THE EFFECT OF NASONOV BASED DISPENSERS ON HONEY BEE BEHAVIOR AND
ON POLLINATION IN BLUEBERRY, APPLE, AND CHERRY
By
Julie Anna Adams

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Entomology - Master of Science
2014

ABSTRACT
THE EFFECT OF NASONOV BASED DISPENSERS ON HONEY BEE BEHAVIOR AND
ON POLLINATION IN BLUEBERRY, APPLE, AND CHERRY
By
Julie Anna Adams
Honey bee attractants are not novel, but the time is right to continue to investigate their
importance in the growing fruit industry.

Due to pollinator declines, potential honey bee

attractants have been used for the purpose of improving and increasing pollination services by
bees. Polynate™ is a potential product manufactured for this purpose that contains a synthetic
mixture similar to that of the Nasonov pheromone in honey bees. In this thesis, two main
objectives are investigated: 1) to determine the ability of Nasonov based dispensers to increase
fruit set and visitation in blueberries, apples and cherries and 2) to quantify behavioral changes
associated with dispenser application. The first objective was accomplished by assessing the
differences in forager visitation rates and fruit yield in crops treated and untreated with Nasonov
based dispensers. The second objective was accomplished by directly observing the responses of
honey bees in a hoop house when exposed to Nasonov based dispensers on artificial feeders.
Dispensers did not increase fruit set or forager activity in blueberries, apples or cherries. A half
rate of Polynate also did not cause a difference between treated and untreated areas. Given a
choice between two identical feeders, one treated with a dispenser and one not, the honey bees
had no visible preference.

Copyright by
JULIE ANNA ADAMS
2014

I dedicate this thesis to my entire family, especially to my fiancé Jason. Thank you all for your
continuous support throughout my entire career especially during my years as a graduate student.
Your patience and listening ears were greatly appreciated.

iv

ACKNOWLEDGMENTS

Larry Gut, Rufus Isaacs and Fred Dyer are thanked for their assistance, knowledge, and
encouragement. Pete McGhee is thanked for his logistical input and encouragement. Nikki
Rothwell is thanked for her contribution in cherries. Chris Adams is thanked for construction of
the hoop house. Katelind Aho, Annie Kirk, Keith Mason, Dana Blanchard, Tommy Gut, and
Mitch Efaw thanked for their assistance.

v

TABLE OF CONTENTS

LIST OF TABLES ....................................................................................................................... viii
LIST OF FIGURES .........................................................................................................................x
CHAPTER 1 Honey Bees, Pollination and the History of Attractants ...........................................1
The Importance of Pollination ................................................................................................1
Measures of Pollination and Bee Efficiency ..................................................................3
Honey Bees and Pollination ....................................................................................................4
Biology...........................................................................................................................4
Colony Health ................................................................................................................5
Role as Pollinators .........................................................................................................6
Honey Bee Visual and Olfactory Cues ...................................................................................9
Scent Taste and Sight .....................................................................................................9
Pheromones and Attractants ........................................................................................10
The Nasonov Gland .....................................................................................................10
The Components of Nasonov Pheromone ...................................................................11
The Functions of Nasonov Pheromone ........................................................................12
Queen Mandibular Pheromone ....................................................................................13
Pollination Enhancement ......................................................................................................14
Nectar Guides...............................................................................................................14
Scent Guides ................................................................................................................15
Project Aims and Objectives.................................................................................................19
CHAPTER 2 Dispensing Components of the Nasonov Gland to Enhance Bee Visitation and
Pollination in Blueberry, Apple and Cherry .........................................................................21
Introduction ...........................................................................................................................21
Materials and Methods ..........................................................................................................25
Blueberry......................................................................................................................26
Apple ............................................................................................................................29
Cherry ..........................................................................................................................30
Stastical Analysis .........................................................................................................31
Results ...................................................................................................................................31
Blueberry......................................................................................................................31
Apple ............................................................................................................................35
Cherry ..........................................................................................................................36
Discussion .............................................................................................................................38
CHAPTER 3 The Response of Honey Bees Reared in a Hoop House to Polynate and SPLAT...43
Introduction ...........................................................................................................................43
Materials and Methods ..........................................................................................................46
Nuclear Colony (Nuc) Rearing ....................................................................................46
vi

Artificial Feeders and Sucrose Solutions .....................................................................47
Experimental Design and Protocols .............................................................................48
Stastical Analysis .........................................................................................................50
Results ...................................................................................................................................51
Foraging Intensity ........................................................................................................52
Feeding Duration .........................................................................................................52
Discussion .............................................................................................................................53
CHAPTER 4 The Use of Polynate and SPLAT in Agriculture ............................................56
APPENDICES ......................................................................................................................60
APPENDIX 1 ...............................................................................................................61
APPENDIX 2 ...............................................................................................................62
BIBLIOGRAPHY.................................................................................................................66

vii

LIST OF TABLES

Table 1. The mean number of blueberry fruit per flower ± SEM at three distances from honey
bee hives in untreated and Polynate treated fields during 2012................................32
Table 2. The mean number of blueberry fruit per flower ± SEM in untreated and half or full
Polynate treated fields during 2013 ..........................................................................32
Table 3. The mean count of honey bees, native bees and other pollinators per minute ± SEM at
three distances from honey bee hives in untreated and Polynate treated fields during 2012
...................................................................................................................................33
Table 4. The mean count of honey bees, bumble bees, other bees and flies in blueberry fields per
minute ± SEM during 2013.......................................................................................33
Table 5. The mean mass (g) ± SEM of blueberries in untreated and full or half Polynate treated
fields during 2012 and 2013 .....................................................................................34
Table 6. The mean mature seed count ± SEM of blueberries in untreated and Polynate treated
fields during 2012 .....................................................................................................34
Table 7. The mean number of apple fruit per flower ± SEM in untreated and Polynate treated
orchards during 2013 ................................................................................................35
Table 8. The mean count of honey bees, native bees and other pollinators per minute ± SEM at
three distances from honey bee hives in untreated and Polynate or SPLAT treated orchards
during 2012 ...............................................................................................................35
Table 9. The mean count of honey bees, bumble bees, other bees and flies per minute ± SEM in
untreated or Polynate treated orchards during 2013 .................................................36
Table 10. The mean number of cherry fruit per flower ± SEM at three distances from honey bee
hives in untreated and Polynate treated orchards during 2012 .................................37
Table 11. The mean number of cherry fruit per flower ± SEM in untreated and Polynate or
SPLAT treated orchards during 2013 .......................................................................37
Table 12. The mean count of honey bees, native bees and other pollinators per minute ± SEM at
three distances from honey bee hives in untreated or Polynate treated orchards during 2012
...................................................................................................................................38
Table 13. The mean count of honey bees, bumble bees, other bees and flies per minute ± SEM in
untreated and Polynate or SPLAT treated orchards during 2013 .............................38

viii

Table 14. The mean count of honey bee foragers per minute and the mean time spent feeding ±
SEM for each treatment. ...........................................................................................52
Table 15. The organization of field sites, treatment plots and sampling in 2012 ..............62
Table 16. The organization of field sites, treatment plots and sampling in 2013 ..............62

ix

LIST OF FIGURES

Figure 1. A diagram of the dorsal side of a honey bees abdomen showing the
Nasonov gland (McIndoo 1914) ...............................................................................11
Figure 2. Example of plot and sampling location arrangement within each block
for field experiments in 2012 ....................................................................................63
Figure 3. Example of plot and sampling location arrangement within each block
for field experiments in 2013 ....................................................................................63
Figure 4. A single SPLAT dollop applied in cherry .........................................................64
Figure 5. A single Polynate dispenser applied in blueberry ..............................................64
Figure 6. The hoop house used in my experiments without the end partitions
(pictured in figure 8) .................................................................................................65
Figure 7. An experimental feeder during a trial .................................................................65
Figure 8. An experimenter videotaping one experimental feeder and the set-up of
the nuc and stock feeder............................................................................................65

x

CHAPTER 1
Honey Bees, Pollination and the History of Attractants
The Importance of Pollination
Most of our food comes from cereals or grains that rely only on gravity or wind to
disperse pollen.

However, about one third of our food comes from plants that need bee

pollination to produce fruit (Free 1970; McGregor 1976). This includes foods like dairy and
meat, which are reliant on plant products that are pollinated by bees. It was estimated that honey
bees could bring as much as 1.6-8.3 billion dollars to agriculture in the United States (Southwick
and Southwick Jr. 1992). While it is true that mankind could probably survive without bee
pollination, many luxury foods that we take for granted such as ice cream or watermelon, would
no longer be available to even the richest of people if bees did not pollinate plants. Around 130
crops grown in the United States require pollination by bees. Thus, much of the specialty crop
agricultural economy in the United States is dependent on bees (McGregor 1976; Delaplane and
Mayer 2000). Some time ago, it was estimated that the bee pollination in the United States was
worth approximately 9 billion dollars. Honey bees specifically were estimated to provide 5.7
billion dollars and non-managed bees approximately 1.6 billion dollars in pollination services
(Robinson, Nowogrodzki, and Morse 1989). In 2000, honey bees brought in $14.6 billion
through the increase of yield and produce quality of crops (Morse and Calderone 2000). It is
reasonable to assume that these numbers would be higher now in 2014.
Crops need varying levels of pollination due to factors such as self-fertility,
parthenocarpy, and plant diversity. Some crops and varieties are self-infertile, meaning that the
ovum cannot be fertilized by pollen from the same plant. In these cases, it is essential for a bee
to visit multiple flowers to facilitate pollen transfer. Parthenocarpy occurs when an ovule is not
fertilized and a seedless fruit is produced. Growers are encouraged to select cultivars that exhibit
1

parthenocarpic traits so that their reliance on pollination is less (Allsopp, de Lange, and
Veldtman 2008). Pollination requirements also differ among crops. For example, apples have a
king bloom, one blossom that produces a fruit that is bigger and better quality than the others. It
is more important for this king bloom to be pollinated than the other blossoms. Blueberries have
flowers that are oriented downward which makes it difficult for some pollinators to collect pollen
from them.

The ovules of cherries start to deteriorate before the flowers open.

Thus,

fertilization must occur shortly after bloom for fruit to set (Delaplane and Mayer 2000).
It is important to note that bees can pollinate certain crops more efficiently than others.
A bee’s ability to pollinate depends on factors such as flower shape, and the orientation of the
stamen and stigma (Delaplane and Mayer 2000). There are many species of bee that provide
pollination services.

However, not all of these bees are at sufficient numbers to sustain

agriculture. For example, bumble bees are extremely efficient pollinators, but are harder to
manage commercially than honey bees. Unfortunately, the protocols for managing other species
of bees are, in most cases, proprietary or not readily accessible. Although honey bees may not be
the most efficient pollinators, they are by far the easiest to manage on a commercial scale
(Delaplane and Mayer 2000).
Of crops that are produced globally, 13 are dependent on pollination for production while
57 are highly to moderately dependent and 28 are only slightly or not dependent (Klein et al.
2007). Pollination of apple, cherry, and blueberry are the primary focus of this thesis. Out of
these crops, apple is most dependent on bee pollination to produce fruit, followed by cherry then
blueberry (Robinson, Nowogrodzki, and Morse 1989; Southwick and Southwick Jr. 1992;
Williams 1994).

2

Measures of Pollination and Bee Efficiency
Since better pollination results in higher fruit quality and quantity, it is important to
determine if we can produce a method to measure pollination and bee efficiency. It can be
challenging to quantify the efficiency of bees in regards to pollination because pollinators have
particular behaviors that can affect pollen transfer. For example, the total amount of time a bee
spends foraging, the number of flowers visited per unit time, and the efficiency of pollen
deposition all affect pollination. It is generally assumed that the longer a bee spends visiting a
flower; the more time she spends collecting pollen. This could result in an increase in the
amount of pollen being transferred from flower to flower as well which may result in increased
fruit set. However, it may be detrimental to the level of pollination achieved if different flower
species are visited during a single foraging trip.
It is even harder to quantify the amount of pollination that a crop is receiving. One way
is to look at the number of seeds and the quality of fruit. Bee pollination significantly improves
the quality of fruit in apple and blueberries (Dulta and Verma 1987; MacKenzie 1997). When a
flower is pollinated, its seeds are fertilized; quite simply, the more seeds the more pollination
that occurred. The number of seeds influences fruit quality. For example, a fruit that develops
from a flower that has been poorly pollinated is usually misshapen and undersized (Delaplane
and Mayer 2000). Another way is to count the number of blossoms during bloom versus the
number of fruit that develop from those blossoms to determine fruit set. In crops without
parthenocarpy, the greater the level of fruit set, the more pollination occurred. In other crops, the
size of the fruit is proportional to the level of pollen deposited.

3

Honey Bees and Pollination
Biology
The European honey bee, Apis mellifera L. (Hymenoptera: Apidae), is not native to the
United States. As the name suggests, it is native to Europe as well as parts of Africa and the
Middle East. Honey bees were brought to the United States (and many other countries) by
European settlers in the 1700’s where they flourished for years due to their high capacity to adapt
(Delaplane and Mayer 2000; vanEngelsdorp and Meixner 2010). Honey bees are perhaps one of
the most studied social insects in the world. There is no question about the numerous services
that they provide us, from pollination of crops to production of wax and honey. They are more
easily managed than other species of bee for our purposes because of their hive structure.
A honey bee hive consists of several castes, the queen, the drones and the workers. The
majority of honey bees that we see are the workers, since they are responsible for collecting
resources as well as maintaining the hive through various tasks.

The queen is the only

reproductive female, though in dire need the workers also have the ability to develop mature
sexual organs and lay eggs. The drones are the only males in the hive and have one purpose; to
mate with queens from other hives.
This distribution of responsibilities makes the worker bees the most important caste from
a pollination standpoint. While the queen regulates much of the activity of the hive through
pheromones, the workers alone are responsible for the decisions that they make outside of the
hive while foraging for resources. There are other factors within the hive that have an influence
on foraging behavior such as quantity of pollen and nectar, presence of brood and egg laying
(Jay 1986). However, since on-site visual and chemical cues are only available for the foragers
to interpret, they are the most influential on foraging based decisions. These cues include nectar

4

quality, distance traveled and floral patterns (Dyer 2002).

Honey bees use memories of

landmarks and the orientation of the sun to locate food sources. The information about food
source location is relayed in the hive by means of the waggle dance (von Frisch 1993), but the
actual cues are only experienced by individual workers.

Colony Health
Declines in honey bee numbers have not only been occurring in the United States, but
worldwide (Ghazoul 2005; Aizen et al. 2008). Overwintering losses on average in the United
States were 30.6% from 2012-2013 which is consistent with previous years (Steinhauer et al.
2014). Due to the recent decline in honey bees, there has been a greater effort to preserve them
and the pollination services that they provide. These declines are not specific to honey bees, but
other bee species as well. Declines of other pollinators however, are intermittent and dependent
on species and location (Ghazoul 2005). In some regions, such as south west China, wild bee
loss is so severe that apple and pear crops must now be hand pollinated (Partap and Ya 2012).
Even though the bee losses in the United States have yet to reach a level that seriously effects
crop yield, there is evidence to suggest that this will soon occur because of an increase in the use
crops that are dependent on bee pollination (Ghazoul 2005; Klein et al. 2007; Aizen et al. 2008).
When managing honey bee colonies for crop pollination, it is important to maintain
colony health. To accomplish this, honey bees need a constant supply of nectar and pollen
during the spring, summer and fall seasons in addition to treatment for diseases. However, crops
do not bloom all year long; most of them only have a short period of time in which they bloom.
Placing a hive in the middle of one crop would not be sufficient to support the colony throughout
their active season. This means that either multiple varieties or crops with overlapping bloom

5

periods must be available to the colonies, or the colonies must be moved to other locations with
blooming flowers once the initial crop’s bloom is over. Currently, beekeepers are paid to do just
that; ship their bees across the country to aid in pollination. Inevitably, whenever a colony is
moved, even at night, it will suffer the loss of many foragers (Delaplane and Mayer 2000). In
2000, more than 2 million out of 2.9 million colonies in the United States were shipped across
the country for the pollination of various crops as well as for harvesting honey and wax (Morse
and Calderone 2000). At this rate it would not be a surprise if there were a correlation between
how much a hive has been moved throughout the summer, and its likelihood to survive to the
next season (Stokstad 2007).
Over the years honey bees have been plagued by pests such as Varroa destructor,
tracheal mites, Nosema, and now a disease with an unknown cause called Colony Collapse
Disorder or CCD (Tew and Ferree 1998; Delaplane and Mayer 2000). Many blame this issue on
pesticides, the arrival of new pests, poor nutrition, and even a small gene pool. However, the
cause of CCD is most likely not due to any one problem but the synergistic effects of all that
plagues honey bees (vanEngelsdorp et al. 2009).

Role as Pollinators
Honey bees are generalist foragers, meaning that they will forage on a wide range of
flowers if they are an adequate source of pollen and nectar. This makes them very versatile
because they can pollinate a variety of crops. However, this also means that they tend to
pollinate other non-target flowers instead of crop flowers. von Frisch (1966) noted that once an
individual honey bee has found a type of flower with good resources, she will consistently revisit the same type of flower over others. His reasoning was that because honey bees must learn

6

how to locate the nectar in a particular flower, it is much easier for them to continue to visit the
same type of flower instead of learning to collect from another (von Frisch 1966).
Undoubtedly, honey bees that are foraging for pollen are better crop pollinators than bees
that are foraging for nectar because nectar collectors have a habit of bypassing the anthers and
stigma of flowers to get to the nectar. Without the bee-to-anther contact, pollen is not transferred
from plant to plant as the bee forages. Even if the bee is covered with pollen, they must also
make contact with flower stigmas for pollination to occur (Jay 1986).
Distance is also a huge factor for honey bee foragers. There has been a debate on the
precise distances that honey bees will travel to forage but the general consensus is that honey
bees prefer to forage closer to their hive than travel longer distances. However, honey bees have
been found to prefer foraging more than three to four miles (Beekman and Ratnieks 2000).
Even though honey bees provide excellent pollination services to a large variety of crops,
their generalist nature can cause them to be less efficient than other bees. For example, Orchard
mason bees, Osmia lignaria Say (Hymenoptera: Megachilidae), are more efficient in pollinating
apple blossoms because of their tendency to land directly on the stigma and anthers of the
flowers instead of the petals like honey bees. Bumble bees, Bombus spp. (Hymenoptera:
Apidae), may also be more efficient pollinators of apple and blueberry (Delaplane and Mayer
2000; Javorek, Mackenzie, and Vander Kloet 2002). Honey bees will sometimes collect nectar
without brushing against the sexual parts of the flower; this results in pollen neither being carried
away nor deposited. Bumble bees, in contrast, almost always make contact with the stigma and
anthers; therefore a higher percentage of flowers that are visited are also pollinated (Delaplane
and Mayer 2000). Blueberry flowers are more efficiently pollinated by bees like bumble bees
that sonicate flowers through buzzing (Javorek, Mackenzie, and Vander Kloet 2002). The

7

vibrations from their buzzing shakes the blueberry pollen loose which will fall onto the visiting
bee (Delaplane and Mayer 2000).

However, bumble bees tend to avoid foraging on less

rewarding flowers (Townsend-Mehler, Dyer, and Maida 2011), and unfortunately, most crop
flowers are neither as rewarding or attractive as other flowers in the surrounding area (Free
1968b). This suggests that bumble bees may be more attracted to the ground cover flowers than
honey bees.
It is no secret that native bees are more efficient pollinators for specific crops than honey
bees. This however, does not mean that honey bees are not capable of adequately pollinating
crops. In fact, crop yield would decrease about 30-50% without the aid of honey bees in crops
such as almond, apple, and cherry (Southwick and Southwick Jr. 1992). In addition, managing
native bee colonies at the scale required by our agricultural needs may not be feasible because
they are difficult to manage. Conversely, the art of honey bee keeping has been practiced for
centuries and is not part of some trade secret like some rearing practices of native bees. The
ability to move honey bee colonies long distances makes them especially versatile as individual
hives can be moved among crops as bloom occurs. It would be more beneficial to use native
bees such as bumble bees, orchard mason bees, soil-nesting bees and leafcutting bees, as a
supplement when honey bee numbers are low or when they are inefficient at pollinating some
varieties (Delaplane and Mayer 2000).

Since it is unrealistic to completely rely on other

pollinators and it is becoming harder for growers to purchase additional honey bee hives, it
would be advantageous to discover a way to increase the efficiency of honey bees.

8

Honey Bee Visual and Olfactory Cues
Scent Taste and Sight
The senses of sight and smell are extremely important to foragers and honey bees in
general (von Frisch 1950; Winston 1987). Foragers need both senses to locate pollen and nectar
resources. Sight is most important during long-range exploration. However, honey bees can see
no more than four colors and can’t make out the exact shape of flowers which makes their sense
of smell paramount during close range exploration (von Frisch 1966; Winston 1987). Bees
trained to a scented box of a specific color containing food tended to investigate other boxes of
the same color, but only entered boxes that emitted the scent; color or not (von Frisch 1971).
This seems to suggest that honey bees use sight to find the general location of food, but use scent
to find the more exact location.
Training bees based on sent cues has been well documented. Artificial feeders that have
an attractive sent associated with them are visited with a higher frequency than sources that are
not (von Frisch 1966). One explanation is that the scent makes it easier for the honey bee to
pinpoint the exact location of the food source.
Contrary to what one may believe, honey bees actually have a poor sense of taste. A 2%
sugar solution, which would taste sweet to humans, is no more attractive to a honey bee than a
solution of pure water. This doesn’t mean that taste is not important in terms of pollination. On
the contrary, it prevents bees from wasting time visiting less-rewarding sugar sources (von Frisch
1966).
For a flower to be the most attractive, it must have strong visual, olfactory, and
somatosensory cues. This is because foragers take cues from all of these senses to ensure that
they are foraging on the most rewarding of flowers.

9

Pheromones and Attractants
Because honey bees are social insects, they require a complex means of communicating
with each other. Pheromones are chemicals released by an individual as a signal to a member of
that individual’s same species (Nordlund and Lewis 1976). Attractants can be pheromones, but
not always. Attractants are materials that draw an organism to it either for the organism’s benefit
or demise. In the case of a bee, attractants include numerous types of scents from flowers, nectar
scents, pheromones, and many others. Attractants may or may not have an effect on the target
organism, but pheromones always elicit either a positive or negative response (Schmidt 1999).
Honey bees use many types of pheromones such as Queen Mandibular Pheromone
(QMP), Nasonov pheromone, mating pheromones, brood pheromones alarm pheromones and trail
pheromones to name a few (Free 1987; Pankiw 2004). All of these chemicals are emitted from
one bee, or even sometimes the whole hive, to communicate with another. Nasonov pheromone
and QMP have been studied most extensively for the purpose of increasing pollination services in
crops.

The Nasonov Gland
The Nasonov or scent gland is located under a honey bees sixth intertergal membrane; inbetween tergas six and seven (McIndoo 1914; Snodgrass 1984). A honey bee releases Nasonov
pheromone when it lifts up its abdomen and exposes a part of the Nasonov gland (Wells et al.
1993). She may also fan her wings so that the pheromone is dispersed. The rest of the gland that
is not exposed is invaginated, forming an internal canal (McIndoo 1914). To the naked eye, the
gland appears to be a small white strip just above the tip of the abdomen.

10

Figure 1. A diagram of the dorsal side of a honey bees abdomen showing the Nasonov gland
(McIndoo 1914)

Generally, honey bees will only release Nasonov pheromone when they are collecting
water and not nectar or pollen. This is because nectar and pollen have their own natural scents to
attract bees and water is usually void of any distinct scent (Free and Williams 1970). However,
since honey bee dances are not precise and recruits are not able to follow exact directions (von
Frisch 1967), pheromone may be used to draw them in. Honey bees that are visiting a food
source that is already marked with Nasonov pheromone will not expose their Nasonov glands to
discharge extra pheromone (Free 1968a). To do so would be a waste of energy. However,
honey bees do expose their Nasonov glands when Nasonov pheromone is present at their hive
entrance (Ferguson and Free 1981).

The Components of Nasonov Pheromone
The components of the honey bee Nasonov pheromone are geraniol (100 parts), nerolic
acid (75 parts), geranic acid (12 parts) E-citral (1 part), Z-citral (1 part), E,E-farnesol (50 parts)
and nerol (1 part) (Boch and Shearer 1962; Boch and Shearer 1964; Shearer and Boch 1966;
Pickett et al. 1980; Free et al. 1984). E,E-farnesol is a sesquiterpenoid while the other six
11

components are monoterpenes. Honey bees will expose their Nasonov glands at their hive
entrance when (E)–citral, geraniol, nerolic acid and geranic acid are individually applied to the
front of the hive. However, (E, E) – farnesol and nerol did not elicit the same response, and
actually decreased the honey bees response to the previous mentioned chemicals (Free, Pickett,
and Ferguson 1983).

The Functions of Nasonov Pheromone
Nasonov pheromone is thought to be the most significant pheromone that bees use for
orientation and organization during swarming (Morse and Boch 1971; Jay 1986). It is also
thought to be the main pheromone that honey bees use during clustering, and may help a
disorientated queen-less swarm re-locate their queen (Morse and Boch 1971; Mautz, Boch, and
Morse 1972).
Wild honeybee swarms found artificial nest sites set up by scientists more often when
accompanied by synthetic Nasonov pheromone when compared to nesting sites without the
pheromone (Schmidt 1994). Again in 1999, Schmidt and colleagues tested the hypothesis that
Nasonov pheromone was indeed a pheromone that is used in part to orient honey bees seeking
nest sites and not is just a simple odor. They found that swarms were almost five times more
likely to choose a nest site including a synthetic Nasonov mixture than a nest site that that had
none. They also tested four other odors, linalool, skatole, clove oil and wax moth pheromone, all
of which attracted far fewer swarms than the nest sites with Nasonov pheromone.

They

concluded that Nasonov is indeed a pheromone that is essential to the nest-seeking behavior
(Schmidt 1999).

12

von Frisch (1923) conducted a study which rejected his hypothesis that the Nasonov
pheromone is attractive to honey bees during foraging. In 1947, he conducted a similar study
and found the exact opposite, so he changed his view and maintained that the Nasonov
pheromone was still important in foraging (Wenner and Wells 1990; Wells et al. 1993). In an
additional experiment in 1993, Wells and colleagues were unable to conclude that foraging bees
are attracted to Nasonov pheromone or its components. However, they also stated that it is
useful to train bees to respond to different motivators (Wells et al. 1993).
There have obviously been some mixed results and opinions regarding the usefulness of
the Nasonov pheromone, its components, and synthetic reproductions in attracting honey bees.
Therefore, more research should be done to clarify their effects.

Queen Mandibular Pheromone
QMP is also a honey bee pheromone that appears to have many functions. It is not the
only pheromone that the queen possesses, but it is by far the most important (Strauss et al. 2008).
Unlike the Nasonov pheromone, QMP mainly functions in the hive level and honey bees rarely
encounter it outside of the hive. It stimulates young workers to perform retinue behaviors such
as feeding and grooming the queen. It also directs workers in colony maintenance and facilitated
comb building (Beggs et al. 2007; Ledoux et al. 2001). Other functions of QMP include
decreasing the chance of another queen being reared, delaying the age of first foraging,
suppressing juvenile hormone, and inhibiting the development of worker ovaries (Pankiw et al.
1998; Hoover et al. 2003; Strauss et al. 2008).
QMP has been hypothesized to be useful to attract honey bee foragers to crops because it
is used to orientate workers to the queen during swarming.

13

However, while testing the

attractiveness of nesting sites to honey bees, researchers demonstrated that the presence of
Nasonov pheromone increased site attractiveness more than the presence of QMP. However, a
combination of Nasonov and QMP attracted the greatest number of swarms (Schmidt, Slessor,
and Winston 1993). The queen scent alone may not be enough to move a swarm and the
combined scents of queens and workers are very attractive to a queen-less swarm (Morse and
Boch 1971).

Pollination Enhancement:
Due to the recent declines in pollinators and the concern that many crops are no longer
being adequately pollinated, there has be an increase in conservation efforts (Ghazoul 2005)
which have also lead to the development of many products geared to increase pollination in
crops. However, this phenomenon is not new; pollinator declines in the past (vanEngelsdorp and
Meixner 2010) also spurred an increase interest in pollination enhancement.

Nectar Guides
Nectar guides are honey bee attractants based on sugar content and smell to a lesser
extent. As mentioned before, honey bees can only recognize sugar solutions that are relatively
strong. This means that they are only attracted to the most rewarding resources (von Frisch
1966). Numerous studies have concluded that spraying sugar water on flowers does not increase
pollination (Free 1965; Jay 1986). While they may increase pollinator activity, they decrease the
number of bees actually pollinating flowers. This is because bees are more likely to collect the
sugar syrup on the leaves rather than the nectar from flowers (Free 1965; Mayer and Johansen
1982).

14

Beelure® and Beeline® are two products that have been used in the past as nectar guides.
Beelure is a colored corn syrup containing strawberry flavoring. Before being sprayed onto
target crops, it is diluted with water. It was found to be ineffective aiding pollination in apple
(Rajotte and Fell 1982). Later in 1984, Tew and Ferree found that Beelure increased the number
of foragers in apple, though not significantly. Instead of visiting flowers, however, bees were
often observed collecting syrup from the tree leaves. This explains why scientists did not also
observe a higher fruit yield (Tew and Ferree 1984).
Beeline is a wettable powder containing lactose, fat, proteins, sucrose, pollens and
vitamins. It was intended to be a food supplement for bees as well as a nectar guide. In red
clover, an increased number of honey bee foragers and a higher yield were observed in untreated
plots compared to plots that were treated with Beeline. However, the researchers observed a
greater number of foragers in the untreated plots even before Beeline was applied. It was
therefore determined that Beeline had no effect on foragers or yield in red clover instead of
having a negative effect (Burgett and Fisher 1979). Beeline also did not increase pollinator
activity in apple or pears. This is consistent with further research (Mayer and Johansen 1982).
Beeline also had no effect on pollination through number of visits or yield in either cucumber or
watermelon (Schultheis et al. 1994).

Scent Guides
Scent guides are attractive to honey bees solely based on smell. Placing bouquets of
flowers in orchards will increase pollinator activity but decrease pollination (Waller 1970).
Again, this is because the bees are more interested in visiting the flowers in the bouquets and
collecting their pollen, than the pollen of the target crop. Scent guides that do not include a

15

pheromone component and have been studied include anise oil, lemon grass extract, and Beehere®.
Sugar added to anise oil did not increase pollinator activity in apples or pears (Mayer and
Johansen 1982). The extract from lemon grass when diluted with water was found to be
attractive to honey bees in sweet citrus (Malerbo-Souza, Nogueira-Couto, and Couto 2004).
Bee-Here when diluted with water and not sugar syrup was found to attract honey bees to sweet
orange orchards (Malerbo-Souza, Nogueira-Couto, and Couto 2004).
Other scent guides include Pollenaid®, Pollenaid-D®, Bee-Scent®, Fruit Boost® and
Polynate™. Pollenaid is part Nasonov pheromone components plus some sugars and attractive
oil (Dag 2011). Pollenaid-D is a source of nutrients consisting of iron, nitrogen and gluconic
acid. It did not increase pollinator activity in apples or pears (Mayer and Johansen 1982). BeeScent was found to have varying percents increase of both honey bee foraging and fruit set in
apple cherry and pear (Mayer, Lunden, and Britt 1989). It appeared to increase foraging seed
content and fruit yield in watermelon (Elmstrom and Maynard 1991). However, a few years
later, it was found to have no effect on number of visits or yield in either cucumber or
watermelon (Schultheis et al. 1994). It also was reported to increase honey bee foraging on
apple, but not fruit set (Tew and Ferree 1998).
The use of QMP and its components for the purpose of increasing pollination has had
mixed results. Spraying QMP in blueberry and cranberry did not increase the number of flower
visits, but foragers did spend a longer time in direct contact with flowers (Higo, Winston, and
Slessor 1995). Fruit Boost®, which is composed of synthetic QMP components, did not increase
pollinator frequency or fruit set in pear or sweet cherry. However, a 7% increase in pear weight
was observed that led to a $400 per hectare increase (Naumann et al. 1994). Fruit Boost also did

16

not increase honey bee visitation or fruit set in watermelon (Ellis and Delaplane 2009). Honey
bees do not normally encounter queens outside of the hive (except in swarms). Thus, products
containing QMP are most likely not going to attract foragers to crops.
Polynate, the main subject of this thesis, is a Nasonov based pollination enhancement
product. It is hand-applied prior to bloom at a rate of 500-100 dispensers per acre. It has shown
promise in the past to increase the pollination services honey bees provide (Gut and Isaacs, 2011
unpublished data). The Polynate dispenser is loaded with Z-citral (neral), E-citral (geranial),
nerol, geraniol, geranic acid and anethole (whereas Nasonov pheromone contains Z-citral, Ecitral, nerol, geraniol, nerolic acid, geranic acid and E,E-farnesol), components that have been
documented as attractants for honey bees and other insects.

Below, I review the relative

attractiveness of each of the components individually to honey bees.
Geraniol in Japanese beetle traps, when combined with anethole and eugenol, is attractive
to bumble bees (Hamilton, Schwartz, and Townshend 1970). Similar results were found again
with a trap containing only a geraniol lure (Ladd, Beroza, and McGovern 1974). However, traps
containing a geraniol lure did not catch significantly more honey bees than un-baited traps
(Ladd, Beroza, and McGovern 1974; Allsopp and Cherry 1991). Geraniol also did not increase
pollinator activity in apples or pears (Mayer and Johansen 1982). However, when ten focal bees
were given a choice between feeding on sucrose solutions marked with several scents (anise,
bay, citral, geraniol and nerol) some bees did prefer the geraniol marked solution (Wells et al.
1993).
Traps containing a citral lure caught a similar number of honey bees as un-baited traps
(Allsopp and Cherry 1991). When ten focal bees were given a choice between feeding on
sucrose solutions marked with several scents (anise, bay, citral, geraniol and nerol), no bees

17

visited the citral marked solution (Wells et al. 1993). However, citral was found to be attractive
to honey bees in sweet orange when diluted with water (Malerbo-Souza, Nogueira-Couto, and
Couto 2004). (E)-citral has specifically been shown to elicit Nasonov gland exposure in honey
bees. The other isomer (Z)-citral also seems to be important in eliciting honey bees to expose
their Nasonov glands at the hive entrance. There is evidence that these two isomers have the
same level of attractiveness to honey bees (Free, Pickett, and Ferguson 1983).
When ten focal bees were given a choice between feeding on sucrose solutions marked
with several scents (anise, bay, citral, geraniol and nerol) no bees significantly preferred the
nerol marked solution. However, the feeder did receive some bee visits (Wells et al. 1993).
Geranic acid has been shown to elicit Nasonov gland exposure in honey bees (Ferguson
and Free 1981; Free, Pickett, and Ferguson 1983; Free et al. 1984).
Anethole, a chemical used in Japanese beetle traps, was found to be attractive to honey
bees. Anethole is not a honey bee pheromone but is thought to act as a kairomone to them since
it is found in anise oil (van Praagh and von der Ohe 1983). When studying a Japanese beetle
trap, scientists observed many honey bees and bumble bees were also being caught (Hamilton,
Schwartz, and Townshend 1970). Traps containing an anethole lure caught significantly more
honey bees and bumble bees than un-baited traps (Ladd, Beroza, and McGovern 1974; Allsopp
and Cherry 1991).
There is evidence that combinations of these potential attractants may work better to
increase pollination than just one attractant alone. However, the presence of a scent will not
stimulate bees to forage and recruit without the presence of a reward. For example, citral,
geraniol and anise were found only to be attractive to bees (alone and in various combinations
with each other) when accompanied by a sucrose solution. Geraniol was found to be more

18

attractive than citral only when combined with sucrose. Without sucrose, citral was the more
potent attractant (Waller 1970).

Artificial hives with lures containing a mixture of neral,

geraniol, nerolic acid and geranic acid were effective in catching Africanized honey bee swarms
(Schmidt and Thoenes 1987).

Project Aims and Objectives
Because honey bees are so beneficial to agriculture, it is important to maintain and
enhance the pollination services that they provide. This is becoming harder as the years pass due
to their declining health and the increased demand for pollination. Development of a pollination
attractant, such as Polynate, could provide improved pollination or more efficient pollination of
hard-to pollinate cultivars.
The many products that have been developed and tested to date have generated some
success, but results have been variable. Even when increased crop quality and yield have been
obtained, the mechanism by which it is achieved is still unknown. Furthermore, any positive
effects have been short-lived, lasting for only a day or two. In Michigan, sprayable products
would be expected to wash off readily during spring rainfall. As early as 1986, it was proposed
that the use of long-release dispensers, which protect the chemical from rain and degradation on
leaf surfaces, would be beneficial for field deployment (Jay 1986).
Polynate is a reservoir-type device that contains the same chemicals that honey bees use
to recruit each other. This plastic dispenser cannot be washed off and protects the components
from degradation. However, it is possible to impregnate the same chemical composition into a
wax paste (SPLAT: Specialized Pheromone and Lure Application Technology). The overall aim
of this project is to test if Polynate and SPLAT can be applied in blueberry, apple or cherry by

19

growers to increase pollination, yield and profit. The specific objectives are: 1) to determine the
attractiveness of Polynate and SPLAT and their chemical components to honey bees and 2) to
determine whether deployment of Polynate and SPLAT dispensers loaded with bee pheromone
and other attractants increases bee visitation and pollination of highbush blueberry, apple, and
cherry.

20

CHAPTER 2
Dispensing Components of the Nasonov Gland to Enhance Bee Visitation and
Pollination in Blueberry, Apple and Cherry
Introduction
Honey bees, Apis mellifera L. (Hymenoptera: Apidae), provide pollination services that
are extremely important to agriculture. In North America, apples and cherries are primarily
pollinated by honey bees. Blueberries have other major pollinators, but honey bees are still
important in producing a higher yield of berries. Due to the recent decline in honey bee
populations because of colony collapse disorder, it is becoming difficult for beekeepers to supply
enough bees to pollinate these fruit crops, as well as the many other pollination-dependent crops
in the United States (McGregor 1976; Delaplane and Mayer 2000). This has resulted in it
becoming harder to rent hives, as well as more costly. On average, renting a honeybee hive cost
about $70.85 in 1999 in the Pacific North West (Burgett 2011). The California Beekeeper’s
Association estimates that the average cost of renting honey bee hives for almond pollination in
2014 will be as much as $170 per hive (“CSBA Pollination Survey Results” 2013). It is
estimated that over 2 million honey bee colonies are required for crop pollination each year
(Morse and Calderone 2000). A possible option for decreasing the demand and therefore the
cost of renting hives would be to maximize the pollination services that each hive provides.
Some varieties of apple have the ability to produce fruit without pollination, but the
quality and quantity of fruit produced is low. About six to seven of the ten ovules must be
fertilized to produce a well shaped fruit (Brault and de Oliveira 1995). Fertilization is more
likely to happen when cross pollination by pollinators occurs.

21

Cross pollination can also

increase the amount of calcium in apples, which maintains apple quality in storage (Volz, Tustin,
and Ferguson 1996).
Highbush blueberries are able to produce a relatively constant number of blueberries
through self-fertilization and parthenocarpy (El-Agamy, Sherman, and Lyrene 1979; MacKenzie
1997). However, as in apples, cross-pollination increases the quality and quantity of the fruit as
well as increasing the rate of fruit maturation through the development of a higher number of
mature seeds. When adequately pollinated, a blueberry can have as many as 65 mature seeds. A
blueberry flower must be pollinated within three days of opening otherwise fruit will not set
(Delaplane and Mayer 2000).
Sour cherry, in contrast to sweet cherry varieties, can self-fertilize their flowers whereas
most sweet cherries need cross-pollination to produce fruit. In both types of cherry, it is very
important that pollination occurs quickly, because they only have one ovary (thus two ovules)
that starts to decay even before the flower opens (Delaplane and Mayer 2000). This can pose a
problem since cherries bloom in the early spring, when it may still be too cold for some
pollinators to forage.
To meet the pollination needs of these crops, many growers buy or rent honey bee hives
in the spring to ensure adequate pollination. Even though there are other types of pollinators,
honey bees are the most versatile because they are generalist feeders, meaning that they visit all
types of flowers to acquire food. The foraging behaviors of honey bees also facilitate a high rate
of cross pollination. They prefer to visit consecutive trees or bushes in a row instead of flying
through rows (Delaplane and Mayer 2000). They also tend to visit multiple flowers on the same
plant before moving down the row to another (Lyrene 1989). This means that not only are honey

22

bees in contact with a generous amount of pollen from each plant, but they are also spreading the
pollen throughout the crop.
The extent that honey bees are successful at pollinating is determined by the flowers’
characteristics. For example, honey bees are less effective at pollinating the Delicious apple
variety due to their tendency to collect nectar from the flowers without brushing against the
pollen covered anthers (Delaplane and Mayer 2000). The shape and downward orientation of the
blueberry flower makes it difficult for honey bees to pollinate them. The pollen containing
anthers are located further back within the flower which decreases the chance that a honey bee
will brush against them while trying to extract nectar. It would be beneficial to find a way to
increase the attractiveness of crops to honey bees so that they are more persistent at flowers that
are more difficult to pollinate.
Polynate™ is a product manufactured by BioGlobal in Australia that is intended to
increase honey bee visitation and thus, pollination (Figure 5). It contains a synthetic mixture of
the compounds found in the Nasonov gland in honey bees. In the wild, this pheromone is used
by foraging honey bees to attract more foragers to food sources with high rewards or water. It is
also used during the swarming process, to keep the swarm together (Free 1987). Other honey
bee attractants have been tested in the past with uncertain success (Burgett and Fisher 1979;
Mayer and Johansen 1982; Rajotte and Fell 1982; Tew and Ferree 1984; Mayer, Lunden, and
Britt 1989; Elmstrom and Maynard 1991; Naumann et al. 1994; Schultheis et al. 1994; Tew and
Ferree 1998; Malerbo-Souza, Nogueira-Couto, and Couto 2004; Ellis and Delaplane 2009). A
major advantage of Polynate over these previous products is that it is applied manually rather
than sprayed on the crop. Thus, the release rate is controlled and it is not susceptible to wash off
by rain.

23

Polynate has shown promise in preliminary trials as a pollination enhancer. Polynate was
found to increase fruit set as much as 10-15% in blueberries, apples and cherries (Gut and Isaacs,
2011 unpublished data). When applied to a target crop, Polynate may cause honey bees to
become more attracted to its flowers. It is expected that Polynate will cause honey bees to visit
crop flowers more frequently and for a shorter period of time (thus visiting a higher number of
flowers throughout time) causing an increase in pollen flow throughout the crop. It has been
shown that visitation rates increase in flower patches with a higher reward (Southwick, Loper,
and Sadwick 1981). This change in bee behavior could be measured as an increase in fruit
production which would increase revenue. However, there is some concern that Polynate would
repel other native foragers that are also important in pollination since it only contains
components of honey bee pheromone and not pheromones of other important pollinators.
The objectives for this study were 1) to determine the potential of Nasonov based
dispensers to increase the number of honey bee foragers in blueberries, apples and cherries; 2) to
determine if Nasonov based dispensers have a repellent effect on other non-honey bee foragers;
3) to determine the potential of Nasonov based dispensers to increase fruit set in blueberries,
apples, and cherries; 4) to determine if Nasonov based dispensers are equally effective at various
distances from honey bee hives; 5) to determine the most effective rate of application for
Polynate in blueberries; and 6) to compare the effectiveness of SPLAT or Specialized
Pheromone and Lure Application Technology (a wax paste impregnated with the same chemical
composition of Polynate in figure 4) to Polynate.

24

Materials and Methods
Experiments were conducted in four blueberry fields located in Grand Junction MI (Van
Buren County), four apple orchards on “The Ridge” in Grand Rapids MI (Kent Co.), and two
cherry orchards located in the southwest near Hartford MI (Van Buren Co.) and two in Traverse
City MI (Leelanau Co.). The efficacy of Polynate dispensers or SPLAT was determined by
comparing fruit set and pollinator counts between plots of treated and untreated blueberry, apple
and cherry. The experimental design was a randomized complete block with fields or orchards at
four sites per crop. Each site was split into pairs of plots with one randomly assigned a potential
attractant and the other designated as the untreated control. Experimental blocks were selected
that had a consistent elevation throughout the block and treatments at each site were positioned
such that pheromone would not be carried by prevailing winds from the treated plots to the
untreated plots. Honey bee hives were supplied by the growers and positioned centrally with
respect to the two treated plots. The aim was to place honey bee hives equidistant from the two
treatments. The total number of hives differed between crop and field site due to adhering to the
grower’s usual practices in renting hives. However, there were approximately 4-8 two-high
hives in apple and cherry for each field site. In blueberry, there was an average of 50 two-high
and 54 three-high per field site. However, there was as many as 44 to 188 hives (both two and
three-high hives) in one field site. Colony health was not assessed to determine the exact
number of foragers. There was a possibility that wild honey bees or other managed bees from
different hives visited the field sites. However, since it was impossible to distinguish bees from
the test hives or other sources, it was assumed that each plot had the potential to be visited
equally in the absence of treatment.

25

Polynate (Figure 5) or SPLAT (Figure 4) dispensers were applied 7-14 days prior to
bloom at the recommended rate of 1,000 units/ha. Polynate dispensers were applied at 1.5-2.0 m
height in the canopy by twisting them onto branches to ensure that they would remain in place
throughout the duration of bloom. SPLAT was applied with a caulking gun to deposit quarter
sized dollops (approximately 1 g) on the trunks of the trees at the same height and potency.
Within the treated and untreated plots, bee activity and fruit set were assessed to
determine treatment effects on pollination. The position and number of sampling locations
differed from 2012 and 2013 and between crops. Details on bee activity and fruit set or quality
measurements are provided below for each crop, as well as summarized in Table 15 and 16. In
general, there were three sampling locations in each crop in 2012: near honey bee hives, in the
middle of the plot and far from the hives (Figure 2). Due to a limited supply of Polynate in 2013,
only a 0.8 ha section in the middle of each block was treated, thus sampling of bee activity and
pollination were restricted to a single location in the center of the treated or untreated plots
(Figure 3).

Blueberry
In 2012, the experiment was conducted in four 1.6 ha Jersey variety blueberry fields. At
each site, a 0.81 ha section was treated with Polynate at the rate of 1000 dispensers/ha by placing
a dispenser on every third bush. The remaining 0.81 ha section was left untreated. Bee activity,
pollination and fruit set were measured in three sampling locations: near the honey bee hives, in
the middle of the plot and far from the hives (Figure 2). The near sampling locations were
located in the first row along the perimeter where the hives were placed, about 10-20 m from the

26

hives, the intermediate locations were approximately 30 bushes into the plot or 30-45 m from the
hives and the far locations were approximately 60 bushes or 60-90 m from the hives.
Timed observations were used to determine the number of honey bees, native bees and
other pollinators visiting blueberry flowers in treated and untreated plots. One minute counts
were conducted on each of 10 bushes randomly selected in the near, intermediate and far
sampling locations. During each one-minute count, the observer would circle around the bush,
or count pollinators on one side of two bushes. Observers counted the number of honey bees,
native bees and other potential pollinators that made contact with the flowers.

Pollinator

observations were only conducted during peak bloom and on days with optimal foraging
conditions (sunny days at approximately 20⁰C with wind blowing no more than 16 kmh).
Observations were conducted in the afternoon, during peak foraging times when honey bees and
other pollinators were visibly active (12:00 to 4:00). Observations were only conducted once
during bloom on the same day at a given site to minimize differences between treatments in
foraging due to weather conditions.
The impact of treatment on fruit set, berry size, and blueberry weight (an indication of the
number of mature seeds) was determined by comparing open pollinated fruit clusters versus
clusters that were enclosed in a fine mesh bag to prevent pollinators from accessing the flowers.
The bags consisted of a sheer fabric and were held in place with an elastic band around the
opening. Prior to bloom (mid-May) two flower clusters on each of four bushes (in the same row)
at each of the three sampling locations per plot were selected for sampling and marked with
flagging tape. The number of buds on each cluster (at least ten buds per cluster) were counted
and subsequently one cluster on each bush was bagged and the other left unbagged. After
bloom, when berries started to develop (June), bags were removed and the number of berries

27

formed on each cluster was counted. A final assessment was made when blueberries were ripe
(July) by removing the clusters and bringing them back to the laboratory to determine berry size.
Approximately ten berries from each cluster were weighed and dissected to determine the
number of seeds per fruit (two sites for seed counts and three sites for weights).
The experimental set-up was modified in 2013 due to a limited supply of Polynate
dispensers. The experiment was conducted in four 6.1 ha Jersey variety blueberry fields. Each
field was divided into three experimental units that were randomly assigned one of three
treatments: 1) Polynate deployed at 1,000 units/ha, 2) half of the Polynate dispenser deployed at
the rate of 1,000 units/ha, and 3) an untreated control. The Polynate dispenser is a twin-tube
design making in easy to split the dispenser in half for the low rate treatment (Figure 5). Only a
0.4 ha section in the center of each block was treated, thus sampling of bee activity and
pollination were restricted to a single location in the center of the treated or untreated plots
(Figure 3). The impact of treatment on fruit set, berry size, and blueberry weight and bee activity
were assessed as described above. However after the fruit were counted, the bags were replaced
to prevent berry loss due to animals or pickers. Fruit counts as previously described were taken
from 16 bushes in a 40 m² area located centrally in each 0.4 ha experimental area. To determine
the berry size all the ripe berries from each cluster were weighed, but seed counts were not taken.
Pollinator counts as described above were taken from 20 random bushes located in the 0.4 ha
experimental area. The number of honey bees, bumble bees, other bees and flies were counted
per minute.

28

Apple
In 2012, experiments were conducted in three 2.4 ha apple orchards. A total of three
treatment plots were used in each orchard. Polynate and SPLAT were applied to 0.81 ha sections
at a rate of 1000 dispensers/ ha. The remaining 0.81 ha was left untreated. Apple varieties varied
across sites and included Jonagold, Golden Delicious, Gala, Honey Crisp, Empire, Paula Red,
Portland and Ida Red. However, varieties were the same in treated and untreated plots at each of
the three sites. Due to a severe early season frost, accurate fruit counts were unobtainable at
harvest in 2012. Pollinator activity was assessed using timed observations as described for
blueberries. Sampling was conducted on 12 trees in each of three sampling locations in treated
and untreated plots. The near sampling locations were located in the closest row to the honey
bee hives, the intermediate locations were approximately 15 trees or 30-40 m from the hives and
the far locations were approximately 30 trees or 60-80 m from the hives.
As with blueberries, the experimental set-up was modified in 2013 due to the limited
supply of dispensers. The experiment was conducted in three 4 ha apple orchards. Varieties
were the same as in 2012. Each orchard was divided into two experimental plots and a 0.6 ha
area within the center of one plot was treated with Polynate at 1000 dispensers/ha and a 0.6 ha
area within the center of the other plot was delineated with flagging tape and left untreated. Fruit
set was assessed by counting flower buds on 25 trees in a 250-300 m² area located centrally in
each 0.6 ha experimental plot. Prior to bloom (May), a pair of branches on each tree that had at
least 30 fruit buds were identified, marked with flagging tape and the total number of buds
counted. Four to five weeks later, when fruit were 10-l5 mm in circumference, the total number
of fruit present on the same branches was counted. The percent of flowers that produce fruit is
expected to be anywhere from 0.2% to 25.9% (Stephenson 1981). Timed observations as

29

described previously were used to determine pollinator activity in treated and untreated plots.
One-minute counts were conducted on each of 24 trees randomly selected in the near,
intermediate and far sampling locations.

Cherry
In 2012, experiments were conducted in three 1.6 ha ‘Montmorency’ cherry orchards. At
each site, Polynate was applied to 0.81 ha section at a rate of 1000 dispensers/ ha. The remaining
0.81 ha was left untreated.

Pollinator activity was assessed using timed observations as

described previously. Sampling was conducted on 20 trees in each of three sampling locations.
The near sampling locations were located in the closest row to the honey bee hives, the
intermediate locations were approximately 15 trees or 30-45m from the hives and the far
locations were approximately 30 trees or 60-90m from the hives. Fruit set was assessed by
counting fruit buds on 10 trees located in the middle of each 0.81 ha plot. Prior to bloom (May),
a suite of four branches on each tree that had at least 100 fruit buds were identified, marked with
flagging tape and the total number of buds counted. Four to five weeks later when set fruit were
present, the total number of fruit present on the same branches was counted.
The experimental set-up was modified in 2013 due to the limited supply of dispensers.
The experiment was conducted in three 4 ha ‘Montmorency’ cherry orchards. Each orchard was
divided into two experimental units and a 0.6 ha area within the center of one plot was treated
with Polynate at 1000 dispensers/ha and a 0.6 ha area within the center of the other plot were
delineated with flagging tape and left untreated. Fruit set was assessed as in 2012 by counting
the number of set fruit from a know number of fruit buds. The expected percentage of flowers
that produce fruit is 23.5% to 50.1% (Stephenson 1981). Timed observations as described

30

previously were used to determine pollinator activity in treated and untreated plots. One-minute
counts were conducted on each of 24 trees randomly selected in the near, intermediate and far
sampling locations.

Statistical Analysis
All data were analyzed using a randomized complete block design in SAS 9.3 (SAS
Institute Inc. 2013). A one-way Analysis of Variance (ANOVA) was performed using PROC
MIXED and treatment as the independent variable. This was used to determine the effect of
treatment on pollinator abundance and fruit yield (i.e. dependent variables).

Pollinator

abundance and fruit set data were transformed using a square root transformation to normalize
the data. Mean separations and comparisons were calculated using Tukey’s Honest Significant
Difference test (HSD) at a 5% confidence level (SAS Institute Inc. 2013).

Results
Blueberry
Fruit set or the number of fruit produced was significantly lower for the bagged than the
un-bagged clusters in 2012. However, there was no significant difference between Polynate
treated or untreated plots for either unbagged or bagged clusters (F = 0.84; df = 1,1; p = 0.53).
The untreated plots tended to have a slightly higher fruit set than the treated plots but there was
no significant difference in fruit set between the three sampling locations (F = 0.42; df = 2,6; p =
0.67) (Table 1).

31

Table 1. The mean number of blueberry fruit per flower ± SEM at three distances from honey
bee hives in untreated and Polynate treated fields during 2012
Sampling Location Bag
Near
Intermediate
Far

Yes
No
Yes
No
Yes
No

Untreated

Polynate

0.17 ± 0.05
0.76 ± 0.07
0.10 ± 0.02
0.84 ± 0.03
0.24 ± 0.04
0.76 ± 0.02

0.24 ± 0.09
0.73 ± 0.18
0.15 ± 0.11
0.80 ± 0.04
0.14 ± 0.05
0.70 ± 0.19

In 2013, fruit set was significantly lower for clusters that were bagged compared to those
not bagged. There was not a significant difference in fruit set between Polynate or untreated
plots (F = 2.75; df = 2,2; p = 0.27), but the blocks treated with a half rate had numerically fewer
fruit per flower compared to bagged and unbagged exposures (Table 2).

Table 2. The mean number of blueberry fruit per flower ± SEM in untreated and half or full
Polynate treated fields during 2013
Full
Half
Bag
Untreated
Polynate
Polynate
Yes
0.33 ± 0.09 0.32 ± 0.09 0.22 ± 0.10
No
0.84 ± 0.02 0.86 ± 0.02 0.74 ± 0.04

The number of pollinators observed per minute visiting blueberry flowers in 2012 was
not significantly different between the Polynate treated or untreated blocks (F = 0.94; df = 1,57;
p = 0.34). However, there was a slight trend for higher counts in the treated blocks. There also
was not a significant difference in pollinator activity between the three sampling locations (F =
0.90; df = 2,6; p = 0.45). Honey bees were the most abundant pollinator in both treatments. The
numbers of bumble bees and other bees were extremely low (Table 3).

32

Table 3. The mean count of honey bees, native bees and other pollinators per minute ± SEM at
three distances from honey bee hives in untreated and Polynate treated blueberry fields during
2012
Sampling
Untreated
Polynate
Location
Near
Honey bees
3.60 ± 0.45
3.25 ± 0.46
Native bees
0.05 ± 0.03
0.05 ± 0.03
Other Pollinators
0.10 ± 0.08
0.03 ± 0.03
Intermediate
Honey bees
3.75 ± 0.36
3.48 ± 0.45
Native bees
0.03 ± 0.03
0.08 ± 0.04
Other Pollinators
0.13 ± 0.09
0.03 ± 0.03
Far
Honey bees
3.73 ± 0.32
3.33 ± 0.43
Native bees
0.13 ± 0.05
0.15 ± 0.05
Other Pollinators
0.10 ± 0.05
0.05 ± 0.03

There was not a significant difference between treatments for pollinator counts per
minute in 2013 (F = 0.11; df = 2,39; p = 0.89) although, plots treated with a full or half rate of
Polynate had numerically higher numbers of flower visitors. Honey bees again were the most
abundant pollinator observed. Counts of other pollinators were very low making it difficult to
measure any differences between the treatments (Table 4).

Table 4. The mean count of honey bees, bumble bees, other bees and flies in blueberry fields per
minute ± SEM during 2013
Organism
Honey
bees
Bumble
bees
Other bees
Flies

Full
Polynate

Untreated

Half
Polynate

2.78 ± 0.77 3.20 ± 0.99 3.43 ± 0.69
0.04 ± 0.02 0.03 ± 0.01 0.00 ± 0.00
0.03 ± 0.03 0.05 ± 0.03 0.01 ± 0.01
0.20 ± 0.06 0.28 ± 0.11 0.26 ± 0.11

33

There was not a significant difference between treatments in berry mass for 2012 (F =
3.52; df = 1,1; p = 0.31) and 2013 (F = 0.09; df = 2,2; p = 0.92). Polynate treated plots had
slightly heavier blueberries in 2012, but slightly lighter blueberries in 2013 (Table 5). There was
not a significant difference between sampling locations in 2012 (F = 0.98; df = 2,2; p = 0.51).
There also was not a significant difference in berry mass between bagged and unbagged clusters
in either year, although the unbagged berries were always heavier.

Table 5. The mean mass (g) ± SEM of blueberries in untreated and full or half Polynate treated
fields during 2012 and 2013
Year Sampling location
Bag
Untreated
Full Polynate Half Polynate
2012
Near
Yes
0.43
0.84 ± 0.15
No
1.01 ± 0.05
1.12 ± 0.13
Intermediate
Yes
0.79 ± 0.20
0.93 ± 0.05
No
1.03 ± 0.06
1.22 ± 0.04
Far
Yes
0.57 ± 0.15
0.62 ± 0.06
No
1.04 ± 0.28
1.14 ± 0.26
2013
NA
Yes
0.83 ± 0.10
0.73 ± 0.08
0.49
No
1.27 ± 0.11
1.05 ± 0.11
0.80 ± 0.21

There was not a significant difference in fertilized seed counts between Polynate or
untreated plots in 2012 (F = 0.10; df = 1,1; p = 0.80). There also was not a significant difference
in seeds between locations (F = 0.05; df = 2,2; p = 0.95) or between bagged and unbagged
clusters (Table 6).

Table 6. The mean mature seed count ± SEM of blueberries in untreated and Polynate treated
fields during 2012
Sampling location
Bag
Untreated
Polynate
Near
Yes
19.00
13.17 ± 9.77
No
44.20 ± 7.10
33.98 ± 2.73
Intermediate
Yes
15.50 ± 1.50
23.00 ± 7.50
No
37.41 ± 4.71
40.98 ± 2.16
34

Far

Table 6. (cont’d)
Yes
13.30 ± 0.70
No
31.70 ± 0.00

25.63 ± 11.63
40.55 ± 3.45

Apple
Due to freezing temperatures in the spring, we were not able to collect reliable data for
apple fruit set in 2012. There was not a significant difference in fruit set between treatments (F =
2.86; df = 1,1; p = 0.34) in 2013. However, the untreated plots had a slightly higher numerical
fruit set than the Polynate treated plots (Table 7).

Table 7. The mean number of apple fruit per flower ± SEM in untreated and Polynate treated
orchards during 2013
Treatment
Untreated
Polynate
Fruit/Flower 0.19 ± 0.04 0.16 ± 0.02
There was not a significant difference between the Polynate, SPLAT or untreated plots in
the number of honey bees, other bees or flies visiting apple flowers in 2012 (F = 0.08; df = 2, 68;
p = 0.92) per minute. There was also not a significant effect of distance from the hive on
pollinator activity (F = 0.69; df = 2,4; p = 0.55) (Table 8).

Table 8. The mean count of honey bees, native bees and other pollinators per minute ± SEM at
three distances from honey bee hives in untreated and Polynate or SPLAT treated orchards
during 2012
Sampling Location
Near
Honey bees
Native bees
Other Pollinators

Untreated

Polynate

SPLAT

1.94 ± 0.54 1.86 ± 0.43 1.61 ± 0.59
0.00 ± 0.00 0.00 ± 0.00 0.03 ± 0.03
0.00 ± 0.00 0.06 ± 0.03 0.06 ± 0.06

35

Table 8. (cont’d)
Intermediate
Honey bees
Native bees
Other Pollinators
Far
Honey bees
Native bees
Other Pollinators

1.44 ± 0.56 0.97 ± 0.34 1.19 ± 0.56
0.06 ± 0.06 0.03 ± 0.03 0.00 ± 0.00
0.06 ± 0.06 0.00 ± 0.00 0.03 ± 0.03
1.44 ± 0.48 1.12 ± 0.70 2.64 ± 0.77
0.00 ± 0.00 0.03 ± 0.03 0.00 ± 0.00
0.06 ± 0.06 0.12 ± 0.03 0.03 ± 0.03

Pollinator counts per minute in Polynate and untreated plots were not significantly
different in 2013 (F = 1.13; df = 1,17; p = 0.30) either. However, the Polynate treated plots
consistently had a higher number of pollinators. Honey bees comprised the majority of flower
visitors in 2012 and 2013; the number of native bees and bumble bees visiting flowers were very
low (Table 9).

Table 9. The mean count of honey bees, bumble bees, other bees and flies per minute ± SEM in
untreated or Polynate treated orchards during 2013
Organism
Untreated
Polynate
Honey
8.40 ± 0.38 10.51 ± 1.13
bees
Bumble
0.00 ± 0.00 1.67 ± 0.07
bees
Other bees 0.40 ± 0.33 0.35 ± 0.27
Flies
0.03 ± 0.01 0.04 ± 0.02

Cherry
There were no significant differences in fruit set between treatments in 2012 (F = 0.74; df
= 1,5; p = 0.43). The untreated plots had a higher numerical fruit set in the near and far
locations, but the opposite was true for the intermediate location. Distance was also not a
significant predictor of the number of fruit per flower (F = 0.98; df = 2,2; p = 0.51) (Table 10).

36

Table 10. The mean number of cherry fruit per flower ± SEM at three distances from honey bee
hives in untreated and Polynate treated orchards during 2012
Sampling
Untreated
Polynate
Location
Near
0.16 ± 0.02 0.08 ± 0.04
Intermediate 0.11 ± 0.04 0.14 ± 0.02
Far
0.09 ± 0.00 0.08 ± 0.05

Overall, treatment was not a significant factor affecting the fruit per flower ratio in 2013
(F = 7.82; df = 2,3; p =0.06). The plots treated with SPLAT had a numerically higher number of
fruit per flower than the Polynate treatment and the untreated plots. However, since SPLAT was
only tested in a single plot the difference in fruit set could be associated with plot differences
rather than the treatment (Table 11).

Table 11. The mean number of cherry fruit per flower ± SEM in untreated and Polynate or
SPLAT treated orchards during 2013
Treatment
Untreated
Polynate
SPLAT
Fruit/Flower 0.16 ± 0.03 0.12 ± 0.02
0.29

There was not a significant difference between treatments in the number of pollinators
observed per minute in 2012 (F = 1.01; df = 1,16; p = 0.33). However, untreated plots tended to
numerically have more flower visitors than treated plots. Honey bees were the most frequently
counted pollinator. There was a numerical trend of more native bees in untreated plots compared
to treated plots. Distance was not found to be a significant factor in predicting pollinator counts
(F = 5.53; df = 2,2; p = 0.15) (Table 12). A greater number of pollinators were observed in the
more distant plots compared to the near plots.

37

Table 12. The mean count of honey bees, native bees and other pollinators per minute ± SEM at
three distances from honey bee hives in untreated or Polynate treated orchards during 2012
Sampling
Untreated
Polynate
Location
Near
Honey bees
5.20 ± 0.40
4.53 ± 1.28
Native bees
0.43 ± 0.28
0.30 ± 0.25
Intermediate
Honey bees
7.68 ± 1.48
6.58 ± 1.63
Native bees
1.03 ± 0.23
0.00 ± 0.00
Far
Honey bees
7.88 ± 0.33
8.18 ± 0.38
Native bees
1.20 ± 0.75
0.65 ± 0.05
There was not a significant difference in pollinator counts per minute between treatments
in 2013 (F = 1.12; df = 2,26; p = 0.34). However, plots treated with SPLAT had numerically
higher number of honey bees. Honey bees were the most frequent pollinator, followed by other
bees (Table 13).

Table 13. The mean count of honey bees, bumble bees, other bees and flies per minute ± SEM in
untreated and Polynate or SPLAT treated orchards during 2013
Organism
Honey
bees
Bumble
bees
Other bees
Flies

Untreated

Polynate

SPLAT

1.18 ± 0.27 1.57 ± 0.57

1.83

0.13 ± 0.06 0.03 ± 0.02

0.13

0.45 ± 0.13 0.44 ± 0.12
0.18 ± 0.06 0.25 ± 0.17

0.54
0.08

Discussion
Overall there was a lack of consistency in measured differences in pollinator activity and
fruit set in plots treated with Polynate or SPLAT and plots left untreated. There were no
significant differences between treatments in either fruit set or pollinator counts. Although there
38

were some trends, they were not consistent between crops or years. Therefore, Polynate and
SPLAT did not have a significant effect on pollinators or fruit crop pollination. The addition of a
Nasonov based dispenser also did not influence the activity of other flower visitors. We did not
expect to see an increase in activity as the compounds present in Polynate or SPLAT are only
known to influence honey bee behavior. Although the number of other pollinators was generally
low, we did not observe any adverse affects on activity of treating with Polynate or SPLAT.
Additional studies in which one or more species of native bees are released in plots are needed to
confirm our findings.
Surprisingly, there were no differences in pollinator counts between the near,
intermediate and far sampling locations. We hypothesized that there would be less honey bee
activity in the sampling locations further from the hives compared to close to the hives leading to
a greater positive effect of Nasonov based dispenser treatment at these locations. Honey bees
have been documented to fly as far as 9.5 km from their hive (Beekman and Ratnieks 2000).
Thus it is probable that the furthest sampling locations were not far enough away from the hives
to result in major differences in foraging activity. Indeed, overall we observed similar numbers
of honey bees in the untreated plots across the three sampling locations. It is also a possibility
that honey bees from other hives on the farm visited the test plots. The relative positions of the
near, intermediate and far sampling locations for other hives could have been different compared
to the experimental hives.
During bloom in commercial apple, blueberry, and cherry, farms used in these
experiments had a substantial number of honey bees and other potential pollinators visiting
flowers. It may be very difficult to attain measurable improvements in pollination under these
circumstances. It is possible that a pollination threshold is reached where flowers receive the

39

maximum amount of pollination possible. This would mean that no matter what is done to
increase honey bee activity there would not be a measurable increase in fruit set and size. The
maximum number of fruit produced is clearly limited to the number of flowers during bloom and
the amount of natural pollination tends to surpass the amount needed to set fruit (Stephenson
1981). Also, it is possible that too much pollen on a stigma can actually cause pollination failure
in some plants (Wilcock and Neiland 2002).
In these trials where high pollinator activity trends were observed, they generally did not
produce significant positive effects on fruit set. This also occurred when queen mandibular
pheromone sprays were applied in cranberry and blueberry (Currie, Winston, and Slessor 1992a)
and when Bee-Scent was applied in apples (Tew, Ferree 1998). Future studies with Nasonov
based dispensers or other potential pollination enhancements should focus on the potential for
increasing fruit set in orchards that are known to be difficult to pollinate and achieve high fruit
set.

Only when pollinator activity does not meet a pollination threshold, can pheromone

attractants become economically feasible (Currie Winston, and Slessor 1992a; Currie, Winston,
and Slessor 1992b). However, it has been suggested that Nasonov pheromone applications may
not aid in honey bee attraction if flowers provide low reward (Free 1987) which may be the case
for some crops that are harder to pollinate. Studies could be conducted in orchards where few or
no honey bee hives are placed near the orchard during bloom.
Polynate and SPLAT were applied at a rate of 1000/ha, primarily based on the
recommended application rate for similar devices containing sex pheromone and used to disrupt
mating of the targeted species (Gut et al 2004). It is possible that this application rate is either
too high or too low to achieve the desired result of pollination enhancement. Nothing is known
about how the chemicals loaded into Polynate or SPLAT disperse and how large or small the

40

resulting plume is. At a high application or release rate, this could result in plumes spreading
into the untreated areas adjacent to treated plots. Under this circumstance, honey bees would be
equally attracted to the treated or untreated blocks.
Moreover, the threshold at which honey bees respond or fail to respond to the compounds
present in Polynate or SPLAT is unknown. The high application or release rate of Polynate and
SPLAT might have overloaded the sensory system of the honey bees. Alternatively, the plumes
emanating from individual dispensers may have been too small.

Perhaps the plume of a

dispenser did not extend past the tree that it was applied and honey bees only sensed its presence
locally rather than from further distances. Honey bee sensitivity to odors is well known and
studied (von Frisch 1950; Winston 1987). We could readily perceive the odor emanating from
Polynate when it was applied, suggesting that high quantities of the volatiles were present. It is
reasonable to assume that foraging honey bees could detect the odors as well (von Frisch 1950;
Kaissling 1971). This suggests that the rates applied were too high rather than too low. There
was no significance difference in the effect of Polynate applied at a full or half rate. Much lower
release rates may be needed to have the desired effects on honey bee foragers.
The effect that Polynate or SPLAT might have on pollinators may be more subtle than
the hypothesized increase in numbers of individuals visiting flowers. Our sampling methods
were designed to maximize measuring any increase in the numbers of flower visitors. Polynate
or SPLAT may have the more subtle effect on the time an individual pollinator spends foraging
in the treated area. Honey bees are known to spend more time feeding at artificial feeders with a
high reward (von Frisch 1950). In the field, this behavior could increase the amount of pollen
carried from the flower to the bee, but might decrease the actual number of flowers visited. In
reality, honey bees tend to spend less time feeding on individual flowers in attractive patches to

41

maximize energy efficiency. The less time a bee spends on one flower, the more flowers she can
visit in one foraging trip (Schmid-Hempel, Kacelnik, and Houston 1985; Delaplane and Mayer
2000). Thus, flower visitation rates and total time spent are increased in flower patches with
higher rewards (Southwick, Loper, and Sadwick 1981). This could cause an increase in pollen
transfer from flower to flower, but this might not cause an increase in fruit production, especially
if a flower’s pollination needs are already met. Since we only counted the number of visits
(defined as every time a pollinator made contact with a flower) and not the number of flowers
visited or the total time each bee spent at a flower, we are unable to discern any effect of
Polynate or SPLAT on these behaviors. Changes in the number of flowers visited or time spent
at a flower may not greatly change the number of fruit produced per flower (fruit set) but they
may increase the quality of fruit, which was not measured in apples or cherries.
Our results demonstrate the complexities of pollination in crops. Due to recent declines,
pollinators and pollination are receiving increased attention from the scientific community and
the general public. This had led to increased effort to enhance honey bee habitat or efficiency to
improve pollination (Potts et al. 2010).

Perhaps lost in the intense effort to address the

pollination problem is the recognition of the complexity of the process by which honey bees and
other insects pollinate crops. Honey bee workers perform pollination services until they die, and
it may not be possible to entice them to work any harder simply by improving the attractiveness
of crops through pheromones. There are many factors that go into a forager’s choices and
behaviors.

Due to these complexities, developing a product that will increase pollination

services may be a difficult task. However, the less than desirable results obtained with Polynate
and SPLAT should not discourage us from continuing to pursue attractant-based approaches to
enhancing pollination.

42

CHAPTER 3
The Response of Honey Bees Reared in a Hoop House to Polynate and SPLAT
Introduction
Honey bee behavior has been heavily investigated throughout the years, starting with the
famous work of Karl von Frisch (von Frisch 1950, 1966, 1967). It is clear that the factors
affecting honey bee behavior are very complex and that they use many cues to make choices.
One of the most significant social cues honey bees use are chemicals called pheromones. There
are many different types of pheromones such as alarm pheromones, brood pheromones, queen
mandibular pheromone and Nasonov pheromone that honey bees use to communicate with each
other within the hive and outside (Free 1987; Pankiw 2004). Pheromones are used to convey
information from one honey bee to the next and can cause changes in the behaviors of individual
bees for the benefit of the hive.
Honey bees do not start to react to Nasonov pheromone until they are about 28 days old;
when they start to forage (Winston 1987). Nasonov pheromone is thought to be the used for
orientation and organization during swarming (Morse and Boch 1971; Jay 1986) and is also
important in nest-seeking behavior (Schmidt 1999) and marking the hive entrance (Winston
1987). Honey bees often expose their Nasonov glands and fan their wings after some type of
disturbance of the hive such as when a beekeeper moves it (Free 1987). There also is evidence
that Nasonov pheromone is used to make rewarding food resources more attractive. More
specifically, Nasonov pheromone may increase recruitment of honey bees by the forager that
encounters it. However, there is mixed support for this theory (Wenner and Wells 1990; Wells et
al. 1993).
It is common for scientists to attempt to manipulate honey bee behavior using
pheromones because of the way the bees naturally use them. One of the more recent examples of
43

this is to use pheromones to attract honey bees to target crops in the hope of increasing fruit
quantity and quality. It has been suggested many times that it could be beneficial to apply
synthetic Nasonov pheromone to crops in need of further pollination (Free 1968; Free 1987). A
specific pheromone laden dispenser (Polynate™, BioGlobal, Australia) that contains a synthetic
version of Nasonov pheromone was applied to blueberries, apples, and cherries and the impacts
on fruit quantity, quality or pollinator activity was quite variable, with no significant increase in
pollination levels (Chapter 2). However, it was unclear whether Polynate or SPLAT had any
effect on the behavior of honey bees specifically. Thus, it is also important to investigate if these
Nasonov based dispensers are attractive to honey bees in a controlled setting and if it has any
subtle effects that were not measured previously. In the field, the number of honey bees that
visited the experimental blocks could not be controlled. Feed time, flower visit duration or
specific behaviors of individual bees were also not measured. In this controlled study, behaviors
were observed directly instead of only quantifying potential outcomes of changed behaviors
(such as increased fruit set).
Honey bees make foraging decisions based on energy profitability and their foraging
background. For example, a resource with a higher quality of food (higher nectar quality) at a
closer distance is thought to me more rewarding, thus more attractive to honey bees (Shafir
2011). It is possible to measure how attractive a food source is to honey bees by using the
number of visits it receives. When a honey bee returns from a foraging flight, she will relay
information about the location of desirable food sources through the waggle dance to other
foragers in the hive. Thus, if a food source is attractive, the number of recruited foragers visiting
that food source will increase exponentially depending on the quality of the food source and
forager availability (von Frisch 1967; Núñez 1970; Dyer 2002). Interestingly, once a honey bee

44

exposes her Nasonov gland, other honey bees are prone to do so as well (Winston 1987). This
behavior could be used to draw additional foragers to a rewarding food source.
While foraging on flowers which present a consistent reward, honey bees don’t
completely fill their crops because flying back to the hive with a full crop takes more energy
(Schmid-Hempel, Kacelnik, and Houston 1985). This crop filling behavior also occurs because
individual flowers do not contain enough nectar to completely fill a honey bee crop. However,
artificial feeders have plenty of sucrose solution. Indeed, von Frish observed that honey bees
completely fill their crops at artificial feeders with a high sucrose concentration but do not at a
feeder with a low sucrose concentration (von Frisch 1950). Since it is likely that completely
filling a crop takes more time than partially filling, this could mean that honey bees feed for a
longer period of time on artificial feeders that they find attractive. Thus, the attractiveness of an
artificial feeder treated with Nasonov based dispensers could be measured by comparing the time
each bee spends feeding at it with another untreated feeder.
Two pollination enhancement products, Polynate and SPLAT, have shown promise in
preliminary field studies as a tool for increasing fruit set and foragers in blueberries, apples, and
cherries (Gut and Isaacs, 2011 unpublished data). However, more recent field studies did not
yield the same positive results (Chapter 2). Due to inconsistencies in field studies, a series of
controlled studies were undertaken to determine more directly how Polynate and SPLAT might
increase the attraction of a food source to foraging honey bees. The specific objectives were to
1) determine if Polynate or SPLAT increase the number of honey bee visits to an artificial feeder
2) determine if Polynate or SPLAT have an effect on the time that honey bees spend feeding in
one foraging trip 3) to observe the general behavior of honey bees around a pollination

45

enhancement dispenser and 4) determine the attractiveness of different concentrations of
Polynate to foraging honey bees.

Materials and Methods
All experiments were conducted in an outdoor hoop house located on the Southern part of
the Michigan State University (MSU) campus and described in Townsend-Mehler et al. (2010).
The hoop house is 35 m long, 5.6 m wide and 2.3 m high and is covered with 30% greenhouse
shade cloth (Pictured in Figure 6, without end partitions). Only about a 27 m area on one end
was used to avoid interfering with experiments being conducted at the other end of the facility.
Previous studies conducted in this particular hoop house concluded that the 30% light reduction
this shade cloth produces has no effect on the behavior of the bees being studied (TownsendMehler et al. 2010).

Nuclear Colony (Nuc) Rearing
Confining a honey bee hive in a small area such as the hoop house on MSU’s campus
presented some complications. Once a honey bee leaves its hive for the first time, she does
something called an orientation flight. This is so that she will know what her hive looks like as
well as its location. She will often expose her Nasonov gland at the hive entrance afterwards
(Hazelhoff 1941; Free 1987). After the orientation flight, she will start to forage and develop an
impression of the outside world. This means that once foragers in a colony have their first
flights, the colony can no longer be confined within a small area. If a honey bee hive is placed
within the confines of a hoop house the bees will become disorientated and form swarms in the
corners, constantly trying to find ways to escape confinement. Therefore, conducting studies in
the hoop house required the use of a hive that was set up inside the house before the bees had
46

their first orientation flights. In other words, the honey bees must have their first flight inside of
the hoop house and perceived the confined space as their entire foraging area.
To establish a small nuclear colony (a four frame hive) inside of the hoop house, four
frames of honey bee brood were taken from a hive maintained at MSU in mid-June and reared in
an incubator kept at 35⁰C until young bees emerged. Newly emerged bees were brushed into the
nuclear hive located within the hoop house with a handheld bee-brush. A queen was removed
from a colony maintained at MSU and put into a queen cage stopped with candy along with
several worker bees. The queen cage consisted of a small hollowed out wooden block about the
size of a thumb with screen for one side. The queen cage was installed into the two-frame hive,
screen side facing outward (so that worker bees could feed her) and left there until the worker
bees chewed through the candy stopper. The honey bees were provided with pollen (sold
commercially and slightly ground up) and gave foraging bees daily access to a sucrose solution
artificial feeder. All other food sources were eliminated.

Artificial Feeders and Sucrose Solutions
Artificial sucrose feeders were used to control the quality and consistency of food that the
honey bees received. Feeders consisted of a baby food jar upended on a clear plastic platform
and filled with a sucrose solution (Figure 7). Because honey bees are generalists, it is easy to
encourage them to forage on artificial feeders if the reward is enough. In fact honey bees tend to
expose their Nasonov glands near artificial feeders more than near flowers (Free and Williams
1972).
It also was important to provide bees with a sugar source that simulates natural nectar
sources. Honey bees prefer sucrose solutions of about 0.88-1.5 M (30-50% concentration).

47

Lower or higher concentrations of sucrose may cause aversion to the sucrose solution. However,
honey bees are less discriminatory if there are more foragers (Waller 1972) or if the number of
food sources in the area is low (von Frisch 1950; Seeley 1989). When testing the attractiveness
of Polynate or SPLAT it was important to keep a balance between providing bees with an
extremely attractive feeder vs. one that caused aversion. If the feeder was too rewarding, then
the attractiveness of the treatment would be irrelevant because the feeder would attract more
foragers regardless of the presence of Polynate or SPLAT. However, if the opposite was true,
then the honey bees may have learned to associate the Polynate or SPLAT odor with a negative
stimulus.
For the first trial, a 1.5 M sucrose solution was used for the two experimental feeders at
either end of the hoop house (this first trial was thrown out in the end due to a camera
malfunction which led to one side not being filmed). The sugar concentration was decreased to 1
M and 0.5 M for subsequent experiments. Sucrose solutions of lower sugar content were not
used because there is evidence that this would have caused an aversion to the feeders (Waller
1972).

Experimental Design and Protocols
Experiments were conducted between the 20th of August and 23rd of September at
approximately 12:00pm in 2013. For all experiments there were a total of three feeders in the
hoop house (Figure 7). One of the feeders, defined as the stock feeder contained a lower sucrose
concentration (0.25 M) and was placed directly in front of the hive. The other two feeders,
defined as the experimental feeders, contained a higher sucrose concentration of 1.0 M or 0.5 M
(relative to the activity of the honey bees). The two experimental feeders were placed on

48

opposite sides of the hoop house (north and south), equidistant from the hive. Each platform had
a crisscross pattern of colored tape that mimicked natural flower patterns so that the bees could
easily see them. The effect of a Polynate or SPLAT dispenser on bee behavior was evaluated by
placing a dispenser on the same platform as one of the feeders while leaving the other feeder
untreated. Experiments also were conducted with blank dispensers (containing no chemicals;
trial 7) or no dispensers at all (trials 8 and 22) at both feeders. These trials served as additional
controls but were not analyzed statistically due to low replication.
Before the start of each trial, the stock feeder was set up and any bee that visited it was
tagged. Colored and numbered tags, slightly larger than the size of a pinhead were glued with
wood glue to the back of 50 bee’s thoraxes using a toothpick. Additional bees were tagged (in
increments of ten) as the original tagged bees died or lost their tags. At the start of each
experiment, 0.25 M stock feeder was placed in front of the hive and bees (tagged and nontagged) were allowed to visit it. Once at least five untagged bees and one tagged bee were
visiting the feeder sucrose solution was switched to a higher concentration and allowed the
unsettled foragers to settle and start feeding again. The stand and the feeder were moved to one
side of the hoop house (north or south) where an observer counted and recorded the number of
untagged bees and the identity of any tagged bees. The experimental feeder containing a 0.5 M
sucrose solution was replaced with another 0.5 M experimental feeder on another stand back in
front of the hive soon after. Once it was again visited with foragers, it was transported to the
other side of the hoop house where another observer was waiting. One side was randomly
chosen to be treated with a Nasonov based dispenser, which was placed on the stand by the
observer once it reached its side. The other side contained the experimental feeder only. The
following pollination enhancement treatments were tested: 1) Full rate of Polynate, 2) a half rate

49

of Polynate in which the twin-tube was split in half, 3) a quarter rate of Polynate in which one
side of the twin-tube was cut in half and 4) a 1 g SPLAT dollop. There was not a significant
amount of time between the placing of each feeder on opposite sides (not more than 1-3
minutes). After both sides received a feeder, the 0.25M stock feeder was placed back in front of
the hive and left it for the bees even after the completion of the trial. This was an alternative
method to train bees to visit the experimental feeders and was done to avoid the possibility that
honey bees wouldn’t find the feeders throughout the duration of the trials.
For 30-45 minutes, each side was videotaped by an observer (Figure 8) using an Aiptek
Handycam camera filming 29 frames/second. The trial was stopped when either of the sucrose
feeders ran out, when the memory on the cameras was full or if honey bee activity was so high,
that the number of visits was too difficult to count. The observer vocalized the feeding behavior
of the tagged bees; specifically when they started and stopped feeding (defined by the duration of
time between the protraction and retraction of the proboscis (Townsend-Mehler, Dyer, and
Maida 2011) and when they arrived and left the feeder. After each trial, each video was watched
and the total number of visits each feeder received was counted (foraging intensity). The total
feeding time in seconds was also counted for each tagged bee. If no tagged bees visited the
feeder throughout the duration of filming, several bees whose activity was easily seen throughout
their entire visit were randomly chosen and the time they spent feeding was recorded.

Statistical Analysis
All data were analyzed using SAS 9.3 (SAS Institute Inc. 2013). A one-way Analysis of
Variance (ANOVA) was performed using PROC MIXED and treatment as the independent
variable. This was used to determine the effect of treatment on foraging intensity and feeding

50

time (i.e. dependent variables). The foraging intensity data were transformed using arcsin
transformation to normalize the data. Mean separations and comparisons were calculated using
Tukey’s Honest Significant Difference test (HSD) at a 5% confidence level (SAS Institute Inc.
2013).

Results
The honey bees reared in the hoop house fed at the stock and experimental feeders and
the colony thrived.

The hive was strong throughout the summer without having to add

supplementary newly emerged bees to it. The queen laid eggs continuously throughout the
summer, but very little brood was raised. The bees were able to maintain an adequate supply of
honey and pollen with the amount of sugar water that was supplied to them through the
experiments.
Forager activity was high in the hoop house in general. However, very few bees foraged
before the continuous stock feeder in front of the hive was re-filled with 0.25 M sucrose solution
prior to the onset of the experiment. Once the stock feeder was refilled, honey bees were visiting
it and other bees were foraging within minutes. After only a few experiments, the bees seemed
to remember the location of the experimental feeders. Once the stock feeder was refilled, many
foragers immediately went to the locations of the two experimental feeders in anticipation of
sucrose being placed there.

During the experiment, when the experimental feeders were

available, the 0.25 M stock feeder was relatively abandoned. Honey bees resumed feeding at the
stock feeder once the other experimental feeders were removed. Forager activity was lower on
days after the experiment had been delayed; like after a spell of bad weather or the weekend for
example.

51

Foraging Intensity
There were no significant differences between the full Polynate and untreated treatments
(F = 3.57; df = 4,3; p = 0.62) the half Polynate and untreated treatments (F = 0.26; df =4,3; p
=1.00) the quarter Polynate and untreated treatments (F = 0.03; df = 4,3; p = 1.00) and the
SPLAT and untreated treatments (F = 3.31; df = 4,3; p = 0.65) in the number of visits/min to the
feeders as shown in table 14. There also were no differences between trials or between the north
and south feeders. The number of foragers trained to each feeder for each trial did not have an
effect on the number of forager visits per minute.

Feeding Duration
There were no significant differences between the full Polynate and untreated treatments
(F = 1.59; df = 4,3; p = 0.87) the half Polynate and untreated treatments (F = 0.55; df = 4,3; p =
0.99) the quarter Polynate and untreated treatments (F = 0.21; df = 4,3; p = 1.00) and the SPLAT
and untreated treatments (F = 0.20; df = 4,3 ; p = 1.00) in the length of time spent feeding per
visit. The location of the feeder (north or south) also did not predict the amount of time a honey
bee spent feeding (Table 14). The foraging intensity count per minute was not significant in
predicting the feeding duration. The time honey bees spent feeding was affected by the trial (day
that the trial was conducted).

Table 14. The mean count of honey bee foragers per minute and the mean time spent feeding ±
SEM for each treatment.
Treatment
Forager Intensity
Feeding Duration (s)
(visits/min)
Untreated
24.01 ± 2.21
57.17 ± 1.85
Full Rate

29.60 ± 7.78

52.10 ± 3.28

Half Rate

20.36 ± 3.48

55.12 ± 2.92

52

Quarter Rate

Table 14 (cont’d)
26.38 ± 4.41

48.25 ± 1.31

SPLAT

17.69 ± 2.50

62.68 ± 5.97

Discussion
None of the Nasonov based dispenser treatments increased the number of honey bee
visits to an artificial feeder or influenced the time each visitor spends feeding. This included
both the Polynate and SPLAT formulations, as well as reduced rates of the Polynate dispenser. It
was expected that a lower treatment rate would be more effective, but there was no significant
variation among the rates tested. The number of visits per minute did not affect the time a bee
spent feeding. It was observed that during times of high activity, a bee was often pushed out of
the way by a newcomer and thus spent less time feeding.
Honey bees are more likely to expose their Nasonov glands to mark water or an artificial
feeder than to mark flowers (Free and Williams 1970; Free and Williams 1972). This means that
even if Polynate was observed to increase the attractiveness of on one feeder over the other, the
same observation may not be made in the field. In one study, honey bees did scent-mark flowers
that they had visited, but this decreased the chance of another honey bee visiting the same
flower. Perhaps this behavior warned other bees that the flower had already been visited and
been depleted of nectar (Giurfa and Núñez 1992). Although the particular ‘scent’ the honey bees
used to mark the flowers was unclear, they may have been referring to Nasonov pheromone
which could mean that it has a repellent nature instead of an attractant one in the field. This
could help explain why Polynate did not increase the attractiveness of an artificial feeder.
However, no difference between the Polynate treated feeder and the control feeder was observed.
It is thought that honey bees can change the concentration of the Nasonov components
during emission, which could in turn alter the reactions to other bees to it (Winston 1987). This
53

seems to be plausible since an earlier study found that honey bees did not show aversion to
flowers that had been recently visited (Ribbands 1949). Since the concentration of the Polynate
dispenser components was fixed for a particular treatment (full, half or quarter rate), it could be a
reasonable explanation as to why honey bees did not respond to the dispenser as they do to their
own emissions.
It is also possible that activity in the hoop house was too high to determine a difference
between the two feeders. Because there were no other food sources besides the three feeders it is
likely that honey bees foraged on them equally simply because there was no other option for
them to forage on. When there is a food shortage, honey bees tend to be less discriminatory
against food sources, visiting both high and low rewarding flowers (von Frisch 1950; Seeley
1989). They simply visit what is there because there is nothing else for them to do. In fact, the
number of foragers is known to alter the attractiveness of a food source (Waller 1972). It would
be interesting to see what would happen if additional food sources were provided or if the whole
apparatus was moved outside of the hoop house. This way, there would still be some control
over the bees, but the set up would be more natural to them so behaviors would more closely
mimic what happens in the field.
It would also be beneficial to replicate this study without training the bees to feeders
before each trial. If Polynate works by attracting foragers, it would make sense to allow the
foragers to find the artificial feeders on their own. It may take a longer period of time for the
bees to find the feeders if they are able to find them at all. In fact, honey bees tend to have a
hard time finding a feeder that is unscented (Tautz and Sandeman 2003; Riley et al. 2005). Thus,
in the absence of training, there may be a more noticeable difference in foraging intensity

54

between treatments. Since honey bees are not trained to a food source in nature, not training
them would more closely replicate what occurs in field conditions.
Through this research, it has been determined that Nasonov based dispensers do not
impact the foraging of honey bees. In fact, these results support my earlier conclusion that
Nasonov based dispensers do not change honey bee behavior such that an increase in fruit set or
forager activity in the field is increased (Chapter two). Polynate is not the only product that has
failed to increase yield in a crop (Chapter one). The biology of honey bees is extremely complex
and they use pheromones differently than many other insect species. Instead of using one
pheromone to elicit one response, they use numerous pheromones in different combinations to
change minute behaviors in the hive. For example, footprint pheromones are thought to increase
the attractiveness of Nasonov pheromone (Winston 1987). These results combined with the
knowledge of honey bee biology leads me to conclude that simply putting one pheromone
containing device in front of honey bees and expecting them to respond in a specific way is not
practical.

Perhaps a combination of pollination attractants and other pollination enhancing

practices will be able to produce the results that growers desire.

55

CHAPTER 4
The Use of Polynate and SPLAT in Agriculture
Throughout this thesis, the effects of potential products for increasing honey bee
attraction to fruit crops were evaluated and determined to be minimal. Polynate™ and SPLAT
had no significant effect on fruit set or pollinator activity in blueberries, apples or cherries. They
also did not increase the number of honey bee visits to an artificial feeder nor alter the amount of
time each forager spent feeding. It is possible that Polynate or SPLAT have effects that were not
measured, but studies provided little evidence that they could produce an outcome that is
economically beneficial for growers.

Despite these discouraging results, there is still a

possibility that another product or attractants produce positive results. However, it may also be
best to pursue other pollination enhancement options.
Thus far, pollination products such as Beelure®, Beeline®, Bee-Here®, Pollenaid®,
Pollenaid-D®, Bee-Scent®, Fruit Boost® have been largely unsuccessful in increasing fruit set
and pollinator counts in numerous different crops (Burgett and Fisher 1979; Mayer and Johansen
1982; Rajotte and Fell 1982; Tew and Ferree 1984; Mayer, Lunden, and Britt 1989; Elmstrom
and Maynard 1991; Naumann et al. 1994; Schultheis et al. 1994; Higo, Winston, and Slessor
1995; Tew and Ferree 1998; Malerbo-Souza, Nogueira-Couto, and Couto 2004; Ellis and
Delaplane 2009).

Additionally, there is evidence to suggest that the addition of volatile

compounds can’t make a completely unattractive flower, such as one that has a poor amount of
nectar for example, attractive. Unfortunately, crops that growers are most interested in applying
attractant pollination enhancement chemicals are ones that have flowers that are considered to be
unattractive to bees. These attractants may increase visitation but an increase in pollinator
activity does not directly translate to an increase in pollination or fruit set (Currie et al. 1992;

56

Delaplane and Mayer 2000). Even if a bee’s pollination efficiency is improved, an increase in
crop yield may not occur because of potential pollination thresholds (Currie et al. 1992). Even if
an increase in crop yield is observed, the number of additional fruit may not be high enough to
increase revenue to a level that would at minimum pay for the application of the attractant.
Polynate and other potential pollination attractants should still be studied however,
especially if they alter pollinator behavior in subtle ways. Since honey bees are so sensitive to
their surroundings, it is possible that Nasonov based dispensers cause behavioral responses that I
was unable to measure in the field or in the controlled study. The same could be true for other
potential attractants as well. However if the behavioral responses are not measureable, and do
not have the potential to increase yield and revenue, it would be hard to justify private industry
pursuing them or to procure funds for research (Currie et al. 1992).
Despite the current findings, there are still avenues of research to be pursued for Polynate
and other potential pollination enhancement devices. Some studies on scent guides have stressed
the importance of placing the test scent in the hive as well as in the field (von Frisch 1950). This
way, foragers will experience the scent of Polynate before they forage thus encouraging them to
seek the same scent in the field. It would be interesting to see if one could cause Polynate to be
attractive simply by placing a couple of dispensers in hives as well as in the field or next to an
artificial feeder. Although this would not guarantee an increase in pollination, it might increase
the forager frequency within the treated areas.
Also, the hoop house study could be replicated providing honey bees with a variety of
natural food sources instead of artificial feeders. It would also be interesting to see how honey
bees would react to the same experimental design, but outside of the hoop house. The results
may also differ without training the bees to feeders before each trial. In nature, not only do

57

honey bees have to choose between numerous different resources, but they have to find them on
their own. Additionally, it was impossible to determine if Nasonov based dispensers had any
adverse effects on other essential pollinators in the field; it would be interesting to perform
another similar hoop house study with colonies of native bees.
In the first chapter of this thesis, I contrasted the activity of honey bees to that of native
bees. While native bees may be more efficient pollinators, they may not been abundant and are
also harder and more expensive to manage than honey bees. Since honey bees are easier to
manage, they also are the most cost effective organism to use during bloom to increase
pollination. It is still possible to manage some native bee species, even if it is more difficult
(Delaplane and Mayer 2000). Native bee habitat conservation is also important in preserving
pollination services. It may be more advantageous to pursue this route instead of developing
more honey bee attractants, or it is entirely possible that the combination of pollination
attractants and native bee conservation would result in the increase of pollination services.
Due to the recent bee declines all over the world, a solution to the corresponding decrease
in crop pollination has been sought. One solution is applying possible pollination attractants like
Polynate in crops. However, the intricacies of honey bee biology are often overlooked. Worker
bees provide pollination services in the most energy efficient ways possible until they die. It is
unlikely that chemical attractants will entice pollinators to work harder simply by making target
crops more attractive. Pollination attractants may work partially but it is impractical to think that
one pheromone based product will solve the pollination problem. There are many decision
factors such as energy profitability and foraging background that go into honey bee choices and
behaviors (Shafir 2011) and their pheromones are not used for any one purpose, but many

58

(Winston 1987). Due to these complexities, pollination attractants may be useful, but will not
likely contribute substantially to solving the pollination crisis as bee numbers decline.

59

APPENDICES

60

APPENDIX 1
RECORD OF DEPOSITION OF VOUCHER SPECIMENS
The specimens listed below have been deposited in the named museum as samples of those
species or other taxa, which were used in this research. Voucher recognition labels bearing the
voucher number have been attached or included in fluid preserved specimens.
Voucher Number: ____2014-06____

Author and Title of thesis:
Author: Julie Adams
Title: The Effect of Nasonov Based Dispensers on Honey Bee Behavior and on
Pollination in Blueberry, Apple, and Cherry
Museum(s) where deposited:
Albert J. Cook Arthropod Research Collection, Michigan State University (MSU)

Specimens:
Family

Genus-Species

Life Stage

Quantity

Preservation

Apidae

Apis mellifera

adult

10

pinned

61

APPENDIX 2
ADDITIONAL TABLES AND FIGURES

Table 15. The organization of field sites, treatment plots and sampling in 2012

Crop
Blueberry
Apple
Cherry

Field sites
N
Area
(ha)

N

Area
(ha)

N Sampling
locations/plot

4
3
2

2
3
2

0.81
0.81
0.81

3
3
3

1.6
2.4
1.6

Plots

Pollinator activity
N sites N trees
4
3
2

10
12
20

N sites

Fruit set
Dimensions

Limbs

4
Na
2

4X1
Na
5X2

2
Na
4

N sites

Fruit set
Dimensions

Limbs

4
2
5

4X4
5X5
5X5

2
2
2

Table 16. The organization of field sites, treatment plots and sampling in 2013

Crop
Blueberry
Apple
Cherry

Field sites
N
Area
(ha)

N

Area
(ha)

N Sampling
locations/plot

4
3
5

3
2
2

0.4
0.6
0.6

1
1
1

6.1
4
4

Plots

Pollinator activity
N sites N trees
4
3
5

62

20
24
24

Figure 2. An example of plot and sampling location arrangement within each block for field
experiments in 2012. Dark grey: treated plot, light grey: untreated plots, white box: sampling
locations, black box: honey bee hives.

Figure 3. An example of plot and sampling location arrangement within each block for field
experiments in 2013. Dark grey: treated, light grey: untreated plots, white box: sampling
locations, dotted white: unused area, black box: honey bee hives.

63

Figure 4. A single SPLAT dollop applied in cherry

Figure 5. A single Polynate dispenser applied in blueberry

64

Figure 6. The hoop house used in my experiments without the end partitions (pictured in figure
8)

Figure 7. An experimental feeder during a trial

Figure 8. An experimenter videotaping one experimental feeder and the set-up of the hive and
stock feeder

65

BIBLIOGRAPHY

66

BIBLIOGRAPHY
Aizen, M. A., L. A. Couto, S. A. Cunningham, and A. M. Klein. 2008. “Long-Term Global
Trends in Crop Yield and Production Reveal No Current Pollination Shortage but
Increasing Pollinator Dependency.” Current Biology 18 (20): 1572–75.
doi:10.1016/j.cub.2008.08.066.
Allsopp, and R. H. Cherry. 1991. “Attraction of Apis Mellifera L. (hymenoptera: Apidae) to
Volatile Compounds.” Australian Journal of Entomology 30 (3): 219–20.
doi:10.1111/j.1440-6055.1991.tb00416.x.
Allsopp, de Lange, and Veldtman. 2008. “Valuing Insect Pollination Services with Cost of
Replacement.” PLoS ONE 3 (9): e3128.
Beekman, M., and F. L. W. Ratnieks. 2000. “Long-Range Foraging by the Honey-Bee, Apis
Mellifera L.” Functional Ecology 14 (4): 490–96. doi:10.1046/j.13652435.2000.00443.x.
Beggs, K. T., K. A. Glendining, N. M. Marechal, V. Vergoz, I. Nakamura, Keith N. Slessor, and
Alison R. Mercer. 2007. “Queen Pheromone Modulates Brain Dopamine Function in
Worker Honey Bees.” Proceedings of the National Academy of Sciences 104 (7): 2460–
64. doi:10.1073/pnas.0608224104.
Boch, R., and D. A. Shearer. 1962. “Identification of Geraniol as the Active Component in the
Nassanoff Pheromone of the Honey Bee.” Nature 194 (4829): 704–6.
doi:10.1038/194704b0.
Boch, R., and D. A. Shearer. 1964. “Identification of Nerolic and Geranic Acids in the Nassanoff
Pheromone of the Honey Bee.” Nature 202 (4929): 320–21. doi:10.1038/202320a0.
Brault, A., and D. de Oliveira. 1995. “Seed Number and an Asymmetry Index of `McIntosh’
Apples.” HortScience 30 (1): 44–46.
Burgett, M. 2011. “Pacific Northwest Honey Bee Pollination Economics Survey 2010”. Oregon
State University.
Burgett, M., and G. Fisher. 1979. “An Evaluation of Beeline As a Pollinator Attractant on Red
Clover.” American Bee Journal 119 (5): 356–57.
“CSBA Pollination Survey Results.” 2013. California State Beekeeper’s Association.
Currie, R. W., M. L. Winston, and K. N. Slessor. 1992a. “Effect of Synthetic Queen Mandibular
Pheromone Sprays on Honey Bee (Hymenoptera: Apidae) Pollination of Berry Crops.”
Journal of Economic Entomology 85 (4): 1300–1306.

67

Currie, R.W., M. L. Winston, and K. N. Slessor. 1992b. "Effect of Synthetic Queen Mandibular
Pheromone Sprays on Pollination of Fruit Crops by Honey Bees (Hymenoptera: Apidae)."
Journal of Economic Entomology 85 (4): 1293-1299.
Dag, A. 2011. “Crop Pollination In Modern Agriculture.” In All Flesh Is Grass, edited by Z.
Dubinsky and J. Seckbach, 163–81. Cellular Origin, Life in Extreme Habitats and
Astrobiology 16. Springer Netherlands.
Delaplane, K. S., and Mayer. 2000. Crop Pollination by Bees.
Dulta, P. C., and L. R. Verma. 1987. “Role of Insect Pollinators on Yield and Quality of Apple
Fruit.” Indian Journal of Horticulture 44 (3and4): 274–79.
Dyer, F. C. 2002. “The Biology of the Dance Language.” Annual Review of Entomology 47 (1):
917–49. doi:10.1146/annurev.ento.47.091201.145306.
El-Agamy, S. Z A., W. B. Sherman, and P. M. Lyrene. 1979. “Effect of Mating System on Time
of Ripening and Fruit Weight in Blueberry.” Proceedings of the ... Annual Meeting of the
Florida State Horticultural Society 92: 258–59.
Ellis, A., and K. Delaplane. 2009. “An Evaluation of Fruit-Boost as an Aid for Bee Honey Bee
Pollination under Conditions of Competing Bloom.” Journal of Apicultural Research and
Bee World 48 (1): 15–18. doi:10.3896.
Elmstrom, G. W., and D. N. Maynard. 1991. “Attraction of Honey Bees to Watermelon with Bee
Attractant.” Proceedings of the ... Annual Meeting of the Florida State Horticulture
Society 103 (June): 130–33.
Ferguson, A. W., and J. B. Free. 1981. “Factors Determining the Release of Nasonov Pheromone
by Honeybees at the Hive Entrance.” Physiological Entomology 6 (1): 15–19.
doi:10.1111/j.1365-3032.1981.tb00255.x.
Free, J. B. 1965. “Attempts to Increase Pollination by Spraying Crops with Sugar Syrup.”
Journal of Apicultural Research 4: 61–64.
Free, J. B. 1968a. “The Conditions under Which Foraging Honeybees Expose Their Nasonov
Glands.” J. Apic. Res.
Free, J. B. 1968b. “Dandelion as a Competitor to Fruit Trees for Bee Visits.” Journal of Applied
Ecology 5 (1): 169–78. doi:10.2307/2401281.
Free, J. B. 1970. "Insect Pollination of Crops". Academic Press, London & New York.
Free, J. B. 1987. “Pheromones of Social Bees.” Chapman and Hall. IBSN 0-412-24740-02

68

Free, J. B., J. A. Pickett, and A. W. Ferguson. 1983. “Effect of the Components of the Nasonov
Pheromone on Its Release by Honeybees at the Hive Entrance.” Journal of Apicultural
Research 22 (3): 155–57.
Free, J. B., J. A. Pickett, A. W. Ferguson, J. R. Simpkins, and C. Williams. 1984. “Honeybee
Nasonov Pheromone Lure.” Bee World v. 65(4) p. 175-181
.
Free, J. B., and I. H. Williams. 1970. “Exposure of the Nasonov Gland by Honeybees (Apis
Mellifera) Collecting Water.” Behaviour 37 (3/4): 286–90.
Free, J. B., and I. H. Williams. 1972. “The Role of the Nasonov Gland Pheromone in Crop
Communication by Honeybees (Apis Mellifera L.).” Behaviour 41 (3/4): 314–18.
Ghazoul, J. 2005. “Buzziness as Usual? Questioning the Global Pollination Crisis.” Trends in
Ecology & Evolution 20 (7): 367–73. doi:10.1016/j.tree.2005.04.026.
Giurfa, M., and J. A. Núñez. 1992. “Honeybees Mark with Scent and Reject Recently Visited
Flowers.” Oecologia 89 (1): 113–17. doi:10.2307/4219856.
Hamilton, D. W., P. H. Schwartz, and B. G. Townshend. 1970. “Capture of Bumble bees1 and
Honey Bees1 in Traps Baited with Lures to Attract Japanese Beetles2,3,4.” Journal of
Economic Entomology 63 (5): 1443–45.
Hazelhoff, E. H. 1941. “Luchtverversching in de Bijenkast in de Zomer.” Bijenhouden, January.
Higo, H. A., M. L. Winston, and K. N. Slessor. 1995. “Mechanisms by Which Honey Bee
(Hymenoptera: Apidae) Queen Pheromone Sprays Enhance Pollination.” Annals of the
Entomological Society of America 88 (3): 366–73.
Hoover, S. E. R., C. I. Keeling, M. L. Winston, and K. N. Slessor. 2003. “The Effect of Queen
Pheromones on Worker Honey Bee Ovary Development.” Naturwissenschaften 90 (10):
477–80. doi:10.1007/s00114-003-0462-z.
Javorek, S. K., K. E. Mackenzie, and S. P. Vander Kloet. 2002. “Comparative Pollination
Effectiveness Among Bees (Hymenoptera: Apoidea) on Lowbush Blueberry (Ericaceae:
<I>Vaccinium angustifolium</I>).” Annals of the Entomological Society of
America 95 (3): 345–51. doi:10.1603/0013-8746(2002)095[0345:CPEABH]2.0.CO;2.
Jay, S. C. 1986. “Spatial Management of Honey Bees on Crops.” Annual Review of Entomology
31 (1): 49–65. doi:10.1146/annurev.en.31.010186.000405.
Kaissling, K. 1971. “Insect Olfaction.” In Olfaction, edited by L. M. Beidler, 351–431.
Handbook of Sensory Physiology 4 / 1. Springer Berlin Heidelberg.
Klein, A. M, B. E. Vaissière, J. H Cane, I. Steffan-Dewenter, S. A. Cunningham, C. Kremen, and
T. Tscharntke. 2007. “Importance of Pollinators in Changing Landscapes for World
69

Crops.” Proceedings of the Royal Society B: Biological Sciences 274 (1608): 303–13.
doi:10.1098/rspb.2006.3721.
Ladd, T. L. R., M. Beroza, and T. P. McGovern. 1974. “Attraction of Bumble bees and Honey
Bees to Traps Baited with Lures for the Japanese Beetle.” Journal of Economic
Entomology, no. 2 (April): 307–8.
Ledoux, M. N., M. L. Winston, H. Higo, C. I. Keeling, K. N. Slessor, and Y. LeConte. 2001.
“Queen and Pheromonal Factors Influencing Comb Construction by Simulated Honey
Bee (Apis Mellifera L.) Swarms.” Insectes Sociaux 48 (1): 14–20.
doi:10.1007/PL00001738.
Lyrene, P. M. 1989. “Pollen Source Influences Fruiting of ‘Sharpblue’ Blueberry.” Journal of
the American Society for Horticultural Science v. 114(6) p. 995-999 (November).
MacKenzie, K. E. 1997. “Pollination Requirements of Three Highbush Blueberry (Vaccinium
Corymbosum L.) Cultivars.” Journal of the American Society for Horticultural Science
122 (6): 891–96.
Malerbo-Souza, D. T., R. H. Nogueira-Couto, and L. A. Couto. 2004. “Honey Bee Attractants
and Pollination in Sweet Orange, Citrus Sinensis (L.) Osbeck, Var. Pera-Rio.” Journal of
Venomous Animals and Toxins Including Tropical Diseases 10 (2): 144–53.
doi:10.1590/S1678-91992004000200004.
Mautz, D., R. Boch, and R. A. Morse. 1972. “Queen Finding by Swarming Honey Bees.” Annals
of the Entomological Society of America 65 (2): 440–43.
Mayer, D. F., and C. A. Johansen. 1982. “Field Evaluation of Chemical Pollinator Attractants on
Tree Fruits.” American Bee Journal 122 (4): 287–89.
Mayer, D. F., J. D. Lunden, and R. l. Britt. 1989. “Evaluation of BeeScent as a Honey Bee
Attractant.” American Bee Journal 129 (1): 41–42.
McGregor. 1976. Insect Pollination of Cultivated Crop Plants. USDA.
McIndoo, N. E. 1914. “The Scent-Producing Organ of the Honey Bee.” Proceedings of the
Academy of Natural Sciences of Philadelphia 66 (2): 542–55.
Morse, R. A., and R. Boch. 1971. “Pheromone Concert in Swarming Honey Bees (Hymenoptera:
Apidae).” Annals of the Entomological Society of America 64 (6): 1414–17.
Morse, R., and N. Calderone. 2000. “The Value of Honey Bees as Pollinators of U.S. Crops in
2000.” Bee Culture.

70

Naumann, K., M. L. Winston, K. N. Slessor, and M. J. Smirle. 1994. “Synthetic Honey Bee
(Hymenoptera: Apidae) Queen Mandibular Gland Pheromone Applications Affect Pear
and Sweet Cherry Pollination.” Journal of Economic Entomology 87 (6): 1595–99.
Nordlund, D. A., and W. J. Lewis. 1976. “Terminology of Chemical Releasing Stimuli in
Intraspecific and Interspecific Interactions.” Journal of Chemical Ecology 2 (2): 211–20.
doi:10.1007/BF00987744.
Núñez, J. 1970. “The Relationship between Sugar Flow and Foraging and Recruiting Behaviour
of Honey Bees (Apis Mellifera L.).” Animal Behaviour 18, Part 3 (August): 527–38.
doi:10.1016/0003-3472(70)90049-7.
Pankiw, T. 2004. “Cued in: Honey Bee Pheromones as Information Flow and Collective
Decision-Making.” Apidologie 35 (2): 217–26. doi:10.1051/apido:2004009.
Pankiw, T., Z. Y. Huang, M. L. Winston, and G. E. Robinson. 1998. “Queen Mandibular Gland
Pheromone Influences Worker Honey Bee (Apis Mellifera L.) Foraging Ontogeny and
Juvenile Hormone Titers.” Journal of Insect Physiology 44 (7–8): 685–92.
doi:10.1016/S0022-1910(98)00040-7.
Partap, U., and T. Ya. 2012. “The Human Pollinators of Fruit Crops in Maoxian County,
Sichuan, China.” Mountain Research and Development 32 (2): 176–86.
doi:10.1659/MRD-JOURNAL-D-11-00108.1.
Pickett, J. A., I. H. Williams, A. P. Martin, and M. C. Smith. 1980. “Nasonov Pheromone of the
Honey bee,Apis Mellifera L. (Hymenoptera: Apidae).” Journal of Chemical Ecology 6
(2): 425–34. doi:10.1007/BF01402919.
Potts, S. G., J. C. Biesmeijer, C. Kremen, P. Neumann, O. Schweiger, and W. E. Kunin. 2010.
“Global Pollinator Declines: Trends, Impacts and Drivers.” Trends in Ecology &
Evolution 25 (6): 345–53. doi:10.1016/j.tree.2010.01.007.
Rajotte, E. G., and R. D. Fell. 1982. “A Commercial Bee Attractant Ineffective in Enhancing
Apple Pollination.” HortScience 17: 230–31.
Ribbands, C. R. 1949. “The Foraging Method of Individual Honey-Bees.” The Journal of Animal
Ecology 18 (1): 47. doi:10.2307/1581.
Riley, J. R., U. Greggers, A. D. Smith, D. R. Reynolds, and R. Menzel. 2005. “The Flight Paths
of Honeybees Recruited by the Waggle Dance.” Nature 435 (7039): 205–7.
doi:10.1038/nature03526.
Robinson, W. S., R. Nowogrodzki, and R. A. Morse. 1989. “The Value of Honey Bees as
Pollinators of U.S. Crops. II.” American Bee Journal v. 129(7) p. 477-487 (July).

71

Schmid-Hempel, P., A. Kacelnik, and A. I. Houston. 1985. “Honeybees Maximize Efficiency by
Not Filling Their Crop.” Behavioral Ecology and Sociobiology 17 (1): 61–66.
doi:10.1007/BF00299430.
Schmidt, J. O. 1994. “Attraction of Reproductive Honey Bee Swarms to Artificial Nests by
Nasonov Pheromone.” Journal of Chemical Ecology 20 (5): 1053–56.
doi:10.1007/BF02059741.
Schmidt, J. O. 1994. 1999. “Attractant or Pheromone: The Case of Nasonov Secretion and
Honeybee Swarms.” Journal of Chemical Ecology 25 (9): 2051–56.
doi:10.1023/A:1021084706241.
Schmidt, J. O., K. N. Slessor, and M. L. Winston. 1993. “Roles of Nasonov and Queen
Pheromones in Attraction of Honeybee Swarms.” Naturwissenschaften 80 (12): 573–75.
doi:10.1007/BF01149276.
Schmidt, J. O., and S. C. Thoenes. 1987. “Swarm Traps For Survey and Control of Africanized
Honey Bees.” Bulletin of the ESA 33 (3): 155–58.
Schultheis, J. R., J. T. Ambrose, S. B. Bambara, and Wyatt A. Mangum. 1994. “Selective Bee
Attractants Did Not Improve Cucumber and Watermelon Yield.” HortScience 29 (3):
155–58.
Seeley, T. D. 1989. “Social Foraging in Honey Bees: How Nectar Foragers Assess Their
Colony’s Nutritional Status.” Behavioral Ecology and Sociobiology 24 (3): 181–99.
doi:10.1007/BF00292101.
Shafir, S. 2011. “Bee Cognition And Crop Pollination: Proven And Potential Applications.” In
All Flesh Is Grass, edited by Z. Dubinsky and J. Seckbach, 183–98. Cellular Origin, Life
in Extreme Habitats and Astrobiology 16. Springer Netherlands.
Shearer, D. A., and R. Boch. 1966. “Citral in the Nassanoff Pheromone of the Honey Bee.”
Journal of Insect Physiology 12 (12): 1513–21. doi:10.1016/0022-1910(66)90041-2.
Snodgrass, R. E. 1984. Anatomy of the Honey Bee. Cornell University Press.
Southwick, E. E., G. M. Loper, and S. E. Sadwick. 1981. “Nectar Production, Composition,
Energetics and Pollinator Attractiveness in Spring Flowers of Western New York.”
American Journal of Botany 68 (7): 994–1002. doi:10.2307/2443231.
Southwick, E. E., and L. Southwick. 1992. “Estimating the Economic Value of Honey Bees
(Hymenoptera: Apidae) as Agricultural Pollinators in the United States.” Journal of
Economic Entomology 85 (3): 621–33.
Steinhauer, N. A., K. Rennich, M. E. Wilson, D. M Caron, E. J. Lengerich, J. S. Pettis, R. Rose,
et al. 2014. “A National Survey of Managed Honey Bee 2012-2013 Annual Colony
72

Losses in the USA: Results from the Bee Informed Partnership.” Journal of Apicultural
Research 53 (1): 1–18. doi:10.3896/IBRA.1.53.1.01.
Stephenson, A. G. 1981. “Flower and Fruit Abortion: Proximate Causes and Ultimate
Functions.” Annual Review of Ecology and Systematics 12: 253–79.
Stokstad, E. 2007. “The Case of the Empty Hives.” Science. 316: 970-972.
Strauss, K., H. Scharpenberg, Robin M. Crewe, F. Glahn, H. Foth, and R. F. A. Moritz. 2008.
“The Role of the Queen Mandibular Gland Pheromone in Honey Bees (Apis Mellifera):
Honest Signal or Suppressive Agent.” Behavioral Ecology and Sociobiology 62: 1523–
31.
Tautz, J., and D. C. Sandeman. 2003. “Recruitment of Honeybees to Non-Scented Food
Sources.” Journal of Comparative Physiology A 189 (4): 293–300. doi:10.1007/s00359003-0402-6.
Tew, J., and D. Ferree. 1984. “The Effects of Bee Lure on Honey Bee (Apis Mellifera)
Pollination of Apples.” Ohio Agricultural Research and Development Center, no. 283: 2.
Tew, J., and D. Ferree. 1998. “The Influence of a Synthetic Foraging Attractant, Bee-ScentTM,
on the Number of Honey Bees Visiting Apple Blossoms and on Subsequent Fruit
Production.” Ohio State University Bulletin Fruit Crops: A Summary of Research.
Townsend-Mehler, J. M., F. C. Dyer, and K. Maida. 2011. “Deciding When to Explore and
When to Persist: A Comparison of Honeybees and Bumblebees in Their Response to
Downshifts in Reward.” Behavioral Ecology and Sociobiology 65 (2): 305–12.
doi:10.1007/s00265-010-1047-4.
van Praagh, and von der Ohe. 1983. “The Role of Scents in Pollination by the Honeybee (Apis
Mellifera).” Acta Horticulturae 139: 65–68.
vanEngelsdorp, D., Jay D. Evans, C. Saegerman, C. Mullin, E. Haubruge, B. K. Nguyen, M.
Frazier, et al. 2009. “Colony Collapse Disorder: A Descriptive Study.” PLoS ONE 4 (8):
e6481. doi:10.1371/journal.pone.0006481.
vanEngelsdorp, D., and M. D. Meixner. 2010. “A Historical Review of Managed Honey Bee
Populations in Europe and the United States and the Factors That May Affect Them.”
Journal of Invertebrate Pathology 103, Supplement (January): S80–S95.
doi:10.1016/j.jip.2009.06.011.
Volz, R. K., D. S. Tustin, and I. B. Ferguson. 1996. “Pollination Effects on Fruit Mineral
Composition, Seeds and Cropping Characteristics of ‘Braeburn’ Apple Trees.” Scientia
Horticulturae 66 (3–4): 169–80. doi:10.1016/S0304-4238(96)00934-X.

73

von Frisch, K. 1950. Bees; Their Vision, Chemical Senses, and Language. Cornell University
Press.
von Frisch, K. 1966. The Dancing Bees. Methuen and Co. Ltd.
von Frisch, K. 1967. The Dance Language and Orientation of Bees. Cambridge (Mass.);
London: Harvard university press.
Waller, G. D. 1970. “Attracting Honeybees to Alfalfa with Citral, Geraniol and Anise.” J Apicult
Res, no. 1: 9–12.
Waller, G. D. 1972. “Evaluating Responses of Honey Bees to Sugar Solutions Using an
Artificial-Flower Feeder.” Annals of the Entomological Society of America 65 (4): 857–
62.
Wells, P., H. Wells, V. Vu, N. Vadehra, C. Lee, R. Han, K. Han, and L. Chang. 1993. “Does
Honey Bee Nasanov Pheromone Attract Foragers?” Bulletin of the Southern California
Academy of Sciences 92 (2): 70–77.
Wenner, A. M., and P. H. Wells. 1990. Anatomy of a Controversy: The Question of a Language
Among Bees. Columbia University Press.
Wilcock, C., and R. Neiland. 2002. “Pollination Failure in Plants: Why It Happens and When It
Matters.” Trends in Plant Science 7 (6): 270–77. doi:10.1016/S1360-1385(02)02258-6.
Williams, I. H. 1994. “The Dependence of Crop Production within the European Union on
Pollination by Honey Bees.” Agricultural Zoology Reviews no. 6.
Winston, M. L. 1987. The Biology of the Honey Bee. Harvard University Press.

74