Microsatellite Marker Fingerprinting Reveals Paternity Predicament in Pear Rootstocks

Abstract

Early in the 20th century, a collection of fire blight resistant pears from around the world was assembled in southern Oregon in efforts to develop improved rootstocks. ‘Old Home’ and ‘Farmingdale’ are two cultivars from Illinois that exhibited fire blight resistance and useful horticultural traits. Both became important as interstem stocks and as parents in the development of new rootstocks. In the 1950s, an Oregon nurseryman collected seed from an ‘Old Home’ tree in British Columbia pollinated by ‘Farmingdale’ and hundreds of numbered selections of this cross ‘Old Home’ x ‘Farmingdale’ (OHxF) were evaluated. Several OHxF selections are now used as rootstocks worldwide, and 45 unique OHxF selections are maintained at the USDA-ARS National Clonal Germplasm Repository (NCGR), in Corvallis, Oregon. Simple Sequence Repeat (SSR) or microsatellite-based profiles were generated for ‘Old Home’, ‘Farmingdale’, 8 OHxF selections, and standard pear cultivars at NCGR using a standard fingerprinting set developed by the European Cooperative Programme for Plant Genetic Resources. ‘Farmingdale’ is thought to be a seedling of ‘Beurr d’Anjou’, and shared at least one SSR allele at each locus tested, confirming this parental relationship. All OHxF selections shared an allele with ‘Old Home’ at each locus, with one allele carrying a suspected pair of base deletions. However, based on the SSR results, it is impossible for ‘Farmingdale’ to be the pollen parent for any of the OHxF selections examined. Evaluation of the world pear collection at NCGR with this fingerprinting set will help identify the actual pollen parent of these rootstocks.

Full text

157
Journal of the American Pomological Society 67(3): 157-167 2013
OH˟ F Paternity Perplexes Pear Producers
JosEpH postMan1, daEil kiM2, and naHla bassil1
1 USDA Agricultural Research Service, National Clonal Germplasm Repository, Corvallis, Oregon, USA.
2 Department of Horticultural Science, Chungbuk National University, Cheongju, Chungbuk 361-763, South Korea.
Abstract
Early in the 20th century, a collection of re blight resistant pears from around the world was assembled in
southern Oregon in an effort to develop improved rootstocks. ‘Old Home’ and ‘Farmingdale’ are two cultivars
from Illinois that exhibited strong re blight resistance and useful horticultural traits. Both became important as
interstem stocks and as parents in the development of new rootstocks. In the 1950s, an Oregon nurseryman col-
lected seed from an ‘Old Home’ tree in British Columbia purportedly pollinated by ‘Farmingdale’ and hundreds
of numbered selections of this cross of ‘Old Home’ × ‘Farmingdale’ (OH×F) were evaluated. Several OH×F
selections are now valued as rootstocks worldwide, and 45 unique OH×F selections are maintained at the USDA-
ARS National Clonal Germplasm Repository (NCGR), in Corvallis, Oregon. Simple Sequence Repeat (SSR)
or microsatellite-based proles were generated for ‘Old Home’, ‘Farmingdale’, 8 OH×F selections, and several
reference pear cultivars at NCGR using a standard ngerprinting set developed by the European Cooperative
Programme for Plant Genetic Resources. ‘Farmingdale’ is thought to be a seedling of ‘Beurré d’Anjou’. Our
study showed that ‘Farmingdale’ shared at least one SSR allele with ‘Anjou’ at each locus tested, conrming this
parental relationship. Our studies showed that all OH×F selections shared an allele with ‘Old Home’ at each locus,
with one allele carrying a suspected pair of base deletions. However, based on our SSR results, it is impossible for
‘Farmingdale’ to be the pollen parent for any of the OH×F selections examined. Evaluation of the world pear col-
lection at NCGR with this ngerprinting set established the cultivar ‘Bartlett’ as the actual pollen parent of these
rootstock clones. Fruit and leaf morphology is also consistent with ‘Bartlett’ and not ‘Farmingdale’ as a parent of
OH×F rootstock selections. The highly re blight resistant ‘Farmingdale’ is apparently very under-represented in
the pedigrees of current pear rootstocks, and deserves renewed consideration.
Early in the 20th century, Oregon State Uni-
versity (OSU) professor Frank Reimer began
scouring the world for Pyrus germplasm with
resistance to the important bacterial disease
re blight (Erwinia amylovora (Burrill) Win-
slow et al.), which made its rst appearance
in the Rogue River Valley in 1906 (Reimer,
1925). Reimer was the rst superintendent
of the OSU Southern Oregon Experiment
Station near Medford. From 1917 to 1919
his travels took him to China, Korea, Man-
churia, and Japan where he collected scions
from many local pear cultivars and seed from
pear wild relative species (Reimer, 1919;
Fig. 1). He also collected seed from wild and
hedgerow perry pears in France, and clones
of hundreds of the leading pear cultivars of
Western Europe and North America. Seed-
ling populations were generated and clones
were grafted and established at the OSU
station in Talent, Oregon, where the diverse
germplasm collections were evaluated for
fruit quality, disease resistance and as poten-
tial rootstocks (Reimer, 1919; Reimer, 1925;
Fig. 1. Frank C. Reimer traveling while investigating
pears in northern China in 1919.

158 Journal of the american Pomological Society
Hartmann, 1957). Two of Professor Reimer’s
most valuable discoveries came from a 1915
trip to visit fruit grower Benjamin Buck-
man in Farmingdale, Illinois. One was the
cultivar ‘Farmingdale’, an open pollinated
seedling Buckman had found near a ‘Beur-
ré d’Anjou’ tree on his farm, and the other
was ‘Old Home’, a seedling that had come
from a nursery in Paris, Illinois some years
earlier. Reimer observed both of these trees
to be completely free of re blight and he
brought scions from ‘Old Home’ back to Or-
egon when he returned. ‘Farmingdale’ scions
were sent to him several years later (Hum-
mer, 1998; Westwood and Brooks, 1963;
Westwood, 1967). After more than a decade
of observing natural infections, and perform-
ing articial inoculations in southern Oregon
with E. amylovora, Reimer found many pear
species selections and Asian cultivars with
strong re blight resistance, but only three
European cultivars: ‘Farmingdale’, ‘Long-
worth’, and ‘Old Home’ had excellent blight
resistance as trunk stocks (Reimer, 1925).
The many valuable pear selections identied
or developed by Reimer and his successors
were later transferred to the National Clonal
Germplasm Repository at Corvallis, Oregon
when the USDA Agricultural Research Ser-
vice established this genebank in 1981 (Post-
man et al., 2006; Westwood, 1982).
‘Farmingdale’. The origin and character-
istics of ‘Farmingdale’ are discussed in con-
siderable detail by Reimer (Reimer, 1925;
Reimer, 1950). It is apparently a pure form
of P. communis which originated as a chance
seedling in the orchard of Benjamin Buck-
man, at Farmingdale, Illinois. The fruit size
is medium to large, and it resembles ‘Anjou’
in form and coloration. Flesh is white, fairly
ne, buttery, moderately juicy, and quite free
of grit. It is reasonably sweet but somewhat
lacking in desirable avor characteristics,
and ripens in midseason. The original seed-
ling was discovered close to an ‘Anjou’ tree,
which was presumed to be the parent. Hart-
mann (1957) noted “while this pear is of little
value for its fruit, it is quite remarkable in
other characteristics. Its tree is vigorous, well
formed, fairly productive, and is the most
blight resistant of all the P. communis cul-
tivars tested at the Southern Oregon Branch
Experiment Station.”
‘Old Home’. ‘Old Home’ is also appar-
ently a pure form of P. communis that origi-
nated as a chance seedling with B.O. Curtiss,
at Paris, Illinois (Hartmann, 1957) and was
growing in Benjamin Buckman’s trial or-
chard in Farmingdale prior to 1907 (Ragan,
1908). The fruit of ‘Old Home’ is small,
round, slightly truncate, with no neck. Skin
is yellow-green, becoming yellow when
ripe. Flesh is white to yellow, ne textured,
becoming soft and tender when ripe with
few stone cells, medium dry, subacid and
with poor avor (Howlett, 1957). While its
fruit is of no consequence, the variety was
widely used as a blight resistant trunk and
framework stock, and as a very good source
of blight resistant seedlings. When crossed
with ‘Farmingdale’, a high percentage of the
seedlings were very blight resistant as well
as vigorous and uniform in growth (Hart-
mann, 1957). Reimer (1925) found ‘Old
Home’ to have very good graft compatibility
with quince and it has since been used as an
interstem bridge for incompatible pear cul-
tivars. Westwood (1967) showed that ‘Old
Home’ is also graft compatible on hawthorn
(Crataegus sp.) rootstock. When the disease
Pear Decline spread through the pear grow-
ing areas of British Columbia in the 1940s
and through Washington, Oregon and Cali-
fornia in the 1950s and 1960s, it was found
that trees with ‘Old Home’ root or trunk
stocks were much more resistant to pear de-
cline as well as to re blight. ‘Old Home’ was
extremely difcult to propagate as a clonal
rootstock, however, while many OH×F se-
lections rooted much easier from hardwood
cuttings (Westwood and Brooks, 1963).
‘Old Home’ and ‘Farmingdale’ in breed-
ing. In the mid 1920s, following the death of
Benjamin Buckman, the original ‘Farming-
dale’ and ‘Old Home’ trees in Illinois were
destroyed, and the trees established at South-

159
ern Oregon Experiment Station became the
primary source of nursery stock (Fig. 2;
Brooks, 1984; Hummer, 1998; Westwood,
1967). Reimer (1950) found in the 1930s that
when ‘Farmingdale’ was used as a pollen
parent in crosses with other blight resistant
selections, a high percentage of the resulting
seedlings exhibited a high resistance to blight.
This was especially true when ‘Old Home’
was the seed parent. Although a number of
crosses between other blight resistant parents
resulted in >90% seedlings that were resis-
tant to blight following inoculation, many
of those resistant seedlings became infected
when a susceptible cultivar such as ‘Bartlett’
or ‘Beurré Bosc’ was grafted onto them. In
those cases, re blight could spread from an
infected cultivar across the graft union into
the rootstock. However, the OH×F seedlings
were resistant even to the spread of re blight
from infection of a grafted cultivar (Reimer,
1950). Based on Reimer’s work, ‘Old Home’
and ‘Farmingdale’ became important parent
pEar
Fig. 2. HortScience cover in 1967 showing F. Reimer
by original Old Home tree recently transplanted to Tal-
ent, Oregon.
stocks for the development of both rootstock
and edible pear cultivars. Several important
Canadian cultivars obtained their re blight
resistance from ‘Old Home’, and at least two
US fruit cultivars have ‘Farmingdale’ in their
pedigrees (Table 1). By far, the most im-
portant use of ‘Farmingdale’ and especially
‘Old Home’ has been in the development of
pear rootstocks. ‘Old Home’ in particular has
been a valuable parent in rootstock breeding
programs in France, UK, Germany, and else-
where (Table 1; Elkins et al., 2012).
OH×F selections. One of Reimer’s goals
was to establish a mother block of ‘Old
Home’ and ‘Farmingdale’ at the Southern
Oregon Experiment Station to generate seed
for producing blight resistant seedling root-
stocks. Oregon nurseryman Lyle Brooks was
concerned about the variability of pear culti-
vars grafted onto OH×F seedling rootstocks,
and he set out to develop clonal rootstocks
from this cross. In 1952, he obtained half a
kilogram of seed from an isolated block of
‘Old Home’ trees planted with ‘Farmingdale’
pollenizers at the Canada Research Station in
Summerland, British Columbia. Of the thou-
sands of resulting seedlings, he planted 516
in a nursery block and evaluated them for
ease of propagation from cuttings and other
traits (Brooks, 1984; Westwood and Brooks,
1963). Propagation by hardwood cuttings
was successful, and thirteen of the more eas-
ily propagated OH×F numbered selections
(OH×F 18, 34, 51, 69, 87, 97, 112, 198, 217,
230, 267, 333, and 361) were extensively
evaluated at OSU for disease and insect re-
sistance, environmental tolerance, anchor-
age, dwarng and fruit production of grafted
cultivars (Lombard and Westwood,1987).
More than 40 OH×F selections, including the
13 above, are preserved at the USDA gene-
bank in Corvallis (Table 2). Some continue
to be propagated worldwide and are in high
demand for new plantings even though they
lack the size control and precocity demanded
for high density orchards (Elkins et al., 2012).
OH×F 333 was selected for its resistance to
multiple diseases and moderate size control

160 Journal of the american Pomological Society
to be the standard clonal rootstock for the
thousands of pear clones maintained at the
USDA genebank (Postman, 2008a; 2008b).
A new generation of seedlings generated
from open pollinated (OP) seed collected
from an isolated group of select OH×F clones
(OH×F 40, 51, 87, 333 and 339) resulted in
the Horner rootstock series (Table 2; Mielke
and Smith, 2002; Mielke and Sugar, 2004).
Horner 4 is under evaluation in the 2005 NC-
140 collaborative pear rootstock trial (Elkins
et al., 2011), and a trial comparing Horner 4,
Horner 10 and OH×F 87 was established in
2009 in the states of Oregon and Washington
(Einhorn, personal communication).
DNA ngerprints. Bassil and Postman
(2009) developed expressed sequence tag
(EST)-simple sequence repeat (SSR) mark-
ers and evaluated them for usefulness in a
diverse collection of individuals from the
three most commonly cultivated pear spe-
cies: P. communis, P. pyrifolia and P. ussu-
riensis at the USDA genebank. They found
that ‘Farmingdale’ shared one SSR allele
with ‘Anjou’ at each of 10 loci, suggesting
that ‘Anjou’ was very likely the maternal par-
ent of ‘Farmingdale’, as Buckman and Re-
imer had suspected nearly a century earlier
(Bassil and Postman, 2009; Hartmann, 1957;
Reimer, 1925). OH×F 333 was the only se-
lection included in that study and appeared
to share one allele with either ‘Old Home’ or
‘Farmingdale’, thus conrming its reported
parentage. This paper reports on the pheno-
Table 1. Pear selections reported to have either ‘Old Home’ or ‘Farmingdale’ in the pedigree, and USDA gene-
bank accession numbers.
Plant name Pedigree Accession no. Origin
Farmingdale in Pedigree
P. betulifolia-2 x Farmingdale P. betulifolia-2 x Farmingdale PI 541785 United States
Rogue Red Comice x (Seckel x Farmingdale sdlg. 122) PI 541252 United States
Vistica Nectar Farmingdale x Burkett none United States
Old Home in Pedigree
AC Harrow Gold (HW 616) Harvest Queen x Harrow Delight none Canada
BU 2/33 (Pyro II) Old Home x Bonne Louise d’Avranches PI 617679 Germany
Harrow Delight (HW 603) Bartlett x (Early Sweet x Old Home) PI 541431 Canada
Harrow Sweet (HW 609) Bartlett x (Old Home x Early Sweet) PI 617562 Canada
Harrow Sweet (Old Home x Early Sweet) x HW
Harrow selection HW 623 605 none Canada
Harrow selection HW 624 Harrow Sweet x NY10353 none Canada
Nijisseiki x Old Home Nijisseiki x Old Home PI 541767 United States
OH 11 (Pyriam) Seedling of Old Home CPYR 2700 France
OH 20 re blight resistant Old Home seedling selection PI 541237 United States
rootstock
OH 50 re blight resistant Old Home seedling selection PI 541238 United States
rootstock
Old Home x P. betulifolia-1 Old Home x P. betulifolia-1 PI 541787 United States
Old Home x P. betulifolia-1 Old Home x P. betulifolia-1 PI 541788 United States
Pyrodwarf Old Home x Bonne Louise d’ Avranches PI 617654 Germany
QR 708-02 BP-1 x Old Home PI 617680 United Kingdom
QR 708-12 BP-1 x Old Home CPYR 2704 United Kingdom
QR 708-36 BP-1 x Old Home CPYR 2705 United Kingdom
Nijisseiki x Old Home (seedlot) Nijisseiki x Old Home W-6 PI 541753 United States
Old Home x Bartlett (seedlot) Old Home x Bartlett PI 541380 United States

161
pEar
Table 2. Pear selections reported to have both ‘Old Home’ and ‘Farmingdale’ in pedigree, and USDA genebank ac-
cession numbers.
Plant name Pedigree Accession no. Origin
Harrow selection (Anjou x Farmingdale) x Harrow Delight none Canada
HW 621
Horner 4 O.P. from population of OHxF 40, 51, 87, 333 & 339 CPYR 2955 United States
Horner 10 O.P. from population of OHxF 40, 51, 87, 333 & 339 CPYR 2956 United States
Horner 51 O.P. from population of OHxF 40, 51, 87, 333 & 339 PI 657931 United States
OHxF 1 Old Home x Farmingdale seedling selection PI 541408 United States
OHxF 2 Old Home x Farmingdale seedling selection PI 541413 United States
OHxF 4 Old Home x Farmingdale seedling selection PI 541409 United States
OHxF 5 Old Home x Farmingdale seedling selection PI 541410 United States
OHxF 7 Old Home x Farmingdale seedling selection PI 541399 United States
OHxF 9 Old Home x Farmingdale seedling selection PI 541421 United States
OHxF 18 Old Home x Farmingdale seedling selection PI 541397 United States
OHxF 23 Old Home x Farmingdale seedling selection PI 541411 United States
OHxF 29 Old Home x Farmingdale seedling selection PI 541398 United States
OHxF 34 Old Home x Farmingdale seedling selection PI 541412 United States
OHxF 40 Old Home x Farmingdale seedling selection PI 541402 United States
OHxF 51 Old Home x Farmingdale seedling selection PI 541369 United States
OHxF 58 Old Home x Farmingdale seedling selection PI 541454 United States
OHxF 69 Old Home x Farmingdale seedling selection PI 665738 United States
OHxF 87 Old Home x Farmingdale seedling selection PI 541415 United States
OHxF 97 Old Home x Farmingdale seedling selection PI 541370 United States
OHxF 101 Old Home x Farmingdale seedling selection PI 541416 United States
OHxF 109 Old Home x Farmingdale seedling selection PI 541417 United States
OHxF 112 Old Home x Farmingdale seedling selection PI 541418 United States
OHxF 130 Old Home x Farmingdale seedling selection PI 541406 United States
OHxF 132 Old Home x Farmingdale seedling selection PI 541404 United States
OHxF 198 Old Home x Farmingdale seedling selection PI 541419 United States
OHxF 217 Old Home x Farmingdale seedling selection PI 541371 United States
OHxF 226 Old Home x Farmingdale seedling selection PI 541372 United States
OHxF 230 Old Home x Farmingdale seedling selection PI 541420 United States
OHxF 247 Old Home x Farmingdale seedling selection PI 541422 United States
OHxF 257 Old Home x Farmingdale seedling selection PI 541396 United States
OHxF 259 Old Home x Farmingdale seedling selection PI 541423 United States
OHxF 261 Old Home x Farmingdale seedling selection PI 541401 United States
OHxF 266 Old Home x Farmingdale seedling selection PI 541424 United States
OHxF 267 Old Home x Farmingdale seedling selection PI 541400 United States
OHxF 280 Old Home x Farmingdale seedling selection PI 541425 United States
OHxF 282 Old Home x Farmingdale seedling selection PI 541426 United States
OHxF 288 Old Home x Farmingdale seedling selection PI 541373 United States
OHxF 319 Old Home x Farmingdale seedling selection PI 541427 United States
OHxF 333 Old Home x Farmingdale seedling selection PI 541405 United States
OHxF 340 Old Home x Farmingdale seedling selection PI 541403 United States
OHxF 361 Old Home x Farmingdale seedling selection PI 541374 United States
OHxF 377 Old Home x Farmingdale seedling selection PI 541428 United States
OHxF 501 Old Home x Farmingdale seedling selection PI 541407 United States
OHxF 512 Old Home x Farmingdale seedling selection PI 541414 United States
OPR-005 OHxF C-44 Bartlett x (Old Home x Farmingdale) PI 617519 United States
OPR-013 OHxF C-34 Old Home x Farmingdale PI 541368 United States

162 Journal of the american Pomological Society
typic verication of the ‘Farmingdale’ and
‘Old Home’ clones growing at the USDA
genebank, and the expanded SSR ngerprint-
ing of these cultivars and several OH×F root-
stock selections using a universal ngerprint-
ing set previously described by Evans et al.
(2009) that is made up of more polymorphic
genomic SSR loci.
Materials and Methods
Plant material. Young, actively growing
leaves were collected in the spring of 2007
from the clonal genebank accessions of ‘An-
jou’, ‘Old Home’, ‘Farmingdale’, OH×F 51,
OH×F 69 (3 sources), OH×F 87, OH×F 97,
OH×F 230, OH×F 333 and several standard
cultivars in the NCGR orchard collection in
Corvallis, Oregon (44.554° lat., -123.222°
long.). The OH×F 69 clone in the gene-
bank collection had come from OSU with a
note in its inventory record about a possible
identity question. In 2008, additional clones
of OH×F 69 that trace back to the original
Daybreak Nursery tree were obtained from
Fowler Nurseries (Newcastle, CA) and North
American Plants (McMinnville, Oregon) for
comparison.
DNA extraction. Thirty to 50 mg of leaf
tissue was placed in a cluster tube (Corn-
ing, Tewksbury, MA) containing a 4 mm
stainless steel bead (McGuire Bearing Com-
pany, Salem, OR). The samples were frozen
in liquid nitrogen and stored at -80°C, until
extraction. Grinding was performed in the
Retsch MM301 Mixer Mill, (Retsch, Inc.,
Hann, Germany) rapidly at a frequency of 30
cycles·s-1 using three 30 s bursts. DNA was
extracted with the Qiagen protocol described
in detail by Gilmore et al. (2011).
SSR genotyping. Twelve microsatellite
primers that make up a standard primer set
proposed by the European Cooperative Pro-
gram for Plant Genetic Resources (ECPGR)
(Evans et al., 2009) were used. A standard
set of six reference cultivars (Table 3) was
included to allow harmonization of genetic
data when comparing results generated in
different laboratories and using different sep-
aration platforms. The Type-it Microsatellite
Multiplex PCR Kit (Qiagen Inc., Valencia,
CA) was used to amplify SSRs in two reac-
tions of 15 µL total volume. The 15 µL PCR
reaction mix contained: 8.3 µL of 2x Type-it
Multiplex PCR Master Mix (Qiagen), 1.7 µL
of a 10x multiplex primer mix containing 2
µM of each primer, 1.7 µL Q solution, and
3.3 µL of 3ng·µL-1 template DNA. Touch-
down amplication was performed with a
Tetrad thermocycler (MJ Research, Inc., Wa-
tertown, MA) using an initial step of 95oC for
5 min, followed by 10 cycles of 95oC for 30
s, annealing temperatures starting at 62oC for
90 s (decreasing by 1oC/cycle), and 72oC for
30 s for extension. This step was followed
by 28 cycles of 95oC for 30 s, 52oC for 90 s,
and 72oC for 30 s, with a nal extension step
of 60oC for 30 min. Success of the PCR was
conrmed by 2% agarose gel electrophore-
sis. Fragment analysis followed separation
on a Beckman CEQ 8000 capillary genetic
analyzer (Beckman Coulter, Fullerton, CA).
Allele sizing and visualization were per-
formed using the fragment analysis module
of the CEQ 8000 software. Alleles were
scored by tting the peaks into bins less than
one nucleotide. For unweighted pair group
method with arithmetic mean (UPGMA)
cluster analysis, individuals were scored for
the presence or absence of each allele and
PowerMarker (Liu and Muse, 2005) was
used for cluster analysis.
Paternity testing. Paternity analysis for the
six OH×F selections included in this study
(Table 3) was carried out using a likelihood
approach with the CERVUS 3.0 software
(Kalinowski et al., 2007). Using the geno-
type le at the 12 loci, CERVUS was used
to determine allele frequency, to simulate pa-
ternity analysis and calculate critical values
of likelihood ratios. The paternity analysis
module (unknown sexes) used the twelve
cultivars included in the study as candidate
parents. Other parameters set for CERVUS
were 0.99 for the proportion of loci typed and
0.022 for rate of genotyping error.
Phenotype verication. Fresh fruits and

163
pEar
Table 3. Genotypic information at 12 SSR loci for ‘Old Home’, ‘Farmingdale’, 6 OH×F selections, and 10 pear cultivars available at USDA-ARS, NCGR. Alleles
present in Farmingdale but not observed in any of the OHxF selections are highlighted. Allele mismatches are underlined.
CH05c06 EMPc117 GD147 EMPc11 CH04e04 CH01f07a CH03g08 CH01d09 CH03d13 CH02b11 CH01d10 GD97
Abbe Fetela 97/115 117/119 125/125 146/152 181/199 182/192 246/250 289/297 111/115 128/128 155/157 145/153
Anjou 97/115 117/121 125/125 142/152 181/181 182/196 252/260 283/285 113/127 132/138 155/157 159/175
Bartletta 91/95 89/117 125/125 152/152 181/207 178/186 230/246 243/279 111/127 122/128 153/161 159/175
Comicea 91/91 117/117 125/131 152/156 181/199 184/186 232/236 279/285 111/115 134/138 155/161 145/153
Conferencea 91/101 119/121 125/125 142/152 181/207 182/194 230/260 279/285 111/127 124/128 161/161 171/199
Farmingdale 91/115 99/121 125/125 142/142 181/181 182/190 230/260 283/285 111/113 138/138 157/161 159/175
Harrow Sweet 91/97 113/117 125/125 152/156 181/181 186/192 246/260 243/285 111/127 122/122 153/157 153/159
Hosuia 87/109 95/107 135/137 148/148 189/189 182/204 256/262 283/283 101/101 124/134 159/159 175/175
OH×F 51 95/95 89/119 125/125 152/152 181/181 178/182 230/246 279/279 111/127 122/132 135/153 159/167
OH×F 69 91/91 89/119 125/127 152/156 181/181 186/192 230/260 279/281 115/127 128/128 135/161 159/167
OH×F 87 91/91 117/117 125/127 152/152 181/207 186/192 230/246 243/281 111/115 122/132 135/153 159/173
OH×F 97 91/91 117/119 125/125 152/156 181/207 186/192 230/246 279/281 111/115 128/132 135/153 159/167
OH×F 230 95/95 89/119 125/127 152/156 181/181 178/192 230/246 279/279 111/127 128/132 135/153 159/173
OH×F 333 91/95 117/119 125/127 152/152 181/181 182/186 230/230 279/279 115/127 122/132 153/157 159/173
Old Home 91/95 119/119 125/127 152/156 181/181 182/192 230/260 279/281 111/115 128/132 135/159 167/173
Passe Crassanea 91/111 101/117 125/129 152/152 181/181 182/182 230/246 279/285 127/127 132/132 157/161 159/175
Pyrodwarf 91/91 119/121 125/125 152/156 181/181 182/192 260/260 279/285 111/115 124/128 135/161 167/199
Rogue Red 91/95 117/121 125/127 146/152 181/181 186/210 236/260 279/279 111/111 138/138 161/161 145/159
a Reference accessions recommended by Evans et al. (2009) to allow data comparison across collections.

164 Journal of the american Pomological Society
leaves were photographed, or photos were
examined from the genebank photo archives
for ‘Anjou’, ‘Farmingdale’, ‘Old Home’, and
several OH×F selections. Source histories of
the genebank cultivars were reviewed and
fruit characteristics were compared to pub-
lished descriptions.
Results and Discussion
DNA ngerprints. Each of the 12 cultivars
and OH×F selections were uniquely identi-
ed except for the three OH×F 69 selections
obtained from different sources which had
identical ngerprints (Fig. 3). All OH×F se-
lections shared an allele with ‘Old Home’ at
each of the 12 loci with two exceptions: OH×F
87 was homozygous for 117 at EMPc117 and
did not have the 119 allele from ‘Old Home’;
and OH×F 333 did not have the 135 or 159
alleles present in ‘Old Home’ at CH01d09
(Table 3). Neither of the two alleles at four
SSR loci (EMPc117, EMPc11, CH01d08
and CH02b10) in ‘Farmingdale’ was found
in any of the OH×F selections indicating that
‘Farmingdale’ is highly unlikely to be the
pollen parent for these selections (Table 3).
‘Farmingdale’ shared an allele with ‘Anjou’
at each locus (Table 3) conrming the latter
as one of the likely parents as previously re-
ported (Bassil and Postman, 2009).
Six of the cultivars selected as references
(Evans et al., 2009) were included in this
study to standardize microsatellite allele
scoring and allow data comparison across
collections and separation platforms in the
future (Table 3). ‘Bartlett’ appeared to share
one allele at each locus with the OH×F se-
lections (Table 3). Paternity analysis us-
ing CERVUS (Kalinowski et al., 2007) was
conducted to conrm this possibility. It con-
rmed ‘Old Home’ and ‘Bartlett’ as the most
likely parent pair with the highest likelihood
at the relaxed condence level (80%) for
OH×F 69, 87 and 333 and at the strict con-
dence level (95%) for OH×F 51, 97 and
230. No allele mismatches were observed for
Fig. 3. Unweighted pair group method with arithmetic mean (UPGMA) cluster analysis of the evaluated pear
cultivars and selection based on 12 SSR loci.

165
all but OH×F 87 and OH×F 333 which had
a single mismatch. The mismatch of one al-
lele at one locus (described above) for these
two selections is most likely due to mutation,
common at microsatellite sites. This result
reveals ‘Bartlett’ as the original pollen parent
for the OH×F clones included in this study.
Based on the phenotypes of the 41 OH×F
clones in the genebank collection, it is prob-
able that ‘Bartlett’ was the pollen parent for
all of them.
Cultivar source histories and phenotypes.
Genebank records show that the ‘Farming-
dale’ and ‘Old Home’ clones were obtained
as scions from the original trees that Reimer
had established at the Southern Oregon Ex-
periment Station (USDA-ARS, 2012) and
identity verication notes indicate that pre-
vious visual examination of these cultivars
match those of the original trees as described
by Hartmann (1957). ‘Farmingdale’ fruit has
a very distinctive shape, not unlike a swollen
and slightly elongated ‘Anjou’ (Fig. 4) with a
similar ne texture and good fruit quality. The
peduncle is unusually short, thick and swol-
len where connected to the fruit. ‘Old Home’
is distinctively and unusually round for a Eu-
ropean pear, with a slender peduncle of me-
dium length (Fig. 4). The eating quality of
‘Old Home’ fruit is remarkably poor. The fruit
as grown at NCGR clearly matches the pho-
tograph from the variety collection in Ohio
taken in the mid 20th century (Howlett, 1957).
The various ‘Bartlett’ and ‘Anjou’ clones in
the genebank collection produce fruit that
exactly matches those cultivars in fruit qual-
ity, ripening times, and disease susceptibility
as they are known from U.S. commercial or-
chards. We are condent that the ‘Farmingda-
le’, ‘Old Home’, ‘Anjou’, ‘Bartlett’ and other
ve standard reference clones represented at
the USDA genebank and used in this study are
correctly identied based on their phenotypes.
DNA ngerprints of ‘Farmingdale’ and the six
pEar
Fig. 4. Fruit and leaves of 'Old Home', 'Farmingdale', 'Bartlett' and 3 OH˟ F selections.

166 Journal of the american Pomological Society
reference cultivars (Table 3) obtained from the
National Fruit Collections (Brogdale Collec-
tions) in the UK also exactly match those from
the NCGR collection (Evans, personal com-
munication).
All OH×F clones produce fruit that is
somewhat similar in size, shape and color
(Fig. 4) with crisp, juicy texture and sweet
sprightly avor; although, they differ in ma-
turity dates. All of the OH×F selections re-
semble ‘Old Home’, but not ‘Farmingdale’,
in fruit shape and especially in peduncle size.
The foliage of the various OH×F selections
is similar in leaf length/width ratio, pedi-
cel length and thickness, and pubescence,
the latter of which is quite distinct. They all
bear some resemblance to the foliage of ‘Old
Home’, but not to ‘Farmingdale’ which has
characteristically lanceolate leaves and less
pubescence (Fig. 4).
While ‘Bartlett’, rather than the more
blight resistant ‘Farmingdale’, may be the
pollen parent of the OH×F selections now
being used commercially, these OH×F selec-
tions have exhibited excellent blight resis-
tance, as well as resistance to pear decline
and good fruit production when used as
rootstocks for our most important European
pear cultivars (Elkins et al., 2012; Lombard
and Westwood, 1987). The highly re blight
resistant ‘Farmingdale’ is apparently very
under-represented in the pedigrees of pres-
ent day pear rootstocks as well as in the par-
ent clones currently being used in rootstock
breeding programs. ‘Farmingdale’ as a par-
ent is not likely to instill dwarng or pre-
cocity traits in a rootstock, however if re
blight resistance is to be an important genetic
component of future pear cultivars and root-
stocks, the results of Frank Reimer (1925,
1950) should inspire the reconsideration of
‘Farmingdale’ germplasm in breeding.
Genetic testing of fruit varieties has be-
come increasingly important in improving
the efciency of breeding new varieties (Iez-
zoni et al., 2010) and has the additional ben-
et of clarifying pedigrees of existing variet-
ies (Sitther et al., 2012). Kimura et al. (2003)
used SSR markers to conrm and revise the
paternity of several Japanese pear cultivars.
Discovering the paternity of important or
historic varieties like the OH×F rootstocks
or ‘Bing’ cherry (Prunus avium L.) not only
makes interesting news (Warner, 2013), but
helps to avoid inbreeding and aids in the
identication of sources of useful horticul-
tural and adaptive traits. Future completion
of DNA ngerprinting in the USDA pear
genebank will help eliminate redundancy in
the collection, streamline the introduction
and conservation of unique and valuable
germplasm accessions, and help breeders
better understand the paternity of parents
when making crosses to develop improved
varieties.
Acknowledgements
We wish to thank David Hunter and Rich-
ard Bell for providing pedigree information
for pear selections developed from ‘Old
Home’ or ‘Farmingdale’. We also thank Kate
Evans and Todd Einhorn for their critical re-
views of this manuscript and April Nyberg
for her laboratory expertise in ngerprinting
these accessions.
Literature cited
Bassil, N.V. and J.D. Postman. 2009. Identication of
European and Asian pears using EST-SSRs from Py-
rus. Genetic Resources and Crop Evol. 57:357-370.
Brooks, L.A. 1984. History of the Old Home × Farming-
dale pear rootstocks. Fruit Var. J. 38:126-128.
Elkins, R., R. Bell, and T. Einhorn. 2012. Needs as-
sessment for future US pear rootstock research di-
rections based on the current state of pear produc-
tion and rootstock research. J. Amer. Pomol. Soc.
66(3):153-163.
Elkins, R., S. Castagnoli, C. Embree, R. Parra-Queza-
da, T. Robinson, T. Smith, and C. Ingels. 2011.
Evaluation of potential rootstocks to improve tree
precocity and productivity. Acta Hort. 909:184-194.
Evans, K.M., F. Fernandez-Fernandez, and C.L. Go-
van. 2009. Harmonizing ngerprinting protocols to
allow comparisons between germplasm collections
– Pyrus. Acta Hort. 814:103-106.
Gilmore, B., K. Hummer, and N. Bassil. 2011.
DNA Extraction protocols from dormant buds of
twelve woody plant genera. J. Amer. Pomol. Soc.
65(4):201-207.

167
pEar
Hartmann, H. 1957. Catalog and evaluation of the pear
collection at the Oregon Agricultural Experiment
Station. Technical Bulletin 41. 80 pp.
Howlett, F.S. 1957. Preliminary evaluation of new and
uncommon pear varieties. North Central Regional
Bulletin 75 and Ohio Agricultural Experiment Sta-
tion Research Bulletin 790.131 pp.
Hummer, K. 1998. ‘Old Home’ and ‘Farmingdale,’ the
Romeo and Juliet of pear rootstocks: an historical
perspective. Fruit Var. J. 52:38-40.
Iezzoni, A., C. Weebadde, J. Luby, C. Yue, E. van de
Weg, G. Fazio, D. Main, C.P. Peace, N.V. Bassil,
and J. McFerson. 2010. RosBREED: Enabling
marker-assisted breeding in Rosaceae. Acta Hort.
859:389-394.
Kalinowski, S.T., M.L. Taper, and T.C. Marshall. 2007.
Revising how the computer program CERVUS ac-
commodates genotyping error increases success in
paternity assignment. Mol. Ecol. 16:1099-1006.
Kimura, T., Y. Sawamura, K. Kotobuki, N. Matsuta, T.
Hayashi, Y. Ban, and T. Yamamoto. 2003. Parent-
age analysis in pear cultivars characterized by SSR
markers. J. Jap. Soc. Hort. Sci. 72:182-189.
Liu, K. and S.V. Muse. 2005. Powermarker: integrated
analysis environment for genetic marker data. Bio-
informatics 21:2128–2129.
Lombard, P.B. and M.N. Westwood. 1987. Pear root-
stocks pp. 145-183 in R.C. Rom and R.F. Carlson
(eds.). Rootstocks for Fruit Crops. John Wiley &
Sons.
Mielke, E.A. and L. Smith, 2002. Evaluation of the
Horner rootstocks. Acta Hort. 596:325-330.
Mielke, E.A. and D. Sugar. 2004. Initial seven-year
evaluation of thirteen Horner pear rootstocks. Acta
Hort. 658:513-517.
Postman, J., K. Hummer, E. Stover, R. Krueger, P.
Forsline, L.J. Grauke, F. Zee, T. Ayala-Silva, and
B. Irish. 2006. Fruit and nut genebanks in the US
National Plant Germplasm System. HortScience
41(5):1188- 1194.
Postman, J.D. 2008a. The USDA quince and pear
genebank in Oregon, A world source of re blight
resistance. Acta Hort. 793:357-362.
Postman, J.D. 2008b. World Pyrus collection at USDA
genebank in Corvallis, Oregon. Acta Hort. 800:527-
534.
Ragan, W.H. 1908. Nomenclature of the pear; a cat-
alog-index of the known varieties referred to in
American publications from 1804 to 1907. USDA
Bureau of Plant Industry Bulletin No. 126.
Reimer, F.C. 1919. Report of a trip to the Orient to
collect and study Oriental pears. Station report for
the Southern Oregon Experiment Station and the
USDA Ofce of Foreign Seed and Plant Introduc-
tion.
Reimer, F.C. 1925. Blight Resistance in Pears and
Characteristics of Pear Species and Stocks. Oregon
Agricultural College, Experiment Station Bulletin
214, Corvallis, Oregon.
Reimer, F.C. 1950. Development of blight resistant
pear stocks. Oregon Agricultural Exp. Sta. Bulletin
485.
Sitther, V., D. Zhang, D. Harris, W.R. Okie, S. Dhek-
ney, and A. Yadav. 2012. Cultivar identication,
pedigree verication, and diversity analysis among
peach (Prunus persica L. Batsch) cultivars based
on Simple Sequence Repeat markers. J. Amer. Soc.
Hort. Sci. 137:114-121.
USDA-ARS. 2012.Germplasm Resources Informa-
tion Network - (GRIN). http://www.ars-grin.gov/
cgi-bin/npgs/acc/search.pl?accid=PI+541188
(Farmingdale); http://www.ars-grin.gov/cgi-bin/
npgs/acc/search.pl?accid=PI+541456 (Old Home)
(December 6, 2012).
Warner, G. 2013. Paternity test implicates Napoleon:
scientists used genetic markers to identify Bing’s
parents. Good Fruit Grower (Jan. 1) 64(1):17-19.
Westwood, M. N. 1967. About our cover: the quest for
re blight resistance in pear. HortScience 2(3):93.
Westwood, M.N. 1982. Pear germplasm of the new
National Clonal Repository: its evaluation and uses.
Acta Hort. 124:57-66.
Westwood, M.N. and L.A. Brooks.1963. Propagation
of hardwood pear cuttings. Proc. Int. Plant. Prop.
Soc. 13:261-268.