre at maturity, although
selections have been described that release readily.
The species is native to the Balkan Peninsula, Turkey,
the Caucuses, and northern Iran, growing as scattered
trees in deciduous and mixed coniferous forests. In the
Caucuses, it can be found 840–1,750 m above sea
level in shady, moist deciduous forests with soils
high in organic matter. Nuts are harvested from the
wild and used and sold locally. However, the species
has been more widely used as a source of high quality
timber for construction of buildings, boats, and furni-
ture. It is grown as a low maintenance shade tree in
Europe and the US and has also been used as a non-
suckering rootstock for C. avellana (with limited use
today) for nut production and for ornamental Corylus,
such as the contorted form C. avellana “Harry
Lauder’s Walking Stick” and the weeping form
C. avellana “Pendula”. Descriptions were derived
from Smolyaninova (1936), Kudasheva (1965), Kasa-
pligil (1972), Duke (1989), and Mehlenbacher (1991a,
2003).
Corylus jacquemontii: Indian tree hazel. Plants are
single trunk trees 12–15 m tall. Bark is thinner and less
corky than C. colurna. Leaves range from 15–24 cm
in length and are obovate in shape, with the base
shallowly lobed and margins sharply serrate. Nuts
develop in clusters of 1–5 in involucres up to three
times the length of the nut, similar in appearance to
C. colurna but less fleshy with non-glandular pubes-
cence. The small (up to 1.5 cm), thick-shelled nuts are
generally more easy to remove from the involucre than
C. colurna. The species is distributed across Northeast
Afghanistan, northern India, northern Pakistan, and
western Nepal at 1,900–3,000 m above sea level.
Nuts are harvested from the wild and sold in local
markets. Descriptions were derived from Mehlenbacher
(1991a,2003), Thompson et al. (1996), and Farris
(2000).
Corylus chinensis: Chinese tree hazel. Plants are
large, single-trunk trees, 20–40 m tall, with trunk
diameters up to 2 m. Young shoots and petioles
are sparsely villous, stipitate glandular to glabrescent.
Bark is considerably thinner and smoother than
C. colurna. Leaves range from 8–18 cm in length
and are ovate, ovate-elliptic, or obovate-elliptic in
shape, with an obliquely cordate base and mucronate
or shortly caudate apex. Leaf margins are irregularly
and doubly serrate. Nuts develop in clusters of 2–12,
each in a fleshy, tube-like pubescent to glabrous invo-
lucre longer than the nut and tightly constricted after
the nut with a forked or toothed apex. Nuts are ovoid-
globose and 1.0–1.5 cm in diameter with thick shells
that are strongly attached to the involucre at maturity.
The species is distributed across southern China in
parts of the Shangxi, Sichuan, Hubei, Hunan, Yunnan,
and Guizhou provinces where it is found as scattered
trees on moist, forested mountain slopes. Its timber has
been used in China for furniture and paneling. The
species is now considered threatened due to its scarcity
(Sun 1998). Descriptions were derived from Duke
(1989), Mehlenbacher (1991a), and eFloras 2009.
Corylus fargesii: Paperbark tree hazel. Plants are
single-trunk trees up to 25 m tall. Young shoots and
petioles are pubescent. The peeling bark of older stems
and the trunk is similar to that of river birch (Betula
nigra L.), which is a major distinguishing characteristic
separating the species from C. chinensis. Leaves range
from 6–9 cm in length and are oblong-lanceolate,
obovate-oblong, or lanceolate in shape, with the base
cordate or sub-rounded and the apex acuminate. Leaf
margins are coarsely and irregularly doubly serrate.
Nuts develop in clusters of 2–4 each enclosed in a
tubular involucre that is tightly constricted after the
nut and measuring 2.0–5.0 cm long, with its apex divi-
ded into triangular-lanceolate lobes.Nuts are ovoid-
globose and 1.0–1.5 cm in diameter. The species is
distributed throughout mountain valleys of the Henan,
Sichuan, Hubei, Shangxi, Gangsu, and Guizhou pro-
vinces of China. Descriptions were derived from
Aiello and Dillard (2007), Farris (2000), and eFloras
(2009). Corylus fargesii has not been officially
placed in subsection Colurnaea due to poor represen-
tation in taxonomic studies. However, morphological
2Corylus 19
examination of recent introductions in the US suggests
its likely inclusion.Also, C. fargesii may be a syno-
nym for C. papyraceae Hickel; further investigation is
needed to clarify its taxonomic position.
2.2.4 Subsection Acanthochlamys
Corylus ferox: Tibetan hazel and Himalayan tree hazel.
Plants are single-trunk trees 6–9 (20) m tall. Young
shoots are pubescent, sometimes stipitate glandular or
glabrescent. Petioles are densely pilose when young,
glabrescent later. Leaves are 5–15 cm long and are
ovate-oblong, obovate-oblong, obovate, or elliptic in
shape, with the base obliquely rounded or subcordate
and the apex long acuminate to caudate-acuminate.
Leaf margins are sharply and doubly mucronate serrate.
Nuts develop in clusters of 3–6 in spiny, cup-shaped
involucres unlike any of the other Corylus species. The
clusters are very similar to spiny chestnut (Castanea L.)
burrs. Nuts are ovoid-globose to slightly compressed
and 1.0–1.5 cm in diameter. The species is native to
forested mountain slopes 1,700–3,800 m above sea
level in the eastern Himalayan Mountains from Bhutan,
Northeast India, northern Myanmar, and Nepal to parts
of the Yunnan, Sichang, and Xizang provinces of
China. The botanical variety, C. ferox var. thibetica
(Batal.) Franch., is recognized to differ from C. ferox
by having less spines (or pubescence) on the base of the
involucre and lacking pubescence on vegetative buds
and young shoots. It is found in the Gansu, Guizhou,
Hubei, Ningxia, Shaanxi, Sichuan, Xizang, and Yunnan
provinces of China in mixed forests at 1,500–3,600 m
above sea level. Descriptions were derived from Kasa-
pligil (1972), Mehlenbacher (1991a), Thompson et al.
(1996), Farris (2000), and eFloras (2009).
2.3 Conservation of Wild Corylus
Genetic Resources
Essentially no work has been done to investigate
population structure, genetic diversity, and possible
genetic erosion (loss of genetic resources) of wild
Corylus species. Nearly all efforts have been focused
on cultivated forms of C. avellana largely to better
understand their origin, to fingerprint germplasm
accessions, and to evaluate genetic diversity present
in collections (Bassil et al. 2005a,b,2009; Boccacci
et al. 2006,2008; Boccacci and Botta 2009,2010;
Go
¨kirmak et al. 2009;G
€
urcan et al. 2010). The most
recent survey of Corylus in North America was
conducted by Drumke (1964), though extensive land
development since this time limits the usefulness of
this work. Sun (1998) reported that C. chinensis was
becoming scarce in China, leading to its threatened
status. It is possible that genetic resources of other
Corylus species are in danger of being lost, especially
in highly populated countries or regions that have
undergone widespread deforestation.
2.3.1 World Germplasm Collections
In general, world Corylus germplasm collections con-
sist primarily of cultivated forms of C. avellana and
are located in regions where production occurs. Major
collections include the US Department of Agricul-
ture (USDA), Agriculture Research Service (ARS),
National Clonal Germplasm Repository (NCGR) in
Oregon, US, with a backup collection at Parlier,
California (Hummer 2001; Bassil et al. 2009); the
Hazelnut Research Institute in Giresun, Turkey;
Institut de Recerca I
˙Technologia Agroalimenta
`ries
(IRTA) in Reus, Spain; Institut National de la
Recherche Agronomique (INRA) in Ponte-de-la-
Maye, France; the University of Torino and the Insti-
tute Sperimentale per Frutticoltura in Italy; the (VIR)
Breeding Station, Maykop, and the Russian Academy
of Agricultural Sciences Institute of Floriculture and
Subtropical Cultures, Sochi, Russia; and the Nikita
Botanical Gardens in Yalta, Ukraine. These and a num-
ber of smaller collections are listed in Mehlenbacher
(1991a), Koksal (2000), and Bacchetta et al. (2009).
Throughout nearly all of these collections, wild
Corylus species are very poorly represented, including
uncultivated forms of C. avellana from across its
native range. In recent years, the NCGR and Oregon
State University have increased efforts to collect
cultivated and wild accession of Corylus, and their
collections now total more than 700 accessions
between them representing all major Corylus species
(G€
urcan et al. 2010); however, a number of species are
still lacking, especially when considering their wide
20 T.J. Molnar
geographic range. Continued collection and evaluation
efforts are still needed (Table 2.1).
In addition to specifically designated Corylus col-
lections, many botanical gardens and arboreta around
the world hold specimens of the genus. These speci-
mens represent a valuable, yet largely untapped
resource. Unfortunately, some trees in these settings
appear to be mislabeled as to species or cultivar, and
diversity across institutions may be limited due to
direct sharing of germplasm and/or collaborative col-
lection expeditions. Efforts are needed to identify,
better characterize, and catalog Corylus plants existing
in these settings to make them available for conserva-
tion, research, and breeding efforts. In the US, small
but diverse collections can be found at the Morris
Arboretum in Pennsylvania, the Brooklyn Botanical
Garden in New York, the Dawes and Holden Arboreta
in Ohio, the Arnold Arboretum in Massachusetts, the
Morton Arboretum in Illinois, and likely others, both
public and private.
2.3.2 Germplasm Preservation
Because Corylus seeds cannot be stored much longer
than one year without losing viability, germplasm pres-
ervation is based on tree accessions grown in collec-
tion orchards. The expense of maintaining these orchards
can be prohibitive, as exemplified by the land needed
to grow trees of C. colurna and C. chinensis, which
can reach well over 20 m tall. While a useful means to
capture and preserve the genetic diversity of wild
species is as trees derived from seeds of local origins,
the outcrossing, highly heterozygous nature of Corylus
requires selected genotypes (cultivars) be propagated
by asexual methods to retain their genetic identity.
Clonal propagation adds to the expense and challenge
of obtaining, accurately maintaining, and distribu-
ting Corylus germplasm. This has been traditionally
accomplished by layering, an effective yet inefficient
practice, and to a lesser extent grafting. Grafting pre-
sents problems due to the suckering growth habit of
the rootstocks typically used, excluding C. colurna,
which has been used but is not widely available, partly
due to its slow seed germination and poor ability to be
transplanted in a nursery setting (compared to C. avel-
lana). This problem is especially acute in germplasm
collection settings where rootstock shoots can be con-
fused with the original scion accession. Also, grafted
plants do not have the capability to re-grow after
mechanical injury and disease, like those on their
own roots.
A micropropagation system for hazelnuts was
recently developed that is effective for C. avellana,
as well as interspecific hybrids of C. avellana with
C. colurna,C. americana, and C. heterophylla (Yu
and Reed 1995; Nas and Read 2004; Bacchetta et al.
2008; Gao et al. 2008). Micropropagation allows for
very efficient clonal propagation and it is currently
being used by commercial nurseries in the US. Since
plants are on their own roots, they provide the same
benefits as layered plants, but through a much more
efficient propagation system. Micropropagation also
provides a means for space-efficient in vitro preser-
vation of germplasm. Currently, 70 accessions are
backed up at the NCGR in the form of tissue culture
plantlets, consisting mostly of economically important
Table 2.1 Species accessions held at the US Dept of Agriculture, Agricultural Research Service, National Clonal Germplasm
Repository in Corvallis, Oregon as of July 2010 (NCGR 2010). Subspecies and botanical varieties are included under the species
headings
Species Number of accessions Countries of origin
Corylus americana 45 2
Corylus avellana 464 32
Corylus californica 49 1
Corylus chinensis 12 4
Corylus colurna 21 12
Corylus cornuta 19 2
Corylus fargesii 83
Corylus ferox 52
Corylus heterophylla 63 5
Corylus jacquemontii 64
Corylus sieboldiana 42 6
2Corylus 21
C. avellana cultivars and C. avellana selections from
the wild, with only a few other Corylus species avail-
able at present (Joseph Postman, personal communi-
cation). The aseptic process of tissue culture can also
provide essentially disease- and virus-free plant mate-
rial, which can aid in meeting quarantine regulations
to facilitate sharing of genetic resources and improved
cultivars.
For long-term germplasm preservation, a method to
store hazelnut embryonic axes in liquid nitrogen was
developed. Excised, dehydrated embryos of hazelnut
seeds previously treated to a period of moist-chilling
(stratification) survived freezing in liquid nitrogen to
be thawed and grown successfully in tissue culture
(Normah et al. 1994; Reed et al. 1994). Based on this
work, embryonic axes of seeds from C. americana,
C. colurna,C. heterophylla, and C. sieboldiana were
cryo-preserved in liquid nitrogen and are stored at the
National Seed Storage Laboratory in Fort Collins,
Colorado (Reed and Hummer 2001). Cryopreservation
of Corylus pollen is also possible (Craddock 1987),
and pollen of 53 C. avellana cultivars has been pre-
served in liquid nitrogen for long-term storage at the
NCGR. A clonal cryo-preservation technique is still
needed, however. Research is underway at the NCGR
to develop a method to preserve dormant vegetative
buds. A major challenge to developing the clonal cryo-
preservation of hazelnuts has been grafting the buds,
once thawed, to successfully regenerate new plants
(Joseph Postman, personal communication, 2009).
2.4 Genetic Studies and Genomic
Resources of Wild Corylus
Similar to the germplasm collection efforts discussed
in the previous section, nearly all Corylus research –
including genetic studies and genomic resources –
has been focused on C. avellana. These advances
have increased breeding efficiency and contri-
buted to knowledge of pollen-stigma incompatibility
(Mehlenbacher 1997), as well as helping to clarify
the genetic control of many traits that should prove
applicable to future studies and improvement efforts
utilizing wild Corylus and interspecific hybrids.
The inheritance of qualitative traits characterized for
C. avellana include red leaf color (Thompson 1985),
chlorophyll deficiency (Mehlenbacher and Thompson
1991), cut-leaf habit (Mehlenbacher and Smith 1995),
pollen color (Mehlenbacher and Smith 2002), style
color (Mehlenbacher and Thompson 2004), contorted
growth habit (Smith and Mehlenbacher 1996),
non-dormancy (Thompson et al. 1985), self-compati-
bility (Mehlenbacher and Smith 2001,2006), and
resistance to eastern filbert blight (EFB), a destructive
fungal disease caused by Anisogramma anomala Peck
E. M€
uller that has severely limited hazelnut cultiva-
tion in eastern North America (Fuller 1908; Thompson
et al. 1996), from C. avellana “Gasaway”, “Zimmer-
man”, “Ratoli”, OSU 408.040, and OSU 759.010
(Mehlenbacher et al. 1991; Chen et al. 2005; Lunde
et al. 2006; Sathuvalli 2007). The inheritance of numer-
ous quantitative traits has also been examined, which
should prove applicable in interspecific improvement
efforts. These include EFB resistance (Coyne et al.
1998,2000), bud mite resistance (Thompson 1977a),
pellicle removal (Mehlenbacher and Smith 1988), nut
and kernel defects (Mehlenbacher et al. 1993), and
other morphological and developmental characteris-
tics related to commercial nut production (Thompson
1977b; Yao and Mehlenbacher 2000).
Recent studies have used sequence data from the
nuclear ribosomal DNA internal transcribed spacer
(ITS) region, 5S rRNA, chloroplast matK, and ribulose-
bisphosphate carboxylase (rbcL) gene regions to
discern phylogenic relationships in Corylus and within
the Betulaceae family (Chen et al. 1999; Erdogan
and Mehlenbacher 2000b; Forest and Bruneau 2000;
Whitcher and Wen 2001). Sequences of these genes
for most wild species are accessioned and available
through GenBank (2010); however, the numbers of
genotypes from which the sequence data have been
derived are limited. While these conserved gene
regions are useful to distinguish between species,
they tend to provide low intraspecific resolution,
which limits their use in population studies. Fortu-
nately, microsatellite or simple sequence repeat
(SSR) markers that provide a higher level of resolution
have recently been developed and used to successfully
study genetic diversity and relationships of cultivated
C. avellana (Bassil et al. 2005a,b; Boccacci et al.
2006;Go
¨kirmak et al. 2009), with many markers
(about 200) placed on a saturated genetic linkage
map (Mehlenbacher et al. 2006; Mehlenbacher 2009;
G€
urcan et al. 2010). Bassil et al. (2005b) showed some
SSR markers developed for C. avellana provide cross
amplification in C. americana,C. heterophylla,
22 T.J. Molnar
C. chinensis,C. colurna, and C. californica, although
no wild Corylus genetic diversity studies have yet to
be completed. Work, however, is underway at Oregon
State University, in Corvallis, Oregon, to assess the
genetic diversity of C. americana accessions held in
their collection, the NCGR, and other locations in
the US, using SSR markers (Shawn Mehlenbacher,
personal communication). Hopefully the development
of SSR markers will open doors for similar studies
with additional Corylus species in the near future.
Marker-assisted selection (MAS) techniques have
been developed to identify seedlings carrying the
“Gasaway” single dominant gene for EFB resistance,
as well as resistance genes from “Ratoli”, OSU
408.040, and OSU 759.010, using random amplified
polymorphic DNA (RAPD), amplified fragment
length polymorphisms (AFLP), and other PCR-based
DNA marker systems (Mehlenbacher et al. 2004;
Chen et al. 2005; Sathuvalli et al. 2009). These tools
greatly increase the efficiency of breeding for resis-
tance and should be useful for advanced-generation
interspecific hybrids derived from these breeding
lines. In addition, MAS has been developed for iden-
tifying desired sporophytic self-incompatibly S-allele
genotypes, although to date RAPD markers have
only been identified linked to the S1 and S2 alleles
(Pomper et al. 1998). With further development, this
system will allow breeders to identify S-alleles at a
much earlier stage, saving valuable time and resources
that are currently expended on growing seedlings
until they flower to test incompatibility reactions
against known tester genotypes using florescence
microscopy (Mehlenbacher 1997). Very recent work
involves the development of a bacterial artificial chro-
mosome (BAC) library for the new C. avellana release
“Jefferson” (OSU 703.007), which is intended to
be used for map-based cloning of the “Gasaway”
EFB-resistance gene and the sporophytic incom-
patibility locus (Mehlenbacher 2009; Sathuvalli
and Mehlenbacher 2009). In addition, efforts are in
place to begin sequencing of the genome of C. avel-
lana (S. Mehlenbacher, personal communication,
2010). Once complete, these critical advances will
provide great opportunities to improve the understand-
ing, utilization, and conservation of Corylus genetic
resources.
2.5 Role of Wild Corylus
in Crop Improvement
Plants of Corylus (despite its self-incompatibility sys-
tem) are typically very amenable to breeding com-
pared to nearly all of the other temperate nut crops
of world importance. Generation times are shorter
(3–5 years to maturity), the plant size is smaller in
stature, female flowers stay receptive to pollen for
several weeks or longer, flowers are easy to isolate
from foreign pollen, pollen is readily collected and
stored for up to one year, and hand pollinations can
yield large numbers of hybrid seeds (Thompson et al.
1996). All Corylus species produce edible nuts of
relatively similar quality, although size, shell thick-
ness, amount of fiber on the pellicle, and other char-
acteristics may vary. Furthermore, the interspecific
hybridization potential within Corylus is high, as
discussed by Mehlenbacher (1991a), Thompson et al.
(1996), and Erdogan and Mehlenbacher (2000a).
In general, successful hybrids can be readily created
between members of the same subsection (Corylus,
Siphonochlamys, and Colurnaea), with crosses
between the different subsections more limited, but
in many cases, also possible. It should be noted that
reciprocal differences have been observed in many
crosses, and crosses with C. avellana have been gen-
erally more successful when the species was used
as the staminate (pollen) parent. Interestingly, when
C. californica was used as a pistillate parent, it was
compatible with all other Corylus species tested, sug-
gesting that it may have value as a bridge species
(Erdogan and Mehlenbacher 2000a). Erdogan and
Mehlenbacher (2001) also suggest that sporophytic
incompatibility may exist in wild Corylus species,
similar to that reported for C. avellana, where incom-
patibility is controlled by a single S-locus with multi-
ple alleles (Thompson 1979; Mehlenbacher 1997),
although many undetermined S-alleles and the pres-
ence of other barriers to hybridization appear to be
involved. While further work is needed to clarify
the incompatibility system in wild Corylus, many
desirable and economically useful characteristics are
present that can be accessed through interspecific
hybridization in wild species, which are lacking
in cultivated C. avellana. Traits of interest include
2Corylus 23
non-suckering growth habits, tolerance of alkaline
soils, extreme precocity, early maturing nuts, extreme
cold hardiness, drought tolerance, attractive fall color
and other ornamental attributes, rare incompatibility
alleles, and novel sources of resistance to EFB. Most
genetic improvement efforts will likely continue to be
centered on cultivated forms of C. avellana due to
its superior nut quality, yield, and other commercial
production characteristics, with wild species used as
donor parents to contribute desired traits by following
some form of a modified backcross program. A num-
ber of examples are discussed later in this section
where interspecific hybrids, sometimes difficult to cre-
ate (as in C. colurna C. avellana), are more
easily backcrossed to either parental species, suggest-
ing barriers to hybridization are reduced in advanced-
generation hybrids.
Table 2.2 presents current breeding objectives and
standards of C. avellana seedling evaluation in the
OSU hazelnut breeding program, which is the largest
and longest running hazelnut breeding program in the
world (Mehlenbacher 2009). This table provides min-
imum selection criteria of seedlings that are necessary
to develop improved cultivars to meet the demands
of the international kernel market (Mehlenbacher
2003). Hybrid hazelnuts developed for production in
new regions will not only require enhanced adapta-
tion capabilities, including resistance to local pests
and diseases – especially EFB if grown in North
America – but they will also need to meet kernel
quality standards to be competitive in the world
market. Nearly all genetic studies of hazelnut have
centered on cultivated C. avellana. As such, efforts
are needed to characterize and better understand the
inheritance of traits from wild species to enhance the
efficiency of interspecific breeding efforts. It is espe-
cially important to recognize that the use of wild
species in breeding may be accompanied by varying
levels of undesirable traits, some which are unstudied
and may be linked to characteristics of interest. Fortu-
nately, a number of first-generation hybrids exist from
early breeding efforts, as discussed below, which
should prove useful in breeding more widely adapted
cultivars. However, to maintain and enhance genetic
diversity, breeders will also need to make additional
first generation interspecific hybrids to widen the
germplasm base and to better exploit the potential of
Corylus. To be most effective, much more extensive
collection of wild germplasm – and subsequent evalu-
ation of these collections – must be made to further
understand and best use existing genetic variability.
Table 2.3 presents desirable breeding attributes,
potential limitations, and known interspecific compat-
ibilities of the 11 most widely recognized Corylus
species. Their history of use, if available, and a thor-
ough description of their possible roles in genetic
improvement are discussed below.
2.5.1 Corylus avellana
Due to its outcrossing, highly heterozygous nature,
substantial genetic diversity can be found in the pool
of existing C. avellana cultivars. This wide diversity,
Table 2.2 Objectives and standards of seedling selection in the Oregon State University hazelnut (Corylus avellana) breeding
program, adapted from Mehlenbacher (2003). In addition to these requirements, all plants developed for North America should
express a very high level of resistance to infection by Anisogramma anomala, which causes eastern filbert blight. For many
locations, cold hardiness of plants, especially the male flowers (catkins), is also vital
Objective Age at evaluation Minimum standard (cultivar) Ideal standard (cultivar)
Bud mite resistance 4 ‘Casina’, ‘Clark’ (moderate resistance) ‘Barcelona’ (highly resistant)
Round nut shape 4–5 ‘Tonda Gentile delle Langhe’ ‘Willamette’
High percent kernel 4–5 ‘Tonda Romana’ (48%) Casina (56%)
Precocity 5 At least 35 nuts produced –
Yield (total, consistency) 4–16 ‘Barcelona’ ‘Lewis’
Kernel blanching 5–8 ‘Barcelona’ ‘Negret’
Few nut and kernel defects 4–16 ‘Barcelona’ –
Early maturity 5–8 ‘Barcelona’ ‘Tonda Gentile delle Langhe’
Free-falling nuts at harvest 5–8 ‘Casina’ (70% are free falling) ‘Barcelona’ (95% are free falling)
24 T.J. Molnar
Table 2.3 Summary of positive breeding attributes, possible limitations, and hybridization potential of the eleven most widely recognized Corylus species. Compatibility is based
on Erdogan and Mehlenbacher (2000a) and Thompson et al. (1996)
C.
avellana
C.
americana
C.
heterophylla
C.
cornuta
C.
californica
C.
sieboldiana
C.
colurna
C.
jacquemontii
C.
chinensis
C.
fargesii
C.
ferox
Positive attributes Diversity of cultivated forms ••
Large nut and kernel size ••
Early maturing nuts • •• ••
High yield potential/
productive selections
•• • •
Cold hardiness • •• •• •• •• ••
Drought tolerance • • • • •
Heat tolerance • • •
Small growth habit for high
density planting
•• •
Stoloniferous habit
a
•
Non-suckering tree form • • • •
Releases nuts from
involucres
•• •
Precocious (produce nuts
at young age)
•••
Ornamental attributes
b
•• • • • •
Resistance to EFB • •• ¤ • • ¤ • ? ¤ ¤ ?
Possible limitations Small, thick-shelled nuts • • •• • •• • •• •• •• •
Late maturing nuts • • •
Cold sensitive • • • • •
Husk retains nut at maturity •• • • • • • • •• •• •
Involucres covered with
irritating hairs
•• • ••
Suckering growth habit • •• •• •• •• ••
Stoloniferous growth habit
a
•
Susceptible to EFB •• ¤ ? ?
Not precocious •• • •
Limited germplasm • • • •• •• •• ••
Compatibility (as
female parent)
C. avellana •• •• •• • • ? ? ?
C. americana •• •• • • ? ? ?
C. heterophylla ••• ? ??
C. cornuta ¤••••••••? ??
C. californica •• •• •• •• •• •• •• ? • ? ?
C. sieboldiana ¤•• •••• •• ? ??
C. colurna •• • • •• ? •• ? ?
(continued)
2Corylus 25
Table 2.3 (continued)
C.
avellana
C.
americana
C.
heterophylla
C.
cornuta
C.
californica
C.
sieboldiana
C.
colurna
C.
jacquemontii
C.
chinensis
C.
fargesii
C.
ferox
C. jacquemontii ?? ? ?? ? ?•• ? ??
C. chinensis •• •• •• ? •• ? ?
C. fargesii ?? ? ?? ? ?? ? ••?
C. ferox ?? ? ?? ? ?? ? ?••
• Characteristic observed (cross is compatible)
•• Characteristic very prominent (high level of compatibility)
¤ Characteristic observed on rare occasion but more evaluation needed (compatibility reported but not regular)
? Unknown, has not been evaluated
a
Stoloniferous growth habit is listed as both a positive attribute and a potential limitation, as it can be useful for soil reclamation and planting on marginal sites, but is not
advantageous for commercial production
b
Ornamental attributes vary between species. C. avellana: contorted and weeping stems, red/purple or yellow leaves, dissected leaves; C. americana: pink/red fall color, frilly
involucres, small growth habit; C. heterophylla: lobed and truncated leaf habits; C. fargesii: attractive peeling bark
26 T.J. Molnar
expressed as morphological, phenological, and DNA
sequence variability, has been discussed by nume-
rous authors, including more recently Mehlenbacher
(1991a,b,1997), Thompson et al. (1996), Erdogan
(1999), Boccacci and Botta (2009,2010), Boccacci
et al. (2006,2008), Biodiversity et al. (2008), Go
¨kirmak
et al. (2009), and G€
urcan et al. (2010). World germ-
plasm collections hold numerous cultivars, many of
which are well characterized and, in recent years, have
become more readily available for use in breeding
and research efforts (Bacchetta et al. 2009;NCGR
2010). Access to rapid international mail services
makes it possible to share pollen and scion wood to
facilitate germplasm exchange and breeding efforts
between and within many countries. In addition, effi-
cient breeding techniques have been developed
(Thompson et al. 1996) and many traits vital to
improved nut quality and nut production, such as per-
cent kernel, kernel weight and nuts per cluster, have
been shown to be moderately to highly heritable,
as discussed by Thompson (1977b), Mehlenbacher
(1991a), Thompson et al. (1996), Mehlenbacher et al.
(1993) and Yao and Mehlenbacher (2000). These facts,
in combination with the paucity of modern breeding
efforts and wide genetic diversity of Corylus,haveset
the stage for rapid gains in the genetic improvement of
hazelnut, which is exemplified by several new EFB-
resistant cultivars released from the Oregon State Uni-
versity breeding program. These improved cultivars,
which are the result of systematic breeding efforts
over the past 30 years, will save the US hazelnut indus-
try considerable production costs due to their signifi-
cantly reduced fungicide requirements and other
disease management constraints (Julian et al. 2009a),
and increased yields of high-quality kernels per hectare,
compared to the traditional standard in Oregon, “Bar-
celona” (Mehlenbacher et al. 2007,2008,2009).
While cultivated forms of C. avellana typically
express better nut quality and production characteris-
tics than their wild counterparts, most are not reliably
yielding outside of Mediterranean-like climates. Many
lack cold hardiness, especially of staminate flowers, as
well as other traits necessary for wide adaptation and
consistent yields of nuts. Breeders will need to search
for genetic resources outside of cultivated forms when
intending to significantly expand the regions where
hazelnuts can be produced commercially. Fortunately,
cold-hardy, wild C. avellana exist that grow as far
north as coastal Scandinavia, as well as in the Ural
Mountains of Russia, the Carpathian Mountains of
Poland, and other inland areas of Europe with conti-
nental climates periodically exposed to extreme tem-
peratures. Through intraspecific hybridization, these
cold-hardy wild forms represent a plausible means
to substantially improve the climatic adaptability of
cultivated hazelnuts. Select plants should provide a
more rapid means to achieve this goal then by using
other Corylus species. This is because, barring specific
incompatibility alleles, wild C. avellana is fully com-
patible with the cultivated forms, which would allow
for larger numbers of hybrid seed from controlled
crosses, and the resulting progeny would be fully
fertile, unlike some interspecific hybrids (Erdogan
and Mehlenbacher 2000a). In addition, wild C. avel-
lana exist that express traits amenable to commercial
production that are not widely found in other Corylus
species, such as short involucres that release the
nuts on maturity, thin shells, high percent kernel,
and upright growth habits. Other traits of C. avellana
worthy of exploration include resistance to diseases
and pests, such EFB and big bud mite (Phytoptus
avellanae Nal. and Cecidophyopsis vermiformis Nal.)
(Thompson et al. 1996; Lunde et al. 2000; Molnar
et al. 2007; Chen et al. 2007; Sathuvalli et al. 2010).
Unfortunately, wild C. avellana is poorly represen-
ted in world germplasm collections, especially from
its most northern range. Collection and evaluation
efforts remain necessary to access its full potential in
breeding.
The first breeding work using wild C. avellana to
develop cultivated hazelnuts for colder regions began
in the early 1900s. The most notable was done in the
Michurinsk and Moscow provinces of Russia, as dis-
cussed below, with minor efforts made in Ukraine
(Slyusarchuk and Ryabokon 2001,2005), Belarus
(Volovich and Chripach 1998), and Estonia (Kask
1998,2001). Commercial hazelnut production is cur-
rently limited to southern regions of the former Soviet
Union near Sochi, Russia, the Caucasus Mountains
along the Black Sea, and southern Crimea, Ukraine.
2.5.1.1 I.V. Michurin
In the early 1900s, under the direction of the famous
Russian plant breeder I.V. Michurin, attempts at
developing more cold-hardy hazelnuts were started at
an independent genetics lab in the Tambov province of
2Corylus 27
Russia, which was then part of the Agricultural Acad-
emy of Science of the Soviet Union (the current name
is the All Russian Scientific Research Institute of
Genetics and Breeding of Fruit Species named after
I.V. Michurin). Michurin and his colleagues I.S.
Gorshkov and S.K. Chaplyaev hybridized southern
cultivars with cold-hardy wild C. avellana from the
Tambov region of Russia, with goals of developing
plants that combined the high quality and large nut
size of the southern cultivars with the cold hardiness
of local hazelnuts (Pavlenko 1957). The southern cul-
tivars “Adygeisky” and “Panakhesky”, which were
widely grown in the Krasnodar region of Russia at
the time, were used as pistillate parents for the first
generation hybrids with the northern C. avellana.
From these crosses, they planted a reported 200,000
hybrid seedlings for evaluation at the breeding station
in Michurinsk (Denisova 1975). They assessed this
very large population for cold hardiness, nut yield,
and kernel quality, and selected only cold-hardy plants
that produced nuts with more than 40% kernel by
weight for further evaluation. Only 1% of this large
population met these criteria and these plants were
then used to create a second generation of progeny
with breeding objectives to combine yield and cold
hardiness with resistance to weevils (likely Curculio
nucum L). The southern cultivars “Gigantsky Gallsky”
and “Barcelona” were used as recurrent parents for
this second generation of hybrids. Significant improve-
ment was reported from this generation, as 20% of the
progeny expressed desirable characteristics. From this
work, Pavlenko (1957) described ten selections that
were productive in regions where temperatures drop to
34C. In the years that followed, a third generation
of seedlings was grown, largely from open-pollinated
seed collected from improved selections, with 25% of
the resulting plants expressing improved nut charac-
teristics (Denisova 1975). Breeders eventually selected
53 plants that expressed improved cold hardiness and
very consistent annual yields. However, even in these
advanced generations, it remained a challenge to
obtain high yields. Denisova (1975) describes the top
two selections from the Michurinsk program as very
cold hardy, productive, and pest resistant: Selection
4–24, derived from the open-pollination of an unre-
ported parent, has nuts that are 22 15 15 mm and
45% kernel, which is 53.2% oil by weight; Selection
5–10, also derived from the open-pollination of an
unreported parent, has nuts that are 20 13 12 mm
and 50% kernel, which is 63.2% oil by weight. While
breeding work has been terminated, the Michurinsk
Institute hazelnut collection currently holds more than
50 cold-hardy cultivars and forms (Director N.I. Save-
lev, personal communication, 3 July 2009). Breeding
selections from the Michurinsk breeding efforts,
although not widely available, represent valuable
genetic resources inherently useful for continuing
efforts to develop further improved cold-hardy
cultivated hazelnuts.
2.5.1.2 A.S. Yablokov
A program similar to the one in Michurinsk was
started in the 1930s by A.S Yablokov at the All
Union Scientific Research Institute of Forestry and
Mechanization in Moscow province (now called the
All Russian Scientific Research Institute of Forestry
and Mechanization). After unsuccessful attempts at
growing seedlings of many southern cultivars in
Moscow province to identify cold hardy individuals,
in 1933–1935 a new approach was taken. Mother trees
were selected from local wild C. avellana and were
crossed with pollen collected from cultivars in Sochi
(largely “Barcelona”, but also “Kudryavchik”,
“Cherkesskii II”, “Yevgenia”, and “Brunsvik”).
Reciprocal crosses were also made in Sochi on similar
cultivars using pollen from select wild northern hazel-
nuts. The resulting seedlings were germinated at the
agricultural experiment station in Moscow province
and were field planted in 1938 and 1939 for evalua-
tion. Open-pollinated seedlings of “Barcelona” from
Sochi were planted for comparison. All of the “Barce-
lona” seedlings expressed poor growth and were killed
by frost in their first years of life (Yablokov 1962;
Kudasheva 1965). Around 50% of the hybrid progeny
also perished due to the cold, although they persisted
longer than the “Barcelona” seedlings. Of the remain-
ing progeny, a small proportion continued to grow but
was continually damaged by frost each year and pro-
duced little fruit, while the rest were winter hardy,
vigorous, and productive. From these, several superior
selections were identified as being exceptionally cold-
hardy, producing staminate flowers tolerant of very
cold test winters in Russia with improved nut size
over the wild type. Other selections were found with
nuts similar in size to southern cultivars, although their
hardiness was not remarked upon.
28 T.J. Molnar
Yablokov also collected open-pollinated seeds
from select plants growing at the Michurinsk breeding
station and grew them in Moscow province. From this
plant material, promising hazelnuts were also selected.
In 1948, several of the best hybrid plants were used
in second-generation crosses, mostly with other selec-
tions from the Moscow institute. In addition, large
populations were grown from seeds derived from
open pollination of the best selections. From these
populations, additional improved selections were
identified, with a large focus on plants with red leaves
(Kudasheva 1965).
2.5.1.3 R.F. Kudasheva
In 1954, R.F. Kudasheva continued Yablokov’s breed-
ing and evaluation work at the All Union Scientific
Research Institute of Forestry and Mechanization. In
addition to continuing evaluations of the breeding
nurseries, in 1957 she made controlled crosses with
additional southern cultivars growing in Azerbaijan
with pollen from advanced hybrid selections from
Moscow and wild Moscow plants. From 1957 to
1964, she created and evaluated more than 16,000
hybrids, with plant material grown and evaluated in
Tambov, Moscow, and Krasnodar regions of Russia.
Breeding efforts were largely discontinued in the late
1960s, but a number of notable cold-hardy cultivars
were released, including “Tambovsky pozdniy”,
“Tambovsky rannii”, “Moskovsky rubin”, “Mos-
kovsky rannii”, and “Severnii 42”, several of which
were recommended for planting on the central cherno-
zem region of the central (former) USSR (Pavlenko
1985). Today, a collection of plant material derived
from this early work remains at the All Russian Scien-
tific Research Institute of Forestry and Mechanization.
It holds around 350 wild hazelnut selections from the
Tambov region, as well as around 500 hybrid selec-
tions from the past breeding efforts. Of these, 150 have
red leaves (Eugene Momonov, personal communica-
tion, 2003). A number of the Moscow selections are
also held at the All Russian Scientific Research Insti-
tute of Genetics and Breeding of Fruit Species named
after I.V. Michurin in Michurinsk (Director N.I. Savelev,
personal communication, 3 July 2009). Around 50
selections were imported from the Moscow institute
in 2003 and are under evaluation at Oregon State
University and Rutgers University in New Jersey.
A number of these Moscow plants were found to
be EFB-resistant in Oregon (Sathuvalli et al. 2010).
Select genotypes will be used in breeding and eventu-
ally preserved at the NCGR.
2.5.2 Corylus americana
The wild American hazelnut, Corylus americana, can
be found growing across a wide range of climates and
soils throughout much of eastern North America. Sev-
eral C. americana selections have been identified and
named that produce relatively large size nuts with
good quality (“Rush”, “Winkler”, and “Littlepage”);
however, most plants produce tiny nuts with thick
shells that are of little commercial value. Wild hazel-
nuts were collected more widely in the past for home
consumption and local sale, with few collected today.
Corylus americana is the natural host of the fungus
Anisogramma anomala, which causes EFB disease.
While C. americana can vary in its response to the
fungus, it is typically highly tolerant of the disease,
with infected plants developing only tiny cankers, or
none at all. Alternatively, the European hazelnut,
C. avellana, is highly susceptible to this disease,
which causes severe stem cankering, die back, and
death of most plants within 4–10 years after exposure
(Johnson and Pinkerton 2002). While the European
hazelnut was likely brought to eastern North America
as early as the first settlers from Europe (Rosengarten
1984), its production never became established there,
largely due to EFB and compounded by the harsher
climate of the northeastern US, compared to Europe
(Fuller 1908; Thompson et al. 1996). Alternatively,
European hazelnut production thrived in the Pacific
Northwest for almost 100 years due its Mediterranean-
like climate and lack of EFB in this region. Unfortu-
nately, this situation changed in the late 1960s, with
the introduction of EFB into southwest Washington
(Davison and Davidson 1973). After first devastating
most of the production orchards in Washington, EFB
spread south and is now found throughout the entire
Willamette Valley where its control measures (ample
fungicide applications, pruning, etc.) add much expense
to current production (Julian et al. 2009a).
Fortunately, the typically EFB-tolerant Corylus amer-
icana hybridizes readily with C. avellana (Erdogan and
Mehlenbacher 2000a) and resistant progeny can be
2Corylus 29
recovered from the cross, although transmission of
resistance has not been well studied and appears to
be controlled by multiple genes, as well as single
dominant genes in some genotypes (Thompson et al.
1996; Molnar et al. 2009). It should also be noted
that any hazelnut cultivar developed for North
America should express a high level of tolerance or
resistance to EFB in order to be economically and
environmentally sustainable. A number of attempts
have been made in the past to exploit C. americana
to develop better-adapted, cultivated forms of hybrid
hazelnuts, as discussed in more detail later in this
section. While progress has been made over the past
century to develop improved hybrids, inadequate fund-
ing, intermittent efforts, lack of knowledge about
EFB, and poor access to diverse genetic resources
has limited the achievements of these breeding efforts.
However, the potential for C. americana in breeding
remains very promising, especially in advanced gen-
erations backcrossed to superior forms of C. avellana.
Currently, germplasm collections of C. americana
can be found at the NCGR and OSU, as well as the
USDA National Resource Conservation Service Plant
Materials Center in Elsberry, Missouri, with few
accessions in other collections worldwide. Based on
its extensive native range, existing collections do not
fully represent the genetic diversity present in the
species. Therefore, larger collection and evaluation
efforts are needed to assess and utilize its full potential
for breeding and the conservation of genetic resources.
In addition to cold hardiness and resistance to EFB,
selections made from its southern distribution may add
heat tolerance or low-chill requirements needed to
grow hazelnuts for production in more southern lati-
tudes. Furthermore, select forms of C. americana and
hybrids with C. avellana have shown high yield poten-
tial (Hammond 2006), and their smaller growth habit
may be amenable for developing high-density plant-
ings, which would provide earlier economic returns
than more widely spaced orchards (Julian et al.
2009b). Small-statured plants also open up the possi-
bility of mechanically harvesting crops directly from
the bushes using modified versions of machines simi-
lar to those used for harvesting blueberries (Vaccinium
spp.) or grapes (Vitis spp.). This method of harvesting
would be opposed to collecting nuts from the orchard
floor, as is now done in most commercial settings
outside of Turkey, where hazelnuts remain harvested
from the bushes by hand (Thompson et al. 1996).
Mechanically harvesting the nuts directly from the
bushes would reduce or eliminate the need for clean
cultivation of the orchard floor, which would allow
more sustainable planting on sloping land due to its
reduced opportunity for soil erosion. This method of
harvesting would also lessen the need for new culti-
vars to release their nuts cleanly from the involucre
upon maturity – a trait lacking in most wild Corylus
that is necessary to meet current mechanical harvest
methods. Corylus americana also has value as an
ornamental, as many selections express striking red
and pink fall color, a trait not typically found in others
of the genus. Also, their oversized, frilly involucres
provide an additional ornamental attribute, especially
when hybridized with purple-leaf C. avellana, as the
purple color remains expressed in these tissues late
into the summer even as the leaves typically fade to
dark green.
2.5.2.1 J.F. Jones
After a thorough study of introduced cultivars,
J.F. Jones of Lancaster, Pennsylvania believed there
was little chance production of European hazelnuts
could succeed in the US, outside of the Pacific coast.
To remedy this situation, in 1917 he began attempts to
hybridize cultivars of C. avellana with a locally
selected wild hazelnut, C. americana “Rush”, which
was well-adapted, high-yielding, and produced rela-
tively large-sized nuts for the species (Reed 1936;
Crane and Reed 1937). Jones’ apparent goal was to
combine the cold hardiness of the native species with
the larger nut size and thin shell of the European
cultivars. He was unsuccessful in acquiring hybrid
seed from his crosses for 2 years during which he
used “Rush” solely as the staminate (pollen) parent.
Finally, in 1919 he used “Rush” as the pistillate parent
and subsequently developed the first reported interspe-
cific hybrids of the two species. He used C. avellana
“Barcelona”, “Cosford”, “Daviana”, “Italian Red”,
and “Duchilly” as staminate parents and grew and
evaluated many hybrid progeny. Unfortunately, he
passed away in 1928 before final evaluations of his
hybrids could be made. However, two plants that stood
out early as being superior were released from his
estate in the 1930s. They were named “Bixby” (“Rush”
“Italian Red”) and “Buchanan” (“Rush” “Barce-
lona”). While neither proved to warrant commercial
30 T.J. Molnar
planting, they were described as being productive,
cold-hardy, and suitable for home cultivation (Reed
1936; Crane and Reed 1937; ASHS Press 1997).
While “Bixby” appears to be lost from cultivation, C.
americana “Rush” and “Buchanan” are available from
the NCGR.
2.5.2.2 C.A. Reed
The use of “Rush” as a parent in interspecific hybridi-
zations was continued by C.A. Reed of the Bureau
of Plant Industry, US Department of Agriculture in
Beltsville, Maryland. From 1928 to 1932, Reed made
hybrids with “Rush”, using C. avellana staminate par-
ents similar to Jones’, as well as using “Hall’s Giant”,
“Kentish Cob”, “Red Aveline”, and several others. He
also made crosses, although to a much lesser extent,
with C. americana “Winkler” from Iowa and “Little-
page” from Indiana, as well as intraspecific crosses
between various C. avellana cultivars growing at the
Bixby nursery in eastern Long Island, New York
(Reed 1936; Crane and Reed 1937). From Reed’s
breeding effort, around 2,000 plants were provided
for evaluation at the USDA experiment station in
Maryland. Seedlings were also sent for evaluation to
the New York State Agricultural Experiment Station
in Geneva, New York. No additional hybrids were
made by Reed; however, the resulting progeny were
evaluated for many years to follow. While the pure
C. avellana plants showed little increase in adaptation
over their parents, the hybrids with “Rush” showed
promise. In 1951, two superior plants were selected
and released: “Potomac”, a hybrid of “Rush”
“Duchilly”, and “Reed”, a hybrid of “Rush”
“Hall’s Giant”. While both were described as cold-
hardy and productive under eastern conditions, neither
proved to be of commercial value (Crane and McKay
1951; Reed and Davidson 1958; ASHS Press 1997).
“Potomac” was Later reported to be resistant to eastern
filbert blight (EFB), while “Reed” was found to be
susceptible (Lunde et al. 2000). Both cultivars are
available from the NCGR.
2.5.2.3 G.H. Slate
Hazelnut research was initiated in 1925 at the
New York State Agricultural Experiment Station in
Geneva, New York by G.H. Slate. The early goal was
to collect and evaluate a wide range of C. avellana
cultivars for production in New York; at its largest, the
collection held about 120 cultivars imported from
Europe, as well as several C. americana selections
and interspecific hybrids. It became evident that most
clones of C. avellana were poorly adapted to
New York conditions, with staminate flowers and
wood proving to be only marginally hardy (Slate
1935,1936,1947). Efforts to develop better adapted
hazelnuts were initiated in 1930 when Slate first made
crosses with “Rush” plants growing in Ithaca,
New York, with pollen collected from several C. avel-
lana cultivars held in the Geneva collection. Addi-
tional crosses with “Rush” were made in Ithaca in
1931 and 1933. In 1932, intraspecific crosses were
made at Geneva using “Barcelona” and, to a much
lesser extent, “Duchilly” as pistillate parents, crossed
with several C. avellana cultivars that showed promise
in the collection. No additional crosses were made by
Slate (1936,1947). The same year, however, 535
hybrid seedlings were planted that had been derived
from crosses made by Reed of the Bureau of Plant
Industry in Maryland. In total, nearly 2,000 hybrid
seedlings were planted and thoroughly evaluated at
the experiment station, of which 1,232 were offspring
of “Rush”. Plants were evaluated for nut characteris-
tics including size, shell color, kernel quality, and
yield, as well as plant growth habit and cold hardiness.
By 1947, 340 of the 2,000 seedlings showed merit
and were retained for further observation, with 52
selected for propagation and testing in a new orchard.
Nearly all were progeny of “Rush”, with only a few
progeny of “Barcelona”, and no “Rush” “Winkler”
or “Rush” “Littlepage” hybrids (Slate 1947). In
1961, Slate reported that the performance of the
selected hybrids moved to the new orchard, which
was more exposed to winds and on poorer soil,
was much less satisfactory than in the original plant-
ing. Winter injury of the wood and catkins was more
serious and crop yields were light, which made eva-
luations challenging. Furthermore, bud mites infested
the orchard to further reduce crops and limit opportu-
nities to evaluate the selections (Ourecky and
Slate 1969). By 1980, only 24 of Slate’s original
selections remained at the experiment station, of
which 23 were progeny of “Rush” (Reich 1980).
Today, nothing remains of the hazelnut breeding
efforts in Geneva; however, a number of Slate’s most
2Corylus 31
promising breeding selections are available at the
NCGR where they can contribute to future breeding
efforts. The drawbacks of using a limited diversity of
C. americana parents are present in the body of work
published by Slate. In addition to bud mite suscepti-
bility inherited from “Rush”, most did not release the
nuts from the husk on maturity, many were not suffi-
ciently cold-hardy, and they generally produced low
numbers of catkins (Slate 1961). After several decades
of work, no cultivars were released from Slate’s efforts.
However, useful EFB-resistant hybrids were developed
and identified, and several have been propagated today
for backyard use (Earnest Grimo, personal communica-
tion, 2010). Several of Slate’s selections have been used
as EFB-resistant parents in private and public breeding
efforts, including advanced generation hybrids at Ore-
gon State University (Molnar et al. 2010). A number of
Slate’s selections (designated with a New York identi-
fication number, i.e. NY616, NY398, etc.) are currently
available from the NCGR.
2.5.2.4 S.H. Graham
Also in the 1930s, S.H. Graham of Ithaca, New York
continued work pioneered by J.F. Jones. He grew and
evaluated hundreds of plants from open-pollinated
seeds of Jones’ first generation hybrids, including
seeds collected from “Bixby” and “Buchanan” grow-
ing in close proximity to one another. Graham consid-
ered these plants second-generation hybrids, expecting
to see an improvement over Jones’ work in this appar-
ent next generation. He also grew seedlings of
C. americana “Winkler” and made his own interspe-
cific hybrids using “Winkler”, “Rush”, and unnamed
interspecific hybrids (Graham 1936). Graham’s plant-
ing was in a colder area of New York than Geneva, and
his trees experienced significant winter injury there.
His plantings were also infected by EFB, which was
still not present in the research plots at Geneva as late
as 1952 (Slate 1952). While Graham (1936) found that
a majority of his progeny proved inferior to their
parents in nut quality, and it was reported that he lost
most of his hybrids to EFB (Slate 1969), two cultivars
were named and released from his efforts: “Morning-
side” (“Rush” C. ave. “Duchilly”) in 1945 and
“Graham” (“Winkler” C. ave. “Longfellow”) in
1950. “Morningside” is reported to have been lost
due to EFB (ASHS Press 1997). The status of
“Graham” is currently unknown.
2.5.2.5 C. Weschcke
Carl Weschcke also worked to hybridize C. americana
and C. avellana in the 1930s and 1940s. He was a
very ambitious nurseryman in River Falls, Wisconsin,
interested in northern nut trees, who got his start with
hazelnuts in 1921 by ordering one hundred “Rush”
hazelnut plants from J.F. Jones’ nursery in Pennsylva-
nia. The plants ended up being seedlings of “Rush”,
instead of clones, and many appeared to be hybrids
between C. americana and C. avellana (Weschcke
1954). The diversity reported in this planting likely
sparked Weschcke’s interest in hazelnut breeding. In
1927 he planted “Winkler” hazelnuts purchased from
a nursery in Iowa, and the following year planted
additional Jones hybrid hazels (seedlings of Jones’
interspecific hybrids). Then, in 1929, he planted clones
of twelve different C. avellana cultivars purchased
from a New York nursery. Over the next couple of
years, most of the Jones hybrids and all the of cultivars
of C. avellana, which included “Italian Red” and
“Medium Long” reported to be cold-hardy by Slate
(1959), were killed by cold temperatures, demonstrat-
ing the harsh climate of Wisconsin. To continue his
project, in 1932, Weschcke planted C. avellana selec-
tions from J.U. Gellatly of West Bank, British Columbia,
which also suffered from winter injury, although they
survived for several years. They were of value
to Weschcke when he found an exceptional wild
C. americana plant growing in the woods on his farm.
In 1934, he crossed the wild plant with pollen from a
surviving Gellatly hazelnut; four cold-hardy hybrid
plants were the result, which Weschcke called “hazil-
berts.” Three hybrid plants, possibly from this first
cross, were released several years later named
“Carlola”, “Delores”, and “Magdalene”. All were
listed as having the staminate parent C. avellana
“Brag” developed by J.U. Gellatly (ASHS Press
1997). None of these plants are known to exist today.
In 1939, Weschcke made crosses between “Wink-
ler” and various C. avellana parents. He also crossed
his surviving Jones hybrids with pollen from cold-
hardy C. avellana seedlings from Gellatly. Later,
Weschcke found several more exceptional wild hazel-
nuts in his area that expressed traits such as high
32 T.J. Molnar
yields, large size nuts, early maturity, and thin shells,
which he used as female parents in his crosses; from
1942–1945, hundreds of hybrids between C. americana
and C. avellana were produced. Pollen was obtained
from other hazelnut growers in the US, or from
C. avellana surviving on his farm. Staminate parents
included “Barcelona”, “Duchilly”, “Red Aveline”,
“White Aveline”, “Purple Aveline”, “Italian Red”,
“Daviana”, and several others. By 1945, he had around
2,000 plants under evaluation and, by 1952, had accu-
mulated extensive data on 650 of them. Weschcke
(1954) described that, although there were likely sev-
eral plants worthy of release in this group, he would
prefer to see what is found in the next generation of
1,000 plants. Nearly 10 years later, he reported that his
C. americana C. avellana hybrids were reliable
croppers under all conditions and were bred so that
EFB would not be a problem. However, he stated that
there was not yet any individual commercially valu-
able plant, but he expected to recover this in the next
10,000 seedlings he brought into bearing (Weschcke
1963). This expectation never came to fruition, as
Weschcke (1970) later reported that most of his
hybrids were killed by EFB after all, although he
declared that not all of the plants died. He passed
away in 1973. After several decades of breeding, no
cultivars were released from his work, although his
efforts generated better-adapted and EFB-resistant
hybrid plant material that has been used in more recent
breeding efforts.
2.5.2.6 P. Rutter
Philip Rutter of Canton, Minnesota expanded on
Weschcke’s work by collecting seeds from select
trees of the thousands surviving in Weschcke’s
overgrown, EFB-infected orchards in River Falls,
Wisconsin. In an attempt to identify reliably produc-
tive parent plants to initiate his own mass selection
program, Rutter collected seeds from EFB-resistant
plants of apparent C. americana C. avellana origin
that were bearing nuts in a year when most plants did
not produce crops. He grew the resulting plants on his
hilly, windswept farm in south-eastern Minnesota.
Rutter later added seedlings originating from various
interspecific hybrids developed by G.L. Slate,
J.U. Gellatly, and C. Farris, as well as some of his
own collections of wild C. americana and C. cornuta.
Inferior plants were eliminated by the harsh climate of
Minnesota, the low-maintenance nurseries in which
they were grown, and the presence of EFB. In addition
to cold hardiness and resistance to EFB, the main
breeding objectives were to select plants expressing
high kernel productivity and increased cropping
potential. Adapted, high-yielding seedlings were iden-
tified from these early plantings and their nuts were
harvested to plant successive generations to undergo
similar evaluations. Pollinations were controlled to
some degree by emasculating inferior plants prior to
anthesis (Rutter 1987). Today, several generations and
over 50,000 seedlings have been cycled through eva-
luations by Rutter. From this work, he has identified
plants expressing EFB resistance and cold hardiness
that are segregating for increased nut yields and qual-
ity, with mass selection efforts continuing. While no
cultivars have been released to date, improved seed-
ling plants have been widely distributed. Experimental
plantings of his hybrid seedlings have been established
in numerous parts of Minnesota, Nebraska, Wisconsin,
and other states. From this germplasm, Hammond
(2006) identified several consistently high-yielding
selections out of over 5,000 seedlings grown at the
Arbor Day Farm in Nebraska City, Nebraska. Based
on single-plant estimates, the 4-year average of the
highest yielding selection was 4 ton/ha of dried, in-
shell nuts. While single-plant extrapolations can be
misleading, Hammond’s work was done in a tightly
spaced orchard with no irrigation, fertility, or pest
management, which provides evidence for the poten-
tial of select C. americana hybrids to produce abun-
dant crops in the Midwest US with little inputs.
Overall, the past work hybridizing C. americana
and C. avellana provided useful genetic resources that
are currently available, as well as insight into the
direction a plant breeding program may need to go to
make the best, most efficient use of this hybrid combi-
nation. While the genetic resources used in the early
work were limited, some cold-hardy, EFB-resistant
plants with relatively large, high-quality nuts were
developed. However, not all hybrids were resistant to
EFB, and unreliable and occasionally low yields did
not warrant commercial production in the east. Most
breeding efforts stopped at the first generation, or
putative second generation crosses were made by
hybridizing within the best of the first generation
hybrids, sometimes with no or little control of pollen
flow. This approach is largely inefficient and may have
2Corylus 33
further narrowed the genetic diversity present in the
hybrid germplasm when plants were grown under the
high selection pressure (bottle neck) of severe envi-
ronmental stress and/or disease pressure. Based on the
results of past efforts, a different approach is suggested
here. To maintain the high-quality, large kernels, and
high yields of C. avellana, with the wider adaptation
of C. americana, a diversity of select C. americana
parents must be used (parents that compliment their
C. avellana counterparts) in a systematic, multi-
generational breeding effort. Large hybrid progenies
must be evaluated in the proper environment to iden-
tify the rare recombinants that express the highest
levels of desirable traits of each species. The best
individuals must then be clonally propagated and
tested in multiple locations to identify those with the
highest potential for consistent production and for use
as improved breeding parents. Advanced generation
hybrids must then be made by backcrossing the supe-
rior first-generation hybrids to improved, complimen-
tary, and unrelated C. avellana. From here, the cycle
will likely need to be continued for at least one to
several generations to combine the wide adaptation
of C. americana with the nut qualities of C. avellana –
in the opinion of the author, a very laudable, but feasible
goal.
2.5.3 Corylus heterophylla
TheSiberianhazel,C. heterophylla, can be found grow-
ing across a wide range of climates and soils in Korea,
Japan, China, and the Russian Far East. Corylus hetero-
phylla crosses readily with C. avellana and C. ameri-
cana, although success of seed set depends on the
choice of parental clones (Cho 1988; Weijian et al.
1994; Erdogan and Mehlenbacher 2000a). Corylus het-
erophylla is analogous to C. americana in many char-
acteristics, including its potential for breeding widely
adapted hybrid cultivars, and plants exist that drop the
nuts from the involucres at maturity – a rare occurrence
in C. americana. Compared with cultivated C. avellana,
the nut is smaller and thicker-shelled and the yield is
generally lower. However, plants from its northern
range are extremely cold-hardy and drought-tolerant,
some being adapted to regions of northeast China that
have snowless winters and temperatures dropping
below 30C. Seedlings of the species have been
reported to be extremely precocious, sometimes flow-
ering in only 1 or 2 years from seed (Thompson et al.
1996). Cho (1988) reports that this characteristic is
also expressed in hybrids of C. heterophylla C.
avellana. In addition, selections of C. heterophylla
and its hybrids have also been found to be EFB resis-
tant (Coyne et al. 1998; Chen et al. 2007; Molnar et al.
2010), and Corylus het. var. yunannensis is adapted to
alkaline soils (Thompson et al. 1996). Furthermore,
many selections of C. heterophylla have distinct,
truncated, and variable leaf shapes, which may
enhance other ornamental attributes like purple leaf
color when used to develop hybrid ornamental land-
scape plants. The success of the Chinese breeding
program, as discussed below, suggests that C. hetero-
phylla may be one of the more useful wild Corylus
species for enhancing the climatic adaptation of com-
mercial hazelnuts. Based on its wide native range,
outcrossing nature, and adaptation to varieties of
soils and stressful climates, it is expected that
considerable genetic diversity exists in the species.
Unfortunately, Corylus germplasm collections outside
of China and Korea contain only a very limited repre-
sentation of C. heterophylla.
2.5.3.1 Economic Forest Research Institute
of Liaoning Province, Dalian, China
Traditionally, nuts of C. heterophylla have been col-
lected from wild stands for home consumption and
local sale across its native range. In the 1960s and
1970s, forest management, including tree thinning,
weeding, and pest control, was done in wild hazelnut
stands in China to increase production (Weijian et al.
1994). To better meet a growing demand, open-
pollinated seeds of several cultivars of C. avellana
were introduced from Bulgaria, Albania, and Italy
between 1972 and 1975, with the resulting plants
grown at the Economic Forest Research Institute of
Liaoning Province in Dalian, China. The goals were to
evaluate the potential of C. avellana in China and
to select improved individuals suited for commercial
production in Liaoning Province. A total of 203 seed-
lings from these collections were planted. They were
quickly found to be sensitive to winter injury in
Dalian, which routinely reaches 15 to 20C during
winter months, compounded by high wind and low
air humidity in the winter. The poor performance of
34 T.J. Molnar
C. avellana turned breeders’ attentions to the prospect
of improving wild C. heterophylla growing in northern
China.
In 1980, a program was undertaken to evaluate
very large numbers of wild C. heterophylla growing
in Liaoning province for their production potential
and nut characteristics. Over the next 5 years, out of
many thousands of plants, 31 superior C. heterophylla
strains were identified that expressed qualities such as
high yield, improved nut quality, and thin shells. The
most useful plants were propagated and placed in the
first Corylus germplasm collection in China, which by
1991 contained six species and over 100 cultivars and
lines (Weijian et al. 1994). No individual C. hetero-
phylla selection was shown to be ideal for commercial
production, causing breeders to focus their efforts
on creating interspecific hybrids between C. hetero-
phylla and C. avellana. The first interspecific hybrids
were made in 1980. Breeding goals were to develop
high-yielding plants that produced the large nuts,
high-quality kernels, and thin shells of cultivated
C. avellana, while expressing the cold hardiness
and adaptability of C. heterophylla. In their crosses,
ten strains of select C. heterophylla and 20 of the best
C. avellana seedlings identified from earlier efforts
were used. The pollen of different strains within a
species was mixed before making hybridizations, to
help ensure seed set. While it was possible to use
C. avellana as the pistillate parent, the compatibility
was much higher when using C. heterophylla (Weijian
et al. 1994).
From 1980 to 1986 more than 2,300 hybrid progeny
were produced and grown in Dalian. Over the next
10 years plants were evaluated for cold hardiness, nut
quality, and yield. From these evaluations, around
40 hybrid plants were identified that were superior
to the selected strains of pure C. heterophylla and
were much better adapted than C. avellana (Weijian
et al. 1994). In 1990, the best 12 of these plants
were placed into replicated yield trials in Dalian,
Anshan, and Shenyang. From these trials, five superior
interspecific hybrid plants were named and released in
1999 for production: “Pingdinghuan”, “Bokehong”,
“Dawei”, “Jinling”, and “Yuzui” (Ming et al. 2005).
Recognizing a need for improved quality nuts, breed-
ing efforts were continued in Dalian. In 2001 and
2003, attempts at second-generation hybrids were
made between advanced hybrid selections and several
select C. heterophylla plants,including collaborations
with researchers in Italy (Ming et al. 2005). Although
none of these second-generation plants have yet to be
released, several more cultivars from the original
crosses were released in 2007 and 2008: “Liaozhen
1”, “Liaozhen 2”, “Liaozhen 3”, and “Liaozhen 4”
(Ming et al. 2007,2008; JinLi et al. 2007,2008). The
cultivation of hazelnut in China has increased due to
the success of the interspecific hybridization program.
Currently, around 1,200 ha of hybrid hazelnuts have
been planted in northern China, with production
continuing to expand (Liang et al. 2008; FAOstat
2010).
2.5.3.2 Breeding Efforts in Korea
A similar program to that in China was initiated in
Korea in 1975 at the Rural Development Administra-
tion, as discussed by Mehlenbacher (1991a). Large
numbers (more than 40,000) of native C. heterophylla
and C. sieboldiana were evaluated for immediate pro-
duction or for use in a hybridization program with
C. avellana. From this work, 35 C. heterophylla and
10 C.sieboldiana selections were made, several of
which are now held at the NCGR, including C. hetero-
phylla “Ogyoo” and several numbered selections.
A hybridization program was undertaken to combine
the adaptation of the native species with the larger,
high-quality nuts of C. avellana (Cho 1988), resulting
in the release of “Poongsil” (C. heterophylla
C. avellana “Butler”), and “Gaeam No. 1” (“Ogyoo”
C. avellana “Butler”). The current status of this pro-
gram is unknown. Hazelnut research, including a
germplasm collection, has also been undertaken at
the Institute of Forest Genetics in Korea (Mehlenba-
cher 1991a;Lee2002), although recent details are
lacking in Western literature.
2.5.3.3 C. Farris
Cecil Farris, a private hazelnut breeder from Lansing,
Michigan, was one of the first to hybridize Corylus
heterophylla with C. avellana in the US. He used a
single accession of the species C. heterophylla var.
sutchuensis obtained from western China in crosses
with pollen of C. avellana “Holder” in 1971–1973
(Farris 1974). Farris grew out several dozen seedlings,
of which some were dwarf and stunted and others were
2Corylus 35
vigorous and healthy. The progeny appeared to Farris
to be true hybrids based on plant morphology; he
selected the five best plants and named them Estrella
hybrids, numbered 1 through 5. While some of the five
selections appeared to have sterility issues, he success-
fully crossed “Estrella #2” with pollen from C. avel-
lana “Royal”. All of the offspring appeared to grow
normally, demonstrating the ability to backcross the
hybrids to C. avellana. He also successfully
crossed “Estrella #2” with pollen from “Faroka”, a
C. colurna C. avellana interspecific hybrid devel-
oped by J.U. Gellatly. “Estrella #1” and “Estrella #2”
are available from the NCGR; “Estrella #1” was
shown to be resistant to EFB (Chen et al. 2007).
2.5.4 Corylus cornuta
Corylus cornuta is the most cold-hardy Corylus spe-
cies in North America. It grows wild across much of
the northern US and southern Canada into regions that
reach 50C. It is also believed to be highly EFB-
resistant as there are no reports of this disease occur-
ring on C. cornuta even though its range significantly
overlaps that of C. americana, the native host of
A. anomala. Coyne et al. (1998) confirmed this resis-
tance through greenhouse inoculations. Corylus cornuta
also has very early maturing nuts, a trait that is benefi-
cial to regions like Oregon where a rainy season
begins in autumn that can significantly interfere with
the harvest of late maturing cultivars. This trait is also
needed to grow plants for production in northern
regions with short growing seasons. Corylus cornuta
has a very stoloniferous, spreading-growth habit and
can form dense thickets, which may prove useful in
soil reclamation or for providing wildlife habitat. Con-
versely, this trait would be a negative attribute in most
orchard settings. While Erdogan and Mehlenbacher
(2000a) were only able to cross C. cornuta with
C. californica,C. sieboldiana, and to a limited degree
C. heterophylla, Gellatly (1950,1956) reported suc-
cess making crosses with C. avellana to develop his
“Filazel” hybrids.
2.5.4.1 J.U. Gellatly
Gellatly (1950) collected extremely cold-hardy C. cor-
nuta from the Peace River District of Alberta, Canada,
and hybridized it with his own selections of cold-hardy
C. avellana, including “Craig” and others. His goal
was to combine the extreme cold hardiness and early
nut maturity of C. cornuta with the large nut size of
cultivated C. avellana. While he did not use hand
pollinations to make the crosses, he was able to select
hybrid offspring by identifying seedlings expressing
characteristics that were intermediate between the
two species (Gellatly 1950; Farris 1990; Thompson
et al. 1996). Gellatly distributed seed and plants and
released a number of selections of this hybrid cross
that include “Peoka”, “Manoka”, “Fernoka”,
“Farioka”, and “Myoka”. “Myoka” is available at the
NCGR (2010), while the availability of others is
unknown. Mehlenbacher (1991a) and Farris (1990)
successfully crossed Filazel #45, a numbered hybrid
(C. cornuta C. avellana) selection of Gellatly’s,
with pollen of C. avellana. Farris (1990) also reported
that pollen of Filazel #45 set nuts on C. avellana
“Ennis” and the advanced-generation C. colurna
hybrid “Grand Traverse” and that the early maturity
of this selection was expressed in its offspring. While
Erdogan and Mehlenbacher (2000a) did not have suc-
cess with crossing C. cornuta and C. avellana in either
direction, it is possible that when using a wider diver-
sity of germplasm and larger numbers of crosses,
poorly understood barriers to interspecific hybridiza-
tion between Corylus species may be overcome and
successful hybrid offspring could be generated. In the
case of C. cornuta, this could strongly assist in devel-
oping commercial quality hybrid plants adapted to
very cold climates and short growing seasons.
2.5.5 Corylus californica
Corylus californica can be found growing in the
coastal mountains of the Pacific Northwest. It has
only been utilized to a minor extent due to its small,
thick-shelled nuts, which like its very close relative,
C. cornuta, are very early maturing. This trait, plus its
less-stoloniferous growth habit and shorter husks, may
make it more useful for breeding, if cold hardiness is
not a primary objective. Since the inadvertent intro-
duction of EFB into the Pacific Northwest, it has
been observed that C. californica growing adjacent
to infected orchards remained free of disease. Coyne
et al. (1998) confirmed the presence of resistance in
the species through greenhouse inoculations. While
selections of pure C.californica remained free of
36 T.J. Molnar
EFB, small cankers developed on several C. californica
C. avellana hybrids. Further supported by additional
unpublished work, the performance of
C. californica C. avellana hybrids indicate that
C. californica expresses quantitative resistance rather
than complete resistance to EFB (Shawn Mehlenbacher,
personal communication, 2009). Erdogan and Mehlen-
bacher (2000a) reported that C. californica, when used
as a female parent, was able to hybridize with all other
species used in their study (Table 2.3), suggesting it
may have use as a bridge species to create advanced
generation interspecific hybrids.
2.5.6 Corylus sieboldiana
Corylus sieboldiana grows across much of eastern and
northern Asia, including the Russian Far East. It is
generally very cold-hardy, with many similar traits to
C. cornuta and C. californica, although it has late
maturing nuts (Erdogan 1999). Coyne et al. (1998)
showed that accessions from Korea were resistant to
greenhouse inoculations. Erdogan and Mehlenbacher
(2000a) showed that it hybridized readily with
C. cornuta and C. californica and to a much lesser
extent with C. americana. Reports from Korea,
described in Mehlenbacher (1991a), claim that it was
hybridized with C. avellana to develop better-adapted
cold-hardy plants with larger nuts, although little
recent information is available on hybrids and few
exist in the West for study. A substantial collection
of C. sieboldiana is available at the NCGR (Table 2.3),
which will prove useful to better study the breeding
potential and genetic diversity of this species.
2.5.7 Corylus colurna
Corylus colurna, the Turkish tree hazel,is found natu-
rally occurring in the Balkan Peninsula, Turkey, and
the Caucuses, but it has been grown widely as an
ornamental shade tree in many parts of Europe and
the US for centuries. In the landscape, C. colurna
naturally forms an attractive pyramidal crown and
displays interesting scaly, corky bark, and heavily
textured leaves. It has been shown to be cold-hardy
and exceedingly drought and stress tolerant, with Dirr
(1998) suggesting its use as a street tree, even under
city conditions. Its small, thick-shelled nuts with high-
quality kernels have been collected from the wild and
consumed and sold in its native areas, but trees are
usually more valued for their excellent timber.
While C. colurna remains a single-trunk tree, the
shrubby natured C. avellana produces suckers from
the base of the plant throughout the growing season.
To maintain grafted trees of C. avellana and to facili-
tate mechanized production in commercial orchards,
suckers must be removed numerous times per year,
which requires a significant amount of expense and
time. As such, a non-suckering rootstock would be
very beneficial to commercial production. Seedlings
of C. colurna were used as rootstocks for C. avellana
in the past, including for ornamental forms. However,
poor germination rates and transplanting problems due
to its less-fibrous tap root system, as well as decre-
ased performance of older nut orchards grafted on
C. colurna, have reduced its value for such applica-
tions (Lagerstedt 1976). Nevertheless, C. colurna still
holds promise for use in interspecific hybridization
programs, especially in terms of developing vigor-
ously growing, stress-tolerant, non-suckering clonal
rootstocks (Lagerstedt 1975,1990).
It is very difficult to hybridize C. colurna and C.
avellana; however, a limited number of fertile hybrids
have been created in the past, and are discussed below.
More recently, Erdogan and Mehlenbacher (2000a)
were able to recover a small number of hybrid seed-
lings when making large numbers of crosses between
C. colurna and C. avellana. It was shown to hybridize
readily with C. chinensis, and less successfully with
C. heterophylla and C. californica. They suggest when
attempting to make the cross with C. avellana, bree-
ders should perform large numbers of pollinations and
expect only a small number of hybrid seedlings. In
addition to their growth habit and adaptation attri-
butes, select C. colurna and C. colurna hybrids have
also been reported to be resistant to EFB and big bud
mites (Coyne et al. 1998; Farris 1978,2000; Lunde
et al. 2000; Chen et al. 2007), suggesting its direct
usefulness for breeding improved Corylus hybrids.
2.5.7.1 J.U. Gellatly
J.U. Gellatly of West Bank, British Columbia reported
some of the first hybrids of C. colurna C. avellana,
2Corylus 37
naming the hybrid plants as a group “trazels” (Gellatly
1964,1966). His plants resulted from the open-
pollination of C. colurna trees grown in close proxim-
ity to trees of C. avellana, reportedly “Craig”, “Holder”,
and “Brag”. Gellatly grew large seedling populations
of C. colurna, presumably for rootstocks for his nurs-
ery business, and he identified hybrids based on their
appearance (Farris 2000). He then evaluated the appar-
ent hybrids for cold hardiness and nut characteristics.
After 30 years of work, several superior hybrids and
numbered selections were named and released by this
method, including “Morrisoka”, “Faroka”, “Karloka”,
and “Eastoka” (Gellatly 1966; Farris 1978). These
plants were considered to be true hybrids based
on their morphology (Erdogan and Mehlenbacher
2000a). These “trazels”, as well as several numbered
selections of Gellatly’s believed to be hybrids between
C. colurna C. avellana, are held in the collection at
the NCGR. Recently, Chen et al. (2007) found several
of Gellatly’s C. hybrids to be highly resistant to EFB,
including Chinese Trazel Gellatly #6 and #11 and
Turkish Trazel Gellatly #3. While Gellatly refers
to the Chinese trazels as hybrids with C. chinensis
(Gellatly 1966), Chen et al. (2007) believe they are
instead C. colurna, based on morphology.
2.5.7.2 C. Farris
A continuation of Gellatly’s “trazel” work was under-
taken by C. Farris of Lansing, Michigan. Farris, a self-
trained plant breeder, grew and evaluated most of
Gellatly’s hybrid selections and considered “Morri-
soka” and “Faroka” the best of the group. While the
hybrid cultivars were not suitable for commercial pro-
duction, he believed they would make superior breed-
ing parents based on their cold hardiness, productivity,
and high nut quality. His goals were to develop non-
suckering, EFB-resistant, cold-hardy plants that pro-
duced high-quality, large kernels. Throughout the
1970s and early 1980s, Farris, used controlled hand
pollinations to successfully cross “Morrisoka” and
“Faroka” with pollen of C. avellana “Royal”, which
has large size nuts, to create advanced-generation
hybrid progeny. Pollen of “Faroka” was also used
successfully on C. colurna, demonstrating the cross-
compatibility of Gellatly’s first-generation hybrids
(Farris 1978,1982). It should be noted that this fact
was reinforced by Thompson et al. (1996) with the
successful crossing of C. avellana “Willamette” with
pollen of “Faroka”, as well as successfully making the
reciprocal cross with a mixture of C. avellana pollens.
From his crosses, Farris identified a number of
advanced-generation hybrids useful for further breed-
ing or eventual release (Farris 2000). The most widely
recognized was his EFB-resistant “Grand Traverse”,
which he described as a cross between “Faroka”
“Royal” (Farris 1989), although the identity of the
C. avellana parent is somewhat unclear, based on
incompatibility alleles (Lunde et al. 2000). “Grand
Traverse” was recently shown to be completely resis-
tant to more than 12 different A. anomala isolates
collected around the eastern US (Molnar et al. 2010).
It was also shown to transmit its resistance to about
25% of its offspring (Molnar et al. 2009), and work is
underway to better understand inheritance of the resis-
tance. In 2008, “Grand Traverse” was successfully
backcrossed to C. colurna and several C. avellana,
with high cluster set and typical germination
(T. Molnar, unreported). Farris (1990) later released
“Lisa”, an open-pollinated selection of “Grand
Traverse”, believed to be backcrossed to C. avellana.
It was also shown to be resistant to EFB, as well as bud
mites (Chen et al. 2007). “Grand Traverse” and several
others of Farris’ selections are available at the NCGR.
“Lisa” is held in the collection at OSU.
2.5.7.3 H.B. Lagerstedt
In search of non-suckering rootstocks for use in com-
mercial hazelnut orchards in Oregon, H.B. Lagerstedt
initiated a sizable rootstock evaluation and breeding
program at Oregon State University, starting in 1968.
Lagerstedt organized a collection of 19 of Gellatly’s
hybrid trazels, as well as five non-suckering selections
of possible rootstock potential from Farris. He also
included apparent hybrid plants from O. Jemtegaard,
who selected the open-pollinated hybrid “Filcorn” by
similar means as Gellatly. Lagerstedt grew more than
1,000 open-pollinated seedlings of “Filcorn”, which
was growing in a C. avellana orchard far away from
other C. colurna, and from this isolation were assumed
to be advanced-generation backcross hybrids. From
these, he selected 70 plants for further testing
(Lagerstedt 1975,1976). Wide collections were also
made from botanical gardens, arboreta, and nurseries
in the US, Canada, and Europe (Thompson et al.
38 T.J. Molnar
1996). Seedlings were evaluated for vigor and non-
suckering growth habit. The challenge and expense of
identifying an appropriate non-suckering rootstock is
substantial. This is due to the need for the plant, once
identified as a potential candidate, to first be repro-
duced on its own roots by layering, which can take
several years. Then, it must be evaluated in replicated
clonal yield trials, grafted to known cultivars, in com-
parison to those cultivars growing on their own roots
to prove there is not a reduction of yield as a conse-
quence of its use. Finally, in 1990, two rootstock
cultivars were released by OSU: “Newberg” (tested
as USOR 1-71) and “Dundee” (tested as USOR 15-71).
They were selected from a nursery planting in 1971
that contained several thousand open-pollinated seed-
lings of C. colurna (Lagerstedt 1990). Unfortunately,
both are highly susceptible to EFB.
Lagerstedt also had a goal to incorporate red leaf
color into a non-suckering rootstock that would facili-
tate maintenance and help differentiate the scion por-
tion from the rootstock portion of a grafted tree. While
working on this objective, he developed and released
the ornamental hybrid “Ruby”. “Ruby” was derived
form a controlled cross between Chinese trazel #4
(believed to be a C. colurna C. avellana hybrid)
and the red-leaf C. avellana “Fusco Rubra”. “Ruby”
was selected because it retained its red color longer
into the summer than other available red-leaf orna-
mental hazelnuts (Lagerstedt 1990). In 1984, a red-
leaf full-sibling of “Ruby”, USOR24-82, was crossed
with a red-leaf breeding selection, C. avellana OSU
A-28, by M. Thompson and D. Smith. From the result-
ing progeny, an advanced-generation, ornamental,
interspecific hybrid “Rosita” was selected and released
in 1999 (Smith and Mehlenbacher 2002).
2.5.8 Corylus jacquemontii
Corylus jacquemontii is very poorly represented in
western germplasm collection, research, and breeding
efforts. Its genetic diversity and potential for breeding
have not been evaluated. The species’ inclusion in
recent taxonomic studies has been based on only one
or two genotypes available in western collections
(Erdogan and Mehlenbacher 2000b; Forest and
Bruneau 2000; Whitcher and Wen 2001). Sharma and
Kumar (2001) describe efforts to evaluate purported
selections of C. jacquemontii growing in northwestern
India, stressing its underutilized nature. Based on its
growth habit, potential uses of this species are as a
non-suckering rootstock and as an ornamental shade
tree. The species was shown to be susceptible to EFB
by Coyne et al. (1998), with all seven accessions
succumbing to the disease. Farris (2000) described
C. jacquemontii succumbing to EFB in Michigan;
however, these findings were based on a very limited
sample of plant material. In 2008, a seedling tree at
the Dawes Arboretum in Newark, Ohio (accession
D1997-0030.002) was observed by the author growing
completely free of disease while adjacent to several
heavily EFB-infected C. avellana plants.
2.5.9 Corylus chinensis
Corylus chinensis is also very poorly represented in
Western germplasm collections and breeding efforts,
and it is considered endangered in China (Sun 1998).
Based on its growth habit, it may have direct use as
a rootstock or in the development of interspecific
hybrids to develop improved non-suckering root-
stocks. Based on experience with small seed lots
imported from China, seedlings grow more vigorously
and with a more fibrous root systems than C. colurna
(an issue limiting the production of C. colurna in the
nursery trade). However, based on its region of origin,
the species is likely less cold hardy. It would also have
use as an ornamental shade and timber tree due to its
vigorous growth habit; large mature size; oversized,
attractive leaves; and interesting involucres. Seedling
C. chinensis trees growing at Rutgers University in
New Jersey appear to express a high level of resistance
to EFB, being disease-free after more than 6 years of
exposure (T. Molnar, unpublished). Interestingly, they
also appear to be highly resistant to feeding damage
done by Japanese beetles (Popillia japonica New-
man), a common, and often severe, pest problem of
hazelnuts grown in the eastern US.
While accessions of C. chinensis were limited,
Erdogan and Mehlenbacher (2000a) successfully
crossed it with C. colurna,C. avellana,C. americana,
and C. californica. The cross with C. avellana,
although only with C. avellana as the pollen parent,
was notable in that it produced the most vigorous
offspring out of all the various hybrid combinations
2Corylus 39
in their comprehensive study. This finding strongly
suggests that hybrids of C. chinensis and C. avellana
may be useful for developing vigorous non-suckering
rootstocks.
2.5.10 Corylus fargesii
Corylus fargesii, also known as C. papyraceae,is
native to southern China and has only recently been
introduced to the West. Its genetic diversity and poten-
tial for breeding has not been fully evaluated. Wider
collection efforts and preservation in germplasm repo-
sitories and botanical gardens is urgently needed. The
earliest reported introduction to the US was made by
Cecil Farris in 1982 from southern Gansu Province,
China. Farris imported 50 seeds, but only recovered
three seedlings (Farris 1995,2000). More recent seed
introductions were made by members of the North
American China Plant Exploration Consortium in
1996 and 2005, leading to the establishment of
C. fargesii at a number of US arboreta, the NCGR,
and Rutgers University (Aiello and Dillard 2007). Its
unique qualities include a single-trunk habit, vigorous
growth, and very attractive peeling bark, which resem-
bles that of river birch. Aiello and Dillard (2007) believe
that, with improved propagation techniques, C. fargesii
has merit to become a valuable ornamental shade tree in
the central and eastern US. It may also hold value as a
non-suckering rootstock, as seedlings grow very rapidly,
are easily transplanted, and appear graft-compatible
with C. avellana. Furthermore, Farris reports his intro-
ductions were resistant to EFB after many years of
exposure (Farris 1995). Trees located at the Morris
Arboretum in Philadelphia, Pennsylvania and Rutgers
University in North Brunswick, New Jersey, also appear
resistant to EFB, as plants are healthy and without
symptoms while growing adjacent to heavily infected
C. avellana for many years (Aiello and Dillard 2007;T.
Molnar, unpublished).
Farris reported successfully crossing C. fargesii
with “Morrisoka” (C. colurna C. avellana)in
1992, with all offspring inheriting the peeling bark
characteristic (Farris 2000). No further reports of use
in breeding have been made, although investigations
are underway at OSU and Rutgers. Corylus fargesii is
native to relatively warm areas of southern China, so
its cold hardiness may be questionable. It also produces
small, thick-shelled nuts that are tightly enclosed in the
involucre. Therefore, its use in genetic improvement
may be limited to developing vigorous non-suckering
rootstocks and attractive ornamentals with peeling
bark for warmer climates, unless cold-hardy compli-
mentary breeding parents are used. The mode of inher-
itance of EFB resistance in C. fargesii has not yet been
investigated.
2.5.11 Corylus ferox
Corylus ferox and C. ferox var. tibetica are poorly
represented in Western germplasm collections and
research efforts. It is a small, single-trunk tree native
to high elevations with mild climates in the eastern
Himalayan Mountains across to parts of Yunnan,
Sichang and Xizang provinces in China. The most
distinctive feature of this species is its chestnut-like
involucre. Farris (2000) suggested that the involucre
could protect the nuts from bird and rodent predation
until the nuts are mature and ready to fall. He grew a
very limited number of C. ferox var. tibetica acces-
sions in Michigan and found it sensitive to cold dam-
age. He was able to maintain the plants in large pots
in his garage to survive the winter. He reportedly
crossed C. ferox with “Lisa”, an advanced-generation
C. colurna C. avellana hybrid. However, the fate of
the hybrid seedlings is unknown. Based on rDNA ITS
sequence data, C. ferox was separated from all the
other Corylus taxa (Erdogan and Mehlenbacher
2000b). A much wider diversity of C. ferox needs to
be collected and evaluated for breeding and research,
especially for studies of genetic diversity and the origin
and evolution of Corylus and the family Betulaceae,
based on its possible ancestral taxonomic position.
2.6 Alternative Uses of Wild Corylus
Access to an increased diversity of wild germplasm
and its systematic use in an interspecific hybridization
program should lead to the development of widely
adapted plants that can produce crops over a much
greater area. This increased production would support
the development of new market applications and
opportunities, including ornamental plants, feedstock
40 T.J. Molnar
for biodiesel or other oleochemicals, value-added
health and food products, animal feed, and other
potential areas like soil reclamation, biomass produc-
tion, and others not yet investigated.
2.6.1 Hazelnut Oil
Hazelnut kernels have a high oil content by weight,
with most containing over 60% oil and some up to
70%. The oil is rich in monounsaturated fatty acids,
especially oleic acid (around 75–80%), and to a lesser
degree linoleic acid (Botta et al. 1994; Ebrahem et al.
1994; Benitez-Sa
´nchez et al. 2003; Xu et al. 2007).
Recent work at the University of Nebraska, Lincoln,
has demonstrated an alternative potential of hybrid
hazelnuts as a low-input feedstock for the production
of biodiesel and other valuable oleochemicals. Based
on the 3-year average production of their highest-
yielding hybrid hazelnut selections, Hammond
(2006) estimated that an equivalent of 4-ton/ha of in-
shell nuts could be produced. Based on the kernel oil
content and shelling percentage of these selections, an
oil yield of 1,000 kg/ha – nearly double the yield per
hectare of soybean oil (around 500 kg/ha) (FAOStat
2010) – could be realized (Xu et al. 2007; Xu and
Hanna 2009,2010). While single-plant extrapolations
can be unreliable, the estimates made by Hammond
(2006) and Xu et al. (2007) are not far from the oil
yields that could be produced commercially in the
Willamette Valley of Oregon, under current produc-
tion systems, if the kernels were processed for oil.
Calculations suggest that nearly 900 kg of oil per
hectare could be produced, based on the past 10-year
average hazelnut yield data using the cultivar
“Casina”, which has a shelling percentage of 56%
and kernel oil content of 65.3% (Ebrahem et al.
1994; Mehlenbacher 2003; FAOStat 2010).
In addition to its high yield potential, hazelnut oil has
a unique fatty acid composition (high monounsaturated
fatty acids and small percentages of saturated and poly-
unsaturated acids), thermal stability, and low tempera-
ture properties that should increase its value over
soybean oil for a number of applications (Xu et al.
2007; Xu and Hanna 2009). Xu and Hanna (2009)
synthesized and characterized biodiesel from hybrid
hazelnut harvested in Nebraska and found it to be an
excellent feedstock for making the fuel. Similar findings
were reported by Gumus (2008)whenhesynthesized
hazelnut-oil based biodiesel and examined its perfor-
mance in diesel engines in Turkey.While the economics
need to be fully examined before it is suggested that
hazelnuts be grown as a sustainable source of biodiesel,
it should be mentioned that high-yielding, well-adapted,
early-generation interspecific hybrids that may not have
nut qualities acceptable for the kernel trade (round,
high-quality, well-blanching kernels) may find a direct
role in oil production where these characteristics are not
important. Plus, they can be grown on sloping and
marginal land not suitable to annual oil crops.
Besides industrial applications, hazelnut oil, which
is very similar in composition to olive oil (Benitez-
Sa
´nchez et al. 2003), can likely play a larger role in the
diets of humans. A comprehensive review of hazelnut
(C. avellana) kernel composition and their reported
health effects can be found in Alasalvar et al.
(2009a). In general, C. avellana hazelnut kernels by
weight are 58–64% fat, 15–18% carbohydrate, and
10–16% protein. They also contain a total of 24 miner-
als (essential and non-essential), with potassium the
most abundant, as well as the fat soluble vitamins
A, E, and K and water-soluble vitamins thiamin, ribo-
flavin, niacin, pantothenic acid, folate, and several
others. The kernels are especially high in vitamin E,
biotin, and folate, which may be attributed to their
reported health-promoting effects. Their fatty acid
composition contains a small amount of saturated
fatty acids (7–9%) and consists primarily of the mono-
unsaturated fatty acids – oleic acid (77–83%) and
linoleic acid (7–14%). This ratio of low saturated
fatty acids to high unsaturated fatty acids has also
been attributed to the health-promoting effects of
hazelnuts, especially on human plasma lipid profiles.
In addition, Erener et al. (2007) found that the con-
sumption of hazelnut oil by broiler chickens increased
oleic acid content of the meat, compared to those fed
soybean oil, and that the ratio of saturated fatty acid to
mono-unsaturated fatty acid was decreased. While
more study is needed, these results suggest hazelnut
oil could be used as an animal feed supplement to
produce healthier meat products containing a higher
level of oleic acids important for human diets. Addi-
tionally, the protein meal remaining after oil extraction
may add substantial secondary value to oil production,
as animal feed or in other products such as baked goods
or supplement bars. However, the use of hazelnut
protein in this manner has not yet been evaluated.
2Corylus 41
2.6.2 Potential Phytochemicals
and Other Products
There are a number of potentially valuable by-pro-
ducts that would become more widely available with
the increased production of hazelnuts production that
would become more widely available with increased
production. A majority of the world’s crop (90% or
more) is cracked and sold as some form of raw or
roasted kernels. This leaves behind the shell (50% or
more of the crop weight), as well as the leafy involucre
that surrounds the nuts, especially in Turkey where
nuts still in the involucre are harvested by hand. Shells
are commonly used directly as a fuel source, many
times to aid in the kernel drying process, and shells
and involucres can also be used as a compost material.
O
¨zcelik and Peks¸en (2007) demonstrated the useful-
ness of the involucres as a component of a substrate to
cultivate shiitake mushrooms [Lentinula elodes
(Berk.) Pegler]. Production also results in a significant
amount of biomass from orchards in the form of tree
prunings and leaves. Finding higher value use of the
by-products of production would provide further eco-
nomic incentives to grow this low-input crop. While
little work has been done to examine possible phyto-
chemicals and other useful compounds in wild Cory-
lus, Alasalvar et al. (2009b), in a comprehensive
review of existing studies, showed that C. avellana
kernels, pellicle, shell, involucre, and leaves are a
rich source of proanthocyanidins, phenolic com-
pounds, total antioxidant activity, and flavonoids.
The pellicle of hazelnut showed the highest level of
antioxidant activity compared to other tissues exam-
ined and was considered to be a potential industrial
source of antioxidants. Furthermore, Oliveira et al.
(2007) demonstrated antimicrobial activities of
C. avellana leaf extracts, suggesting they may be a
good candidate as an agent to control bacteria that
cause gastrointestinal and respiratory tract infections
in humans. In addition, hazelnut by-products also con-
tain low concentrations of paclitaxel and other taxanes
(compounds in the widely used cancer drug Taxol)
(Hoffman et al. 1998; Hoffman and Shahidi 2009).
While concentrations of taxanes in C. avellana appear
uneconomically low compared to the bark of Pacific
yew (Taxus brevifolia Nutt) from which it is currently
extracted, it is possible that these valuable compounds
may be found in higher levels in other Corylus species
and interspecific hybrids that have yet to be assayed
for the compounds. This point can also be applied to
all of the other compounds studied for C. avellana that
have not yet been investigated in wild species.
2.6.3 Ornamental Landscape Plants
A number of highly ornamental traits exist in C. avellana,
including the weeping habit of “Pendula” and the con-
torted growth of “Contorta” (also known as “Harry
Lauder’s Walking stick”). “Rote Zeller”, “Fusco
Rubra”, “Purple Aveline”, and “Syrena” exhibit dark
red/purple leaves in the spring and early summer,
while the bright yellow leaves of “Aurea” and the
highly dissected leaves of “Cutleaf” also have
promising ornamental value. Unfortunately, all of
these cultivars are highly susceptible to EFB, which
significantly limits their use in North America.
Excluding the weeping habit, the inheritance of these
ornamental traits has been studied and most appear to
be simply inherited (Thompson et al. 1996). By devel-
oping ornamental plants with EFB resistance, whether
through C. avellana or other species, hazelnuts could
be much more widely utilized in the landscape as
ornamentals or even as ornamental garden plants as
the plants could still produce nuts. Incorporating genes
for extreme cold hardiness would further increase their
range of usefulness. A number of wild species also
express traits, which would make them directly useful
as ornamentals or as parents in interspecific breeding
programs. For example, some selections of C. ameri-
cana express bright pink or red fall color that is absent
in most other species. While the inheritance of fall
color in Corylus is not well understood, some hybrids
of C. americana C. avellana express fall color, and
through use of select parents, it has been possible to
combine the red spring and summer leaf color of C.
avellana with the EFB-resistance and bright red fall
color of C. americana (T. Molnar, unpublished). The
truncated and variable leaf shapes of C. heterophylla
would further add to the ornamental attributes of such
an interspecific hybrid. The tree hazels offer opportu-
nities to develop improved shade trees that would add
interest and diversity to current landscape designs,
particularly if incorporated with attractive leaf shape
and color traits. Interspecific hybridization offers a
means to develop a wide diversity of plant shapes
and forms, including those with attractive corky and
peeling bark as in C. colurna and C. fargesii,
42 T.J. Molnar
respectively. Similar to breeding for nut production,
the lack of breeding efforts in Corylus, especially for
ornamental attributes, as well as wider access to
genetic resources and the increased understanding of
interspecific hybridization potentials and inheritance
of traits, should provide opportunities to develop a
wide variety of useful, multi-purpose, attractive, land-
scape plants well into the future.
2.7 Recommendations for Future
Conservation, Research, and Genetic
Improvement
The diversity and richness of the Corylus genus, its
usefulness to man, and its importance to natural eco-
systems are substantial. Increased efforts should be
made to preserve, study, and utilize Corylus genetic
resources for the betterment of future generations. In
general, hazelnuts are a very low-input, high-value
crop adapted to a wide variety of climates and soils,
the production of which has many economic and eco-
logical benefits. Furthermore, recent epidemiological
and clinical studies have provided strong evidence that
frequent tree nut consumption, including hazelnuts, is
associated with favorable plasma lipid profiles and a
reduced risk of heart disease, cancers, strokes, inflam-
mation, and other chronic health issues (Alasalvar and
Shahidi 2009). These positive economic, environmen-
tal, and health factors are driving increased production
and market demand worldwide. World production
acreage has increased almost 14% over the past 10
years (Fideghelli and De Salvador 2009). The devel-
opment of more widely adapted cultivars would pro-
vide greater options for farmers to help meet this
increasing demand. As discussed at length in this
chapter, Corylus genetic resources are highly under-
utilized and underrepresented in research studies and
conservation efforts, germplasm collections, and
breeding programs. Aside from cultivated forms of
C. avellana, little is known of their genetic diversity
and population structure, with possible unchecked
genetic erosion occurring due to overdevelopment,
deforestation, and other causes. Molecular biology
tools are now available for Corylus, including a multi-
tude of effective SSR markers (G€
urcan et al. 2010),
that can be utilized to assess genetic diversity in wild
species and also fingerprint accessions to reduce dupli-
cation in germplasm collections and show gaps in
collections. A more thorough and accessible catalog
of wild Corylus germplasm existing at research insti-
tutions, botanical gardens, and arboreta – similar to
that currently conducted by the “Safenut” project in
Europe for cultivated C. avellana (Bacchetta et al.
2009) – should be compiled and made available to
hazelnut researchers worldwide. Besides obvious spe-
cies deficits in Western collections such as C. jacque-
montii,C. ferox, and C. fargesii, collections are also
deficient in plants of the more common species origi-
nating from their most northern and southern ranges.
Collections from these areas would be extremely valu-
able for expanding production of cultivated hazelnuts
into more stressful climates. Efforts are needed to
collect and evaluate this germplasm, with the most
useful selections preserved and made widely accessi-
ble to world germplasm banks.
In the past, Corylus genetic improvement efforts
demonstrated that significant progress can be made
through breeding; moving forward, however, a multi-
generational approach should be followed, and the
enhancement of genetic diversity in breeding lines
should be stressed. Molecular biology tools must be
used in concert with breeding efforts to ensure this is
the case. The lack of diversity used in the interspecific
hybrids made with C. americana and C. avellana in
the first half of the twentieth century exemplifies this
point. While C. americana “Rush” produced large size
nuts for the species, many of its negative attributes
were also transmitted to its offspring, which included a
lack of significant cold hardiness, high susceptibility
to bud mites, nuts that are retained in the husk at
maturity, and a reduced numbers of catkins (Slate
1961). Besides the lack of diversity, most crosses
were also limited to first-generation hybrids. The
recent breeding progress at OSU clearly demonstrates
the potential for substantial improvement in the sec-
ond and third generation of breeding (Mehlenbacher
et al. 2007,2008,2009). Selecting a wide diversity of
the best, complimentary C. avellana parents available
is especially important when using a modified back-
cross program to incorporate characteristics from wild
species. Fortunately, much wider access to cultivated
forms of C. avellana is available now than in the past.
Additionally, the ability to ship pollen overnight from
almost anywhere in the world makes it is possible
to use parents in breeding efforts that would not
2Corylus 43
typically grow in the location of the breeding program,
providing a very efficient means to exchange genetic
resources. For example, pollen from EFB-susceptible,
C. avellana cultivars and breeding selections with
high-yields and excellent nut quality is routinely
shipped from OSU and the NCGR to use in controlled
crosses at Rutgers University in New Jersey, to
develop improved, locally adapted selections. Many
of these parents would not survive long enough in New
Jersey to use in crosses due to EFB and the colder
climate. In addition, pollen carries few viruses or dis-
eases, reducing concerns related to the importation of
seeds or clonal material. For a description on pollen
collection and handling see Thompson et al. (1996).
Exciting opportunities now exist to study and more
widely utilize the Corylus genus. Rapid genetic gains
are expected in breeding, based on its highly heterozy-
gous nature, the ability to hybridize numerous species,
and very limited prior breeding efforts. In the opinion
of the author, the wild species of most immediate
value for breeding improved interspecific cultivated
forms (backcrossed to C. avellana) are C. americana
and C. heterophylla. These both cross readily with
C. avellana and are adapted to a wide climatic range
with those from the northern areas being extremely
cold-hardy. Selections of C. americana and C. hetero-
phylla are also resistant to EFB, although inheritance
is not well understood in these species, and some
plants are very precocious and high yielding. A num-
ber of first-generation hybrids already exist in the US
and China that can play an integral role in developing
the foundation for developing advanced-generation
hybrids. The collection and evaluation of a larger
variety of wild germplasm will likely lead to the iden-
tification of more diverse improved selections to be
used in long-term breeding efforts. Other Corylus spe-
cies merit much wider collection and study for the
conservation of genetic resources and for use in breed-
ing, to enhance genetic diversity in cultivated forms,
and to donate specific traits of interest such as extreme
cold hardiness, drought tolerance, non-suckering
growth habit, ornamental attributes, disease and pest
resistances, and other characteristics that arise as more
is learned about the wild species and as market
demands dictate.
Acknowledgments I would like to gratefully acknowledge the
contributions to this manuscript of J Capik, C Leadbetter,
X Ming, R Funk, and S Mehlenbacher.
References
Aiello AS, Dillard S (2007) Corylus fargesii: a new and
promising introduction from China. Combined Proc Int
Plant Propagators Soc 57:391–395
Alasalvar C, Shahidi F (2009) Tree nuts: composition, phyto-
chemicals, and health effects. CRC, Boca Raton, FL
Alasalvar C, Shahidi F, Amaral JS, Oliveira PP (2009a) Com-
positional characteristics and health effects of hazelnut (Cor-
ylus avellana L.): an overview. In: Alasalvar C, Shahidi F
(eds) Tree nuts: composition, phytochemicals, and health
effects. CRC, Boca Raton, FL, pp 185–214
Alasalvar C, Hoffman AM, Shahidi F (2009b) Antioxidant
activities and phytochemicals in hazelnut (Corylus avellana
L.) and hazelnut by-products. In: Alasalvar C, Shahidi F
(eds) Tree nuts: composition, phytochemicals, and health
effects. CRC, Boca Raton, FL, pp 215–248
ASHS Press (1997) The Brooks and Olmo register of fruit & nut
varieties, 3rd edn. American Society of Horticultural
Science Press, Alexandria, VA
Bacchetta L, Aramini M, Bernardi C (2008) In vitro propagation
of traditional Italian hazelnut cultivars as a tool for the
valorization and conservation of local genetic resources.
HortScience 43:562–566
Bacchetta L, Avanzato D, Botta R, Boccacci P, Drogoudi P,
Metzidakis I, Rovira M, Silva AP, Solar A, Spera D, Aramini
M, Di Giovanni B (2009) First results of “Safenut”: a Euro-
pean project for the preservation and utilization of hazelnut
local genetic resources. Acta Hort 845:55–60
Bassil NV, Botta R, Mehlenbacher SA (2005a) Additional
microsatellite markers of the European hazelnut. Acta Hort
686:105–110
Bassil NV, Botta R, Mehlenbacher SA (2005b) Microsatellite
markers in hazelnut: isolation, characterization, and cross-
species amplification. J Am Hortic Sci 130:543–549
Bassil NV, Postman J, Hummer K (2009) SSR fingerprinting
panel verifies identities of clones in backup hazelnut collec-
tion of USDA genebank. Acta Hort 845:95–98
Bennet MD, Smith JB (1991) Nuclear DNA amounts in angios-
perms. Phil Trans R Soc Lond B 334:309–345
Benitez-Sa
´nchez PL, Leo
´n-Camacho M, Aparicio R (2003)
A comprehensive study of hazelnut oil composition with
comparisons to other vegetable oils, particularly olive oil.
Eur Food Res Technol 218:13–19
Biodiversity, FAO and CIHEAM (2008) Descriptors for hazel-
nut (Corylus avellana L.) Biodiversity International, Rome,
Italy; Food and Agriculture Organization of the United
Nations, Rome, Italy; International Center for Advanced
Mediterranean Agronomic studies, Zaragoza, Spain
Boccacci P, Akkak A, Botta R (2006) DNA typing and genetic
relations among European hazelnut (Corylus avellana L.)
cultivars using microsatellite markers. Genome 49:598–611
Boccacci P, Botta R, Rovira M (2008) Genetic diversity of
hazelnut (Corylus avellana L.) germplasm in northeastern
Spain. HortScience 43:667–672
Boccacci P, Botta R (2009) Investigating the origin of hazelnut
(Corylus avellana L.) cultivars using chloroplast microsatel-
lites. Genet Resour Crop Evol 56:851–859
44 T.J. Molnar
Boccacci P, Botta R (2010) Microsatellite variability and
genetic structure in hazelnut (Corylus avellana L.) cultivars
from different growing regions. Sci Hort 124:128–133
Botta R, Gianotti C, Richardson D, Suwanagul A, Sanz CL
(1994) Hazelnut variety organic acids, sugars, and total
lipid fatty acids. Acta Hort 351:693–699
Bozog
˘lu M (2005) The situation of the hazelnut sector in
Turkey. Acta Hort 686:641–648
Buckman RE (1964) Effects of prescribed burning on hazel in
Minnesota. Ecology 45:626–629
Boufford DE (1997) Corylus. In: Flora of North America Edito-
rial Committee (ed) 1993+. Flora of North America, North
of Mexico. 12+ vols, vol 3. Oxford University Press,
New York, pp 535–538
Chen H, Mehlenbacher SA, Smith DC (2005) AFLP markers
linked to eastern filbert blight resistance from OSU 408.040
hazelnut. J Am Soc Hortic Sci 130:412–417
Chen H, Mehlenbacher SA, Smith DC (2007) Hazelnut acces-
sions provide new sources of resistance to eastern filbert
blight. HortScience 42:466–469
Chen Z, Manchester SR, Sun H (1999) Phylogeny and evolution
of the Betulaceae as inferred from DNA sequences, mor-
phology, and paleobotany. Am J Bot 86:1168–1181
Cho H (1988) Potential use of Korean filbert (Corylus spp.)
landraces. In: Suzuki S (ed) Crop genetic resources of East
Asia. Proceeding of the international workshop on crop
genetic resources of East Asia. International Board for
Plant Genetic Resources, Rome, Italy, pp 169–181; Acces-
sion No: 306003 Fiche No: 305975–6029, ISBN
92–9043–123–7
Coyne CJ, Mehlenbacher SA, Smith DC (1998) Sources of
resistance to eastern filbert blight. J Am Soc Hortic Sci
124:253–257
Coyne CJ, Mehlenbacher SA, Johnson KB, Pinkerton JN, Smith
DC (2000) Comparison of two methods to evaluate quanti-
tative resistance to eastern filbert blight in European hazel-
nut. J Am Soc Hortic Sci 125:603–608
Craddock WJH (1987) Cryopreservation of pollen. MS Thesis,
Oregon State University, Corvallis, OR
Crane HL, Reed CA (1937) Nut breeding. In: 1937 USDA
Yearbook of Agriculture Washington, DC, pp 827–889
Crane HL, McKay JW (1951) Preliminary report on growth,
flowering, and magnesium deficiency of Reed and Potomac
filbert varieties. Annu Rep Northern Nut Growers Assoc 42:
50–55
Davison AD, Davidson RM (1973) Apioporthe and Monchaetia
canker reported in western Washington. Plant Dis Rep 57:
522–523
Deacon J (1974) The location of refugia of Corylus avellana L.
during the Weichselian glaciation. New Phytol 73:1055–1063
Denisova FN (1975) Breeding filbert at the Central Genetics
Laboratory. Sadovodstvo 12:25 (in Russian)
Dirr MA (1998) Manual of woody landscape plants, 5th edn.
Stipes Publishing, Champaign, IL
Drumke JS (1964) A systematic survey of Corylus in North
America. PhD Diss, Univ of Tennessee, Knoxville, TN
Duke JA (1989) Handbook of nuts. CRC, Boca Raton, FL
Ebrahem KS, Richardson DG, Tetley RM, Mehlenbacher SA
(1994) Oil content, fatty acid composition, and vitamin E
concentration of 17 hazelnut varieties, compared to other
types of nuts and oil seeds. Acta Hort 351:685–692
eFloras (2009) Flora of China, Corylus. MO Bot Gard, St. Louis,
MO & Harvard University Herbaria, Cambridge, MA, USA:
http://www.efloras.org.Accessed 13 June 2009
Erdogan V (1999) Genetic relationships among hazelnut
(Corylus) species. PhD Diss, Oregon State University, Cor-
vallis, OR
Erdogan V, Mehlenbacher SA (2000a) Interspecific hybridiza-
tion in hazelnut (Corylus). J Am Soc Hortic Sci 125:489–497
Erdogan V, Mehlenbacher SA (2000b) Phylogenetic relation-
ships of Corylus species (Betulaceae) based on nuclear
ribosomal DNA ITS region and chloroplast matK gene
sequences. Syst Bot 25:727–737
Erdogan V, Mehlenbacher SA (2001) Incompatibility in wild
Corylus species. Acta Hort 556:163–169
Erener G, Ocak N, Garipoglu AV (2007) the influence of dietary
hazelnut kernel oil on the performance and fatty acid com-
position of broilers. J Sci Food Agric 87:689–693
FAOStat (2010): http://faostat.fao.org/site/567/default.aspx#ancor.
Accessed 13 June 2010
Farris CW (1974) An introduction to the stars – a new family of
filbert hybrids. Ann Rep Northern Nut Growers Assoc
67:80–82
Farris CW (1978) The trazels. Annu Rep Northern Nut Growers
Assoc 69:32–34
Farris CW (1982) A progress report on the development of
F2 hybrids of Corylus colurna C. avellana. Ann Rep
Northern Nut Growers Assoc 73:15–17
Farris CW (1989) Two new introductions: the ‘Grand Traverse’
hazelnut and ‘Spartan Seedless’ grape. Annu Rep Northern
Nut Growers Assoc 80:102–103
Farris CW (1990) Hazelnut cultivar development: a progress
report. Annu Rep Northern Nut Growers Assoc 81:118–119
Farris CW (1995) The paper barked hazel of China. Annu Rep
Northern Nut Growers Assoc 86:76–77
Farris CW (2000) The hazel tree. Northern Nut Growers Asso-
ciation, East Lansing, MI
Fideghelli C, De Salvador FR (2009) World hazelnut situation
and perspectives. Acta Hort 845:39–51
Forest F, Bruneau A (2000) Phylogenetic analysis, organization,
and molecular evolution of the nontranscribed spacer of 5 S
ribosomal RNA genes in Corylus (Betulaceae). Int J Pant Sci
161:793–806
Fuller AS (1908) The nut culturist. Orange Judd, New York
Gao XH, Liu JN, Ling Q (2008) Tissue culture propagation of
hybrid hazelnut (Corylus heterophylla C. avellana). Acta
Hort 771:207–211
Gellatly JU (1950) Description of Filazel varieties. Annu Rep
Northern Nut Growers Assoc 41:116–117
Gellatly JU (1956) Filazels. Annu Rep Northern Nut Growers
Assoc 47:112–113
Gellatly JU (1964) Filazels. Annu Rep Northern Nut Growers
Assoc 55:153–155
Gellatly JU (1966) Tree hazels and their improved hybrids.
Annu Rep Northern Nut Growers Assoc 57:98–101
GenBank (2010) Corylus nucleotide search. National Center for
Biotechnology Information (NCBI): http://www.ncbi.nlm.
nih.gov/sites/nuccore. Accessed 13 July 2010
Gleason HA, Cronquist A (1998) Manual of vascular plants of
Northeastern United States and adjacent Canada. The
New York Botanical Gardens, Bronx, NY
2Corylus 45
Go
¨kirmak T, Mehlenbacher SA, Bassil NV (2009) Characteri-
zation of European hazelnut (Corylus avellana) cultivars
using SSR markers. Genet Resour Crop Evol 56:147–172
Graham SH (1936) Notes on an experimental planting in central
New York. Annu Rep Northern Nut Growers Assoc 27:
64–67
Gumus M (2008) Evaluation of hazelnut kernel oil of Turkish
origin as alternative fuel in diesel engines. Renew Energ
33:2448–2457
G€
urcan K, Mehlenbacher SA, Botta R, Boccacci P (2010)
Development, characterization, segregation, and mapping
of microsatellite markers for European hazelnut (Corylus
avellana L.) from enriched genomic libraries and usefulness
in genetic diversity studies. Tree Genet Genomes. DOI
10.1007/s11295-010-0269-y. Published online: 10 Feb 2010
Hammond E (2006) Identifying superior hybrid hazelnut plants
in southeast Nebraska. MS Thesis, University of Nebraska-
Lincoln, Lincoln, NE
Hoffman A, Khan W, Worapong J, Strobel G, Griffen D, Arbo-
gast B, Barofsky D, Boone R, Li N, Zheng P, Daley L (1998)
Bioprospecting for Taxol in angiosperm plant extracts. Spec-
troscopy 13:22–32
Hoffman A, Shahidi F (2009) Paclitaxel and other taxanes in
hazelnut. J Funct Foods 1:33–37
Hummer KE (2001) Hazelnut genetic resources at the Corvallis
repositiory. Acta Hort 556:21–24
JinLi Z, Ming X, Weijian L, Zhongguan J, daoMing W (2007)
Breeding report of new hazelnut cultivar ‘Liaozhen 3’. Chin
Fruits 4:5–7 (in Chinese)
JinLi Z, Ming X, Weijian L, Zhongguan J, daoMing W (2008)
Breeding report of new hazelnut cultivar ‘Liaozhen 4’. Chin
Fruits 6:6–8 (in Chinese)
Johnson KB, Pinkerton JN (2002) Eastern filbert blight.
In: Teviotdale BL, Michailides TJ, Pscheidt JW (eds) Com-
pendium of nut crop diseases in temperate zones. Am Phy-
topathol Soc Press, St. Paul, MN, pp 44–46
Julian J, Seavert C, Olsen JL (2009a) An economic evaluation of
the impact of eastern filbert blight resistant cultivars in
Oregon, U.S.A. Acta Hort 845:725–732
Julian J, Seavert C, Olsen JL (2009b) Establishing and produc-
ing hazelnuts in the Willamette Valley – standard verses
double density. Acta Hort 845:769–774
Kask K (1998) European filberts in Estonia. Annu Rep Northern
Nut Growers Assoc 89:155–156
Kask K (2001) Nut quality and wild European hazelnut in
Estonia and attempts at hazelnut breeding. Acta Hort 556:
37–40
Kasapligil B (1972) A bibliography on Corylus (Betulaceae)
with annotations. Annu Rep Northern Nut Growers Assoc
63:107–162
Koksal AY (2000) Inventory of hazelnut research, germplasm
and references. REU Technical series 56. Food and Agricul-
ture Organization of the United Nations, Rome, Italy: http://
www.fao.org/docrep/003/x4484e/x4484e00.htm#Toc
Kudasheva RF (1965) Propagation and breeding of wild and
cultivated hazel nuts. Lesnaya Promishlennost, Moscow (in
Russian)
Lagerstedt HB (1975) Filberts. In: Janick J, Moore JN (eds)
Advances in fruit breeding. Purdue University Press, West
Lafayette, IN, pp 456–488
Lagerstedt HB (1976) Development of rootstock for filberts.
Annu Rep Northern Nut Growers Assoc 65:161–165
Lagerstedt HB (1990) Filbert rootstock and cultivar introduc-
tions in Oregon. Annu Rep Northern Nut Growers Assoc
81:60–63
Lee SW (2002) Forest genetic resources conservation in the
Republic of Korea. In: Palmberg-Lerche C, Iversen PA,
Sigaud P (eds) Forest genetic resources, vol 30. FAO,
Rome, Italy, pp 40–43
Liang W, Dong D, Wand G, Dong F, Liang L (2008) Studies on
hazelnut hybridization breeding of C. heterophylla
C. avellana in China. Abstracts of the 7th international
congress on Hazelnut. ISHS, Viterbo, Italy
Lunde CF, Mehlenbacher SA, Smith DC (2000) Survey of
hazelnut cultivars for response to eastern filbert blight inoc-
ulation. HortScience 35:729–731
Lunde CF, Mehlenbacher SA, Smith DC (2006) Segregation for
resistance to eastern filbert blight in progeny of ‘Zimmer-
man’ hazelnut. J Am Soc Hortic Sci 131:731–737
Mehlenbacher SA (1991a) Hazelnuts. In: Moore JN, Ballington
JR (eds) Genetic Resources in temperate fruit and nut crops.
Acta Hort 290, pp 789–836
Mehlenbacher SA (1991b) Chilling requirements of hazelnut
cultivars. Sci Hort 47:271–282
Mehlenbacher SA (1994) Genetic improvement of the hazelnut.
Acta Hort 351:23–38
Mehlenbacher SA (1997) Revised dominance hierarchy for
S-alleles in Corylus avellana L. Theor Appl Genet
94:360–366
Mehlenbacher SA (2003) Hazelnuts. In: Fulbright DW (ed)
A Guide to nut tree culture in North America, vol 1. North-
ern Nut Growers Assoc, pp 183–215
Mehlenbacher SA (2009) Genetic resources for hazelnut: state
of the art and future perspectives. Acta Hort 845:33–38
Mehlenbacher SA, Thompson MM (1991) Inheritance of a
chlorophyll deficiency in hazelnut. HortScience 26:
1414–1416
Mehlenbacher SA, Smith DC (1988) Heritability of ease of
hazelnut pellicle removal. HortScience 23:1053–1054
Mehlenbacher SA, Smith DC (1995) Inheritance of the cut leaf
trait in hazelnut. HortScience 30:611–612
Mehlenbacher SA, Smith DC (2001) Parital self-compatibility
in ‘Tombul’ and ‘Montebello’ hazelnuts. Euphytica 56:
231–236
Mehlenbacher SA, Smith DC (2002) Inheritance of pollen color
in hazelnut. Euphytica 127:303–307
Mehlenbacher SA, Thompson MM (2004) Inheritance of style
color in hazelnut. HortScience 39(3):475–476
Mehlenbacher SA, Smith DC (2006) Self-compatible seedlings
of the cutleaf hazelnut. HortScience 41:482–483
Mehlenbacher SA, Thompson MM, Cameron HR (1991) Occur-
rence and inheritance of resistance to eastern filbert blight in
‘Gasaway’ hazelnut. HortScience 26:410–411
Mehlenbacher SA, Smith DC, Brenner LK (1993) Variance
components and heritability of nut and kernel defects in
hazelnut. Plant Breed 110:144–152
Mehlenbacher SA, Brown RN, Davis JW, Chen H, Bassil NV,
Smith DC, Kubisiak TL (2004) RAPD makers linked to
eastern filbert blight resistance in Corylus avellana. Theor
Appl Genet 108:651–656
46 T.J. Molnar
Mehlenbacher SA, Brown RN, Nouhra ER, Go
¨kirmak T, Bassil
NV, Kubisiak TL (2006) A genetic linkage map for hazelnut
(Corylus avellana L.) based on RAPD and SSR markers.
Genome 49:122–133
Mehlenbacher SA, Azarenko AN, Smith DC (2007) ‘Santiam’
hazelnut. HortScience 42:715–717
Mehlenbacher SA, Smith DC, McCluskey RL (2008) ‘Sacajawea’
hazelnut. HortScience 43:255–257
Mehlenbacher SA, Smith DC, McCluskey RL (2009) ‘Yamhill’
hazelnut. HortScience 44:845–847
Ming X, Zheng J, Radicati L, Me G (2005) Interspecific hybri-
dization of hazelnut and performance of 5 varieties in China.
Acta Hort 686:65–67
Ming X, JinLi Z, ShuChai S, SuJuan G, ZhiXia H (2007)
Breeding of a new hazelnut cultivar ‘Liaozhen 1’. Chin
Fruits 5:8–10 (in Chinese)
Ming X, JinLi Z, ShuChai S, SuJuan G, ZhiXia H (2008)
Breeding of ‘Liaozhen 2’ hazelnut cultivar. Chin Fruits
1:11–13 (in Chinese)
Molnar TJ, Goffreda JC, Funk CR (2005) Developing hazelnuts
for the eastern United States. Acta Hort 686:609–618
Molnar TJ, Mehlenbacher SA, Zaurov DE, Goffreda JC (2007)
Survey of hazelnut germplasm from Russia and Crimea for
response to eastern filbert blight. HortScience 42:51–56
Molnar TJ, Capik JM, Goffreda JC (2009) Response of proge-
nies from known resistant parents to Anisogramma anomala
in New Jersey, USA. Acta Hort 845:73–81
Molnar TJ, Goffreda JC, Funk CR (2010) Survey of Corylus
resistance to Anisogramma anomala from different geo-
graphic locations. HortScience 45:832–836
Nas MN, Read PE (2004) Improved rooting and acclimation of
micro propagated hazelnut shoots. HortScience 39:
1688–1690
NCGR (2010) US Department of Agriculture ARS, National
Clonal Germplasm Repository, Corvallis, Oregon, USA:
http://www.ars.usda.gov/main/site_main.htm?modecode¼53-
58-15-00.Accessed 01 July 2010
Normah MN, Reed BM, Yu X (1994) Seed storage and cryoex-
posure behavior in hazelnut (Corylus avellana L. cv. Barce-
lona). Cryo Lett 15:315–322
Oliveira I, Sousa A, Valenta
˜o P, Andrade PB, Ferreira ICFR,
Ferreres F, Bento A, Seabra R, Estevinho L, Pereira JA
(2007) Hazelnut (Corylus avellana L.) leaves as a source of
antimicrobial and antioxidative compounds. Food Chem
105:1018–1025
Ourecky DK, Slate GL (1969) Susceptibility of filbert varieties
and hybrids to the filbert bud mite, Phytoptus avellanae Nal.
Annu Rep Northern Nut Growers Assoc 60:89–91
O
¨zcelik E, Peks¸en A (2007) Hazelnut husk as a substrate for the
cultivation of shiitake mushroom (Lentinula edodes). Bior-
esour Technol 98:2652–2658
Palme AE, Vendramin GG (2002) Chloroplast DNA variation,
postglacial recolonization and hybridization in hazel, Cory-
lus avellana. Mol Ecol 11:1769–1779
Pavlenko FA (1957) Hazelnut. In: Anonymous (ed) Culture of
nut species. Gosselkhozdat, Moscow, pp 5–126, in Russian
Pavlenko FA (1985) Hazelnut. In: Anonymous (ed) Nut, forest,
and orchard species, 2nd edn. Agroparomidaz, 123, p 99, in
Russian
Pomper KW, Azarenko AN, Bassil N, Davis JW, Mehlenbacher
SA (1998) Identification of random amplified polymorphic
DNA (RAPD) markers for self-incompatibility alleles in
Corylus avellana L. Theor Appl Genet 97:479–487
Reed CA (1936) New filbert hybrids. J Hered 27:427–431
Reed CA, Davidson J (1958) The improved nut trees of North
America. Devin-Adire, New York
Reed BM, Normah MN, Yu X (1994) Stratification is necessary
for successful cryopreservation of axes from stored hazelnut
seed. Cryo Lett 15:377–384
Reed BM, Hummer KM (2001) Long-term storage of hazelnut
embryonic axes in liquid nitrogen. Acta Hort 556:177–180
Reich JE (1980) Geneva filbert research – past and present.
Annu Rep Northern Nut Growers Assoc 71:110–111
Rosengarten F (1984) The book of edible nuts. Walker Publish-
ing, New York
Rutter PA (1987) Badgersett research farm – plantings, projects,
and goals. Annu Rep Northern Nut Growers Assoc 78:
173–186
Sarraquigne JP (2005) Hazelnut production in France. Acta Hort
686:669–673
Sathuvalli VR (2007). DNA markers linked to novel sources of
resistance to eastern filbert blight in European hazelnut
(Corylus avellana L.). MS Thesis, Oregon State University,
Corvallis, OR
Sathuvalli VR, Mehlenbacher SA (2009) A hazelnut BAC
library for map-based cloning of a disease resistance gene.
Acta Hort 845:191–194
Sathuvalli VR, Mehlenbacher SA, Smith DC (2009) New
sources of resistance to eastern filbert blight and linked
markers. Acta Hort 845:123–126
Sathuvalli VR, Mehlenbacher SA, Smith DC (2010) Response
of hazelnut accessions to greenhouse inoculation with Ani-
sogramma anomala. HortScience 45:1116–1119
Sharma SD, Kumar K (2001) Preliminary evaluation of hazelnut
seedling trees native to India. Acta Hort 556:29–35
Slate GL (1935) Suggestions for the breeding of nut trees. Annu
Rep Northern Nut Growers Assoc 26:36–41
Slate GL (1936) The filbert breeding project at Geneva. Annu
Rep Northern Nut Growers Assoc 27:62–63
Slate GL (1947) Some results with filbert breeding at Geneva,
NY. Annu Rep Northern Nut Growers Assoc 38:94–100
Slate GL (1952) Filbert varieties. Annu Rep Northern Nut
Growers Assoc 43:53–62
Slate GL (1959) Winter injury of filberts at Geneva, N.Y.
1958–1959.Annu Rep Northern Nut Growers Assoc 50:75–76
Slate GL (1961) The present status of filbert breeding. Annu Rep
Northern Nut Growers Assoc 52:24–26
Slate GL (1969) Filberts-including varieties grown in the east.
In: Jaynes RA (ed) Handbook of North American nut trees.
Northern Nut Growers Assoc, Knoxville, TN, pp 287–293
Slyusarchuk VE, Ryabokon AP (2001) Hazelnut in Ukraine.
Acta Hort 556:137–140
Slyusarchuk VE, Ryabokon AP (2005) Ukrainian hazelnuts: cul-
tivars, agrotechniques, perspectives. Acta Hort 686:603–608
Smith DC, Mehlenbacher SA (2002) ‘Rosita’ ornamental hazel-
nut. HortScience 37:1137–1138
Smith DC, Mehlenbacher SA (1996) Inheritance of contorted
growth in hazelnut. Euphytica 89:211–213
Smolyaninova LA (1936) Hazelnut. In: Vavilov NI (ed) Nut
species. Cultivated flora of the USSR, vol 17. Gosudarstven-
noe Izdatelstvo Sovkhoznoi i Kolkhoznoi Literatury,
Moscow-Leningrad, pp 126–205 (in Russian)
2Corylus 47
Sun W (1998) Corylus chinensis. In: IUCN 2010. IUCN
Red List of Threatened Species. Version 2010.2: www.
iucnredlist.org.Downloaded 13 July 2010
Thompson MM (1977a) Inheritance of bud mite susceptibility in
filberts. J Am Soc Hortic Sci 102:39–42
Thompson MM (1977b) Inheritance of nut traits in filbert.
Euphytica 26:465–474
Thompson MM (1979) Genetics of incompatibility in Corylus
avellana L. Theor Appl Genet 54:113–116
Thompson MM (1985) Linkage of the incompatibility locus and
red pigmentation genes in hazelnut. J Hered 76:119–122
Thompson MM, Smith DC, Burges JE (1985) Nondormant
mutants in a temperate tree species, Corylus avellana L.
Theor Appl Genet 70:687–692
Thompson MM, Lagerstedt HB, Mehlenbacher SA (1996)
Hazelnuts. In: Janick J, Moore JN (eds) Fruit breeding, vol
3, Nuts. Wiley, New York, pp 125–184
Tombesi A (2005) World hazelnut situation and perspectives:
Italy. Acta Hort 686:649–658
Tous J (2005) Hazelnut production in Spain. Acta Hort 686:
659–664
Volovich PI, Chripach PI (1998) Diversity and abundance of
cultivated varieties of filberts in Belarus. Annu Rep Northern
Nut Growers Assoc 89:157–158
Weijian L, Ming Z, Wanying X (1994) Hazelnut breeding in
northern China. Acta Hort 351:59–66
Weschcke C (1954) Hazels and filberts. In: Growing nuts in the
north. Webb, St. Paul, MN, pp 24–38
Weschcke C (1963) Forty-three years of active work in nut
growing. Annu Rep Northern Nut Growers Assoc 54:63–65
Weschcke C (1970) A little nut history. Annu Rep Northern Nut
Growers Assoc 61:113–116
Whitcher IN, Wen J (2001) Phylogeny and biogeography of
Corylus (Betulaceae): inferences from ITS sequences. Syst
Bot 26:283–298
Xu YX, Hanna MA (2009) Synthesis and characterization of
hazelnut oil-based biodiesel. Ind Crop Prod 29:473–479
Xu YX, Hanna MA (2010) Evaluation of Nebraska hybrid
hazelnuts: nut/kernel characteristics, kernel proximate com-
position, and oil protein properties. Ind Crop Prod 31:84–91
Xu YX, Hanna MA, Josiah SJ (2007) Hybrid hazelnut oil
characteristics and its potential oleochemical application.
Ind Crop Prod 26:69–76
Yablokov AS (1962) Selection of woody species. Selkhozgiz,
Moscow, in Russian
Yao Q, Mehlenbacher SA (2000) Heritability, variance compo-
nents and correlation of morphological and phenological
traits in hazelnut. Plant Breed 119:369–381
Yoo K, Wen J (2002) Phylogeny and biogeography of Carpinus
and subfamily Coryloideae (Betulaceae). Int J Plant Sci
163:641–650
Yu X, Reed BM (1995) A micropropagation system for hazel-
nuts (Corylus species). HortScience 30:120–123
Zohary D, Hopf M (2004) Domestication of plants in the old
world. Clarendon, Oxford
48 T.J. Molnar
http://www.springer.com/978-3-642-21249-9