An immunomarking method to determine the foraging patterns of Osmia cornifrons and resulting fruit set in a cherry orchard

Abstract

The foraging patterns of Osmia cornifrons (Radoszkowski) (Megachilidae, Hymenoptera) were determined with an immunomarking method and correlated with fruit set in a commercial tart cherry orchard in Pennsylvania. Adults of O. cornifrons were self-marked with chicken egg-white protein powder from a dispenser nest box placed at the center of the study orchard at early bloom. Flower samples were collected from randomly selected trees (n=30) located at different distances from the nest box. Flowers were analyzed for the presence of immunomarker protein with enzyme-linked immunosorbent assay. Foraging patterns were determined by measuring the distance and direction of marked flowers from the nest box. While marked flowers were found out to 55 m (maximum distance sampled), most marked flowers were found within 35 m from the nest and the percentage of marked flowers declined rapidly beyond that distance. Fruit density per limb cross-sectional area (cm 2) in the study orchard was significantly higher than in the orchard without O. cornifrons, indicating the value of O. cornifrons as pollinators in increasing fruit yield in tart cherries. immunomarking / foraging patterns / Osmia cornifrons / self-marking procedure / pollination / pollinator / mason bees

Full text

An immunomarking method to determine the foraging
patterns of Osmia cornifrons and resulting fruit set
in a cherry orchard
David J. BIDDINGER
1,2
, Neelendra K. JOSHI
1,2
, Edwin G. RAJOTTE
2
,
Noemi O. HALBRENDT
3
, Cassandra PULIG
1
, Kusum J. NAITHANI
4
, Mace VAUGHAN
5
1
Fruit Research and Extension Center, Entomology, Pennsylvania State University, 290 University Dr, Biglerville
17307, PA, USA
2
Department of Entomology, Pennsylvania State University, 501 ASI Building, University Park 16802, PA, USA
3
Fruit Research and Extension Center, Plant Pathology, Pennsylvania State University,
290 University Dr, Biglerville 17307, PA, USA
4
Department of Geography and Intercollege Graduate Degree Program in Ecology, Pennsylvania State University,
University Park 16802, PA, USA
5
Xerces Society for Invertebrate Conservation, 628 NE Broadway Ste 200, Portland 97232, OR, USA
Received 11 June 2012 – Revised 15 May 2013 – Accepted 27 June 2013
Abstract – The foraging patterns of Osmia cornifrons (Radoszkowski) (Megachilidae, Hymenoptera) were
determined with an immunomarking method and correlated with fruit set in a commercial tart cherry orchard in
Pennsylvania. Adults of O. cornifrons were self-marked with chicken egg-white protein powder from a
dispenser nest box placed at the center of the study orchard at early bloom. Flower samples were collected from
randomly selected trees (n=30) located at different distances from the nest box. Flowers were analyzed for the
presence of immunomarker protein with enzyme-linked immunosorbent assay. Foraging patterns were
determined by measuring the distance and direction of marked flowers from the nest box. While marked
flowers were found out to 55 m (maximum distance sampled), most marked flowers were found within 35 m
from the nest and the percentage of marked flowers declined rapidly beyond that distance. Fruit density per
limb cross-sectional area (cm
2
) in the study orchard was significantly higher than in the orchard without O.
cornifrons, indicating the value of O. cornifrons as pollinators in increasing fruit yield in tart cherries.
immunomarking / foraging patterns / Osmia cornifrons / self-marking procedure / pollination / pollinator /
mason bees
1. INTRODUCTION
Osmia cornifr ons (Radoszkowski) (Megachilidae,
Hymenoptera), commonly known as the horn-
faced bee or the Japanese orchard bee, is a
univoltine and solitary bee species. It is a manage-
able and commercially available alternative pollina-
tor of rosaceous tree fruit species (Bosch and Kemp
2002; Mader et al. 2010; Maeta and Kitamura 1974;
Sekita and Yamada 1993). In Japan, the develop-
ment of O. cornifr ons as a fruit pollinator started
in the 1930s and increased in use from 10 % of
thetotalappleproductionareain1981toover
80 % by 1996 (Maeta 1990; Sekita et al. 1996;
Batra 1998). Introduced in the eastern United
States from Japan in 1977 (Batra 1979, 1998)as
an orchard pollinator, it has become naturalized
in several states including Pennsylvania where
natural emergence coincides with red maple and
apricot bloom and continues through the end of
Corresponding author: D.J. Biddinger , djb134@psu.edu;
N.K. Joshi, nkj105@psu.edu
Manuscript editor: Peter Rosenkranz
Apidologie
Original article
* INRA, DIB and Springer-Verlag France, 2013
DOI: 10.1007/s13592-013-0221-x

apple bloom 3–4 weeks later (Biddinger et al.
2009a; Biddinger et al. 2011a, 201 1b). It has more
recently been introduced as a commercial pollina-
tor of fruit in Korea and China (Xu et al. 1995).
While this species nests in abandoned wood
beetle galleries and other cavities in the wild,
O. cornifrons can also be managed by providing
nest boxes full of nest cavities (e.g., cardboard
tubes, bamboo segments, pre-drilled wooden
blocks, etc.) (Mader et al. 2010). These nest boxes
can be held at cold temperatures to delay adult
emergence and synchronize with later flowering
target crops, such as apple and blueberry that
bloom towards the end of the natural emergence
period (Bosch et al. 2008; Mader et al. 2010).
The maximum distance from which nesting
bees are able to find their nests, also known as
homing range, is believed to be a good estimate
of a bee’s maximum foraging range and potential
habitat size (Gathmann and Tscharntke 2002;
Greenleaf et al. 20 07; Guedot et al. 20 09).
However, homing range depends on various
factors, including availability of floral resources.
In the case of O. cornifrons, it was estimated to be
approximately 500 m (Guedot et al. 2009)with
effective pollination range of only 40 to 60 m in
large orchards (Maeta and Kitamura 1974;
Kitamura and Maeta 1969;Yamadaetal.1971).
For a similar orchard pollinator, Osmia cornuta,
adult bees foraged up to 400 m from the nesting
sites (Maccagnani et al. 2003). In a similar study
on O. cornuta foraging in a pear orchard, over
75 % of samples of O. cornuta hadpearpollenin
the fecal matter (Monzon et al. 2004), suggesting
strong foraging activity within the orchard where
nesting shelters were placed. In the case of O.
cornifrons, in spite of considerable research in
Japan on its ecology and management in com-
mercial tree fruit orchards (Yamada et al. 1971;
Yamada et al. 1984, Maeta 1978, Maeta and
Kitamura 1974), its foraging behavior in the fruit
growing regions of the eastern USA or elsewhere
is not well known. Estimating foraging distance of
O. cornifrons in tree fruit species (e.g., cherry and
apple) and their impact on fruit set could be
helpful in determining nest density and placement
of managed nest aggregations (multiple nest
blocks in a nest shelter) in commercial orchards,
as well as the pollination limitations of wild
populations flying into commercial orchards from
adjacent habitat.
In the past, researchers have used a variety of
techniques to quantify the movement of differ-
ent insect species. Most of these techniques
(e.g., mark–release–recapture using trace ele-
ments) had numerous drawbacks, particularly
related to their high cost of implementation or
extensive technical expertise needed for their
operation (Akey et al. 1991). Moreover, some of
these techniques require capturing and killing
foraging bees. In large colonies of social bees
using non-repr oductive workers as foragers
consequences to the colony are likely insignif-
icant, but the capturing and killing of forag-
ing solitary bee females also means killing
potential offspring. Recently, a highly effec-
tive protein-based marking (immunomarking)
method for studying dispersal of insect fauna
was developed (Hagler et al. 1992;Hagler
1997; Hagler and Jackson 1998). In this
method, experimental insects or their habitat
were marked with a vertebrate-specific pro-
tein marker, and the recaptured insects were
examined for the presence of the marker
antibodies using a highly sensitive enzyme-
linked immunosorbent assay (ELISA) procedure.
In the early phase of research and development of
this marking technique, costly rabbit IgG and
chicken IgG proteins were widely used as marker
proteins in mark–release–recapture studies
(Hagler et al. 1992;Hagler1997;Blackmeretal.
2004; Hagler and Naranjo 2004; Buczkowski and
Bennett 2006). These markers were costly, how-
ever, so recently, inexpensive protein markers
such as chicken egg whites, bovine milk
protein, and soy milk protein have been
employed for mark–release–recapture studies
(Jones et al. 2006; Hagler and Jones 2010).
The immunomarking (protein-based) mark–
release–recapture method has been used in
determining dispersal and movement of a wide
range of insect pests of fruit crops including
adult pear psylla (Cacopsylla pyricola), codling
moth (Cydia pomonella) (Jones et al. 2006), and
glassy-winged sharpshooter, Homalodisca
vitripennis (Blackmer et al. 2006).
D.J. Biddinger et al.

Hagler et al. (201 1a, 2011b) used a self-marking
mechanism to apply unique fluorescent powders or
protein markers to Apis mellifera as they left the
hive to determine foraging ranges by capturing
individual bees. Our study used this technique to
mark O. cornifr ons females as they foraged from a
nest shelter placed in the center of a tart cherry
orchard. O. cornifrons are much more wary and
energetic in the field than A. mellifera (Biddinger ,
personal observation) and are capable of visiting up
to 15 flowers per minute (Maeta and Kitamura
1974). This makes them difficult to capture in the
field. So, rather than capturing the bees, we
collected flowers at set distances from the nest
box to test for the presence of the marker to see
where the bees had visited. By looking at the
percentage of flowers marked, we hoped to
quantify pollination range limits at which a single
nest box of O. cornifr ons could effectively pollinate
the crop (without killing the bees or disrupting their
behavior).
The information related to foraging ranges of
O. cornifrons is critical in developing manage-
ment guidelines (e.g., how far apart to place nests)
for using O. cornifrons as a managed pollinator of
tree fruits. O. cornifrons is abundant in forest
edges adjacent to fruit orchards in some parts of
Pennsylvania and many fruit growers are relying
on it and other unmanaged populations of on non-
Apis bees to pollinate cherries, apples, and other
tree fruit (Ritz et al. 2012; Biddinger et al. 2010,
2011a; Joshi et al. 2011). Understanding its
foraging range in flowering orchards is also
necessary to understand how much and how far
pollination services may be provided by bees
nesting in wooded areas adjacent to orchards.
Specifically, we investigated (1) the appropri-
ateness of a self-marking system for O.
cornifrons; (2) the relationship between the
proportion of immunomarked flowers, distance
and direction from the nest and, thus, the foraging
patterns of O. cornifrons in a tart cherry orchard
during early stage of bloom; and (3) the relation-
ship between the proportion of marked flower
samples and the fruit density per limb cross-
sectional area (cm
2
) of sampled trees as a measure
of yield over the marked foraging area and as a
measure of pollination efficiency when these trees
were compared to cherry trees of a control orchard
block without O. cornifrons.
2. MATERIALS AND METHODS
2.1. Study orchards
Field and laboratory studies were conducted in
Adams County, Pennsylvania in a commercial and a
research tart cherr y, Prunus cerasus L., orchard
located adjacent to and on the Pennsylvania State
University Fruit Research and Extension Center
during the 2011 growing season. The nest box with
the protein marker dispenser was placed in a 7.4-ha
commercial Montmorency cherry orchard on
Mahaleb rootstock that was planted in 2004 at a
spacing of 4.9 m by 7.2 m (287 trees/ha). A 0.25-ha
control orchard planted in 2006 with the same tart
cherry variety, on the same rootstock and with the
same row spacing was located on the Penn State Fruit
Research and Extension Center. The control block
was 600 m away from the commercial orchard and
from the nearest Osmia nesting habitat. Timed counts
(30 min) of bees visiting cherry flowers in each
cherry block found equivalent levels of A. mellifera
in both blocks with commercial hives located
equidistant from both blocks. The timed counts of
bees were performed by randomly walking through
the whole area of the study and control orchard
blocks. In the year of the study, honey bees were not
brought into the station until after cherry bloom for
the beginning of apple bloom, so neither cherry block
had stocked honey bees. Other pollinator species
(several species of Andrena) were found infrequently
along the borders of the commercial cherry block, but
not in the control block on the research station.
During the study period, meteorological data (wind
speed, wind direction, and air temperature) were
recorded by a Weather Station of the Penn State Fruit
Research and Extension Center located adjacent to
the study orchard.
2.2. Study insects
The overwintering adults of O. cornifr ons (1,200
loose cocoons extracted from nest tubes of which a
subsample reared to eclosion in cages, indicated a 59:41
ratio of males to females, and about 10 % mortality in the
Foraging patterns of Osmia cornifrons in a cherry orchard

cocoon phase) used in this study were obtained by trap-
nesting wild populations of O. cornifrons from several
nearby Adams County orchard s the previous spring
using Osmia BinderBoards® (Pollinator Paradise,
Parma, ID, USA). Binderboards® are laminated wooden
blocks with 49 holes (8 mm diameter and 150 mm
depth) lined with paper straws. The nests were left in the
orchards all season from mid-April through October and
then gathered into a screened insectarium in October to
overwinter under ambient conditions. The number of O.
cornifrons released for t his study was based on
recommended number of females needed to pollinate
1 ha of apples in Japan (Yamada et al. 1971).
2.3. Nest placements
A nest box with dispenser (for the powdery chicken
egg white protein formulation), previously tested to use
O. cornifr ons as a delivery system for a powdered
formulation of Bacillus subtillus for the biological control
of fireblight in apple and pear orchards (Biddinger et al.
2009b, 2010), allowed the exiting bees to self-mark
(Figure 1). This simple wooden structure was modified
from a design by Maccagnani et al. (2006) who used it to
test fireblight biological control using Bombus as the
dispersing agent, but we modified it to accommodate the
much smaller O. cornifrons. Three empty BinderBoard®
wooden blocks, each with 49 holes (8 mm) lined with
paper straws for nesting were placed within the main
body of the nes t dispenser. Loose O. cornifrons
overwintering cocoons (n=1,200) were placed inside
the nest box shelter in two emergence cardboard boxes
each with two 10 mm emergence holes. A transparent
plastic exit ramp with a removable piece of 10-mm-thick
plexiglass with six shallow channels (4 mm deep) was
loaded manually with the fine egg white powder in the
early morning before flight and again at mid-day. To
force the bees to move through the channels and pick up
the egg white powder rather than directly flying out of the
ramp opening, a piece of clear acetate, used for overhead
projectors, was taped over the opening, sloping down to
the edge of the ramp to force bees to exit through the
constricted opening formed by the acetate and grooved
plexiglass plate. This plexiglass plate could be moved
freely in and out of a groove formed by aluminum
channel plates on each side and rested upon another piece
Figure 1. Osmia nestboxusedinthestudy(Source:PennFruitNews).JOB Japanese orchard bee (O. cornifrons).
D.J. Biddinger et al.

of plexiglass of the same dimensions that was perma-
nently affixed to the outside of the nest box. This allowed
quick and easy removal and filling of the grooves with
the egg white powder without disturbing the bees. The
entrance tubes at the bottom of the nest box were made of
PVC tubing with an inside diameter of 9 mm and were
marked with different dark colors to help females
“remember” the correct entrance hole from their initial
orientation flight and return directly to a specific tube
opening. In the fireblight biocontrol study, Biddinger et
al. (2010) observed that 95 % of the bees exited the nest
properly using the ramp and were exposed to powder, but
only about 50 % returned to enter the lower tubes
properly. The other half used the exit grooves to re-enter .
Proper exit behavior and contact with the biocontrol
product was by far the most important aspect of the
dispenser. In the present study, the nests with O.
cornifrons cocoons were placed in the study orchard
approximately 2 weeks prior to immunomarking in
order to allow for emerg ence, mating, and nest estab-
lishment prior to cherry bloom. Adult bees were
observed foraging on weeds in the ground cover prior
to cherry bloom.
2.4. Marking of bees
O. cornifrons adults were self-marked with chicken
egg whites (Deb-El Foods Corporation, Elizabeth, NJ,
USA). Egg white marker was placed in the grooves of
the plexiglass plate 30 h prior to flower sample
collection. Early bloom and the short exposure period
was chosen to maximize the possibility of bees moving
to fresh flowers, for flowers to be sampled before
senescence, and to minimize secondary movement of
the marker to new flowers by other pollinators such as
honey bees and the few Andrenid bees that were
present. Excluding night hours when the bees do not
fly, flight was possible during the 19 h of daylight, but
3 h were below the 10 °C threshold for flight. Of the
remaining 16 h, wind speed above 10 km/h with wind
speed gusts of over 24 km/h minimized flight for about
10 h, so actual optimal foraging time before the flower
samples were collected was approximately only 6 h.
2.5. Sample collection and handling
Sampling points (n=30) at different distances were
randomly selected in three concentric zones (15, 35, and
55 m radius from the Osmia nest box) in the study
orchard (Figure 2). The closest distance sampled was
4.95 m from the nest box. In each zone, ten randomly
selected trees were sampled comprising single-tree
replications. From each replicate tree we randomly
collected 50 samples (five cherry blossoms/sample)
totaling 250 blossoms per tree. Thus a total 1,500
samples (7,500 blossoms) (500 samples/concentric zone)
were collected from 30 trees for analysis. In all samples,
each blossom was individually plucked from trees with
the help of a flat-tip straight forceps (made of stainless
steel). Blossoms were randomly selected and were
collected from the interior and exterior canopy of sample
trees. Forceps were cleaned in alcohol (70 %) when
moving to new trees and only the base of the flower was
touched while collecting to avoid cross-contamination.
All samples of cherry blossoms were collected
into plastic sample bags with mesh dividers (Agdia
Inc., IN, USA), and were immediately placed in a
cooler containing ice in field. These samples were
brought to the laboratory and were kept in a −80 °C
storage facility (Thermo LabSystems, MA, USA).
ELISA was used to determine the presence of protein
marker chicken egg albumin (CEA).
2.6. Laboratory assays
Standardizatio n of bioassays for O. cornifrons
marked flowers was conducted in a set of preliminary
laboratory studies, and the ELISA procedure (Jones et al.
2006; Hagler and Jones 2010) described below was
optimized for extracting and quantifying the
immunomarker from flower samples collected from
cherry trees.
Extractions of CEA from cherry blossoms were
performed with 10 mL of TBS with 1 % EDTA at 27 °C
for 3 min. Eighty microliters of sample were then
transferred to ELISA plates (Plates, Nunc-Immuno Plate
MaxiSorp Surface (NU NC Brand Products, A/S,
Roskilde, Denmark)). All samples and controls were run
in triplicate. Negative controls were prepared from
unmarked cherry blossoms sampled from a control
orchard. Incubation was done for 1 h at 37 °C. Plates
were washed 5× with PBST using a Dynex multi plate
washer (Thermo LabSystems). Plates were blocked with
200 μLofPBS–BSA (1% BSA) at pH 7.4 (Sigma) and
incubated overnight at 4 °C. After blocking, plates were
washed 5× with PBST. All reagents were prepared in
Foraging patterns of Osmia cornifrons in a cherry orchard

filtered double-distilled water (Thermo Nanopure system).
Anti-CEA prepared in rabbit at 1:8,000 dilution (Sigma
No.C-6534) prepared in PBS–BSA (1 %)+1.3 ppm
Silwet L77 (1.3 μL/mL) was added per well and incubated
for 1 h at 37 °C. Plates were washed 5× with PBST. Goat
anti-rabbit IgG (Sigma No. A-6154) conjugated to
hydrogen peroxidase prepared 1:12,000 dilution in PBS–
BSA (1%)+(1.3 μL/mL) 80 μl per well was added and
incubated for 2 h at 37 °C. Plates were washed 5× with
PBST. Eighty microliters per well of substrate (TMB
Substrate) were incubated at 15 and 30 min and read with
microplate reader (Multiskan Ascent, Thermo
Labsystems) at 650 nm. Blossom samples were considered
marked when the absorbance reading was twice the
negative check.
2.7. Horticultural measurements in study
orchard
2.7.1. Measurement of percent bloom
Estimates of percent bloom were determined at the
time of immunomarked O. cornifrons release as well as
at the time of flower sample collection. Percent bloom
on the day of release of immunomarked O. cornifr ons
was determined by counting fully opened flowers and
Figure 2. Map of the study area where red triangle is O. cornifrons nest dispenser box and circles represent
sampled trees. Diameter of each circle represents protein marking (%) found on that tree ranging from 0.01–
100 % and X represents zero protein marking. Arrows indicate the wind direction and associated number is
averaged duration of wind in that direction. Red arrow is showing the bee entrance direction. Sample points in
dark gray, blue, and red color represent sampled trees in concentric zone of 15 m (inner), 35 m (middle), and
55 m (outer) radius, respectively from the nest box.
D.J. Biddinger et al.

flower buds until n=100 was reached and then
calculating the proportion of fully opened flowers on
five randomly selected limbs of all 30 sample trees.
Similarly, percent bloom at the time of flower collection
was calculated from all trees in the same manner.
2.7.2. Fruit density measurement in the study
orchard
The fruit density per limb cross-sectional area
(cm
2
) of sample trees in the study orchard was
compared with sample trees (n=30) of a control
orchard block at the time of fruit harvest. Sample
trees in control orchards were randomly selected from
all directions from the center and in similar concen-
tric zones as study orchard. At that time, three limbs
were randomly selected from each sample tree, the
total number of fruit per limb was counted and the
limb cross-sectional area was measured (Schupp and
Greene 2002). Limb cross-sectional area and fruit
density were determined as follows:
Limb cross‐sectional area¼
Limb circumferenceðÞ
2
4π
ð1Þ
Fruit density¼
Total number of fruits
Limb cross‐sectional area cm
2
ðÞ
ð2Þ
2.8. Statistical analysis
The relationship between distance from the point
of release (i.e., nest box) of O. cornifrons and the
proportion of immunomarked cherry flower samples
collected from different sampling points (i.e., cherry
tree) was determined by using a linear regression
analysis. This regression analysis was also used to
analyze the data related to the percent bloom at the
time of sample collection and distance from the nest
box. A polynomial regression analysis (SigmaPlot®
11, Systat Software, Inc., Chicago, IL) determined the
relation between the distance of sample trees from the
nest box and the fruit density per limb cross-sectional
area (cm
2
). A linear regression analysis was used to
determine the relation between fruit density per limb
cross-sectional area (cm
2
) and the proportion of
protein marked samples. A t test compared fruit
density per limb cross-sectional area (cm
2
) of study
orchard and control orchard block.
3. RESULTS
3.1. Foraging of O. cornifrons in cherry
orchard at the time of bloom
Percent bloom at the time of sample collec-
tion was evenly distributed throughout the study
orchard and did not show any significant
relationship with distance from O. cornifrons
nest box (P=0.42, R
2
=0.024, slope=0.038,
intercept=6.76). During the study, the average
wind direction was northwest for 22 h of the 30-h
foraging period (Figure 2).
The proportion of protein marking decreased
with increasing distance from O. cornifrons nest
box (illustrated by circles whose diameter
corresponds to the proportion of protein mark-
ing in Figure 2). Trees at the medium distance
(16–35 m radius) from the nest box of O.
cornifrons showed greatest variability in protein
marking (Figure 2). The proportion of protein-
marked flowers from trees at increasing dis-
tances from the nest linearly decreases (R
2
=
0.35, slope=1.52, inte rcept=80.38, P=0.0006;
Figure 3). There was a sharp decrease in the
proportion of protein markings beyond 35 m
from the nest box (Figure 3).
3.2. Relationship between distance from nest
box, pro portion of immunomarked
flower samples, and fruit density
There was no statistically significant declining
trend (P=0.20, R
2
=0.11) in the mean fruit density
(in terms of mean number of fruit per limb cross-
sectional area) with increasing distance from O.
cornifrons nest box (Figure 4). There was no
trend in average fruit density within 40 m from O.
cornifrons nest box; however, a decline in fruit
density after 40 m from Osmia nest box was
recorded (Figure 4). The mean fruit density of
sampled trees in the study orchard did not show
Foraging patterns of Osmia cornifrons in a cherry orchard

any trend with proportion of protein markings on
sampled flowers from sampling trees at each
sampling distance point (R
2
=0.02, intercept=
57.65, slope=0.076, P=0.42).
Figure 3. Foraging of O. cornifrons as quantified from the proportion of protein marking in flower samples
collected from 30 trees (sampling points) at different distances within a 55-m radius area from O. cornifrons
nest placed at the center of a commercial tart cherry orchard in Adams County, Pennsylvania. Inner solid lines
represent mean and 95 % confidence interval and the outer dashed lines represent the prediction band.
Figure 4. Relationship between the mean fruit density [number of fruits/limb cross-sectional area (cm
2
)] and
different distances (sampling points) within 55 m radius area from O. cornifrons nest box. Inner solid lines
represent mean and 95 % confidence interval and the outer dashed lines represent the prediction band.
D.J. Biddinger et al.

3.3. Fruit yield in O. cornifrons pollinated
orchard and control orchard block
The study orchard with the O. cornifrons nest
box showed significantly greater fruit density
on the trees that were sampled for protein
marking (mean=60.57 fruits/cm
2
of tree limb,
SD=19.89, P=0.000, t=7.146, n=30 trees) than
the control orchard block without O. cornifrons
nest (mean=29.11 fruits/cm
2
of tree limb, SD=
13.63, n=30 trees).
4. DISCUSSION
The results of this study showed that the
foraging range and behavior of O. cornifrons
adults in a cherry orchard can be quantified by
using an immunomarker. In a manner similar to
studies which have successfully used Osmia sp.
as a vector to disperse bacterial biological control
agents of fire blight (Biddinger et al. 2009a, 2010;
Maccagnani et al. 2006), O. cornifrons adults
were able to acquire the fine egg white particles on
their bodies and disperse these particles to cherry
flowers at levels high enough to be detected. Wind
direction appears to influence the distribution
pattern of immunomarking in this study. Wind
plays a crucial role in the dispersal of insects in
field environment (Schneider 1962), especially
for a relatively small-bodied bee attempting to
return to a nest flying into a stiff wind. Though, in
our study, the O. cornifrons nest was sheltered
from the wind with a plastic shelter from all
directions except the front (facing southeast), that
wind direction and speed affect the foraging
behavior of O. cornifrons in an orchard is evident
from the majority of the marked blossoms being
detected downwind from the nest box (Figure 2).
Average wind duration for N-W direction was
22 h of the 30-h foraging period with most of
the foraging taking place in the 4 h that wind
speed was below 10 kmh. Therefore, effect of
wind speed and direction on O. cornifrons
dispersal as quantified in terms of the proportion
of immunomarker found in the flower samples in
field environment needs further investigation.
Our results suggest that the protein marking
decreased with incr easing distance from O.
cornifrons nest box (Figure 2). Results from
the marked bloom samples show a decline in
fruit yield at those same distances after 40 m
and indicate the effective foraging range of
approximately 500 O. cornifrons females in a
commercial tart cherry orchard to be about 35–
40 m under the relatively short foraging period
allowed i n this experiment and under the
relatively cool, windy conditions that prevailed.
This range estimate correlates well with visual
observations made for O. cornifrons in Japanese
apple orchards of 40–60 m from the nests, as
well as the rec ommendation to place nests
100 m apart (Yamada et al. 1971; Maeta and
Kitamura 197 4). Several other meth ods of
studying dispersal tend to overestimate foraging
distances, but the immunomarking method may
underestimate them, as the number of flowers
visited in a foraging bout increases, the amount
of immunomarking on the bee's body will
decrease. So these bees may visit other flowers
in the study orchard beyond 35–55 m, while no
longer depositing immunomarking powder.
However, this immunomarking dilution effect
needs further investigations.
The slightly lower foraging range we found
compared to the Japanese studies may be due to
the limited foraging time we allowed our bees
during this study compared to the entire bloom
period for the Japanese studies or due to differ-
ences in cherry and apple as a crop. Such short
range foraging of O. cornifrons in cherry orchard
could be due to the presence of more blossoms per
tree than other fruit trees such as apple. In general,
Osmia bees may fly long distances while foraging,
but when sufficient pollen resources/host flowers
are present near their nesting sites (such as in this
study) they may prefer foraging close to their
nesting sites instead of flying longer distances
(Zurbuchen et al. 2010, Radmacher and Strohm
2010). Besides number of blossoms on a tree,
differences between the aforementioned studies
could be due to tree planting/density and bee
shelter location within the orchards.
In previous studies, foraging distance was
estimated by calculating homing ranges, the
maximum distance from the nest that a bee can
find its way home. However, this overestimates
Foraging patterns of Osmia cornifrons in a cherry orchard

foraging range. Greenleaf et al. (2007) reviewed
96 published records for 62 bee species that
evaluated maximum foraging/homing distances,
and found a highly significant positive relation-
ship between the bee body size and various
methods of measuring homing/foraging limits.
Foragingrangeforabee,however,isdependent
on multiple factors such as the spatial and temporal
distribution and availability of resources during its
life cycle. For example, while the homing range of
O. cornuta (Latreille) was estimated at 1,800 m,
when placed in an orchard during peak bloom with
abundant floral resources, most females forage
within 100–200 m of their nests (Vincens and
Bosch 2000). The homing range of the smaller O.
cornifrons (Radoszkowksi) in Japanese apple
orchards was likewise found to be approximately
500 m, but likewise, its effective pollination range
in large orchards was estimated to be only 40 to
60 m from the nesting site (Maeta and Kitamura
1974; Kitamura and Maeta 1969; Guedot et al.
2009). The foraging ranges of several stingless bee
species have likewise been shown to be approxi-
mately 300 m shorter than their homing distance
(van Nieuwstadt and Ruano Iraheta 1996).
The O. cornifrons nest box remained in the
orchard throughout the 7-day bloom period thus
the yield data should have been an estimate of the
full pollination potential of that nest beyond the
30-h test period. Our study found the number of
cherries per limb cross-sectional area to be over
twice as high on trees in the Osmia test orchard
than in a control orchard very similar in age, size,
and management and where equivalent levels of
honey bees were present. While comparing these
two orchard blocks, any minor differences be-
tween blocks should have been minimized by the
horticultural method of relating fruit number to
the scaffold limb cross-sectional area (Schupp and
Greene 2002). This control block was isolated by
distance from habitat containing wild populations
of O. cornifrons and none had been released, so
the effects on fruit number is directly attributable
to the additive effect of having O. cornifrons
supplementing the level provided by honey bees
alone. Similarly, a related species, Osmia lignaria
Say doubled cherry yields in Utah over a 5-year
period over that of A. mellifera Utah (Bosch and
Kemp 2006). They attributed this to shorter
foraging range that forced it to concentrate on fruit
trees within the targeted orchard, a greater fidelity
to fruit trees when foraging, longer foraging period
due to its lower light and temperature thresholds,
and in its greater effectiveness in transferring
pollen (Torchio 1981; Kuhn and Ambrose 1984).
Other factors contributing to higher fruit
density (yield) in the study orchard could be the
abundance of additional native pollinators, but net
collections of bees in both the test and control
orchard during bloom found that while honey bee
numbers were fairly abundant and similar in both
blocks, solitary bees and bumble bees were not
found in the areas sampled of either cherry block.
This study demonstrates an immunomarking
method to mark bees and quantify where they have
been by measuring the residues they leave on
flowers is viable and a powerful tool in under-
standing the foraging behavior and limits of bees.
Our data based on the marking and yield data
suggest that the effective pollination range of 500
O. cornifr ons females from a single nest box is
only about 40 m rather than the 500 m estimate
based on homing range. This translates into an
effective foraging/pollinating area of slightly more
than 0.5 ha for each nest and we would recom-
mend placing nests at a slightly closer range of
80 m apart rather than 100 m apart as recommend-
ed for apple pollination currently in Japan. Future
research into optimizing release rates of female O.
cornifr ons in nests may find the optimal number of
bees needed for commercial pollination of various
fruit crops in the eastern USA. This number may
vary by crop. For example, since tart cherry needs
20–75 % of flowers set for a commercial crop and
has more flowers per tree than apple, which needs
only 2–8 % set, the number of Osmia necessary for
pollination might be much higher in cherry
(Chaplin and Westwood 1980).
ACKNOWLEDGMENTS
The authors sincerely thank the USDA NIFA for a
SCRI grant on sustainable fruit pollination (PEN04398)
and the State Horticultural Association of Pennsylvania
for their financial support of this study. The authors also
thank Kathy Wholaver, Edwin Winzeler, Amanda Ritz,
D.J. Biddinger et al.

Maryann Frazier, and numerous summer interns/assis-
tants for their help in conducting this study, Doug Lott
for the use of the commercial orchard, and several
anonymous reviewers for their constructive comments
on a previous draft of this manuscript.
Technique d’immunomarquage pour déterminer les
modes d’approvisionnement d’Osmia cornifrons et la
production de fruits en résultant, dans un verger de
cerisiers
Technique de marquage / appr ovisionnement / processus
d’auto-marquage / pollinisation/ guêpe maçonne /
pollinisateur
Eine Immuno-Markierungsmethode zur Bestimmung des
Sammelverhaltens von Osmia cornifrons und der daraus
resultierende Fruchtansatz in einer Kirschplantage
Immunomarkierzng / Sammelverhalten / Osmia
cornifrons / Selbst-Markierung / Bestäubung / Bestäuber
REFERENCES
Akey, D.H., Hayes, J.L., Fleischer, S.J. (1991) Use of
elemental markers in the study of arthropod movement
and trophic interactions. Southwest. Entomol Suppl.
14,1–87
Batra, S.W.T. (1979) Osmia cornifrons and Pithitis
smaragdula, two Asian bees introduced into the United
States for crop pollination. Maryland Agricultural
Experiment Station, Special Miscellaneous Publication
1,307–312
Batra, S.W.T. (1998) Hornfaced bees for apple pollina-
tion. Am. Bee J. 138, 364–365
Biddinger , D.J., Mullin, C., Frazier, J., Frazier, M., Ashcraft,
S., Rajotte, E. (201 1a) Assessing the pesticide suscep-
tibility of native bees in Pennsylvania orchards. Penn
Fruit News 91,49–55
Biddinger, D., Rajotte, E., Joshi, N., Ritz, A. (2011b) Wild
bees as alternative pollinators. Fruit Times 30,1–4
Biddinger, D.J., Ngugi, H., Frazier, J., Frazier, M.,
Leslie, T., Donovall, L.R. (2010) Development of
the mason bee, Osmia cornifrons, as an alternative
pollinator to honey bees and as a targeted delivery
system for biological control agents in the manage-
ment of fire blight. Penn Fruit News 90,35–44
Biddinger, D., Frazier, J., Frazier, M., Rajotte, E.,
Donovall, L., Leslie, T. (2009a) Solitary bees as
alternative pollinators in Pennsylvania fruit crops.
Penn Fruit News 89 ,84–94
Biddinger , D.J., Ngugi, H., Frazier, J. (2009b) Development
of the mason bee, Osmia cornifr ons,asatargeted
delivery system for biocontrol agents in the manage-
ment of fire blight. Penn Fruit News 89,95–100
Blackmer, J.L., Hagler, J.R., Simmons, G.S., Canas,
L.A. (2004) Comparative dispersal of Homalodisca
coagulata and Homalodisca liturata (Homoptera:
Cicadellidae). Environ. Entomol. 33,88–99
Blackmer, J.L., Hagler, J.R., Simmons, G.S., Henneberry,
T.J. (2006) Dispersal of Homalodisca vitripennis
(Homoptera: Cicadellidae) from a point release site in
citrus. Environ. Entomol. 35,1617–1625
Bosch, J., Kemp, W.P. (2002) Developing and establishing
bee species as crop pollinators: the example of Osmia
spp. (Hymenoptera: Megachilidae) and fruit trees.
Bull. Entomol. Res. 92,3–16
Bosch, J., Kemp, W.P. (2006) Bee population returns and
cherry yields in an orchard pollinated with Osmia
lignaria (Hymenoptera: Megachilidae). J. Econ.
Entomol.
99,408–413
Bosch, J., Kemp, W., Trostle, G. (2006) Bee population
returns and cherry yields in an orchard pollinated with
Osmia lignaria (Hymenoptera: Megachilidae). J. Econ.
Entomol. 99, 408–413
Bosch, J., Sgolastra, F., Kemp, W.P. (2008) Life cycle
ecophysiology of Osmia mason bees used as crop
pollinators, pp. 83–103. In R. R. James and T. L.
Pitts-Singer [eds.], Bee pollination in agricultural
ecosystems. Oxford University Press, New York.
Buczkowski, G., Bennett, G. (2006) Dispersed central-
place foraging in the polydomous odorous house ant,
Tapinoma sessile as revealed by a protein marker.
Insectes Soc. 53,282–290
Chaplin, M.H., Westwood, M.N. (1980) Relationship of
nutritional factors to fruit set. J. Plant Nutr. 2,477–505
Gathmann, A., Tscharntke, T. (2002) Foraging ranges of
solitary bees. J. Anim. Ecol. 71, 757–764
Greenleaf, S.S., Williams, N.M., Winfree, R., Kremen,
C. (2007) Bee foraging ranges and their relationship
to body size. Oecologia 153, 589–596
Guedot, C., Bosch, J., Kemp, W.P. (2009) Relationship
between body size and homing ability in the genus
Osmia (Hymenoptera: Megachilidae). Ecol. Entomol.
34,158–161
Hagler, J.R. (1997) Field retention of a novel mark-release-
recapture method. Environ. Entomol. 26,1079–8
Hagler, J.R., Jackson, C.G. (1998) An immunomarking
technique for labeling minute parasitoids. Environ.
Entomol. 27,1010–16
Hagler, J.R., Jones, V.P. (2010) A protein-based approach
to mark arthropods for mark-capture type research.
Entomol. Exp. Appl. 135,177–192
Hagler, J.R., Naranjo, S.E. (2004) A multiple ELISA
system for simultaneously monitoring intercrop
movement and feeding activity of mass-released
insect predators. Int. J. Pest Manag. 50, 199–207
Hagler, J.R., Cohen, A.C., Dradley-Dunlop, D., Enriquez,
F.J. (1992) New approach to mark insects for feeding
and dispersal studies. Environ. Entomol. 21,20–25
Foraging patterns of Osmia cornifrons in a cherry orchard

Hagler, J., Mueller, S., Teuber, L.R., Van Deynze, A.,
Martin, J. (2011a) A method for distinctly marking
honey bees, Apis mellifera , originating from multi-
ple apiary locations. J. Insect Sci. 11:143 available
online: insectscience.org/11.143
Hagler, J.R., Mueller, S., T euber , L.R., Machtley, S.A., Van
Deynze, A. (2011b ) Foraging range of honey bees, Apis
mellifera, in alfalfa seed production fields. J. Insect Sci.
1 1:144 available online: insectscience.org/1 1.144
Jones, V.P., Hagler , J.R., Brunner, J.F., Baker , C.C., W ilburn,
T.D. (2006) An inexpensive immunomarking technique
for studying movement patterns of naturally occurring
insect populations. Environ. Entomol. 35, 827–836
Joshi, N. K., Biddinger, D., Rajotte E.G. (2011) A survey of
apple pollination practices, knowledge and attitudes of
fruit grower s in Pennsylvania. 1 0th International
Pollination Symposium, Puebla, Mexico.
Kitamura, T., Maeta, Y. (1969) Studies on the pollination of
apple by Osmia: (III) Preliminary report on the homing
ability of Osmia cornifrons (Radoszkowski) and O.
pedicornis Cockerell. Kontyu 37,83–90
Kuhn, E.D., Ambrose, J.T. (1984) Pollination of ‘Delicious’
apple by megachilid bees of the genus Osmia
(Hymenoptera: Megachilidae). J. Kans. Entomol. Soc.
57,169–18 0
Maccagnani, B., Ladurner, E., Santi, F., Burgio, G. (2003)
Osmia cornuta (Hymenoptera, Megachilidae) as a
pollinator of pear (Pyrus communis): Fruit- and seed-
set. Apidologie 34,207–216
Maccagnani, B., Bazzi, C., Biondi, E., Tesoriero, D.,
Maini, S. (2006) Potential of Osmia cornuta as a
carrier of antagonistic bacteria in biological control
of fire blight: a comparison with Apis mellifera. Acta
Hortic. 704, 379–386
Mader, E., Spivak, M., Evans, E. (2010) Managing
alternative pollinators: a handbook for beekeepers,
growers and conservationists. Sustainable Natural
Resource and Education Program handbook; 11,
NRAES Cooperative Extension, Ithaca, NY. 162 pp.
Maeta, Y. (1990) Utilization of wild bees. Farming Japan
24,13–22
Maeta, Y. (1978) Comparative studies on the biology of the
bees of the genus Osmia of Japan, with special
reference to their managements for pollinations of
crops (Hymenoptera: Megachilidae). T ohoku National
Agricultural Experiment Station. Bulletin No. 57, 221
Maeta, Y., Kitamura, T. (1974) How to manage the Mame-
ko bachi, Osmia cornifrons (Radodoskowski), for
pollination of fruit crops. Co. Ltd., Ask. 28 p
Monzon, V.H., Bosch, J., Retana, J. (2004) Foraging
behavior and pollinating effectiveness of Osmia cornuta
and Apis mellifera (Hymenoptera: Megachilidae,
Apidae) on 'Comice' pear . Apidologie 35,575–58 5
van Nieuwstadt, M.G.L., Ruano Iraheta, C.E. (1996)
Relation between size and foraging range in stingless
bees (Apidae, Meliponinae). Apidologie 27,219–228
Radmacher, S., Strohm, E. (2010) Factors affecting
offspring body size in the solitary bee Osmia bicornis
(Hymenoptera, Megachilidae). Apidologie 41,169–
177
Ritz, A., Biddinger, D., Rajotte, E., Sahli, H., Joshi, N.
(2012) Quantifying the efficacy of native bees for
orchard pollination in Pennsylvania to offset the
increased cost and decreased reliability of honey bees.
Penn Fruit News 92,6–66
Schneider, F. (1962) Dispersal and migration. Ann. Rev.
Entomol. 7, 223–242
Schupp, J.R., Greene, D.W . (2002) Thinning Mcintosh apple
trees with blossom thinners, with and without post-
bloom NAA: A report to the New England Tree Fruit
Growers Research Committee. Fruit Notes 67,9–12
Sekita, N., Watanabe, T., Yamada, M. (1996) Population
ecology of Osmia cornifrons (Hymenoptera,
Megachilidae) in natural habitats. Bulletin of the
Aomori Apple Experiment Station 29,17–36
Sekita, N., Yamada, M. (1993) Use of Osmia cornifrons
(Radoszkowski) for pollination of apples in Aomori
Prefecture, Japan. JARQ Japan Agricultural
Research Quarterly 26, 264–270
Torchio, P.F. (1981) Field experiments with Osmia lignaria
propinqua Cresson as a pollinator in almond orchards:
II, 1976 studies (Hymenoptera: Megachilidae). J.
Kans. Entomol. Soc. 54,824–836
Vincens, N., Bosch, J. (2000) Nest site orientation and
relocation of populations of the orchard pollinator
Osmia cornuta (Hymenoptera: Megachilidae).
Environ. Entomol. 29,69–75
Xu, H.-L., Yang, L.-I., Kwon, Y.J. (1995) Current status
on the utilization of Osmia bees as pollinators of
fruit trees in China (Hymenoptera: Megachilidae).
Korean J. Apiculture 10, 111–116
Yamada, M., Kawashima, K., Aizu, H. (1984) Population
dynamics of the horn faced bee, Osmia cornifrons
Radoszkowski, with a special reference to the popula-
tion management. Bulletin of the Aomori Apple
Experiment Station 21,23–92
Yamada , Y., Oyama, N., Sekita, N., Shirasaki, S.,
Tsugawa, C. (1971) The ecology of the megachilid
bee Osmia cornifrons and its utilization for apple
pollination [In Japanese]. Bulletin of the Aomori
Apple Experiment Station 26,39–77
Zurbuchen, A., Landert, L., Klaiber, J., Müller, A., Hein,
S., Dorn, S. (2010) Maximum foraging ranges in
solitary bees: only few individuals have the capabil-
ity to cover long foraging distances. Biol. Cons.
143, 669–676
D.J. Biddinger et al.