..

.v, ‘ z.
u t . b...
: J. . Ix!

.I .a

 

“Luv.“
3......

r15

I...) w‘!) . . i’n?
:..LN?J.L«5. . . Snail.

 

:»
. I...
1.1:.

 

fl ‘. n33... .1

 

v.1. It.
.{tl}|.
.rlwv 1 . 4 13%

 

 

THESIS
Z

lllllllllllllllllllllllll‘lllll7 Illllllllllllllllllllll

3 1293 01555 2387

 

 

 

     

This is to certify that the

dissertation entitled

AN EVALUATION OF DOMESTIC CHICKENS AND GEESE AS

BIOLOGICAL CONTROL AGENTS FOR INSECT PESTS AND WEEDS
presented by

MICHAEL SEAN CLARK

has been accepted towards fulfillment
of the requirements for

Ph .D . degree in ENIOMDLDQL

SM 146““,

Major professél

Date 7 MAY 1998

MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN 3511"!" BOX to roman this checkout from your record.
TO AVOID FINES Mum on or baton data duo.

DATE DUE DATE DUE DATE DUE

     

  
  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Li MSU IsAn Aim-mum WIN-l Oppoflufmy Intuition
Wanna-9.1

__________________.._————— _

 

AN EVALUATION OF
DOMESTIC CHICKENS AND GEESE
AS BIOLOGICAL CONTROL AGENTS
FOR INSECT PESTS AND WEEDS

By
Michael Sean Clark

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY

Department of Entomology

I 996

ABSTRACT

AN EVALUATION OF
DOMESTIC CI-HCKENS AND GEESE
AS BIOLOGICAL CONTROL AGENTS
FOR INSECT PESTS AND WEEDS

By
Michael Sean Clark

This dissertation reports a study of free-range domestic chickens and geese as insect
pest and weed biological control agents. The study is comprised of four parts: 1) an
evaluation of the compatibility of the birds in an experimental, non-chemical
agroecosystem; 2) an assessment of the effectiveness of the birds as biological control
agents through an evaluation of their effects on insect pests, weeds, crop damage, and crop
productivity; 3) an assessment of the impact of the birds on beneficial soil invertebrates
including epigeic predators and earthworms; and 4) a survey of small farmers’ views on
using free-range chickens and geese as biological control agents. The four sub-studies
comprise chapters three through six, respectively.

Free-range domestic chickens and geese were evaluated as components of a non-
chemical, potato-intercropped, apple orchard. Both chickens and geese were found to be
compatible in the system but provided different benefits and had different requirements.
Chickens were omnivorous, highly active throughout the day, and dispersed throughout
the available area In contrast, geese were strictly herbivorous, less active, and usually

remained close to their coop and water source. Geese substantially reduced vegetation

biomass under the trees and around the potatoes without damaging either of the crops.
Chickens reduced non-crop vegetation biomass slightly but also were observed to consume
several insect species, including Japanese beetle and Colorado potato beetle, suggesting
that they could be used to control certain insect pests. Factors influencing the feasibility of
integrating domestic chickens or geese into agroecosystems are discussed.

The effects of the chickens and geese on insect pests and weeds were evaluated in
the same experimental, nonchemical agroecosystem. The objective was to assess the
potential of these domestic bird species as biological control agents. Four insect pests were
focused on in this study: plum curculio (Conotrachelus nenuphar), apple maggot
(Rhagoletis pomonella), Japanese beetle (Papillia japonica), and Colorado potato beetle
(Leptinotarsa decemlineata). Chickens were found to feed on several potential crop pests,
including Japanese beetle. Although Japanese beetle abundance on apple trees was reduced
in the presence of chickens, no reduction in the percentage of damaged fruit was found.
Possible explanations for the lack of a crop protection effect are discussed. Furthermore,
chickens were found to have no effect on weed abundance or crop productivity. In
contrast, geese were found to be effective weeders. Their feeding activities reduced weed
abundance and led to greater potato plant growth and yields compared to a minimally-
weeded control. In addition, the activities of geese indirectly resulted in a reduction in
apple fmit damage by plum curculio and a higher percentage of pest-free fruit. The
possible reasons for this effect are discussed. The potential compatibility of geese with
several vegetable crops and the economics of using geese as weed management agents are
assessed.

The effects of the chickens and geese on the abundance of beneficial soil
macroinvertebrates was assessed. Two groups of organisms were studied: epigeic
predators and earthworms. Pitfall traps were used to sample epigeic predators. The
presence of the birds resulted in a reduction in spider (Araneae) and harvestrnen (Opiliones)
activity, but no reduction in ground beetle (Carabidae) or rove beetle (Staphylinidae)

activity. The reason for these differential effects may be attributable to differences in the
diurnal activity patterns of these invertebrates. Overall, the effects of the birds on epigeic
predators were relatively minor in comparison to intercropping practices in the orchard. No
significant effects on earthworm abundance due to the birds was found. However,
earthworm abundance in the orchard and surrounding areas was highly correlated with soil
organic matter (r =0.78) and the relative distance of the sample from an old field which had
never been tilled (r =-O.77). A statistical model based on these two factors accounted for
most of the variance in earthworm abundance.

Six farmers with experience raising domestic birds were interviewed to determine
their perceptions of the feasibility of using chickens and geese as biological control agents.
The farmers generally believed that domestic birds had potential as biological control ‘
agents, however, none of these growers specifically used the birds for this purpose on a
regular basis. The reasons for this non-use were primarily due to a lack of efficient
management systems or grower experience. Many other reasons for keeping domestic
birds were mentioned: egg production, food and waste recycling, aesthetics, meat
production, manure fertilizer source, and general enjoyment. All of the farmers expressed
interest in better utilizing domestic birds for biological control. Future research directions

are suggested based upon these interviews.

ACKNOWLEDGMENTS

I would like to express my gratitude to the members of my graduate committee,
Mark Whalon, Cathy Bristow, Torn Edens, and especially Stuart Gage and Laura DeLind.
Their expertise and experience helped guide me through this research. Furthermore, their
openness to my ideas and trust in my abilities allowed me to "range freely" and accomplish
what I wanted. I owe thanks to the farmers who cooperated with me, Bruce Schultz and
the folks at Celebration Gardens, Markus Held, Quinn and Shelly Cumberworth, and the
farmers who provided me with their perspectives of domestic birds on their farms. The
interest of these farmers in my research and their willingness to give me their time, and in
some cases use their land, was as important to this project as the experiments at the Kellogg
Biological Station. I wish to acknowledge the folks who worked on the project during the
summers, Peter Bingen, Kristen Hobbie, and Hugh Atkinson. Finally, I owe a
tremendous amount of thanks to Stephanie and Niles Clark, not only for their patience in
allowing me to pursue this degree and complete the research, but also for contributing so

much of themselves to the project in so many ways! I could not have done it without them.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................... viii
LIST OF FIGURES .............................................................................. ix
INTRODUCTION ................................................................................. 1
CHAPTER 1 - AN ECOLOGICAL PERSPECTIVE ON MODERN
AGRICULTURE: A LITERATURE REVIEW ....................................... 4
SOIL RESOURCES ...................................................................... 5
WATER RESOURCES .................................................................. 6
ENERGY AND NUTRIENT RESOURCES .......................................... 7
BIOLOGICAL RESOURCES ........................................................... 9
AGRICULTURAL SUSTAINABILITY ............................................. 11
THE ROLES OF DOMESTIC ANIMALS IN AGRICULTURAL
SUSTAINABILITY .................................................................... 14
PEST MANAGEMENT AND AGRICULTURAL SUSTAINABILITY ........ 17
CHAPTER 2 - THE SPECIFIC CONTEXT OF THIS PEST MANAGEMENT
PROBLEM: A LITERATURE REVIEW ............................................. 21
CURRENT STATUS OF APPLE PEST MANAGEMENT ....................... 22
CURRENT STATUS OF POTATO PEST MANAGEMENT ..................... 3O
FREE-RANGE DOMESTIC BIRDS AS BIOLOGICAL CONTROL
AGENTS ................................................................................. 34
CHAPTER 3 - COMPATIBILITY OF DOMESTIC BIRDS WITH A
N ONCHEMICAL AGROECOSYSTEM ............................................. 37
INTRODUCTION ...................................................................... 37
MATERIAL AND METHODS ........................................................ 38
RESULTS ................................................................................ 41
DISCUSSION ........................................................................... 44

CHAPTER 4 - EVALUATION OF FREE-RANGE CHICKENS AND GEESE
FOR MANAGING INSECT PESTS AND WEEDS IN AN

AGROECOSYSTEM ................................................................... 58
INTRODUCTION ...................................................................... 58
MATERIALS AND METHODS ...................................................... 60
RESULTS ................................................................................ 66
DISCUSSION ........................................................................... 71
CHAPTER 5 — EFFECTS OF FREE-RANGE DOMESTIC BIRDS ON THE
ABUNDANCE OF BENEFICIAL SOIL MACROINVERTEBRATES ......... 94
INTRODUCTION ...................................................................... 94
MATERIALS AND METHODS ...................................................... 95
RESULTS ................................................................................ 98
DISCUSSION ........................................................................... 99
CHAPTER 6 - FARMER PERCEPTIONS OF FREE-RANGE DOMESTIC
BIRDS AS BIOLOGICAL CONTROL AGENTS ON SMALL FARMS ....... 110
INTRODUCTION ..................................................................... 110
MATERIALS AND METHODS ..................................................... 1 1 1
RESULTS AND DISCUSSION ..................................................... 112
CHAPTER 7 - SUMMARY AND CONCLUSIONS ....................................... 119
LIST OF REFERENCES ...................................................................... 126

LIST OF TABLES

Table 3-1. A list of food items observed to be consumed by chickens and geese in
the potato-intercropped, apple orchard in 1993 ...................................... 48

Table 4-1 . The abundance of common weed species based on percent groundcover
estimates (mean + SEM) taken on 30 May, 1995. The asterisks indicate
significant reductions in plant species abundance in the bird treatment
compared to the control based on Student’s t-test (**= 0.05 _>_ p > 0.01;

***= p 5 0.01). ......................................................................... 77

Table 4-2. A comparison of labor and material input costs ($U.S.) for nonchemical
weed management with and without weeder geese in a hypothetical 1 ha
apple orchard with intercropped potatoes ............................................. 78

Table 4-3. The diet components of free-range chickens in the experimental system
based on the presence / absence of items in digestive crop dissections from
two sampling periods; 6-19 June (N=10) and 19 July (N=12) .................... 79

Table 4-4. The results of a feeding trial, which tested the potential palatability of
vegetable crops (N=5 geese), and an unreplicated garden trial which
assessed feeding and trampling damage caused by weeder geese during a
four—day period. ......................................................................... 80

Table 5-1. Management histories of the agricultural habitats sampled for earthworm
abundance estimates .................................................................... 102

Table 5-2. Pearson correlations between earthworm abundance and soil texture,

organic matter, nutrient levels, cation exchange capacity, and distance from

an old field. ............................................................................. 103
Table 6-1. General descriptions of the farmers interviewed in this study ................ 116

Table 6-2. Questions used in interviews with farmers on role of domestic birds on
small farms as biological control agents. ............................................ 117

viii

LIST OF FIGURES

Figure 3-1. Map showing the experimental layout. Apple trees are designated P
and R for ‘Priscilla’ and ‘Redfree’ respectively. One randomly-selected
potato alley of each plot was mulched while the other was left unmulched. ..... 49

Figure 3-2. Chicken and goose activity patterns based on the mean percentage of
observations in which birds were observed to be resting, foraging, and
drinking. The time periods are morning (8:00arn-10:59am, N=5), mid-day
(1 1:00am-1:59pm, N=2), afternoon (2:00pm-4z59pm. N=2), and evening
(5:00pm-8:00pm, N=3). Error bars are standard errors of the mean ............. 50

Figure 3-3. Chicken and goose dispersion within the orchard plots as indicated by
the mean percentage of plot sectors occupied by birds at arbitrarily-selected
times of day over a one-month period (N=4 for each mean). Error bars are
standard errors of the mean. ........................................................... 51

Figure 34 Frequency of chicken and goose occurrence in the apple rows and
potato alleys as indicated by the mean percentage of observation periods in
which the birds were observed in these habitats. The time periods are
morning (8:00am-10:59am, N=5), mid-day (11:00am-lz59pm, N=2),
afternoon (2:00pm-4z59pm. N=2), and evening (5:00pm-8:00pm, N=3).
Error bars are standard errors of the mean. .......................................... 52

Figure 3-5. Frequency of chicken and goose occurrence in the plot half with the
coop and the half without the coop as indicated by the mean percentage of
observation periods in which the birds were observed in these areas. The
time periods are morning (8:00arn-10:59am, N=5), mid-day (11:00am-
l:59pm, N=2), afternoon (2:00pm-4z59pm. N=2), and evening (5:00pm-
8:00pm, N=3). Error bars are standard errors of the mean ........................ 53

Figure 3-6. Above-ground vegetation biomass (grams dry thmZ) in the tree rows
under the three treatments. Total biomass is broken down into grass and
broadleaf weeds. Asterisks indicate the level of statistical significance (see
Materials and Methods) of the difference between the Control and each bird
treatment (one-tailed Mann-Whitney test). Error bars are standard errors of
the mean. ................................................................................. 54

Figure 3-7. Above-ground weed biomass (grams dry wt./m2) in mulched and
unmulched potatoes under the three treatments. Asterisks indicate the level
of statistical significance (see Materials and Methods) of the difference
between the control and each bird treatment (one-tailed Mann-Whitney test).
Error bars are standard errors of the mean. .......................................... 55

Figure 3-8. Apple fruit yield (kg fresh wt./tree) for ‘Redfree’ trees under the three
treatments. Asterisks indicate the level of statistical significance (see
Materials and Methods) of the difference between the control and each bird
treatment (one-tailed Mann-Whitney test). Error bars are standard errors of
the mean. ................................................................................. 56

ix

Figure 3-9. Conceptualization of an integrated agroecosystem based on the
experimental agroecosystem in this study. The arrows indicate the potential
beneficial interactions between the components of the system ..................... 57

Figure 4—1. Layout of a single plot showing important features. Four of the six
apple trees within the dotted box in the center of each plot were designated
as sampling trees for all insect pest monitoring ...................................... 81

Figure 4—2. Percentage of apples damaged by plum curculio based on visual
monitoring during June, 1994 (mean + SEM) ....................................... 82

Figure 4-3. Apple maggot abundance based on the number of adults caught on four
red spheres per plot in 1994 (mean + SEM). ........................................ 83

Figure 44. Japanese beetle abundance based upon direct counts on the sampling
trees in 1994 (mean + SEM). .......................................................... 84

Figure 4—5. The percentage of harvested fruit damaged by insects (PC=plum
curculio, AM=apple maggot, JB=Japanese beetle, TPB=tarnished plant
bug, CM=codling moth) and undamaged (Pest-Free) in 1994 (mean +
SEM). Different letters indicate significant differences ( p g 0.06) ............... 85

Figure 4-6. Potato tuber yields (kg / ha) based on sampling 6m of row in 1994
(mean + SEM). Different letters indicate significant differences ( p 5 0.01). . .. 86

Figure 4-7. Colorado potato beetle (CPB) larvae and adult population dynamics in
the control, chicken, and goose treatments in 1994 (mean + SEM) ............... 87

Figure 4-8. Percent groundcover of weeds and potatoes in the control, chicken,
and goose treatments in 1994 (mean + SEM) ........................................ 88

Figure 4-9. The percentage of harvested fruit damaged by insects (PC=plum
curculio, AM=apple maggot, JB=Japanese beetle, TPB=tarnished plant
bug, CM=codling moth) and undamaged (Pest-Free) in 1995 (mean +
SEM). Asterisks indicate significant differences (p g 0.07) ....................... 89

Figure 4-10. Apple maggot abundance based on the number of adults caught on
four red spheres per plot in 1995 (mean + SEM) .................................... 90

Figure 4-11. Japanese beetle abundance based upon direct counts on the sampling
trees in 1994 (mean + SEM). .......................................................... 91

Figure 4-12. Colorado potato beetle (CPB) larvae and adult population dynamics in
the control and bird treatment in 1995 (mean + SEM). ............................. 92

Figure 4-13. Percent groundcover of weeds and potatoes in the Control and Bird
treatment (mean + SEM) ................................................................ 93

Figure 5-2. Plot diagram showing the general layout and the locations of the pitfall
traps (shown with arrows). Plots measured 16.8m X 9.9m and included
three rows of three trees and two inter-row alleys which were intercropped
with potatoes (mulched and unmulched). ........................................... 104

Figure 5-3. The relative abundance of predator taxa collected in pitfall traps in the
orchard from 2 July until 29 July, 1993 ............................................. 105

Figure 5—4. Mean abundance (i SEM) of predatory arthropods collected in pitfall
traps in main treatments (x axis) and subtreatrnents (legend). Different
letters indicate significant differences according to Tukey’s multiple range
test (p s 0.10) .......................................................................... 106

Figure 5-5. The relationship between earthworm abundance (number/m2) and
distance from old field and percent organic matter, based upon samples
taken from the apple orchard and nearby fields. Earthworm abundance was
negatively correlated with distance from old field (r=-O.77, p=0.003) and
positively correlated with organic matter (r=0.78, p=0.003). .................... 107

Figure 6-1. Mean (+SEM) responses by farmers to interview questions 1 and 3
-10. The original questions are summarized into statements (see entire
questions in Table 1) and the statistics were calculated based on ranks (see
text) ...................................................................................... 118

Figure 7-1. Diagram summarizing the detected (solid arrows) and suggested
(dashed arrows) ecological effects of geese and chickens in the experimental
agroecosystem. The effects are indicated as positive (+) or negative (-) for
each of the measured variables ........................................................ 125

xi

INTRODUCTION

Industrial-based agriculture has been both a blessing and burden to human society.
It has allowed, if not driven, rapid population growth in many parts of the world, the
development of specialized occupations other than food procurement, and the creation of
complex and amazing human civilizations. At the same time industrial agriculture has been
accompanied by human overpopulation and caused localized and global environmental
degradation. The current predicament facing modern industrial agriculture resides in the
fact that resource- and energy-intensive food production methods developed in recent
decades bring with them more than higher yields. The external costs have included
contamination of ground and surface waters with nutrients and pesticides, ecosystem and
wildlife destruction, human exposure to carcinogens, soil loss to erosion, and renewable
and non-renewable resource depletion. At the same time, the exponentially growing human
population will require still greater rates of food production. Can methods be developed
and utilized to meet the growing human food demand without wreaking havoc on our
global life support system?

The goal of this thesis project is to contribute to the development of ecologically-
based food production methods. The primary differences between industrial and ecological
approaches to food production are that the former is based upon the idea that problems in
agriculture can be solved with various types of manufactured inputs, such as pesticides and
fertilizers, while the later is based upon the idea that problems can be solved through an
understanding of ecological systems, through resource conservation, and by living within
reasonable limits. The industrial agricultural paradigm is clearly dominant in the United
States. The trend toward greater labor and capital efficiency and higher yields has resulted

in a system where less than 2% of the population grows the food for everyone and there is

2

still surplus to trade on the world market. No other country in the world has reached this
level of labor efficiency.

Ecological agriculture differs from the industrial approach in a fundamental way and
is typically exemplified by “organic farmers.” Although the primary goal is still to produce
food, other important goals include the replacement of purchased, off-farm inputs with
knowledge of farm ecology, systems integration for holistic resource management, and
minimization of waste and environmental degradation and pollution. Barriers to this type
of farming are social and economic as well as ecological in nature. Therefore, in addition
to a purely ecological analysis, I have attempted to integrate some social, economic, and
philosophical elements into my approach toward this research problem. The redesign of
agroecosystems for sustainable food production requires a multidisciplinary approach.
Although the primary focus of this research project has been entomological in nature, effort
has been made to give reasonable attention to the rest of the system as well so that the
results can be interpreted within a broader context than is generally possible with strictly
disciplinary research.

The overall objective of this research was to evaluate domestic birds as components
of ecologically-based agricultural systems. Domestic birds might serve several functions in
human-managed ecosystems including biological control agents, nutrient cyclers, and
scavengers. Like most microlivestock, domestic birds are relatively easy to manage,
require little capital investment, and are important as a food worldwide. Thus, there may
be potential for domestic birds to be used as biological control agents in some types of
agroecosystems. The specific objectives of this research project were:

1) To assess the compatibility of domestic birds in a diverse, ecologically-managed food
production system.
2) To measure the direct and indirect effects of domestic chickens and geese on insect

pests, weeds, and crops in an experimental integrated agroecosystem.

3

3) To evaluate the potential effects of domestic chickens and geese on beneficial non-target,
soil-dwelling invertebrates.
4) To assess the potential for domestic birds play a role in biological pest management

through interviews with experienced, small farmers.

CHAPTER 1

AN ECOLOGICAL PERSPECTIVE ON MODERN AGRICULTURE:
A LITERATURE REVIEW

Agriculture is dependent upon a variety of biotic and abiotic environmental
characteristics as well as local, regional, and global ecosystem processes. Climate is
probably the most important factor affecting agricultural systems over large spatial and
temporal scales (Rice and Vandermeer, 1990). Climate, or long-term rainfall and
temperature patterns, determine the possible range of farming systems for an area. Human
technology can alter this range but only within the limits of available renewable and non-
renewable resources. Soil characteristics are highly dependent upon climatic history.
Water, in its liquid and solid form, physically and chemically breaks down geological
parent material to create soil. Water also degrades soil by leaching nutrients from it. For
example, the soils of the tropics are much older (100,000 to 1 million years old) and less
fertile than the soils of the temperate region (10,000 to 40,000 years old) which were
formed mostly during the last glaciation period that ended 10,000 years ago. Water has
leached most the nutrients from tropical soils (Rice and Vandermeer, 1990).

Biotic factors affecting agriculture are also highly dependent upon climate. The
composition and effects of agricultural pests, including herbivorous insects, crop
pathogens, and native plant competitors, are determined to a large extent by climate. For
example, pest populations in the wet humid tropics are maintained at relatively consistent
levels throughout the year due to the lack of seasonality whereas in the temperate region
they are characterized by relatively consistent seasonal patterns and are only problematic
during the warmer months (Rice and Vandermeer, 1990). Agriculture is influenced by

local biotic landscape features as well. For example, the composition of surrounding

5

vegetation and the proximity of field edges have been commonly shown to influence pest
insect population levels and crop damage (Landis, 1994). Humans have adapted
agriculture to local and regional biotic and abiotic environmental factors through cropping
system modifications, crop and animal breeding, soil amendment additions, irrigation
systems, and pest suppression and management tactics.

In addition to being affected by the environment, agriculture has been the cause of
dramatic local, regional, and global ecological changes. Cultivated land is estimated to
cover approximately 16 million km2 or 10% of the 147 million km2 of global land surface
(Vitousek et al., 1986). Richards (1990) estimates that the amount of global land surface
used for growing crops has increased by 466% since 1700. In North America this increase
is estimated to be 6667%. Although global land conversion into cropland is accelerating,
there is estrnated to be less than 19 x 106 km2 of total land suitable for rainfed agriculture
(Meyer and Turner, 1992). In addition to displacing native ecosystems, including forests,
wetlands, and grasslands, agriculture is having substantial effects on soil, water, energy,

mineral, and biological resources on a global scale (Meyer and Turner, 1992).

SOIL RESOURCES

The negative effects of agriculture on soil are primarily due to water and wind
erosion and occasionally to contamination by agricultural chemicals. The vulnerability of
soils to erosion varies considerably yet it is a serious problem worldwide. The effects of
soil erosion include reduced soil productivity, increased fertilizer costs for compensation,
sediment deposition in surface waters which reduces suitability for food production,
recreation, and drinking, and eutrophication of surface waters (Soule et al., 1990). Even
with state and federal efforts to control soil erosion in the United States, it remains a
serious problem in some areas due to the use of continuous row cropping, fewer rotations
with perennials, price support programs, and the enlargement of farms without an increase

in the number of managers (National Research Council, 1989).

6

Soil contamination is sometimes caused by the accumulation of pesticides or the
products of pesticide breakdown. Most pesticides applied to crops end up in the soil and
may stay there for variable lengths of time depending upon the compound, soil abiotic and
biotic composition, and amount of rainfall. Pesticides are removed from soils through
breakdown by microorganisms and sunlight, water flow through the soil, wind, and
volatization (Edwards, 1993). The effects of pesticides on soil organisms vary
considerably among taxonomic groups and is dependent upon the chemical compound used
and the frequency of application. The effects on ecosystem functions include reductions in

soil respiration, organic matter degradation, and nutrient cycling (Edwards, 1993).

WATER RESOURCES

The effects of agriculture on water resources result partially from inadequate
management of soil resources. Agricultural lands contribute sediments, nutrients,
minerals, and pesticides to surface waters and nutrients and pesticides to groundwater. It is
estimated that 50 to 70% of nutrients, primarily nitrogen and phosphorus, reaching surface
waters are from agricultural lands (National Research Council, 1989). Nitrogen, in the
form of nitrate from fertilizers, is a common contaminant of drinking water and, in addition
to phosphorus, contributes to eutrophication in surface waters worldwide (Soule et al.,
1990; Spalding and Exner, 1993; Sharpley and Meyer, 1994). Phosphorus is usually
present in small quantities in surface waters and is the primary cause of eutrophication
when it is abundant. It generally binds to soil particles and is therefore not readily carried
into surface waters except where soil erosion is high.

Pesticides have been found in surface water and groundwater throughout the United
States (National Research Council, 1989). It is estimated that over 50% of the United
States population obtains its drinking water from groundwater and that 10.4% of
community wells and 4.2% of private wells have detectable levels of pesticides (Curtis et

al., 1993). Pesticides in groundwater are especially problematic because of the difficulty

7

and expense of decontarninating those water sources. Pesticides found in surface waters
are most common in areas with intensive row cropping of corn and soybeans (National
Research Council, 1989). The environmental effects of pesticides in aquatic ecosystems is
not well known. In general, aquatic organisms are more susceptible to pesticides than
those in terrestrial ecosystems (Edwards, 1993).

Irrigation, a seemingly harmless activity, has also resulted in the degredation and
depletion of soil and water resources. Clearly, irrigation has led to substantial increases in
crop productivity and allowed for production in areas that would otherwise be unsuitable.
At the same time it has led to salinization of ground and surface waters in many areas.
When water used for irrigation evaporates the dissolved salts remain and become
concentrated. These salts either remain in the soil or return to rivers and lakes as runoff.
Major rivers in the United States, including the Rio Grande and the Colorado, are severely
salinized by irrigation (National Research Council, 1989; Soule and Piper, 1992).
Depletion of groundwater reserves has also resulted from irrigation. About 25% of
renewable groundwater is being used faster than it naturally replenishes (Soule and Piper,
1992). Some fossil aquifers, which are non-renewable resources, are being rapidly
depleted as well. The familiar Ogallala aquifer, which lies beneath seven western states,
supports 40% of grain fed cattle in the United States. Yet, at current usage rates this
aquifer may be practically exhausted in less than 50 years (Soule and Piper, 1992).

ENERGY AND NUTRIENT RESOURCES

Industrial agriculture is highly dependent upon fossil fuel inputs. Although only 3-
6% of the total energy consumed in developed nations is used in food production, fossil
fuel use has contributed to dramatic yield increases over the last several decades (Pimentel
and Dazhong, 1990; Stout, 1993). In a historical analysis of corn production in the United
States, Pimentel and Dazhong (1990) found that yield has increased 3.5-fold since 1700

while energy inputs increased 15-fold. Thus, energy efficiency (output/input) has

8

decreased from 10.5 in 1700 to 2.5 in 1983. This type of analysis is somewhat misleading
in several respects. Loomis (1984) has criticized this analysis because it neglects to
consider the embodied energy of workers in the calculation of the energy of human labor.
Loomis states that the energy required for overall human care, for example child care,
recreation, and retirement, should be included in such analysis. Furthermore, this analysis
neglects to consider overall production efficiency (including solar energy inputs) and the
consequences for land use. When solar energy inputs are included in the analysis, modern
corn systems are more efficient at converting solar energy into food grain than primitive
systems (Pimentel and Dazhong, 1990). Furthermore, the amount of land needed for a
given yield is less when fossil-based inputs are used (Loomis and Connor, 1992).

The environmental concerns over energy use in agriculture include the effects of
fossil fuel combustion on the atmosphere, hydrosphere, and geosphere and uncertainty
over the future of agriculture as more nations adopt industrial production methods and
fossil fuel reserves are depleted. Although agricultural production utilizes a relatively small
fraction of total energy, food systems use 15-20% of the total energy consumed in
industrialized nations (Pimentel and Dazhong, 1990; Stout, 1993). Processing, packaging,
distribution, and preparation require more energy than production and are largely
responsible for the fact that more than 10 units of commercial energy are used for each unit
of nutritional energy consumed (Giampietro and Pimentel, 1992). In developing nations
only 3-5 units of energy are used for each unit of nutritional energy consumed. It has been
estimated that energy requirements could be cut in half if a greater proportion of the
population worked on farms; however, the cost of human labor has become so expensive
relative to fossil energy that such a change is considered economically impossible (Pimentel
and Dazhong, 1990).

Petroleum production, which peaked in 1979, will likely remain the dominant
source of fuel only for the next 2-3 decades (Stout, 1993). By the middle of the twenty-
first century petroleum reserves will be exhausted (Loomis and Connor, 1992). Although

9

several hundred years-worth of coal exists, it is bulky and causes considerable air pollution
(Loomis and Connor, 1992). Thus, it appears that solar, wind, hydroelectric, and
geothermal energy production will play a larger role in the future. It is also possible that
the relative costs of human labor will become more competitive with fossil fuels and
industrial societies may return to more agrarian-based economies.

Nutrients are added to agroecosystems because they are removed in harvested
products and lost to erosion, leaching, denitrification, and volatilization. Synthetic
fertilizers, animal feed, and animal manure are typically added for nutrient replenishment.
In industrial societies most nutrients eventually end up in landfills or surface waters
although a small fraction is recycled back to farms as sewage sludge (King, 1990).
Environmental degredation due to nutrient runoff has been discussed. However, there is
also concern over the lifetime of resource reserves used for fertilizer production. Nitrogen
fertilizer production is strictly dependent upon fossil fuel reserves. In fact, nitrogen
fertilizer accounts for 30% of the energy consumed in modern corn production (Pimentel
and Dazhong, 1990). Potassium and phosphorus are mined from geological deposits and
therefore also have finite quantities for use in industrial agriculture. Based upon current
usage trends the deposits of both of these minerals may be used up in less than 100 years
(King, 1990).

BIOLOGICAL RESOURCES

The effects of industrial agriculture on biological resources have included wildlife
habitat destruction, extinction of species, loss of crop and domestic animal genetic
diversity, and the selection for resistant pests. While the amount of land used for
agriculture has increased dramatically since 1700, forested land has decreased globally by
about 15%. Current trends indicate that deforestation has stabilized in developed nations

while it continues in the tropics (Meyer and Turner, 1992).

10

The primary reasons for deforestation include the expansion of cultivated and
pasture lands and the extraction of timber and fuelwood (Meyer and Turner, 1992). Thus,
food production and preparation are largely responsible for forest loss. The consequences
of these trends for biological diversity are difficult to determine. Nonetheless, current
extinction rates are estimated to be several orders of magnitude above background levels
(Vitousek, 1992).

The loss of biodiversity includes that of crop plants and domestic animals. Local
landraces or varieties have been developed over hundreds or thousands of years of
selection by humans. Recent advances in plant breeding have resulted in high-yielding
varieties which have replaced local varieties in areas thoughout the world. There is concern
that the loss of local varieties is endangering the global food supply because current
commercial varieties have a narrow genetic base. The Irish potato famine of the mid-18008
provides an example of the potential consequences of low crop genetic diversity. Europe’s
entire potato crop at that time had been developed from only two samples from South
America, both of which were susceptible to late blight, Phytophthora infestans (Soule and
Piper, 1992). Similarly, in 1970 southern corn leaf blight destroyed 15% of the corn crop
in the United States because most major varieties were developed from the same susceptible
genetic source (Luna and House, 1990; Soule and Piper, 1992).

Although agricultural pests are generally not considered to be biological resources,
the evolution of pesticide resistance has clearly demonstrated that susceptible pests should
be treated as a resource. Heavy reliance on chemical pesticides has led to the evolution of
pesticide resistance in rodents, arthropod pests, weeds, and plant pathogens. Today at
least 500 arthropod pest species have evolved resistance to pesticides (Gould, 1991; Kim,
1993). A significant number of weed and plant pathogen species have also evolved
resistance to one or more pesticides. An important consequence of this is that greater
quantities of pesticides have been required to control resistant pests. Although resistance in

some pests has evolved in response to cultural and biological control measures, the vast

11

majority of cases has resulted from complete reliance on chemical pesticides for pest

management.

AGRICULTURAL SUSTAINABILITY

The sustainability of modem agricultural practices and systems has been fiercely
debated for the last three decades. At the global level the central issue is whether or not the
exploding human population can be fed with agricultural methods which do not lead to
environmental degredation and resource depletion. There is ample evidence that if current
trends in agricultural industrialization continue human life-support systems will be
irreparably damaged and lead to greater human suffering (Brown and Kane 1994; Soule
and Piper, 1992; Meadows et al., 1992). At the same time, human population growth
continues to accelerate and is estimated to reach 8.9 billion by 2030 (Brown and Kane,
1994). Based upon recent food production trends, global demand is expected to greatly
exceed production within several decades (Brown and Kane, 1994).

At national, regional, and local levels issues related to agricultural sustainability
differ and are often complicated by diverse interests. Ultimately the concern over
agricultural sustainability is a concern over human welfare. Although physical aspects of
human welfare, such as nutrition and freedom from disease, seem relatively easy to
measure, other aspects, such as happiness and quality of life, do not. This has led, in part,
to the extreme divisiveness of issues related to agriculture and human welfare in the public
forum. Even issues related to the effects of pesticides on human health are far from settled.
Experts continue to argue over the validity of past studies and the relative risks of human
activities and choices on long-terrn human health and societal welfare (York, 1991; Loomis
and Connor, 1992; Merwin and Pritts, 1993; Pimentel et al. 1993; Schuman 1993).
Although a variety of diverse indicators have been used to assess human quality of life, the
most common are economic, such as annual income and gross national product (GNP).

There are certainly many other elements important to human well-being but none are as

12

easy to measure as the economic indicators. Nevertheless, the potential consequences of
agricultural technologies on human welfare need to be considered beyond the short-term
economic effects.

Certainly the primary purpose of agriculture is to produce food, fiber, and other
materials for human use. But, to farmers, their families, and rural communities, it is also a
source of income and a way of life. Many people consider family farms and traditional
farming methods part of this nation’s heritage. However, the production capacities of
traditional farming methods are dwarfed by the immense capabilities of industrial methods.
The use of chemical fertilizers, pesticides, and machinery has resulted in dramatic increases
in farmer productivity and in a steady decline in the size of farming populations and rural
society as a whole. The industrialization of agriculture and food production systems has
led to many changes in rural society and there is considerable disagreement over whether
the results have been desirable or destructive.

According to Pradgett and Petrzelka (1994) the emphasis on increased production
and new technology has led to a system characterized by competition, self-interest, and
short-term visions. Only those farmers who invested in new technologies and enlarged
their operations have been able to survive. The loss of farms has resulted in a shrinking
farming population and the economic collapse of many rural areas. It has also been argued,
however, that the exodus from rural areas to the cities during this century resulted from
dissatisfaction with the tedious and exhausting life on farms, especially prior to the
introduction of new technologies. Urban areas offered a greater chance for prosperity than
many farmers could obtain growing food (Loomis, 1984).

In market economics the values of society are suppose to be reflected in the way
that individuals spend money . Thus, if society valued traditional agriculture and family
farms, theoretically it could maintain them through the market (White et al., 1994).
Similarly, if society wants agriculture to move in a more sustainable direction now it should

be able to bring about such change though the market system. But with the immense

13

diversity of products on the market and the social and physical distancing between producer
and consumers it appears that most consumers are unaware of the processes behind food
production. Furthermore, because market costs do not necessarily reflect true costs,
consumers do not have enough information for making informed decisions. In addition,
many ecologists and at least some economists acknowledge that a market system based
upon perpetual economic growth and consumption is in conflict with the concept of
sustainability (Hall, 1990; Daly, 1993; White et al., 1994). There are few incentives in a
market economy to conserve resources, prevent pollution, or preserve rural communities
(Hardin, 1968). Therefore, in order to achieve more sustainable agricultural systems more
emphasis is needed on community and cooperation as well as a long-terrn vision of the
future (Lacy, 1994; Pradgett and Petrzelka, 1994). Such changes are not likely to occur
without policies which increase the incentives for farming more sustainably.

Agricultural sustainability is a combination of overlapping goals rather than a
specific set of practices. These goals generally include improving environmental, social,
and economic aspects of food production. The relative importance put on particular goals
varies greatly. White et al. (1994) describe sustainability as “an essential but fluid goal that
continually retreats into the future ahead of us.” Although the concept of agricultural I
sustainability is somewhat vague and subjective, environmental degredation, resource
depletion, and the subsequent effects on human welfare are not. Therefore steps must be
taken to move toward generally accepted goals. The more commonly cited goals for
increasing agricultural sustainability include: reducing pesticide use, improving soil
conservation, increasing nutrient conservation and recycling, improving economic stability,
reducing fossil fuel dependence, maintaining or increasing production efficiency, and

enhancing quality of life.

14

THE ROLES of DOMESTIC ANIMALS in AGRICULTURAL SUSTAINABILITY

Animals were important sources of food, clothing, and shelther for humans long
before domestication began about 12,000 years ago. However, with domestication came
an expansion of the roles and importance of animals in human society. During the last
several thousand years animals, including dogs, cattle, horses, goats, sheep, poultry, and
others, have been used for many purposes beyond simply serving the basic human needs.
These have included traction, transportation, hauling, guarding, recreation, herding, and
companionship. Thus, with domestication the relationship between humans and domestic
animals was fundamentally transformed from that of predator and prey to that of
mutualism.

Today, the discussions and debates on sustainability have focused much attention
on the roles of domestic animals in agriculture and the larger food system, particularly in
industrialized nations. The issues of concern can be categorized into three basic areas:
ecological, nutritional, and ethical. Most of the issues are directly or indirectly related to
the industrialization of animal production.

Ecological concerns over domestic animals stem largely from the resource ,
inefficiencies in industrialized production systems and the environmental deterioration that
seems to accompany most production systems. Industrial animal production is extremely
energy intensive (Hall et al., 1992). According to Pimentel (1984), in the United States it
takes 7-88 kcal of fossil energy to produce 1 kcal of animal protein while it takes only 0.7-
3 kcal of fossil energy to produce 1 kcal of plant protein. Animal production is also land
intensive, with nearly one quarter of the earth's land surface used for pasturing animals
(Rifltin, 1992). Worldwide, animal production, particularly that of cattle, has been a cause
of deforestation, desertification, water contamination, soil erosion, and global warming

(Rifltin, 1992; Pimentel et al., 1995).

15

Animal production has also been criticized on nutritional grounds because animal
protein is not a necessary component of the human diet (Pimentel, 1984; Robbins, 1987;
Lappe, 1991) yet most edible grain in industrialized nations is feed to cattle and other
animals. This system results in reduced food production efficiency and the loss of potential
nutritional energy for human consumption. Furthermore, many of the "diseases of
affluence," such as heart disease, some types of cancer, and diabetes, that have become so
common in wealthy, industrialized nations are attributed to diets high in fat and cholesterol
which typically means diets high in animal products (Robbins, 1987; Lappe, 1991).

The ethical considerations are the most difficult to deal with from a scientific
perspective. Nonetheless, they have been central issues in the sustainable agriculture
movement. Essentially there are two levels of concern. The more general or mainstream
concern is animal welfare under human care. Agricultural industrialization has been
accompanied by the removal of animals from relatively natural settings and subsequent
placement into largely artificial environments which, while providing opportunities for
increased production, commonly result in increased stress and behavioral deprivation
(Robbins, 1987; Kiley-Worthington, 1993). Domestic animal welfare has received
attention from scientists and attempts have been made to develop objective methods of
welfare assessment and more humane producition systems (e.g. Dawkins, 1983; Appleby
and Hughes, 1991).

A more difficult ethical concern for agricultural science is the morality of killing
animals for human food. Advocates of vegetarianism often point to the ecological,
nutritional, as well as the ethical points when defending their position. One's stance
however on the ethics of killing other species for human use is a moral question and
science really has little to offer on the issue. Ethical arguments against animal consumption
are at least somewhat flawed from an ecological point of view in that it is impossible to live
without destroying the life or potential life of another organism (Kiley-Worthington, 1993).

Furthermore, the ecological consequences of not slaughtering the domestic animals we now

16

have would be catastrophic. Ironically, the condemnation of animal consumption also
perpetuates the fallacy that humans are somehow separated from nature (Lappe, 1991), a
perception that many argue is one of the root causes of current ecological problems.

Although animal production in agriculture, particularly industrialized agriculture,
has been criticized on a variety of grounds, there are many potential roles for domestic
animals in ecologically-based agriculture and food systems. Although there is no clear
right and wrong way to raise animals, when viewed from an ecological perspective,
animals can perform functions that would otherwise be more difficult or even impossible
without them. In one analysis, Bender (1994) concluded that animals play an important role
in achieving fundamental goals of ecologically-based farming including soil conservation,
pesticide elimination, financial stability, and nutrient cycling. Much of the land that is in
pasture is used as such because it is unsuitable for cropping. Thus, these lands are
maintained in forage plants which provide protection against soil erosion and feed animals.
Animals convert lignocellulosic plant materials such as forage crops, crop residues, and
byproducts, which are essentially non-competitive, renewable resources, into food for
human consumption (Parker, 1990). Furthermore, the proper management of manure
produced by domestic animals aids in nutrient cycling processes on farms and within
localized regions. Thus, the integration of crop and animal systems can provide
complementary and synergistic effects by allowing for the efficient utilization of on-farrn
resources. In addition, animals can be used for vegetation management by grazing in
agroforesty systems and for biological weed control (Parker, 1990).

The suitability of animals in agriculture and food systems will depend upon local
and regional ecological factors, including soils, climate, and energy resources, as well as
the particular socioeconomic conditions and societal values. While animals and their
products will not be appropriate under all circumstances, the integration of crops and
animals into well-managed systems can provide flexibility, diversity, and stability for

agriculture and food systems.

17

PEST MANAGEMENT and AGRICULTURAL SUSTAINABILITY

Although pest management is only one element of agricultural management it
directly or indirectly influences all of the goals stated for improving sustainability. Pest
management practices directly affect pesticide use, crop productivity, fossil fuel use, and
economic performance. Soil and nutrient conservation are indirectly influenced by insect,
weed, and pathogen management practices. The development of sustainable pest
management strategies needs to be integrated with management approaches for soil fertility,
marketing, labor, and other elements of farm management. Probably the two key issues
for pest management are reducing pesticide use while simultaneously maintaining or
increasing productivity. Careful analyses of all economic and noneconomic costs and
benefits are required to assess current and future pest management methods within a
holistic perspective.

It has been estimated that 35% of global crop and livestock productivity is lost to
pests, primarily insects, weeds, and pathogens (Pimentel, 1991). In the United States this
estimate is 37% with insects, weeds, and pathogens accounting for relatively equal
proportions (Pimentel, 1991). Although pesticide use has increased dramatically over the
last half of this century, total crop and livestock losses to pests have remained steady or
increased slightly (Osteen, 1993). Pimentel et al. (1993) estimate that in the United States
about $4 billion are spent annually for chemical control to protect about $16 billion in crops
from insects, pathogens, and weeds. However, the environmental and social costs
associated with pesticide use were estimated to be $8 billion with the most significant costs
attributed to bird losses, groundwater contamination, pesticide resistance in pests, crop
losses, and public health impacts. Thus, it was calculated that for every 3 dollars spent by
society on pesticides 4 dollars are made. In contrast, it has been estimated that for every

dollar spent on biological control $30 to $100 have been made (Hoy, 1992).

18

Reductions in pesticide use now and in the near future will depend primarily on the
development of cultural and biological control methods. However, most integrated pest
management (1PM) programs have focused on making pesticide use more cost effective
through more efficient use, rather than replacement and integration of alternatives. In fact,
global pesticide use is expected to increase from the current 2.5 billion kg to over 3 billion
kg applied annually by 2000 (Pimentel, 1991). North America and Europe currently
account for nearly 80% of all pesticides applied. As more developing nations attempt to
adopt industrial agricultural methods, pesticides will play a larger role in other parts of the
world. Yet, most of the rest of the world currently depends largely on nonchemical pest
management and there are opportunities to slow the current trend toward pesticide use with
a combination of traditional methods and newly developed technologies and knowledge.

The substitution of biological and cultural management methods for pesticides is
based on the idea that ecological knowledge can be substituted for purchased inputs (Luna
and House, 1990). Current crop losses are high in spite of increased pesticide use and
have been attributed to a variety of causes, including lack of crop rotations, more
susceptible varieties, increase in monocultures, reduced crop diversity, loss of natural
enemies from pesticides, reduced sanitation, higher cosmetic standards, and pesticide
resistance (Pimentel et al., 1993). In an analysis of currently available pesticide
alternatives, Pimentel et al. (1993b) concluded that pesticide use could be reduced by 50%
in the United States without a significant loss in production, as is currently being done in
Denmark, The Netherlands, Sweden, and Ontario, Canada. In that analysis, methods for
pesticide reduction included greater use of scouting, crop rotation, resistant and vigorous
varieties, sanitation, insect pathogens, mulch, cultivation, forcasting, spot treatments, and
biological control.

It has been argued by agricultural scientists in a variety of disciplines that industrial
agroecosystems need to be dramatically redesigned to be more sustainable (Edens and
Haynes, 1982; Hill and MacRae, 1992; Edwards et al., 1993; Altieri, 1994). This

l9

argument seems particularly valid in the area of crop protection. Based upon ecological
theory it has been hypothesized that pests could be better managed by increasing biological
diversity in agroecosystems. According to Edwards et al. (1993) the use of pesticides in
industrial food production systems has replaced the functions of biodiversity in traditional
agroecosystems. They suggest that research should be done to “design practical integrated
systems of crops and animals that can be adapted to regions, minimize energy-based inputs
and have long-term sustainability.” Similarly, Hill and MacRae (1992) state that farms
should be “made more ecologically and economically diverse, resource self-reliant and self
regulating.” These authors suggest that designing agroecosystems with greater biological
diversity is required to sustainably manage pest problems. The idea that diversification
reduces pest problems is supported by two current hypotheses (Risch et al., 1983; Andow,
1991). One is that crop plants are less apparent and predictable in more diverse
environments (resource concentration hypothesis) (V andenneer, 1989; Roltsch and Gage,
1990; Andow, 1991) and the other is that more diverse habitats have a greater diversity and
abundance of natural enemies (enemies hypothesis) (Letoumeau, 1987; Russell, 1989).
Neither of these hypotheses hold up universally but both have been demonstrated to be
valid under a variety of cropping systems.

In order to maintain a holistic approach in the analysis of agroecosystem redesign
the views of critics should also be considered. Loomis (1984), for example, claims that
increasing on-farm diversity would lead to production inefficiencies because more
machinery would be needed, greater knowledge and management skills would be required,
and there would be greater risks and smaller returns. This view is based upon an economic
way of thinking which emphasizes the importance of capital, technology, and growth and
largely ignores the importance of biological and physical resources and ecological
processes. Ironically, sustainable agriculture advocates also claim that greater knowledge
and management are required; however, they propose that these can replace purchased

inputs, leading to greater efficiency, and thus, enhanced sustainability.

20

These issues surrounding agroecosystem diversity, stability, and efficiency in pest
management, as well as in other facets of food production, are quite complex and still not
well understood. The knowledge and skills of a farmer are critical to managing diverse
operations. And, assuming a farmer has adequate knowledge and skill, the stability of
diversified systems should be less susceptible to ecological and economic uncertainties.
Whether or not more machinery would be required for diversified systems would depend
on the composition and structure of the farm and larger food system and cannot be easily
generalized. Only if it is assumed that there are adequate soil, water, mineral, energy,
nutrient, and biological resources and an essentially endless assimilative capacity of the
global ecosystem to maintain industrial-based agriculture indefinitely, or at least over the
next few centuries, do Loomis’ claims appear valid. Nelson (1995, p. 148) succinctly
described this disagreement between the ecological and economic points of view in stating
that, “Economists can point to a ZOO-year history of mistaken predictions of food, energy,
timber, and other dire crises sure to occur in the near future. Yet, the fact that these
predictions essentially all proved wrong does not guarantee that they must always be
erroneous.” In this sense, the economic point of view is understandable, but, in a world of
shrinking resources and deteriorating ecosystems, agricultural diversification on farms,
landscapes, and regions seems to be a logical approach to designing more sustainable

agroecosystems, especially in the arena of pest management.

CHAPTER 2

THE SPECIFIC CONTEXT OF THIS PEST MANAGEMENT PROBLEM:
A LITERATURE REVIEW

The experimental agroecosystem focused on in this research project has been
designed for ecologically-based management with several assumptions in mind. The first
assumption is that on-farrn diversity provides production stability in an environment which
is only partially predictable. Secondly, food production should not lead to environmental
degradation. Finally, farming should be a desirable, safe, and stable occupation. These
assumptions may be contentious, however all research, especially in applied science, is
based upon implicit or explicit assumptions and values. It is therefore important to identify
and examine these assumptions prior to conducting research as well as during the
interpretation of results.

The experimental agroecosystem used for this research is located at the Kellogg
Biological Station, Michigan State University. It is a 2-ha apple orchard planted in 1983
with disease-resistant varieties and managed without the use of pesticides. This has
allowed the pasturing of domestic birds and the use of the orchard alleys for intercropping
annuals. Potato has been used as a model intercrop for this research project primarily
because of the ease in producing it, its presumed compatibility with domestic birds, the
current problems with controlling Colorado potato beetle in commercial production, and
evidence in the literature that chickens have been used to control this pest. Under
conventional orchard management pasturing domestic birds and intercropping annual crops

is generally not possible.

21

22

CURRENT STATUS OF APPLE PEST MANAGEMENT

Apples are generally considered to be a high-value crop which need to be blemish-
free to be marketed today. Since 1945 increasing importance has been placed on the
cosmetic appearance of fruits and vegetables, not only by consumers but also by the
USDA, retailers, and wholesalers. Although the general public prefers visually “perfect”
produce, most people are largely unaware of the connection between improved cosmetic
appearance and increased pesticided use (Pimentel et al., 1993c; van Ravenswaay, 1994).
Nevertheless, efforts have been made in recent decades to reduce pesticide use on apples
and other fruits. These efforts have been driven not only by environmental and health
concerns over pesticide use, but also by increasing pesticide costs, evolution of resistance
in many pests, and decreasing availability of pesticides because of fewer registration
renewals (Prokopy and Croft, 1994).

The adoption of IPM practices by apple growers has been increasing since the early
19703 and is currently widespread (Whalon and Croft, 1984), but pesticide use remains
high. According to Prokopy (1994), this is because most commercial growers practice
chemical-based IPM. Prokopy describes four levels of integrated pest management in
apple. With each increasing level there is greater integration, which begins with the
development of management strategies for individual pests and ends with the “blending of
concerns of all those having vital interest in pest management: researchers, extension
personnel, private consultants, industry, growers, processors and distributors, consumers,
neighbors of growers, environmentalists, and local as well as federal government
regulatory agencies.” According to Prokopy, few commercial orchards are beyond the first
level of apple IPM.

According to estimates presented by Pimentel et al. (1993b), 7 million kg of
insecticides are applied to apples in the US. annually with 96% of hectarage being treated.

Studies have shown that when pesticide applications are discontinued in conventionally-

23

managed orchards or when orchards are left unmanaged, insect damage to fruit can be
extremely high (Glass and Lienk, 1971; Madsen and Madsen, 1982; Reissig et al., 1984;
Prokopy, 1991). The most important insect pests in northeastern and north central North
America are codling moth (Cydia pomonella L.), apple maggot (Rhagoletis pomonella
Walsh), and plum curculio (Conotrachelus nenuphar Herbst). These three species can
damage nearly 100% of the fruit during a growing season if unmanaged.

Alternative control measures for these pests have been evaluated and the results are
variable. For example, Pfeiffer et al. (1993) found that pheromone mating disruption can
reduce codling moth damage. However, under high pest pressure, damage levels still may
be unacceptable for many commercial growers. Reissig et al. (1984) concluded that small
blocks of disease-resistant apples could not be grown in northeastern North America
without the use of insecticides. However, they stated that alternatives including insect
growth regulators for intemal-feeding Lepidoptera like codling moth, orchard sanitation
and sticky red spheres for apple maggot, and disease-resistant apple varieties could be used
to reduce the amount of pesticides used. The literature concerning alternatives to pesticides
in apple production suggests that increasing the feasibility of low-chernical or nonchemical
apple production in northeastern and north central North America depends upon not only
maintaining low levels of damage by codling moth, apple maggot, and plum curculio but
also reducing the costs and labor demands of alternative practices and systems (Pimentel et
al., 1983; Prokopy, 1991; Agnello et al., 1994).

Apple production systems which do not depend on pesticides need to maximize
biological control, host plant resistance, cultural management, ecological knowledge, and
efficiency of design. Biological controls for codling moth, plum curculio, and apple
maggot have been investigated but require further attention. A few studies have shown that
indigenous predators and parasitoids can have an impact on the mortality levels of these
pests (Leius 1967; Subinprasert ,1987; Hagley and Allen, 1988; Allen and Hagley, 1990).

However, studies need to be conducted to evaluate how habitat manipulations can be used

24

to enhance predator populations and their effectiveness against pests. Cultural practices,
such as removing apple drops, mulching, pruning, and conserving habitat for beneficial
arthropods and insectivorous birds, may aid in reducing pest pressure and also need further
attention. Methods to manage ground vegetation effectively could involve the use of low-
maintenance ground covers and domestic animal grazing. Fungicides, which are currently
applied to 90% of apple hectarage (Pimentel et al., 1993b), could be virtually eliminated by
using disease resistant varieties (Reissig et al., 1984; Prokopy, 1991). Tactics which take
advantage of pest behavior, such as the use of sticky red spheres and pheromone
disruption, will also need further development to be more efficiently utilized in nonchemical

production.

Apple Pests Considered in this Study
WWW

Plum curculio is native to eastern North America and feeds on a variety of pome
and stone fruits. Its damage to apple fruit includes cresent-shaped ovipostion scars on the
surface, feeding damage from the adults, internal feeding damage from the larvae, and
premature drop of intemally-damaged fruit caused by the release of pectic enzymes and
cellulase released by the larvae (Racette et al., 1992). The actual yield loss due to
premature drop from plum curculio larval damage is not known because its occurrence
coincides with the natural fruit abcission commonly known as “June drop” (Racette et al.,
1992). The ovipostion scars commonly seen on mature fruit are cosmetic injuries. The
presence of such scars at harvest are not indications of larvae within the fruit, but rather of
failed larval survival. The newly emerged larvae are commonly crushed by the rapidly
growing fruit, leaving surface scars without internal damage. Nevertheless, due to the
strict cosmetic standards for fruits, including apples, such surface damage is generally not

tolerated by wholesalers, retailers, processors, or consumers.

25

Plum curculio has been found to overwinter under the fallen leaves of deciduous
forests, especially thick litter layers, and may also overwinter under other types of ground
debris in and around orchards, although evidence for this is lacking (Lafleur et al., 1987;
Racette et al., 1992). In Michigan it usually migrates into apple orchards just prior to or at
bloom and peaks in activity approximately two weeks after petal fall (Brunner and Howitt,
1981; Lafleur and Hill ,1987). This migration may take several days to several weeks
(Racette et al., 1992).

Chouinard et al. (1993) observed the activities and behaviors exhibited by adult
plum curculio once arriving to an orchard. They found that it spends most of its time (65-
90% of observations) resting. Resting usually occurs on the ground (83% of total
observations, n=1433) and is most common from 0000 and 1200 hours. The two most
fiequently observed activities were crawling and feeding. Crawling was observed equally
both in trees and on the ground while feeding was observed mostly in the trees. Activities
which were infiequently observed included mating, aggregating, flying, and egg laying.
Overall activity was positively correlated with an increase in relative humidity and a
decrease in the air saturation deficit, which is the drying capacity of the atmosphere based
on temperature and humidity. Their observations indicated that flying and dropping from
the trees followed by crawling were important means of within-orchard dispersal.

Current control methods for plum curculio are almost exclusively chemical in
commercial orchards (Racette et al., 1992). One possibility for reducing insecticide
applications for this pest is through fruit monitoring to appropriately timed sprays.
Monitoring along orchard perimeters is an efficient means of early detection so that damage
can be minimized (U .S. Department of Agriculture, 1990). Cultural management which
may reduce plum curculio pressure include habitat management, sanitation, cultivar
selection, and mechanical control (Racette et al., 1992). Habitat management includes
establishing orchards as far from forests as possible, in areas of low humidity, and at least

160-200 m away from wild hosts such as Crataegus spp., Prunus spp., and Amelanchier

26

spp. Sanitation basically consists of removing fallen apples from the orchard. It has been
suggested that ground-feeding birds or pigs might be used for this task (Lafleur and Hill,
1987; Racette et al., 1992)

Biological control of plum curculio is currently limited by a lack of knowledge
about the ecology of this insect (Racette et al., 1992; Vincent and Roy, 1992). A variety of
predators, parastoids, and pathogens of plum curculio have been identified, however they
fail to prevent damage to the degree which the market requires. According to a study by
Vincent and Roy (1992) plum curculio damage to fruit in an unsprayed orchard in Quebec
averaged 48% with a peak at 86% over a 13-year period. Glass and Lienk (1971) observed
fruit damage from plum curculio to range from 1%-46% during a 10—year period after
insecticides were discontinued in a small orchard in New York. These studies indicate that
natural control of plum curculio by natural enemies indigenous to orchard systems do not
provide an acceptable degree of control by modern standards. Further research is needed to
develop management methods which enhance the populations of indigenous natural
enemies, identify habitat manipulations which make indigenous natural enemies more
effective, and find which natural enemies might be useful in augmentative or classical

biological control.

W

The apple maggot is an indigenous North American species whose native host is
hawthom (Crataegus spp.). It was first reported to infest apple in the Hudson Valley of
New York in the mid-18003 (Bush, 1992). It has since become one of the most important
insect pests of apple in the northeastem and north central regions of the United States and
in southeastern Canada (Prokopy and Croft, 1994). The damage to fruit results from the
feeding activities of the larva. Eggs are laid by the female fly just under the surface of the

fruit skin. After hatching, the larvae burrow into the fruit flesh, leaving a brown trail.

27

Many of the eggs do not hatch, but soft, dimple-like areas may be left on the fruit surface.
Fruit damaged by apple maggot larvae frequently drop prematurely (Howitt, 1993).

In most commercial orchards, apple maggot numbers are reduced with insecticides;
up to four applications may be needed per growing season. Yellow rectangle and red
sphere traps can be used to monitor adult populations for more precise timing of pesticide
applications (Prokopy and Croft, 1994). Odor attractants can be used as baits to increase
catch rates with both of these traps. Duan and Prokopy (1995) reported that nearly
acceptable commercial-level control of apple maggot was achieved (l-3% damaged fruit)
with pesticide-treated, baited red spheres in commercial orchards in western Massachusetts.
Moreover, Prokopy (1991) has found that apple maggot can be managed without pesticides
(less that 1% damage) in small-scale orchards using unbaited sticky red spheres. This
method, however, is labor intensive because it requires at least one sphere per tree.

Natural enerrrics are considered to be ineffective at controlling apple maggot
populations in pesticide-treated and nonchemical orchards (Vincent and Roy, 1992). Part
of the reason for this may be attributed to the recent shift of this species onto apple as a host
plant. It is possible that this host shift has released this species from its predators and
parastoids and, in effect, provided it with “enemy-free space.” Feder (1995) has found
support for this senario in a study of two braconid parasitoids of apple maggot. He found
parasitism to be considerably less on apple compared to hawthorn. Apple provided more
physical protection for the developing apple maggot larvae because of the larger fruit size.
In addition, the parastoids were not well synchronized with apple maggot on apple. Feder
also found that competition with other fruit infesting insects on hawthorn forced apple
maggot larvae to feed closer to the fruit surface, making it more vulnerable to parasitoids.
The effect of predators on apple maggot has not been well studied. Allen and Hagley
(1990) found that a wide variety of epigeic predators, including carabids, staphylinids,
ants, and spiders, fed upon apple maggot larvae and pupae in the field. However, the

importance of these predators as mortality agents of apple maggot is still not known.

28

I E ICE '1!" ".11 1

Japanese beetle was introduced into the United States sometime before 1916 when
it was discovered in New Jersey (Johnson and Lyon, 1991). It has since spread
throughout most of the eastern half of the United States and southeastern Canada. This
extremely polyphagous species is known to feed on nearly 300 plant species. Although
apple is one of its food plants, Japanese beetle is generally considered to be a minor pest on
this crop (Howitt 1993). In fact, it is not even mentioned as a pest in many apple
management guides (e. g. US. Department of Agriculture, 1990; Johnson and Herr,
1995). Nevertheless, it has been found to be an abundant pest at the Kellogg Biological
Station orchard (Stuart Gage, personal communication), damaging both foliage and fruit.
Although the reasons for its high population densities at this site are unclear, it may be at
least partially due to the lack of insecticide treatments and the relatively early-maturing
varieties present. In commercial orchards, Japanese beetle populations would likely be
held at lower levels by insecticides applied for other pests such as apple maggot.

Japanese beetle adults emerge from the soil in mid-summer and live from 30-45
days. The beetles, which are strong flyers, are most active in warm weather. Males and
females tend to congregate on food plants, where they mate. Females are capable of
producing 40-60 eggs, which are deposited in the soil (Johnson and Lyon, 1991). The
eggs hatch within a few weeks and the larvae (grubs) begin feeding on plant roots. In fall,
larvae burrow down deeper into the soil and form overwintering cells where they remain
until spring (Davidson and Lyon, 1987).

Most efforts at biological control have focused on the larvae as a turf grass pest. A
variety of agents, including bacteria, fungi, protozoans, nematodes, and insects, have been
used. One commonly used biological control agent is Bacillus popilliae Dutky, which
infects the larvae of this and other scarabaeids. This microbial control agent is commonly

termed “milky disease” because of the hemolymph color in infected larvae (Maddox,

29

1994). Parasitic wasps (T iphiidae) and flies (T achinidae), which attack the adult stage have
been introduced in some areas for control (Davidson and Lyon, 1987). Terry et a1. (1993)
found that epigeic predators, such as carabids and staphylinids, consume Japanese beetle
eggs and that when the insecticides carbaryl and isazofos are used, predator abundance and
predation decrease. Some small mammals, birds, and reptiles are also predators of
Japanese beetle (Johnson and Lyon, 1991). Based upon experience in other crops, the
potential for biological management of Japanese beetle in apple appears to be high.

1!! l . E l E l .

The management of weeds or groundcovers in apple orchards is important for
maintaining adequate levels of nutrients and water. Competition between apple trees and
weeds for these resources can result in reduced fruit yields (Rupp and Anderson, 1985;
Raese, 1990). However, in situations where nutrients and water are abundant, controlling
vegetation beneath apple trees may not lead to higher yields or increased growth.

Orchard floors are managed with a variety of practices which include herbicide
applications, clean cultivation, mowing, planting cover crops, and mulching. The practice
or combination of practices used can have important effects on seasonal labor requirements,
nutrient management, and pest and beneficial insect population dynamics. For example,
mulching can be as effective as herbicides in controlling weeds (N iggli et al., 1990) but is
generally very labor intensive, not only because of the mulch application but also because it
usually must be removed for winter to prevent rodent damage (Prokopy, 1991).

The effects of weed and groundcover management on insect pests and beneficials
has recently received increased attention from researchers and growers. Nevertheless, very
few generalizations that can be made. Although weed species differ in their attractiveness
to beneficials and pests, some researchers have tentatively concluded that a high diversity
of plant species on the orchard floor leads to a reduced level of pest damage (Altieri 1994).
A review of the literature by Bugg and Waddington (1994) provides some support for this

30

idea. They re-analyzed data from Leius (1967) and found that in his study apple orchards
with greater plant species richness had increased levels of parasitism on codling moth and
tent caterpillar. Nevertheless, there is concern that some plant species are harmful to apple
production. Coli and Ciurlino (1990) mentioned that legumes such as vetch and alfalfa can
cause increased tarnished plant bug damage and that dandelions can be a source of virus
and compete with apple trees for pollinators. Clearly, more research is needed, but from a
holistic point of view it can be cautiously concluded that a diverse ground cover has
benefits over clean tillage or complete herbicide bumdown. These benefits include soil
conservation and habitat conservation for parasitoids, predators, and soil invertebrates
important in cycling nutrients and maintaining soil structure. The effects of different
practices, such as mowing, mulching, and manuring, may have important effects as well,

and might be integrated with a strip management approach for greater habitat stability.

CURRENT STATUS OF POTATO PEST MANAGEMENT

Industrial potato production methods as practiced in the North America and Europe
are very chemical intensive. Many of the problems which typify commercial potato
production are exacerbated by continuous cropping or short rotations. Vereijken and Van
Loon (1991) consider potato the most environmentally damaging crop grown in the
Netherlands because of the large quantities of insecticides, herbicides, nematicides, and
fungicides used on it. Cunently, insecticides are used on 96%, herbicides on 97%, and
fungicides on 96% of US. potato hectarage (Pimentel et al., 1993b). Problems
encountered in potato production vary considerably with climate, year, season, and
location. In addition, nonchemical or organic growers do not necessarily encounter the
same sets of problems that conventional growers experience (Peacock and Norton, 1990).
Attempts at developing low-chemical or nonchemical practices for potato production have
had varying degrees of success. Pimentel et al. (1983) concluded that organic potato

production would be considerably less energy and labor efficient than conventional

31

production. This is due to the assumption that the replacement of herbicides would require
five tractor cultivations per growing season and that no effective replacements for
insecticides and fungicides existed. Under these constraints 50% of the crop would be lost

to insect pests and disease.

Potato Pests Considered in this Study
C l l E E l [I . 1 1° 5 1

According to a survey presented by Radcliffe et al. (1991) Colorado potato beetle is
considered to be the most important insect pest of potatoes in northeastern North America.
It is believed that Colorado potato beetle inhabited central Mexico and fed upon Solanum
angustr’foliwn Mill. and Solanum rostratum Dunal prior to becoming a pest of potatoes in
the United States (Casagrande, 1987). Potatoes were introduced into Virginia in the early
16003 and spread westward. In 1859, Colorado potato beetle was found on potatoes in
Nebraska. The widespread cultivation of potatoes allowed this pest to spread eastward
reaching the Atlantic coast by 1874. Although a variety of cultural control methods were
used during this time, acceptible control apparently was not obtained until the discovery of
the insecticidal properties of Paris green, a paint pigment. Additional compounds,
including lead arsenate and calcium arsenate, were added to the arsenal. Although there
was little evidence of resistance to these compounds the availability of the relatively
inexpensive compound, DDT, led growers to switch to using it during the 1940s.
Resistance to DDT was soon recognized. New chemicals were developed and became
available but the rate of resistance development increased. During the 30-year period
between 1954 to 1984, 14 new chemicals were introduced for use against Colorado potato
beetle on Long Island, New York, an area particularly hard hit by this pest. According to
Casagrande (1987), Colorado potato beetle evolved resistance to all except one of those

chemicals during that period.

32

New methods of control which are currently being implemented include the use of
tractor-mounted flamers and genetically-engineered potato plants which produce the delta
endotoxin of Bacillus thuringiensis. The widespread effectiveness and utility of these
approaches remains to be seen. Resistance to the B. thuringiensis delta endotoxin has
already been artificially selected for in one laboratory population of Colorado potato beetle.
Whalon et al. (1993) obtained a strain of Colorado potato beetle which was 59 times more
resistant to the B. thuringiensis delta endotoxin than an unselected stain in just 12
generations. If the use of these transgenic plants becomes widespread, selection pressure
for resistance will be intense.

According to some researchers, nonchemical management practices, including crop
rotation, host-plant resistance, and biological control, have been either under-utilized by
growers or underdeveloped by researchers (Casagrande, 1987; Hare, 1990). Prospects for
enhancing populations of endemic natural enemies through cultural practices appear to be
good. Many common generalist predators will feed on at least one stage of Colorado
potato beetle (Casagrande, 1987; Groden et al., 1990; Riechert and Bishop, 1990; Hazzard
et al., 1991; Nordlund et al., 1991). Brust (1994) found that mulching potatoes led to an
increase in soil-dwelling predators, particularly carabids, and an increase in Colorado
potato beetle mortality. Furthermore, he observed greater egg and early instar mortality,
presumably by coccinelids and chrysopids, in mulched compared to unmulched potatoes.
Consequently, defoliation was over two times greater and yields 30% less in unmulched
potatoes compared to mulched potatoes. Similarly, Riechert & Bishop ( 1990) found lower
numbers of Colorado potato beetles, flea beetles, aphids, and leaflroppers and less
defoliation on mulched compared to unmulched potatoes. Spiders, which showed dramatic
increases in population due to the mulch, appeared to be the most important predatory
arthropod group in this study. Stoner (1993) also reported that mulch around potato plants
reduced Colorado potato beetle defoliation, which led to increased yields relative to a

33

control. She, however, did not determine if natural enemies were responsible for these
effects.

Other studies have evaluated augmentative releases of predators for biological
control. Biever & Chauvin (1992) found that releases of two predaceous stinkbugs,
Podisus maculiventris (Say) and Perillus bioculatus (F.), could substantially reduce
Colorado potato beetle populations in the field and lead to yields 65% higher than if left
untreated. Hough-Goldstein and Whalen (1993) also found releases of P. bioculatus to
reduce Colorado potato beetle abundance and defoliation.

The potential to use biological-based strategies to manage Colorado potato beetle
appear to be good. Possible constraints may include the high cost for purchased natural
enemies, inflexibility in cropping practices because of specialized machinery, and a lack of
market incentives to use nonchemical approaches. Nevertheless, from a biological
perspective, practices such as altering the timing of planting and variety, crop rotation,
intercropping, cover cropping, and mulching, provide adequate tools for developing
integrated management systems for Colorado potato beetle (Casagrande, 1987; Vereijken
and Van Loon, 1991).

1!! 1 . E E l .
Weed management in commericial potato production, as in most commercial field
crop and vegetable systems, is primarily dependent upon herbicides (Bellinder and
Wallace, 1991). Potato plants are poor competitors with weeds, especially during the
early part of the season. Weeds not only compete for water, nutrients, space, and light, but
also present problems for harvesting operations. Mechanical cultivation and reduced tillage
are alternative means of weeds management but these practices do not always result in
adequate yields. This is because cultivation causes soil water loss and, over time results in

soil compaction. Reduced-tillage potato systems can effectively reduce weed germination

34

but require hilling to get adequate yields (Bellinder and Wallace, 1991). Hilling, however,
results in weed germination.

The integration of cover cropping practices and mulching holds promise for
reducing input requirements, including pesticides, and reducing soil erosion. Cover crops
and mulches can suppress weeds by reducing soil temperatures and, in some cases,
releasing allelochemicals which inhibit weed germination and growth. For example, the
allelopathic compounds released by rye mulch can reduce growth in redroot pi gweed and
barnyardgrass (Bellinder and Wallace, 1991). In addition to weed control, mulches also
provide habitat for beneficial arthropods (as discussed above). Thus, the integration of
mulching practices may provide dual benefits in potato pest management.

FREE-RANGE DOMESTIC BIRDS as BIOLOGICAL PEST CONTROL AGENTS

There are many potential environmental and economic benefits in free-ranging or
pasturing domestic birds which are not obtainable with high-density, confined poultry and
egg production systems. Many of the problems associated with high-density, confinement
systems, such as pest and disease outbreaks (Axtell and Arends, 1990), antibiotic
resistance in bacteria (Ojeniyi, 1985), waste disposal costs and environmental pollution
risks (Pope, 1991; Sims and Wolf, 1994), and issues related to animal welfare (Appleby
and Hughes, 1991), are either non-existent or of considerably less importance in free-range
production systems. The benefits from free-range production may go beyond simple
avoidance of these problems when the bird production enterprise is integrated with other
agricultural production systems and may include sanitation, soil fertility maintanence, and
insect pest and weed management. Domestic birds, when used in conjuction with modern
rotational grazing technologies, may be especially useful as insect pest and weed
management agents.

Chickens, as meat and eggs, are an important source of food worldwide (National

Research Council, 1991). Production systems range from backyard scavenger flocks to

35

high-density confinement operations. In the United States the small scale poultry and eggs
producers have become inceasingly rare as their ability to compete with large producers has
been diminished. There are exceptions to this rule however, and successful producers
demonstrate that small-scale production is still potentially viable (T raupman, 1990; Salatin,
1991). Three practices which contribute to the profitability of small farming operations
include: 1) reducing input costs; 2) diversifying and integrating production operations; and
3) direct marketing. Based upon these three criteria small-scale poultry and egg operations
can be profitable and environmentally sound, while playing important ecological roles on
diversified farms including nutrient cycling, waste recycling, and biological control.

The effectiveness of chickens as biological control agents has been only minimally
studied. Although it seems to be common knowledge among many older farmers that
chickens eat insect pests there is very little documented evidence for this. One crop pest for
which chickens have been used as biocontrol agents is the Colorado potato beetle
(Leptinotarsa decemlineata). As mentioned, Colorado potato beetle is the most serious pest
of potatoes the United States and has developed resistance to nearly every insecticide used
against it (Radcliffe et al., 1991). During the mid-1800s, shortly after this insect had
become a pest on potatoes, there were reports of successful and unsuccessful attempts at
using chickens to control it. According to one farmer’s testimony chickens could be trained
to eat the beetles (Casagrande, 1987). However, recent research has demonstrated that the
beetles and larvae are distasteful to young chicks of at least one common commerial breed
(Hough-Goldstein etal., 1993).

Another insect pest for which chickens have been used as a biological control agent
is the plum curculio (Conotrachelus nenuphar). As mentioned above, it has been suggested
that ground-feeding domestic birds or pigs could be used to remove pest-infested apple
drops (Lafleur and Hill, 1987; Racette et al., 1992). There is some evidence in the

literature that domestic birds, including chickens, may also directly feed upon plum

36

curculio while it is on the ground (Racette et al., 1992; VABF, 1993). However, there is
still no experimental evidence that chickens can reduce damage by plum curculio.

Domestic geese have historically been used as weeders in a variety of crops. The
most common uses have been in cotton, fruit orchards, and nursury plantings (Hoare,
1928; Mayton et al., 1945; Johnson, 1960; Holderread, 1981; Gillen and Scifres, 1991).
Geese also have been found to be useful as weeders in strawberries (Cramer, 1992) and
may be compatible with a variety of other crops, including some hi gh-value vegetables.
Research is needed to determine which crops and at which growth stages geese may be
compatible and used as weed biocontrol agents. This type of information would be useful
to diversified, small farmers and market gardeners who might be willing and able to

integrate geese into their operation to reduce hand weeding and/or herbicide use.

CHAPTER 3

THE COMPATIBILITY OF DOMESTIC BIRDS
WITH A N ONCHEMICAL AGROECOSYSTEM

INTRODUCTION

Increasing interest in developing sustainable food production methods has
prompted research into a variety of alternative and traditional agricultural practices such as
biodynarnic farming, intercropping, companion planting, mulching, manuring, and crop
rotation. A common characteristic among all of these practices is the integration of multiple
production components, each intended to interact with and benefit the entire agroecosystem
(Dazhong and Pimentel, 1986; Altieri, 1987;Rudd1e and Zhong, 1988; Vandermeer, 1989,
1995; Mollison, 1990; Crews and Gliessman, 1991; Dazhong et al., 1992). The benefits
of such integrated production methods over conventional production methods can include
reduced waste, fewer external input requirements, increased energy efficiency, higher soil
fertility, and greater ecological and economic stability (Lockeretz et al., 1981; Wagstaff,
1987; Edwards, 1989; Smolik et al., 1993; Reganold et al., 1993).

My discussions with farmers in the Great Lakes region of the United States and
Canada have revealed that domestic birds are sometimes integrated into organic orchards.
In addition, information in the scientific, popular, and local farming literature indicates that
domestic birds are used as integral parts of orchard and non-orchard systems (Fuan, 1985;
Rodale Institute, 1990; Traupman, 1990; Salatin, 1991; Cramer, 1992; VABF, 1993).
However, we have found few experimental investigations of the role that birds play in such

systems (Mayton et al., 1945).

37

38

This chapter reports findings on two objectives: 1) to describe the behavior of
domestic chickens and geese in a potato-intercropped, apple orchard and 2) to determine the
impact of the birds on plants within the system, including crops and non-crops. In
addition, I discuss factors which may present opportunities for or barriers to domestic bird

integration into different types of agroecosystems.

MATERIALS AND METHODS
The Orchard System

A 0.5 ha block located in an apple orchard at Michigan State University’s Kellogg
Biological Station in southwestern portion of the state was used for this study. The
orchard was planted in 1983 with the disease resistant varieties ‘ Redfree,’ ‘ Priscilla,’ and
‘ Liberty’ on M.26 rootstocks and has been managed without the use of pesticides since its
establishment. While the orchard has sustained heavy insect damage during some years,
disease pressure has been low. This has allowed the orchard alleys to be used for
intercropping annuals, including potatoes, tomatoes, and beans, and for pasturing domestic
fowl, including chickens and geese. Under conventional orchard management this would
not have been possible due to the intensive spray schedule which requires that the alleys be
used for tractor traffic. The use of pesticides would also have been incompatible with the
birds.

In 1993 Barred Plymouth Rock chickens, a breed of Gallus domesticus, and
African geese, a domestic breed of Anser cygnoides (Silversides et al. 1988), were
evaluated as components of an potato-intercropped, apple orchard. The selection of these
birds for the experiment was based on recommendations from area farmers and on small-

scale test trials with several varieties of domestic birds in the orchard from 1990 to 1992.

39

Experimental Design

Twelve plots were randomly established in the block for three treatments with four
replications. Plots were 16.8m by 9.9m and consisted of three rows of three apple trees
and two inter-row alleys (Fig. 3-1). The trees in two of the rows were ‘Redfree’ and the
third were ‘Priscilla.’ The tree spacings were 3.3 m and the row spacings were 6.5 m. The
alleys were planted in potatoes (‘Red Pontiac’) on 18 May with 30-cm plant spacings and
75-cm row spacings. This allowed three rows of potatoes to be planted within each alley
such that the outside potato rows were not directly beneath the canopy of the apple trees.

In one randomly-selected alley of each plot the potato plants were mulched to a depth of 15
cm with recently cut rye (Secale cereale var. ‘Wheeler’) which was collected from a field
adjacent to the orchard, creating mulched and unmulched subtreatrnents. The potatoes were
handweeded on 11 June and hand weeded and billed on 17 June.

The three main treatments included 1) placing 15 chickens per plot (chicken
treatment); 2) placing 11 geese per plot (goose treatment); and 3) a control with no birds
(control). Plots with chickens or geese were surrounded with 1.5 m-high plastic fencing
with 6 cm mesh. On 2 July, the birds, at 6 weeks old and with an approximately equal
number of males and females, were introduced into the designated plots. On 17 July, 3
geese were removed from each plot of the goose treatment, setting the population at 8 per
plot. The birds were maintained in the plots until 6 August.

The birds were kept in coops (1.2 X 1.2 m) at night and released to free range from
approximately 7:00 am until 9:00 pm (EST). The coops were located on either the eastern
or western perimeter of the plots (Fig.3-1). The birds were provided with small amounts
of cracked corn and poultry starter immediately after being released from the coops in the
moming and approximately 10min before being put into the coops at night. The purpose of
the morning feeding was to train the birds to disperse out into the plots to find food, while

the night feeding was used to get the birds to return to the coop. The coop was closed each

40

night to protect the birds from predators. Water was provided in buckets and watering cans

and replaced as needed (usually 2 times per day).

Behavior Observations

Structured observations of the chickens and geese were conducted on arbitrarily-
selected days at arbitrarily-selected times between 8:00 am and 8:00 pm except during rain.
Bird behavior was sampled by observing the whole flock of birds in a particular plot for 5
min and recording the occurrence of the following activities: foraging, drinking, and
resting. Due to the difficulty in tracking individual birds for 5 min the entire flock within
each plot was treated as the sampling unit rather than individual birds. A total of 12 sets of
four, 5-min observations were collected for each bird treatment. These were grouped into
four, 3-hr time periods: morning (8:00 am - 10:59 am), mid-day (11:00 am -l:59 pm),
afternoon (2:00 pm - 4:59 pm), and evening (5:00 pm - 8:00 pm), which resulted in 5, 2,
2, and 3 observation sets for the morning, mid-day, afternoon, and evening periods,
respectively. The frequencies of activity occurrence were based on the percentage of
observations in which a particular activity was observed to be performed by at least one
bird in the plot. When possible, observers also recorded what foraging birds were eating.

Plot maps, which were used to record bird locations within the plots, were divided
into 25 sectors (Fig. 3-1). This allowed observers to simply indicate the presence or
absense of birds for each sector. These data were used to estimate the extent of bird
dispersion throughout the plots, deterrninine if the birds preferred a particular habitat within
the plot (apple row or potato alley), and measure the tendency of the birds to roam to the
side of the plot distal to the coop . Bird dispersion was determined by calculating the
percentage of plot sectors occupied by the birds at a particular time of day (Fig. 3—1).
Habitat preference was evaluated by comparing the frequency of occurrence of birds in the

tree row versus the potato alleys. The tendency of the birds to roam away from the coop

41

was evaluated by comparing the frequency of occurrence of the birds in the half of the plot

containing the coop versus the half without the coop (Fig. 3-1).

Plant Sampling

Above-ground, non-crop vegetation biomass was sampled in the apple rows and
potato alleys on 22 July by cutting at ground level and removing all vegetation within
randomly-selected 1 m2 quadrats. Weed biomass in the potato alleys was sampled at a
single, randomly-selected location along the middle potato row of each subplot. Ground
cover biomass within the tree rows was sampled in a single, random location in the middle
tree row of each plot. Plants were separated into grasses and broadleafs, dried for 48 hr,
and weighed. Potato tuber fresh weight was sampled on 30 July by digging and weighing
all potatoes in the center 2 m of each row, resulting in 6 m of potato row sampled for each
subplot. Apple fruit fresh weight was sampled by weighing all fruit from each tree at
harvest. Most of the ‘Priscilla’ trees did not bear fruit. Therefore comparsons were limited
to ‘Redfree’ trees only. N on-crop biomass, potato tuber fresh weight, and apple fresh
weight in the chicken and goose treatments were compared statistically to the Control using
the one-tailed Mann Whitney test (Zar, 1984). In the figures the following p—values are
indicated by one, two, and three asterisks, respectively: 0.10 2 p > 0.05; 0.05 2 p > 0.025;
and 0.025 2 p > 0.01.

RESULTS

The observations of chicken and goose activity patterns indicate that the chickens
spent consistently more time foraging than resting and drinking while for the geese this
varied more with the time of day (Fig. 3-2). The chickens were observed to forage
throughout the day whereas the geese foraged most during the morning and evening and
least at mid-day. As may be expected for the geese, resting and drinking were highest at

mid-day. Although systematic observations were not conducted during rain we noticed that

42

chickens tended to seek shelter under trees or in the coop during rain while geese appeared
unaffected.

Chicken dispersion in the plots was consistently greater than goose dispersion (Fig.
3-3). During morning and afternoon observations chickens occupied between 25% and
35% of the plots with peak dispersion at 9:00 am and 10:00 am. While geese dispersion
peaked at the same time, the geese remained in a tightly clustered flock occupying less than
15% of the plot. Chickens were observed to come together into tighter flocks during mid-
day and evening (Fig. 3-3).

Observations indicated that chickens and geese consistently occupied apple rows
throughout the day (Fig. 4). Chickens were also commonly observed in the potato alleys.
In contrast, the geese were observed less frequently in the potato alleys (Fig. 3-4). Geese
also tended to remain in the half of the plot which contained the c00p while chickens were
observed to consistently roam on both sides of the plot, especially during mid-day and
afternoon (Fig. 3-5).

Observations of foraging birds indicated that the chickens were omnivorous,
feeding on insects and plants, while the geese were strictly herbivorous (Table 3-1).
Chickens were seen feeding on some plants but were more frequently observed feeding on
invertebrates on or near the soil surface. Most of the time it was impossible to see what
types of organisms were being preyed upon; therefore the Table 1 is incomplete, likely
representing only a small fraction of the items consumed. Goose forage, because it was
solely plants, was more easily identified and included grasses and broadleaf plants.
Interestingly, geese were observed to actively pursue some flying insects for short
distances (less than 1 m), however a successful catch was not seen. Food preferences
were not quantified but it appeared that the geese preferred grasses (Poaceae spp.),
common dandelion (T araxacum ofiicinale), and common plantain (Plantago major).
Common ragweed (Ambrosia anemisiifolia) and common larnbsquarter (Chenopodium

album) were only consumed when more preferable plants were unavailable. Canada thistle

43

(Cifisiwn arvense) was fed upon only as a seedling. Although chickens and geese were
seen feeding on unearthed potato tubers neither were observed feeding on potato plants.
Occasionally, geese were seen feeding on apple leaves on low-hanging branches and on
recently dropped fruit on the ground. By contrast, chickens were seen feeding on low-

hanging fruit on the trees, however, they were never seen in the trees feeding on fruit.

Groundcover biomass in the tree rows was substantially reduced in the goose
treatment compared to the control (Fig. 3-6). This reduction was primarily due to a
reduction in grass, supporting the observation that grasses were a preferred forage. A
slight reduction in groundcover biomass was also measured in the chicken treatment,
however it was not statistically significant. Weed biomass in the potato alleys followed a
similar trend, being slightly reduced in the chicken treatment and substantially reduced in
the goose treatment compared to the control (Fig. 3—7).

Although weed biomass in the potato alleys varied considerably across the
treatments, potato tuber yields were statistically similar among the treatments. The pooled
mean yield for the three treatments was 6.02 ;I; 0.3 kg per m of potato row (equivalent to
4754 kg/ha [or 4231 lbs/acre] of intercropped orchard). The lack of treatment effect was
probably due to the extensive hand weeding in all treatments prior to the introduction of the
birds into the plots, giving the potato plants in all three treatments a competitive advantage
over the weeds (Bellinder and Wallace, 1991). Colorado potato beetle (Leptinotarsa
decemlineata), the most important potato pest in Michigan, was not present in this
experiment until late July (greater than 60 days after planting) and therefore probably had
no influence on yield (Zehnder and Evanylo, 1989).

Apple yields in all treatments were low in 1993 due to a lack of fruit thinning which
has caused the orchard to alternate biennially with high and low yields. Mean apple fruit
yield was greater in the chicken and goose treatments than in the control, however the

differences were statistically significant only for the goose treatment (Fig. 3-8). This yield

44

increase was apparently due to the reduction in weed competition in the goose treatment

(Rupp and Anderson, 1985; Raese, 1990).

DISCUSSION

If the compatibility of the birds in this agroecosystem is assessed only on ground
cover management and weed control the geese would appear to be ideal components.
However, a variety of other factors, some of which were addressed in this study, need
appropriate consideration. The compatibility of the birds in this or any type of
agroecosystem will depend upon some factors which are general and predictable and others
which are local and unique. In this study, plots were relatively small and the results
obtained are not likely to be representative of larger plots. Mayton et al. (1945) observed
chickens and geese to effectively control nutgrass in small plots but found that they
provided inadequate control in large plots because the birds would not range far enough
from the coop. A problem such as this, however, may be managed by rotational grazing
within smaller paddocks.

According to Gillen and Scifres (1991, p. 370), “the successful use of selective
grazing is more an art than a science, which usually disallows quantifying necessary
procedures to reach an end result except in selected cases.” A number of variables in this
study, including bird density, bird age, paddock size, supplemental feeding, and date of
introduction, could have a substantial influence on the results of experimental grazing
evaluations and it was not feasible to evaluate all of the possible combinations. We chose
values for these variables based on our previous experience and limitations. However,
they are not necessarily the optima for this system or any other.

According to Mollison (1988) the suitability of introducing domestic birds into a
production system depends upon 1) the behavior of the birds, 2) the nutritional
requirements of the birds and the provisions within the system, and 3) the interactions

between the birds and the rest of the system, including the crops, non-crop plants, other

45

animals, and people. If bird compatibility is evaluated according to these criteria then the
differences between the chickens and geese become more obvious than the similarities.
The feeding behavior of the chickens is omnivorous while that of the geese is herbivorous.
The chickens were active throughout the day, stayed fairly well dispersed, and required a
minimal amount of human attention and labor for management. In contrast, the geese
tended to be active mostly during the morning and evening and stayed in a flock which was
less inclined to move away from the c00p. However, when foraging, the geese consumed
large amounts of vegetation. The geese also consumed large amounts of water and this
demand for water required a relatively high level of human attention and labor in this
experimental system.

The relatively spacious and complex environment within the orchard provided a
greater degree of welfare for the birds than would be possible under conventional bird
production systems (Appleby and Hughes, 1991). The trees provided shade for the
chickens and geese and roosting places for the chickens. We did not determine if the food
provisions within the system would be adequate to sustain the birds without supplemental
feeding. To address this question would require knowledge of the specific objectives of
the farmer. If the goal was to maximize the use of the birds as ground cover or weed
management agents then supplemental feeding could be minimized or possibly even
eliminated. If the birds were intended for egg production or to be sold as meat adequate
supplemental feeding would be required.

The interactions between the geese and weeds suggest that geese could be managed
as weed control agents under a variety of circumstances. By contrast, chickens would need
to be stocked at greater densities than were used in this experiment if weed control was the
goal. However, in addition to weed control, chickens may provide biological control for
some insect pests. They were observed commonly to feed upon Japanese beetle (Popillia
japonica) and occasionally feed on Colorado potato beetle. Although chickens have been

described as effective insect predators (Hoare, 1928), there is some evidence that certain

46

insects, including Colorado potato beetle, are unpalatible (Hough-Goldstein et al., 1993).
However, we found that chickens in this study readily consumed Colorado potato beetles,
even when alternative food sources were available.

Assuming that adequate human management is feasible and that a reliable source of
water is available, chickens and geese appear to be highly compatible in this experimental
agroecosystem according to Mollison’s criteria. However, ecological compatiblity alone
does not mean that bird integration is economically feasible under current market
conditions. Although it is difficult to assess the economics of an experimental system such
as this one there are several characteristics of this type of integrated production system
which may make bird introduction feasible. If a rotational grazing system can be used
effectively the birds might eliminate needs for herbicide applications or mowing; Operations
which can represent a substantial portion of the management costs. For example, in a small
low-input apple orchard which relied on mowing and mulching for weed control, these
operations accounted for 25% and 40% of total labor and material costs, respectively
(Prokopy, 1991). Although the birds do require a high level of human attention, they
could be managed to provide marketable products, such as eggs and meat, which could
offset the labor costs. In addition, the birds could play an important role in nutrient
cycling, reducing the need for external fertilizers. Clearly, the scale of the farming
operation will have an important influence on the feasibility of domestic bird introduction.
Domestic birds in orchards or other agroecosystems require relatively close human
management, for protection of the birds and crops and for the efficient use of the birds,
making bird integration more amenable to operations already accustomed to high labor
inputs.

The development of ecologically-based, non-chemical food production systems will
require that agroecosystems be designed so that they are more energy efficient, are
economically stable, produce little or no pollution, require less therapeutic intervention to

maintain pests at acceptable levels, and are compatible with the needs and philosophies of

47

the producer and consumers (Edens and Haynes, 1982; Vereijken, 1989; Hill and MacRae,
1992). Although these criteria have been used in designing this experimental
agroecosystem the relative emphases put on each criterion will undoubtedly differ from
those identified by farmers themselves. The objectives in this study were to start to
understand the interactions between three components: domestic birds, potato plants, and
apple trees. Although this experiment indicates that these components are compatible and
may benefit each other in various ways (Fig. 3-9), the practicality of such a system from a
farmer’s perspective will ultimately depend upon numerous social, economic,

philosophical, and organizational factors both within and beyond the farm.

48

Table 3-1. A list of food items observed to be consumed by chickens and geese in the

potato-intercropped, apple orchard in 1993.

 

 

 

Commune Scientificl‘larm Ema
Chickens

Colorado potato beetle leptinotarsa decemlineara adults, larvae
Japanese beetle Popillia japonica adults

ant Formicidae sp. adults, larvae
ground beetle Carabidae sp. adult

rove beetle Staphylinidae sp. adult

tomato homworm Manduca quinquemaculata larvae

apple Malus pumila fruit on tree or ground
potato Solanum tuberosum unearthed tubers
grass Poaceae spp.

birdsfoot trefoil Lotus comiculatus

earthworm Lumbricidae spp.

Shea:

grass Poaceae spp.

common dandelion Taraxacum ofi‘icinale

common plantain Plantago major

common larnbsquarter Chenopodium album

redroot pigweed Amaranthus retroflexus

common ragweed Ambrosia artemisiifolia

Canada thistle Cirisium arvense seedlings

curly dock Rumex crispus

apple Malus pumila fruit on ground
potato Solanum tuberosum unearthed tubers
bean Phaseolus vulgarus

 

a Indicates the specific life stage or part of the item consumed or the circumstances during

the consumption.

0.5 ha orchard block

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I Fl T +I:L~E' Potato Rows
‘ Coop Potato Alle‘y T ~ ~ ~
‘ 2.7m 3.8m
A‘ 7 ‘III
‘< III III
‘ ‘ III III
" III III
99m ‘ III III
‘ III III
‘ III III
‘ III III
TI P III R Ill
‘ I—l—k m
‘<
16.8m
I
‘ I I | I I
I __'___'___'__'.___'___
\ lplotsector' I I I
‘ I I l I I
'f—""'I---T""l-T---r--—
“ I plot half with coop I l I plot half without coopl
"---l--"T-"—T"'-I--—
‘ I I I I I
"""——|——"'I'—"'T——"|—-—
I I I I I
T l I l l I

 

 

Figure 3—1. Map showing the experimental layout. Apple trees are designated P and R for
‘Priscilla’ and ‘Redfree’ respectively. One randomly-selected potato alley of each plot was
mulched while the other was left unmulched.

50

I Foraging
Chickens I Resting

Z Drinking

é.

Frequency of Activity
(% of observations)

 

     

Morning Mid-day Afternoon Evening

Geese
100‘

   
 
  
   

80‘

60‘

A
G
1

Frequency of Activity
(% of observations)

 

 

 

c 3
P I

Morning Mid-day Afternoon Evening

Figure 3-2. Chicken and goose activity patterns based on the mean percentage of
observations in which birds were observed to be resting, foraging, and drinking. The time
periods are morning (8:00am-10:59am, N=5), mid-day (11:00am-1:59pm, N=2),
afternoon (2:00pm-4z59pm. N=2), and evening (5:00pm-8:00pm, N=3). Error bars are

standard errors of the mean.

51

45 ‘
40 _ —D— ChTCkenS

35 _ . Geese
30'
25 “
20‘
15‘
10‘

Dispersion
(% of Sectors with Birds)

 

 

I ' l

8aml10ar: thoori '2me 4pm” 6pm '8pn

Figure 33. Chicken and goose dispersion within the orchard plots as indicated by the
mean percentage of plot sectors occupied by birds at arbitrarily-selected times of day over a

one-month period (N=4 for each mean). Error bars are standard errors of the mean.

I Chickens in apple rows
2] Chickens in potato alleys

Ehéh

Morning Mid-day Afternoon Evening

Chickens

if

00
O
r

A
G
1

Frequency of Occurrence
(% of observations)
N 6\
°. r

 

 

 

 

G
r

I Geese in apple rows
Geese Z Geese in potato alleys

   
  
  
  
 

 

 

a 100-
EA
gag 80-
o:
“a 601
°'8
E"; 40-
09
2+5 20-
hr
III

0—

 

 

Morning Mid-day Afternoon Evening

Figure 3-4. Frequency of chicken and goose occurrence in the apple rows and potato
alleys as indicated by the mean percentage of observation periods in which the birds were
observed in these habitats. The time periods are morning (8:00am-10259am, N=5), mid-
day (11:00am-1259pm, N=2), afternoon (2:00pm—4z59pm. N=2), and evening (5:00pm-

8:00pm, N=3). Error bars are standard errors of the mean.

Frequency of Occurrence
(% of observations)

Frequency of Occurrence
(% of total observations)

A ex cc 'T
¢ e o 8
I I I I

N
G

G
gr

Chickens

5.

80‘

60‘

40‘

20‘

 

O
r

Geese

 

53

I Plot half with coop
2] Plot half without coop

  
  

  

 

 

 

/A

Mid-day

A

Afternoon

 

 

Morning

Evening

I Plot half with coop
[2 Plot half without coop

  
   
   
  

 

Afternoon

Morning Mid-day Evening

Figure 3-5. Frequency of chicken and goose occurrence in the plot half with the coop and

the half without the coop as indicated by the mean percentage of observation periods in

which the birds were observed in these areas. The time periods are morning (8:00am-

10259am, N=5), mid-day (11:00am-1:59pm,-N=2), afternoon (2:00pm-4z59pm. N=2),

and evening (5:00pm-8200pm, N=3). Error bars are standard errors of the mean.

54

   
 

’I: .
"1% 300 I Control
35 I Chicken
g; - D Goose
a: 200
:3
2::
SE 100-
“r us
$1:
:9 7/

 

     

 

 

   

O
I

Broadleaf Grass Total

Figure 3-6. Above-ground vegetation biomass (grams dry thmZ) in the tree rows under
the three treatments. Total biomass is broken down into grass and broadleaf weeds.
Asterisks indicate the level of statistical significance (see Materials and Methods) of the
difference between the Control and each bird treatment (one-tailed Mann-Whitney test).

Error bars are standard errors of the mean.

55

  
   

60' I Control
I Chicken
50‘ E Goose

 

101

Weed Biomass
(g dry thsquare meter)

6
L

unmulched mulched

Figure 3-7. Above-ground weed biomass (grams dry wt./m2) in mulched and unmulched
potatoes under the three treatments. Asterisks indicate the level of statistical significance
(see Materials and Methods) of the difference between the control and each bird treatment

(one-tailed Mann-Whitney test). Error bars are standard errors of the mean.

56

Apple Yield
(kg fresh wt./tree)

 

 

Control Chicken Goose

Figure 3-8. Apple fruit yield (kg fresh wt./tree) for ‘Redfree’ trees under the three
treatments (1 kg = 2.2 lbs). Asterisks indicate the level of statistical significance (see
Materials and Methods) of the difference between the control and each bird treatment (one-

tailed Mann-Whitney test). Error bars are standard errors of the mean.

57

 

 

 

f pest conIrol \
weed management
manure fertilizer

  
 

 

Apple
Trees

   

Domestic
Fowl

1'

 
  
 
 

shade
habitat for insect prey

      
 
  

  
 

. habitat tor
habitat for insect prey
beneficials

 
 

 

   
  

    
 

shade Vegetables pest control
wind barrier weed management
habitat for beneficials manure fertilizer

 
 

BBB

  

 

 

Outputs
supplemental bird feed apples
vegetable crop seed vegetables
human labor 69$
diesel fuel poultry

Figure 3-9. Conceptualization of an integrated agroecosystem based on the experimental
agroecosystem in this study. The arrows indicate the potential beneficial interactions

between the components of the system.

CHAPTER 4

EVALUATION OF FREE-RANGE CHICKENS AND GEESE
FOR MANAGING INSECT PESTS AND WEEDS
IN AN AGROECOSYSTEM

INTRODUCTION

The integration of animals into cropping systems has the potential to reduce the
need for external farm inputs. Benefits can include nutrient cycling and conservation,
utilization of crop residues, economic stability through market flexibility, and vegetation
and weed management (King, 1990; Parker, 1990; Gillen and Scifres, 1991; Loomis and
Connor, 1992; Bender, 1994). Most efforts at animal integration in the United States,
however, have focused on large livestock, specifically cattle. However, smaller livestock
species and breeds, sometimes termed rnicrolivestock, have the potential to play a variety
of roles on small farms and market gardens. These roles may include insect pest control,
weed management, sanitation, as well as product diversification.

The use of livestock as biological control agents is relatively unusual in comparison
to the use of arthropods and microbes. Reasons for this include the limited choices of
species and lack of research on livestock for such purpose, as well as the relatively large
investment of labor and land resources required in using such organisms. Furthermore, the
use of livestock requires more managerial experience and skill. Nevertheless, small and
large livestock have been successfully employed as biological control agents. Most
examples are in the arena of weed management. These include using geese for weed
control in strawberry (Cramer, 1992; Ware, 1995) and cotton (Mayton et al., 1945;
Johnson, 1960). Recent experiments by Wurtz (1995) showed the potential for using

58

59

geese as part of an integrated weed management strategy during the establishment of tree
seedlings. Sheep and goats also have utility in weed management including grazing in
orchards and woodlots (Shirley, 1992) and brush removal (Olkowski et al., 1991).

Documentation of insect pest management using livestock is extremely limited.
There is anecdotal information suggesting that poultry can reduce the abundance of orchard
pests (Hoare, 1928; Racette et al., 1991; VABF, 1993), Colorado potato beetle on potato
(Casagrande, 1987), and flies in cattle pasture (Salatin 1991). However there is no
experimental evidence to support these claims. Muscovy ducks have been shown
experimentally to substantially reduce house fly populations in dairy and swine operations
and appear to be an economically feasible option for this application (Glofcheskie and
Surgeoner, 1990; 1993).

The behaviors and activities of free-range domestic chickens and geese in a
nonchemical food production system comprised of apple and potato were discussed in the
previous chapter. Chickens were found to be omnivorous, feeding on weed seedlings,
seeds, insects including several pest species, and soil invertebrates. Geese, by contrast,
fed almost solely upon grasses and broadleaf weeds. Both chickens and geese were found
to be compatible with the experimental agroecosystem and showed promise as insect pest
and weed biological management agents in small—scale operations.

The purpose of this chapter is to report the results of experiments intended to
measure the effectiveness of chickens and geese as biological pest management agents. The
objectives were to: 1) measure the effects of free-range chickens and geese on the
abundance of several important insect pests and on weeds in a non-chemical agricultural
production system and; 2) assess the potential of these domestic birds as crop protection
agents based on their effects on crop productivity and pest damage, the economics of their

use, and their potential compatibility with other types of production systems.

60

MATERIALS AND METHODS
Study Site

All research was conducted during 1994 and 1995 in a 2-ha apple orchard at the
Kellogg Biological Station, Hickory Comers, Michigan. The orchard, planted in 1983,
consists of three disease-resistant varieties, ‘Redfree,’ ‘Priscilla,’ and ‘Liberty,’ on serni-
dwarfing rootstocks (M26). The trees are arranged in three blocks of 0.5 to 0.7 ha which
represent high (430 trees per ha), medium (225 trees per ha), and low (110 trees per ha)
tree planting densities. All three blocks have been managed without the use of pesticides
since the establishment of the orchard. The trees are pruned annually and all cuttings are
burned or removed from the orchard. The fruit is not thinned, therefore harvests alternate
biennially between high and low yields. Apple drops have been removed from the orchard
in years of heavy yield. In 1994, codling moth was managed in the orchard with
pheromone disruption. The groundcover throughout the orchard consists primarily of
orchard grass (Dactylus glomerata) and Kentucky bluegrass (Poo pratensis) but more than
twenty other plant species are also present. The alleys of the high density block were
intercropped with potatoes in 1993. Other vegetable crops, including beans and tomatoes,

have been intercropped in smaller areas of the orchard in previous years.

Effects on insect pests and weeds

In 1994 the effects of domestic chickens and geese on insect pests and weeds were
evaluated in the high density block of the orchard. The alleys were planted into potatoes
(var. ‘Red Pontiac’) at 30-cm plant spacings and 75-cm row spacings on 6 May 1994.
This arrangement allowed three rows of potatoes in each alley (Fig. 4-1). Nine plots, each
measuring 28 m by 12 m, were established. Each plot consisted of five rows of four trees.
Twelve of the trees were ‘Redfree’ and the remaining 8 trees were ‘Priscilla’ or a

combination of ‘Priscilla’ and ‘Liberty.’ The plots were randomly allocated to one of three

61

treatments: 1) with 20 Plymouth Barred Rock chickens per plot (chicken treatment); 2) with
7 African geese per plot (goose treatment); and 3) no birds (control). There were three
replications of each treatment. On 14 May, the designated number of 6-week-old birds
were introduced into the plots. The birds were allowed to free-range from 8am until 9pm
each day and put into coops (1.2 m Xl.2 m) located in the center of each plot at night. The
flock within each plot was provided with 0.2 kg of grain-based commercial feed each
morning and evening.

On the night of 18 May, 30 chickens and 11 geese were stolen from the experiment.
The stolen geese were replaced the following day with geese of the same age. Replacement
chickens of the same variety and age were not immediately available. Therefore, chicken
densities were maintained at 10 birds per plot until 14 June when nine, 3-week-old White
Plymouth Rock chickens were added to each plot. This set the densities at 19 chickens per
plot. Over the course of the experiment, further chicken losses were incurred to predators
including a red-tailed hawk (Buteo jamaicensis) and great horned owl (Bubo virginianus).
These losses were monitored but not replaced. Final chicken densities on 5 August were
12 per plot.

The population dynamics of four insect pests, plum curculio (Conotrachelus
nenuphar), apple maggot (Rhagoletis pomonella), Japanese beetle (Popillia japonica), and
Colorado potato beetle (Leptinotarsa decemlineara), were monitored. The first three pests
directly damage apple fruit while the last feeds upon potato foliage. Japanese beetle also
feeds upon apple and potato foliage. These insects were selected because they are serious
economic pests of these crops and have life cycles which include at least one stage which
spends time on the soil surface where it could be vulnerable to chicken predation.

Four of the six trees in the center of each plot were designated for monitoring the
three apple pests (Fig. 4-1). When possible, two of the trees were ‘Redfree’ and two were
‘Priscilla.’ In plots where one of the ‘Priscilla’ trees was missing or dying an additional

‘Redfree’ was monitored. Due to differences in life cycles of the pests and available

62

sampling technologies different monitoring methods were used for each pest. Plum
curculio was monitored indirectly by examining all fruit for oviposition and feeding damage
on four branches per tree, each week during June. Apple maggot was sampled by counting
the number of adults on sticky red spheres covered with adhesive. One sphere was hung
1.5 m high in each sampling tree on 26 June and sampled every 10 (1 until 5 August.
Japanese beetle abundance was monitored directly by counting all beetles on each of the
sampling trees every 5-10 days from 30 June until 5 August.

Colorado potato beetle larvae and adults were counted at eight designated areas in
each plot. Each sub-sampling area consisted of three l-m lengths of potato row, resulting
in 24 m of row per plot. Counts were conducted every 5-10 days from 14 June until 29
July.

Potato growth and weed abundance were monitored by measuring the percent
groundcover of potato and non-potato plants at four locations in each plot. Percent
groundcover was measured using a 1 m2 quadrat with wires forming a grid of 81 points
which was centered over the middle potato row in each alley. The presence or absence of
potato and non-potato foliage was determined at each grid point. The data from the four
locations of each plot were pooled and used to calculate percent groundcover of potato
plants and weeds.

Yields of apple fruit and potato tubers were measured at harvest. Apple fruit yield
and damage assessments were restricted to ‘Redfree’ due to the uneven distribution of the
other two varieties across the experimental site. Only apples on the trees, and not drops,
were considered in yield determinations. On 18 August, the fruit from all trees in each plot
was weighed and five apples from each tree were randomly selected for insect damage
assessment. These were examined for damage by plum curculio, apple maggot, and
Japanese beetle. In addition, damage by codling moth and tarnished plant bug were
recorded. Fruit with no insect damage or blemishes of any kind were considered “pest-

free.” Potato yields were estimated on 3 August by digging and weighing the tubers of 6m

63

of row in two of the four alleys of each plot. Insect pest and weeds abundance in the
treatments were compared graphically with time-series data. Fruit yield and damage and
potato yield were statistically compared using one-way analysis of variance followed by

Tukey’s multiple range test. P-values 5 0.10 were considered to be statistically significant.

Performance of an integrated bird system

In 1995 the performance of an integrated bird system which included free-range
chickens and geese was evaluated in the high density orchard block. This experiment
included two treatments with three replications: 1) free ranging chickens and geese in the
system as needed for vegetation and insect pest management (bird treatment) and 2) using
minimal hand weeding with no birds (control). The bird treatment and control were
assigned to the plots which in 1994 had been used for the goose treatment and control
respectively. Potatoes were planted on 3 May 1995 using the same protocol as described
for the experiment in 1994.

A flock of seven, 5-week old geese was introduced to each plot of the bird
treatment on 8 May. Within three weeks most of the preferred plant species (common
dandelion, common plantain, and grass seedlings) had been eliminated by the geese.
Therefore, on 28 May four geese were removed from each plot setting the density at three
per plot. These geese were kept in the plots until 13 June. Three geese were put back into
the bird treatment plots for five days in early July to eliminate emerging grass seedlings.

Following the removal of the geese on 13 June, a flock of 20, ten-week-old, Barred
Plymouth Rock chickens were introduced into each plot of the bird treatment. (Prior to
introduction the chicken flocks had been maintained in the areas of the orchard block which
had been used for the chicken treatment in 1994.) The chicken flocks were kept in the plots
until 1 August. However, during this time an average of six chickens per plot (30% of
each flock) were lost to predators, primarily hawks and owls. Chicken flocks were

maintained at equal densities by moving birds from undisturbed plots into plots where

64

predation had occurred. In addition, four chickens were removed from each plot on 12
July for the feeding habit evaluation (described below). Thus, at the termination of the
birds treatment on 1 August, each plot had a flock of 10 chickens.

A single hand weeding was performed in the bird treatment on 30 June to remove
unpalatable weeds, primarily daisy fleabane (Erigeron annuus) and curly dock (Rumex
cn’spus). Weed management in the control plots consisted of a single mowing in the tree
rows and two cultivations in the potato alleys. Vegetation in the tree rows was cut to 15 cm
using a gasoline-powered trimmer on 13 June. The first cultivation in the potatoes in the
control plots was accomplished on 15 June using a rototiller in between the rows and a
hand hoe within the rows. The second cultivation was completed on 5 July with a hand
hoe due to wet soil conditions.

All variables related to insect and weed abundance and crop growth, yield, and
damage were measured according to the protocols described for 1994, except: 1) plum
curculio abundance which was not monitored in 1995; 2) percent groundcover of
individual weed species was determined on one sampling date, 30 May; and 3) the weight
of apple drops was determined for the four sampling trees of each plot and compared
between the treatments as the percentage of total yield in order to correct for yield
differences between plots. Potato yields were determined on 2 August and apple yields on
22 August. Insect pest abundance, weed abundance, fruit yield and damage, and potato
yield were analyzed and compared as described for the 1994 experiment. In addition, all
labor and material inputs for pest management in the two treatments were recorded. These

values were sealed for a 1 ha system and used for an economic assessment.

Free-range chicken feeding habits
In the previous chapter I reported that determination of the feeding habits of free-
range chickens by observation was more difficult than that of geese because the diet was

comprised mostly of small insects and other invertebrates on the soil surface rather than

 

65

plants. Thus, to identify items comprising the chicken diet, dissections were done on the
digestive crops of chickens killed by predators and sampled for experimental purpose.
Many of the chickens killed by raptors were decapitated but had little damage to the rest of
the body. Between 6 - 19 June, ten chicken carcasses were found in good condition and
promptly frozen for dissection. In addition, on 19 July, four chickens were removed from
each bird plot and slaughtered. All of the digestive crops were dissected and the contents

identified, usually to the taxonomic family level.

Vegetable crop compatibility with weeder geese

To assess the compatibility of several common vegetable crops with weeder geese a
feeding trial and an unreplicated garden plot study were conducted in 1995. The crops
tested included tomato, pepper, eggplant, broccoli, cauliflower, cucumber, basil, oregano,
and string bean. Plants were 4-6 weeks old at the time of the trial.

The palatability of the crops was initially tested on 22 June with a feeding trial using
five, lO-week-old geese. Each goose was given several common dandelion leaves before
being offered leaves of each crop. After the dandelion leaves were consumed several
leaves of a given crop were offered. If any of the leaves were consumed by at least one
goose the test for that plant was considered positive for palatability.

On 23 June, nine geese were introduced into an experimental garden plot which
consisted of two 28 m by 2.5 m beds, each situated between rows of apple trees. The
garden beds and tree rows, which occupied 350 m2, were surrounded by electric poultry
netting. A total of 223 crop plants were present with each crop being represented by 6 to
61 individual plants. All of the plants were 15-45 cm in height except for the pepper plants
which were 7 cm high. Tomatoes and eggplant plants were staked and cucumber plants
were trellised; all other plants were without artificial support systems. The geese were
maintained in the plot for four days. Each day all of the 223 crop plants were examined for
feeding and trampling damage by the geese and classified as having no, light, or heavy

66

damage. A classification of ‘light’ damage was given if the plant appeared able to recover
without yield loss; ‘heavy’ damage was given if yield loss appeared likely. In addition,
weed abundance was monitored by counting all of the weed seedlings in six, arbitrarily-

selected 1m2 samples in the garden beds.

RESULTS
Effects on Insect Pests and Weeds

In 1994 apple yields were relatively high for this orchard. Mean fruit weight for
“Redfree’ trees in the control, chicken, and goose treatments were 31, 36, and 34 kg/tree
(68, 79, and 75 lbs/tree) respectively. These differences were not statistically significant (P
> 0.10).

Graphical comparisons indicated that the population abundance of plum curculio
(Fig. 4-2), apple maggot (Fig. 4-3), and Japanese beetle (Fig. 4-4) were slightly lower in
the chicken and goose treatments compared to the control throughout monitoring periods.
However, only the proportion of fruit damaged by to plum curculio differed statistically
across treatments (P = 0.03) (Fig. 4-5). In the goose treatment 61% of the fruit had plum
curculio damage compared to 78% in the control . The difference in plum curculio damage
resulted in a significantly higher proportion of pest-flee fruit in the goose treatment (P =
0.06) (Fig. 4-5). The control had a mean of 9% pest-free fruit compared to 21% in the
goose treatment. The chicken treatment had intermediate levels of plum curculio damage
and pest-free fruit. Because geese are almost completely herbivorous, the effect on plum
curculio was considered to be indirect and due to habitat modifications caused by the
grazing and weed trampling. Damage by codling moth and tarnished plant bug were
relatively low in comparison to that of other pests and did not differ among the treatments.
(Pheromone disruption for codling moth was used in the entire orchard in 1994.)

Potato yields were substantially higher in the goose treatment compared to the

chicken treatment and control (P =0.009) (Fig. 4-6). The mean tuber weight in the goose

67

treatment was 5860 kg /ha (equivalent to 14122 kg/ha [or 12569 lbs/acre] in monoculture).
Interestingly, Colorado potato beetle abundance was also highest in the goose treatment
(Fig. 4-7). In fact, only the goose treatment displayed clear generational peaks of Colorado
potato beetle larvae and adults. This apparent contradiction between pest abundance and
yield can be explained with the weed and potato growth data (Fig. 4-8). Potato plants in
the goose treatment had a clear advantage over weeds which were maintained at less than
10% groundcover throughout the season by goose grazing. Potato plant growth rates after
emergence were substantially greater than those of the chicken treatment and control. At
the time of Colorado potato beetle larval emergence (40 days after planting) potato plants in
the goose treatment were more than four times larger than those of the other treatments.
Research has shown that large potato plant size and low weed abundance have a positive

effect on Colorado potato beetle populations (Horton and Capinera, 1987).

Performance of an integrated bird system

Apple yields in 1995 were extremely low due to biennial yield cycling. Mean fruit
weights at harvest for the bird treatment and control were 7.6 and 5.5 kg/tree (or 16.7 and
12.1 lbs/acre], respectively, and did not differ statistically (P > 0.10). The weight of apple
drops as a percentage of total fruit produced averaged 28% for ‘Redfree’ trees and 18 % for
all sampling trees, but did not differ statistically between treatments (P > 0.10). As in
1994, plum curculio damage was most common, however, codling moth damage increased
dramatically to be the second most common type found (Fig. 4-9). Apple maggot
abundance (Fig. 4-10) and damage (Fig. 4-9) were similar between treatments. Plum
curculio abundance was not measured during the season, however, damage was
significantly lower in the bird treatment compared to the control (P = 0.07) (Fig. 4-9).
Once again, this resulted in a greater prOportion of pest-free fruit (P = 0.06). Although
Japanese beetle abundance was consistently lower in the bird treatment compared to the

control (Fig. 4-11) the proportion of damaged fruit did not differ between the treatments.

68

Potato yields in this experiment were very low due to a disease infestation
(probably early blight, Altemaria solam). This was not surprising considering this was the
third year that this soil had been planted in potatoes. Colorado potato beetle abundance
patterns were similar to those observed in 1994. A clear larval abundance peak occurred in
the bird treatment in late June, approximately 50 d after planting (Fig. 4-12). Considerable
variation between plots is evident in the large standard errors of the means. Adult
abundance was greatest in mid-July and peaked in both treatments approximately 75-80 d
after planting (Fig. 4-12).

Weed growth was considerably greater in the control compared to the bird treatment
at the time of potato plant emergence (Fig. 4-13). This difference was accounted for by
substantial reductions in orchard grass (Dactylus glomerata), Kentucky bluegrass (Poo
pratensis), common chickweed (Stellaria media), and common plantain (Plantago major)
(Table 4-1). This resulted in greater potato growth in the bird treatment, especially during
the first several weeks after plant emergence (Fig. 4-13). The first cultivation in the control
was apparently done too late to have an effect on potato growth. Although the potato plants
in the bird treatment were consistently larger than those of the control for the first 70 d after
planting, yield samples taken 92 d after planting showed no statistical differences between
the treatments (P =0.21). Mean tuber weights for the bird treatment and control were 1341

and 874 kg/ha (or 1193 and 778 lbs/acre), respectively.

Economic Assessment

The hypothetical comparison of labor and material inputs for insect pest and weed
management considered only the geese because no clear crop protection benefits were
derived from the chickens (Table 4-2). Based on the densities used in this study it was
assumed that a flock of 75 geese could be used to effectively manage a 1 ha (2.47 acre)

orchard with intercropped potatoes. A 20% loss to mortality agents including sickness and

69

predators was included. In addition, it was assumed that a paddock (0.25 - 0.5 ha) was
available for pasturing the geese during periods of slow weed growth. ‘

Geese would be purchased as l-day-old goslings, put out into the orchard at 5
weeks old, and kept for a total of 150 days. Daily management of geese on 1 ha would
require 0.5 h of labor for feeding, watering, housing, and fencing. It was assumed that the
intercropped orchard would be divided into two equally-sized paddocks and that the geese
would be rotated between these areas every 1-2 weeks. In this situation 300 m of fencing
would be required for a 0.5 ha area. Material costs for housing (coops of wood and wire)
and fencing (electric poultry netting) were given a 10-year depreciation which results in
$110.00 per year. Additional hand weeding to remove unpalatable weeds would require
1 1h.

Without geese, nonchemical weed management for this experimental production
system would require the use of a rototiller for cultivation in the potatoes and a gasoline-
powered weed trimmer for vegetation management in the tree rows. Both of these activities
require a substantial input of human labor. Furthermore, at least some weeding with hand
tools would be required within the potato rows. Based upon the labor needed for the
control it is estimated that 206 h would be required for weed management on 1 ha of
intercropped orchard. Fuel and supplies would be slightly greater without geese.

The results of this hypothetical comparison indicate that the input costs for using
geese are slightly greater than using human labor (Table 4-2). The largest costs associated
with using the geese are the initial purchase of the birds and the necessary feed. Care of the
geese required approximately 0.5 hour of labor per day. However, if it is assumed that the
geese can be sold at the end of the season for $10 per bird, the total weed management cost
for the weeder geese system is less than 55% of the cost of the system based on human

labor (Table 4-2).

70

Free-range chicken feeding habits

Chicken crop dissections indicated that the birds fed upon a wide variety of insects
and other invertebrates and that the diet changed considerably over the season. A total of
33 food items were identified, however only 11 of them were found in both samplings
(Table 4-3). In June, the most common food items found were ants, muscoid flies, dung
beetles, earthworms, ground beetles, grass, and weed seeds. In July, the diet shifted to
include fewer predators, parasitoids, and detritivores and more herbivores. The most
common food items found at this time were Japanese beetles, shield-backed bugs, flea
beetles, shining leaf beetles, muscoid flies, ants, and grass. Japanese beetle was the most
common food item found in the July sampling: 75% of the chicken digestive crops
contained this pest. Other potential fmit and vegetable pests found in the diet included

tarnished plant bugs, flea beetles, leafhoppers, wireworms, and slugs.

Vegetable crop compatibility with weeder geese

The feeding trial indicated that leaves of the two brassica crops were potentially
palatable to the geese (Table 4-4). However, only one of the five geese consumed these
leaves and this occurred only after several minutes of masticating it with its bill. During the
first two days after the geese were introduced into the garden plot relatively little trampling
damage was observed. A thunderstorm which occurred on the third day of the trial resulted
in erratic and excited movement by the geese throughout the plot and subsequently caused
more damage to the plants. The three crops with support systems, tomato, eggplant, and
cucumber, and the shrubby herb, oregano, received no observable trampling damage
during the trial. The other five crops, pepper, broccoli, cauliflower, basil, and string bean,
had 6 - 22% of plants with light or heavy trampling damage by the end of the trial. Only

one crop, string bean, showed any feeding damage by the geese and this was only found

71

on a single plant (Table 44). During the four-day period mean weed seedling densities

were reduced from 116/m2 to 40/m2, a 65% reduction.

DISCUSSION

The purpose to this study was to evaluate experimentally the effectiveness of
domestic chickens and geese as insect pest and weed control agents in an intensively-
managed and integrated horticultural production system. The results support some findings
of the study by Clark et al. (1995) and clearly demonstrate the potential of geese as weed
control agents. In 1994, weeding by geese resulted in substantially higher potato tuber
yields compared to the unweeded control. Similar patterns in weed control and potato
growth were observed in 1995, however, disease severely affected potato plants and tuber
growth in both treatments and no significant differences resulted.

An unexpected finding was the reduction in plum curculio damage which resulted
from flee-ranging geese. A possible explanation for this finding is that vegetation removal
by the geese resulted in a reduction in soil surface humidity, which consequently reduced
plum curculio activity (Racette et al., 1992). While the observed differences in damage in
1994 undoubtedly resulted from the effects of geese, the cause of the patterns in 1995 can
only be presumed. This is because the geese were removed and chickens introduced
during the time period in which plum curculio is typically active. Plum curculio oviposition
scars were already apparent on many fruit at the time of chicken introduction.
Furthermore, plum curculio was not found in any of the digestive crops of chickens
collected during June. Thus, there is no evidence that chicken predation was responsible
for the reduction in plum curculio fruit damage.

Another somewhat unexpected effect of the geese on an insect was the increase in
Colorado potato beetle abundance. The relatively large potato plants in the weedless
environment produced with weeder geese provided an ideal habitat for Colorado potato

beetle (Horton and Capinera, 1987; Hare, 1990). Nonetheless, in this study Colorado

72

potato beetle abundance appeared to be relatively unimportant in influencing yields. The
lack of effect from this pest was most likely due to its low population densities and late
arrival relative to the planting date (Zehnder and Evanylo, 1989).

The hypothetical economic analysis based on the labor and material input data
indicates that the use of geese may be an economically feasible weed management option
for small-scale operations largely dependent upon labor-intensive weeding methods rather
than herbicides and machinery. The costs associated with weed management in organic or
low-chemical fruit and vegetable production are generally a sizable portion of overall costs
because alternatives to pesticides are more labor intensive (Prokopy, 1991; Pimentel,

1993). Weeder geese might be profitably used in some cases to reduce labor requirements.

 

Even in comparison to chemical control, weeder geese have been found to be economically
comparable (Cramer, 1992).

The estimate of 75 geese/ha (30 geese/acre) for this agroecosystem is substantially
greater than other estimates which range from 8 (Johnson, 1960; Bachmann, 1993) to 45
(Ware, 1995) geese/ha. Clearly, required weeder geese densities will depend upon the
crop species, climate conditions, supplemental weeding methods, and management
strategies. This estimate is based upon the requirements during potato plant emergence
from the soil, which tends to coincide with a period of rapid weed growth in the spring.
Furthermore, it was assumed that no supplemental hand weeding would be conducted
during this time. Although weeder geese density requirements decline later in the growing
season, it is the period of most rapid weed growth that determined this estimate.

An assumption of this simplistic economic analysis is that the geese can be easily
marketed at the end of the growing season. However, the Opportunities for marketing
geese are considerably less than for other fowl, such as chickens or turkeys. Therefore, an
alternative to the sale and re-purchase of geese each year is that a portion of the birds be
kept for breeding and flock re-establishment. In this situation the user would incur greater

costs in maintaining the birds because they would need supplemental feed to over winter.

73

The garden trial indicated that weeder geese can be compatible with crops other than
apple and potato. The most important factor determining compatibility was the presence of
a support system. Those crops which were staked or trellised, tomato, eggplant, and
cucumber, and the low-growing, shrubby herb, oregano, were left undamaged during the
trial. If artificial support systems could be used other vegetable crops may also be
compatible with weeder geese. However the need for stakes or trellises in crops otherwise
not requiring them will add to the costs of production and may make the use of weeder
geese less economical. Furthermore, many young crop plants are likely to be at least as
palatable as young weeds. This means that the compatibility of weeder geese is dependent
upon the absolute age of the crop and the relative age compared to the surrounding weeds.
More field evaluation is necessary to understand the relationship between these variables.

The possibility of weeder geese selecting for unpalatable weeds in agroecosystems
was discussed by Wurtz (1995). In her study of weeder geese in spruce seedling plots the
abundance of unpalatable weeds increased by 25 times after two years. Although such a
pattern was not detected in this study our observations during weeding operations in late
June, 1995, indicated that curly dock and daisy fleabane dominated in areas weeded by
geese. These two species were clearly unpalatable and therefore selected for with the
removal of palatable species. These observations support Wurtz’s assertion that the use of
geese be integrated with other weed management methods to prevent selection for
unpalatable weeds.

Chickens provided no clear crop protection benefits in 1994 or 1995. Although
reductions in the abundance of some insect pests were observed in both years no
differences in pest damage or crop yield resulted. Indirect sampling of plum curculio in
June 1994 indicated a small reduction in activity, however, no significant differences in
fruit damage were found. Similar abundance patterns were observed for apple maggot in

1994, yet no differences in fruit damage were found.

74

In 1994 and 1995 Japanese beetle abundance was consistently lower in the
treatments with chickens compared to the control except for the last sampling date in 1994
(5 August). Furthermore, in the analysis of chicken digestive crops Japanese beetle was
the most common food item found in mid-July. Nevertheless, no differences in apple fruit
damage by this pest were found. There are several possible explanations for these results.
The relatively small size of the plots in comparison to the general mobility of this insect
may have allowed it to rapidly colonize available trees during the few weeks in between
chicken removal and apple harvest. Furthermore, fruit which was heavily damaged by
Japanese beetle tended to drop from the trees prior to harvest; this may have caused some
underestimation of damage in the control. Finally, it is likely that damage from wasps was
counted as that of Japanese beetle in some cases. Both of these insects chew into the fruit
and when such damage could not be differentiated it was assumed to be due to Japanese
beetle. This would have inflated estimates of Japanese beetle damage in all treatments,
assuming wasp damage was not affected by treatments. Although chickens showed no
clear benefits in this study the results suggest they still may be potentially useful as part of
an integrated approach for Japanese beetle management. It is possible that higher chicken
densities in conjunction with the use of lures to draw the beetles within range of the birds,
could be used to control this pest without pesticides in apple or other susceptible crops.
Other options may include using mobile, floorless chicken cages, sometimes called chicken
tractors, to remove or reduce apple drops on the orchard floor.

Although chickens have been observed to feed on Colorado potato beetle in the field
(see chapter 3) there was no clear evidence of this based upon the observed population
dynamics patterns or digestive crop analysis in this study. In 1994, Colorado potato beetle
abundance patterns in the chicken treatment mirrored those of the control primarily because
of similar weed and potato growth patterns. Because of the effects of weeds on Colorado
potato beetle abundance an adequate assessment of chicken predation was not possible in

this experiment. In 1995, weed growth between the two treatments differed considerably,

 

75

especially in the early part of the season (Fig. 4-13), confounding Colorado potato beetle
abundance and again preventing an assessment of chicken predation.

Interestingly, Colorado potato beetle survival between the peaks in larval and adult
abundance was substantially less in the treatment with chickens compared to the control in
1995. While the mean proportion of larvae surviving to the adult stage was 73% in the
control it was only 7% in the treatment with chickens. Although greater mortality may be
expected in a denser population, this large difference in survival suggests the possibility of
an effect by chickens. Although the results of the digestive crop analysis offer no evidence
of predation on Colorado potato beetle, the scratching and pecking which typifies chicken
foraging activity may have been enough to cause some level of mortality, especially during
the larval stage. Hough-Goldstein et al. (1993) found Colorado potato beetles to be
unpalatable to broiler chicks but noted a high mortality rate in peeked larvae. There are
differences in foraging and feeding behaviors between chicken breeds and individual
chickens within breeds (M. S. Clark, personal observation; Hough-Goldstein et al., 1993).
Further research is needed on this intra- and inter-breed variability as it relates to the
potential use of chickens as biocontrol agents of Colorado potato beetle.

The results of this study indicate that geese can be effective and economical weed
control agents in some agricultural production systems. Agroecosystems in which weed
management is largely dependent on human labor might integrate weeder geese to reduce
labor requirements. Habitat factors, such as tree cover, vegetation height, plant species
composition, and water availability, will influence the performance of weeder geese and
therefore should be considered when designing management systems. The overall goal of
this research was to evaluate a non-chemical management strategy (free-ranging domestic
birds) to further the development of agroecosystem design. Although the yields in this
experimental system are low by conventional standards they are fairly typical of organic
systems (Reinken, 1986; Stanhill, 1990; Pimentel, 1993). I should reiterate that organic or

nonchemical agricultural methods cannot be widely adopted without substantial changes in

76

public perceptions and values and the re-organization of food systems toward more
localized production and consumption (Merwin and Pritts, 1993; Pimentel et al., 1993; van
Ravenswaay, 1994). Nonetheless, the results of this research demonstrate the biological

feasibility of a food production system without agrochemicals.

77

Table 4-1. The abundance of common weed species based on percent groundcover

estimates (mean -I- SEM) taken on 30 May, 1995. The asterisks indicate significant

reductions in plant species abundance in the bird treatment compared to the control based

on Student’s t-test (**= 0.05 2 p > 0.01; ***= p g 0.01).

 

Cmrncnhlame
Orchard Grass

Kentucky Bluegrass
Common Dandelion
Common Chickweed
Common Plantain
Daisy Fleabane
Curly Dock

Canada Thistle
Winter Cress

Milan:
Dactylis glomerata
Poa pratensis
Taraxacum ofiicinale
Stellaria media
Plantago major
Erigeron annuus
Rumex crispus
Cirsium arvense

Barbarea vulgaris

Eercem groundcover
ngtrgl Bind
19.0 +1.2 1.3 + 0.3
16.7 -I- 2.2 6.7 + 1.2
7.0 + 5.0 O
6.7 + 2.0 0.3 + 0.3
5.7 + 1.5 0
4.0 + 1.2 2.7 + 2.7
2.3 + 0.9 4.7 -I- 1.3
1.0 + 1.0 0
0.7 + 0.7 1.3 + 1.2

 

***

*31‘

**

**

78

Table 4-2. A comparison of labor and material input costs ($U.S.) for nonchemical weed

management with and without weeder geese in a hypothetical 1 ha apple orchard with

 

 

 

 

intercropped potatoes.

Costs ($US)
Inputs We 11me
Geese (95 @ $5 per bird) 475 -
Feed ($72 per month for 5 months) 360 -
Care of Geese(75 h @ $6 per h) 450 -
Housing and Fencing (IO-year depreciation) 110 -
Weeding and Cultivation ($6 per h) 67 1235
Rototiller (IO-year depreciation) - 100
Weed trimmer (5-year depreciation) 40 40
Fuel and Supplies 20 55
Total Input Costs 1522 1430
9111121115
Geese (75 @ $10 per bird) 750 -
Balance

 

Total Costs (Inputs - Outputs) 772 1430

 

79

Table 4-3. The diet components of free-range chickens in the experimental system based

on the presence / absence of items in digestive crop dissections from two sampling periods;

6-19 June (N=lO) and 19 July (N=12).

 

 

Food Items
Herbivore Japanese beetle Popillia japonica
tarnished plant bug Lygus lineolaris
shield-backed bug Scutelleridae
flea beetle Alticinae
shining leaf beetle Criocerinae
wireworrn Elateridae
click beetle Elateridae
caterpillar Lepidoptera
leafhopper Cicindellidae
cicada Cicadidae
Predator ground beetle Carabidae
hover fly larvae Syrphidae
hister beetle Histeridae
soldier beetle Cantharidae
rove beetle Staphylinidae
assassin bug Reduviidae
wolf spider Lycosidae
crab spider Thornisidae
Parasitoid braconid wasp Braconidae
ichneumon wasp Ichneumonidae
Detritivore earthworm Lumbricidae
slug Limacidae
dung beetle Scarabaeidae
muscoid fly Diptera, Muscoidea
sap beetle Nitidulidae
earwi g Forficulidae
Other animals ant Formicidae
hover fly Syrphidae
moth pupae Lepidoptera
caddisfly Trichoptera
cuckoo wasp Chrysididae
Plant grass Poaceae
weed seeds undetermined

Hfl

H

DJ

UI
qummmScoooowooooqoooxrooqocoOOoo

I-‘O\

 

80

Table 4-4. The results of a feeding trial, which tested the potential palatability of vegetable
crops (N=5 geese), and an unreplicated garden trial which assessed feeding and trampling

damage caused by weeder geese during a four-day period.

 

 

 

£3212" Martial" gardenm
talents MW % W'
dams:
light beam item 11m

tomato - 61 0 0 0 0
pepper - 23 0 0 13 0
eggplant - 18 0 0 0 0
broccoli + 30 0 0 10 0
cauliflower + 23 O 0 13 4
cucumber - 6 0 0 0 0
basil - 18 0 0 6 0
oregano - 7 0 0 0 0
string bean - 37 0 3 14 8

 

 

a tomato and eggplant plants were staked; cucumber plants were trellised; all other crop
plants had no support system

b + indicates that at least one goose fed upon the leaves offered; - indicates that none of the
five geese fed upon the leaves offered.

81

alley

appl tree
I potato row

 

12m

 

©0010
0000]:
00:20

O
O
O
O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

,0000

<

 

28m

Figure 4-1. Layout of a single plot showing important features. Four of the six apple trees
within the dotted box in the center of each plot were designated as sampling trees for all '

insect pest monitoring.

 

82

 

 

 

 

 

 

$110
g + Control
4: a 80‘ _*_ Chicken 4%
.“¢ + Goose
E: 60‘
=5
E“ ..
:5 40
g 20-
D:
0 l l r I W l I
2 6 10 14 18 22 26 30

June

Figure 4-2. Percentage of apples damaged by plum curculio based on visual monitoring

during June, 1994 (mean + SEM).

83

 

    

 

 

 

E 15
g —P'— Control
4 a '_’— Chicken
3.2 10: + Goose
fix
a E
a s .
.2 E 5
Q 4
a.
4
0 I

 

6 July 16 July 26 July 7 August

Figure 4-3. Apple maggot abundance based on the number of adults caught on four red
spheres per plot in 1994 (mean + SEM).

84

 

 

 

o

g 20 + C

: ontrol
3 q —'°""' Chicken
g ——o— Goose
cu

m 10 -

cu

8 ,
G

a

D.

a:

H

 

 

G

. I

. 5"
II

20 30 40
Days After 30 June

I—L

Figure 4-4. Japanese beetle abundance based upon direct counts on the sampling trees in
1994 (mean -I- SEM).

85

 

  

 

 

"E“; 90 a b I Control

h " a

E 75 b - ChECI‘en

° 60' I

0

5" 45 -

tan-

: -I

a. 15 Goose
oi

PC AM JB TPB CM Pest-Free

 

Figure 4-5. The percentage of harvested fruit damaged by insects (PC=plum curculio,
AM=apple maggot, JB=Japanese beetle, TPB=tarnished plant bug, CM=codling moth) and
undamaged (Pest-Free) in 1994 (mean + SEM). Different letters indicate significant
differences (p 5 0.06).

86

 

Potato Yields
(kg fresh wt I ha)

 

 

Control Chicken Goose

Figure 4-6. Potato tuber yields (kg / ha) based on sampling 6m of row in 1994 (mean +
SEM). Different letters indicate significant differences (p _<_ 0.01).

87

CPB LARVAE

 

 

 

 

 

 

 

B
e 4
«5 —'3_ Control
5 3 - _’— Chicken
\
o
a
>
1..
a
A
m
A:
0
CPB ADULTS

g «as
'- —'— Control
'5 _°_ Chicken

0.6'
E + Goose
\
:2
=3
'5
a
m
A:
U

 

 

 

'30 40 so 60 70 so 90
Days After Potato Planting

Figure 4-7. Colorado potato beetle (CPB) larvae and adult population dynamics in the

control, chicken, and goose treatments in 1994 (mean + SEM).

88

WEED

cultivation —9— Control
—‘°— Chicken
—'— Goose

 

oo
Q

L

3.

 

Weed % Groundcover
8 3
l l L
/
\
‘\‘L
In"

 

 

 

 

 

 

 

0 I " r I _l
20 30 40 50 60
g POTATO
2 60
3 + Control
'2 —'°_ Chicken
3 40- + Goose
(S . cultivation _
5° 2% /’
3 /..’
3 .
a
A. 0- - .. . , . r .
20 30 40 50 60

Days after Potato Planting

Figure 4-8. Percent groundcover of weeds and potatoes in the control, chicken, and goose

treatments in 1994 (mean + SEM).

89

 

60‘ * I Control
50- B Bird

Percentage of Fruit

 

 

 

PC AM JB TPB CM Pest-Free

Figure 4—9. The percentage of harvested fruit damaged by insects (PC=plum curculio,
AM=apple maggot, JB=Japanese beetle, TPB=tarnished plant bug, CM=codling moth) and
undamaged (Pest-Free) in 1995 (mean + SEM). Asterisks indicate significant differences
(p _<_ 0.07).

90

 

 

 

 

 

 

 

2
3 30-
<H
«o
@—
333- 20-
2r
as 10‘
G:
2'

0-

8 July 18 July 28 July 7 August

Figure 4-10. Apple maggot abundance based on the number of adults caught on four red
spheres per plot in 1995 (mean + SEM).

91

 

 

 

 

 

 

8

g: 80' —9— Control

2 - —"'— Bird

a 60+

3 .

m 40.

I»

3 .

5 20‘

g ..

_B 0 r ' l ' l ' l
10 16 22 30

July

Figure 4-11. Japanese beetle abundance based upon direct counts on the sampling trees in
1994 (mean + SEM).

92

 

 

 

 

 

 

   
   

3 CPB LARVAE
o
J: 8' " —'_ Control
° . —-— Bird
5 6.
: .
a
>
I...
a
A
E
D
3 CPB ADULTS
E 0.6
.— —9— Control
9 —‘— Bird
5 0.41
£3
E
a 0.2'
5'3
0.0‘

 

 

 

30 40 50 60 7 0 80 90
Days After Potato Planting

Figure 4-12. Colorado potato beetle (CPB) larvae and adult population dynamics in the

control and bird treatment in 1995 (mean + SEM).

93

 

      
 

 

 

 

 

 

.. WEED
E 80‘ cultivation —a— COHU‘OI
% I I '_.— Bird
5 60" cultivation
5 -I
e 40‘
3 0 v I , 1 fl l . l u
20 30 40 50 60 70

POTATO

—9— Control
—’— Bird

 

40

30"

10‘

Potato % Groundcover
’=.’

 

 

 

Q

I 1 I

3 0 4 0 5 0 6 0 7 0
Days After Potato Planting

N
6

Figure 4—13. Percent groundcover of weeds and potatoes in the Control and Bird treatment

(mean + SEM).

CHAPTER 5

THE EFFECTS OF FREE-RANGE DOMESTIC BIRDS
ON THE ABUNDANCE OF BENEFICIAL SOIL MACROINVERTEBRATES

INTRODUCTION

Integrating animals into cropping systems can result in a variety of benefits
including nutrient cycling, resource conservation, natural or biological pest management,
and economic stability. This research reported so far has focused on the potential of the
birds as insect pest and weed biological control agents. However, chickens have been
found to feed upon a wide variety of macroinvertebrates from the soil surface including
ground beetles (Carabidae), rove beetles (Staphylinidae), spiders (Araneae), and
earthworms (Lumbricidae). Geese, by contrast, are herbivores and do not directly affect
soil macroinvertebrates. Nevertheless, the removal of vegetation and mild soil compaction
which results from their foraging activities can dramatically alter the soil surface
environment. Furthermore, the manure produced by both of these bird species could
potentially alter food webs and thereby be manifested in changes in macroinvertebrate
abundance.

There are risks associated with introducing new or exotic species into ecosystems
for biological control, or any other purpose, because it is generally not possible to predict
all direct and indirect consequences of such introductions (Howarth, 1992). Although
domesticated animals present relatively little threat because of their manageability, they can
have significant ecological effects. The impact of overgrazing cattle on soil resources is

one well known example (Pimentel et al., 1995);

94

95

The purpose of this study was to evaluate the effects of free-range chickens and
geese on two groups of beneficial soil macroinvertebrates: epigeic predators and
earthworms. The experimental agroecosystem used for the study was a nonchemical apple
orchard with intercropped potatoes. Research has demonstrated or suggested that both of
these macroinvertbrate groups play important roles in these crops. For example, ground
beetles have been found to be predators of some apple pests, including apple maggot (Allen
and Hagley, 1990) and codling moth (Hagley and Allen, 1988; Riddick and Mills, 1994),
and the potato pest, Colorado potato beetle (Hough-Goldstein etal., 1993; Brust, 1994).
Spiders can be natural control agents of a variety of horticultural insect pests (Riechert and
Bishop, 1990). The role of earthworms in maintaining soil fertility and improving soil
structure has been widely demonstrated in agroecosystems (Berry, 1994; Edwards et al.,
1995; and references therein). Furthermore, earthworms can indirectly reduce plant
diseases in orchards by breaking down leaf litter (Raw, 1962; Kennel, 1990). The results
of two objectives are reported: 1) to evaluate the short-term effects of free-range chickens
and geese and intercropping on the activity patterns of epigeic predators in an apple
orchard, and; 2) to measure the effects of free-range domestic bird production on

earthworm abundance and compare these to the observed landscape-level patterns.

MATERIALS AND METHODS
The Study Site

This study was conducted in an experimental apple orchard and surrounding area at
the Kellogg Biological Station of Michigan State University. The orchard was planted in
1983 with the disease-resistant varieties ‘Redfree’, ‘Priscilla’, and ‘Liberty’ and arranged
in three blocks of 0.5 to 0.7 ha which represent high (430 trees/ha), medium (225
trees/ha), and low (110 trees/ha) tree planting densities. All of the research reported in this
paper was conducted in the high density block and adjacent agricultural fields (Fig. 5-1).

The entire orchard has been managed without the use of pesticides since its establishment.

96

Effects on Epigeic Predators

In 1993, twelve plots were randomly established in the block for three treatments
with four replications. Plots were 16.8 m X 9.9 m and consisted of three rows of three
apple trees and two inter-row alleys (Fig. 5-2). The trees in two of the rows were
‘Redfree’ and the third were ‘Priscilla.’ The tree spacings were 3.3 m and the row spacings
were 6.5 m. Prior to potato planting the orchard alleys were cultivated with a chisel plow
and disk. The alleys were planted by hand in potatoes (‘Red Pontiac’) on 18 May with 30-
cm plant spacings and 75-cm row spacings. This allowed three rows of potatoes to be
planted within each alley such that the outside potato rows were not directly beneath the
canopy of the apple trees. In one randomly-selected alley of each plot the potato plants
were mulched to a depth of 15 cm with recently cut rye (Secale cereale var. ‘Wheeler’)
which was collected from a field adjacent to the orchard, creating mulched and unmulched
subtreatments. The potatoes were hand weeded on 11 June and hand weeded and billed on
17 June.

The three main neatments included 1) placing 15 chickens per plot (chicken
treatment); 2) placing 11 geese per plot (goose treatment); and 3) a control with no birds
(control). Plots with chickens or geese were surrounded with 1.5 m-high plastic fencing
with 6 cm mesh. On 2 July, the birds, at 6 weeks old and with an approximately equal
number of males and females, were introduced into the designated plots. The birds were
kept in coops (1.2 m X 1.2 m) located on the perimeter of each plot at night and released
to free range from approximately 7:00 am until 9:00 pm (EST). The birds were provided
with water and small amounts of grain-based feed twice daily. On 17 July, 3 geese were
removed from each plot of the goose treatment, setting the population at 8 per plot. The
birds were maintained in the plots until 6 August.

Ground-dwelling predators were sampled in each plot with three pitfall traps from 2

July until 29 July. One trap was positioned in the center of each of the alleys with potatoes

97

and one trap was placed in the center of the middle apple row (Fig. 5-2). A trap consisted
of 250 ml plastic cup with a 7-cm diameter opening, placed into the ground so that the top
was flush with the soil surface. It was filled with approximately 100 ml of soapy water as
a killing agent. A plastic rain cover (12 cm X 12 cm) was held 5 cm above the cup with
galvanized nails. The traps were emptied weekly and the contents taken to a field
laboratory for specimen identification. The data for the 27—d trapping period were pooled
for each trap and comparisons made with a two-way analysis of variance with the main
treatments (chicken, goose, and control) and subtreatrnents (apple row, mulched potatoes,
and unmulched potatoes) as the two factors, each with three levels. When significant
differerences were found among the main treatments (P s 0.10) means were separated

using Tukey’s multiple range test.

Effects on Earthworms

Earthworms were sampled in the orchard and nearby fields from 28 to 30, July,
1995 (Fig. 5-1). A sample consisted of 20 cm2 cube of soil which was hand-sorted for
earthworms in the field. Pairs of samples were taken from eight areas of the orchard. In
each area one sample was from the apple tree row and the other from the alley. Five of the
eight areas had been used for ranging chickens and geese from 1993 to 1995 at densities of
150-700 birds/ha for 1—3 months each summer. The other three areas had been free of
birds and were used as controls. Two samples were taken in fields near the orchard which
included an old field (never tilled), two corn fields, and an alfalfa field (Fig. 5-1, Table 5-
1). Nearly all earthworms collected were immature; thus no species identifications were
attempted. In addition, three, S-cm deep, soil probe samples were taken at each sample
site, mixed, and a subsarnple removed. These were analyzed by the Michigan State
University Soil and Plant Nutrient Laboratory for nutrient, texture, and organic matter

characteristics.

98

Earthworm abundance, nutrients, texture, and organic matter were compared
between the areas with and without birds and between tree rows and alleys of the orchard
with two-way analysis of variance. The relationships between earthworm abundance and
soil chemical and physical properties and geographic location (linear distance from the old
field) were evaluated with Pearson correlation and regression analyses. For these analyses
each pair of orchard samples (tree row and alley) was pooled for a total of eight orchard
samples and each pair of samples from the nearby fields was pooled to form a single

sample from each of the four fields.

RESULTS
Effects on Epigeic Predators

Approximately 1100 epigeic predators were collected in pitfall traps during the 27-d
period with four taxa repesenting over 99% of all specimens (Fig. 5-3). The abundance of
all four taxa showed similar patterns with respect to the subtreatrnent effect: the traps in the
tree rows collected more predators than those in the intercropped alleys (Fig. 5-4). These
differences were statistically significant for Staphylinidae, Opiliones, and Araneae, but not
for Carabidae. No difference in predator abundance between mulched and unmulched
alleys was observed.

Effects due to the birds were found for Opiliones and Araneae (Fig. 5-4). In both
cases, predator abundance was reduced by the presence of chickens and geese.
Staphylinidae and Carabidae showed no differences in abundance that could be attributed to
the birds.

Effects on Earthworms
No differences were found between orchard areas with and without birds for
earthworms, clay, silt, sand, organic matter, nitrates, phosphorus, potassium, calcium,

magnesium, or cation exchange capacity. Furthermore, no difference in earthworm

99

abundance was found between the tree rows and alleys. However, when samples from the
orchard and surrounding fields were analysed for correlations, earthworm abundance was
found to have significant positive relationships with percent clay, percent organic matter,
and cation exchange capacity and significant negative relationships with percent sand and
distance from old field (Table 5-2).

As expected, cation exchange capacity was correlated with organic matter and
percent clay. Percent clay was relatively constant, ranging from 17 to 23% for all samples
except those from the old field and alfalfa field. The old field had the highest clay content,
33%, while the alfalfa field had the lowest, 9%. Percent sand displayed the opposite
pattern; the old field had the lowest value, 24%, while the alfalfa field had the highest,
71%. The strongest relationships (i.e. those with the highest correlation coefficients and p
values) with earthworm abundance were with organic matter and distance from the old field
(Table 5-2, Fig. 5-5). A multiple regression model with these variables accounted for most
of the variation in earthworm abundance (r2=0.96; p<.0001). The old field had the highest
organic matter content and earthworm abundance. The two corn fields and the alfalfa field
had the lowest organic matter contents, yet corn field #2 had the second highest earthworm
abundance. This apparent contradiction was explained with distance from the old field
(Fig. 5-5). This rather narrow corn field was located between the old field and orchard,
two stable habitats with high organic matter. Thus, movement into the corn field from

these locations would account for the high earthworm abundance found there.

DISCUSSION

The overall effects of domestic birds on beneficial soil macroinvertebrates in this
study were found to be absent or relatively unimportant. The practice of intercropping in
the orchard had a greater overall effect on the activity patterns of predatory
macroinvertebrates than did the presence of chickens or geese. However, the presence of

the birds did result in fewer spiders and harvestrnen in the pitfall traps.

100

In general, ground-dwelling predators are most active at night (Brust et al., 1986);
however, most of the spiders collected were lycosids, a family with many diumally active
species. Furthermore, the most common harvestrnan, Phalangium opilio L. is commonly
found on the soil surface or on plants in the day. In contrast, most carabids and
staphylinids seek shelter under the soil surface or debris in the day, where they are less
likely to be eaten by foraging chickens or disturbed by foraging geese.

The presence of the birds over the 3-yr period had no detectable effect on
earthworm abundance. Moreover, the soil chemical and physical analyses showed no
differences between the areas with and without birds. Earthworm abundance was clearly
correlated with soil organic matter, a relationship which has been widely shown in other
studies (Hendrix et al., 1992; Edwards et al., 1995 and references therein). Soil organic
matter provides a food source for earthworms and when it is lost, due to agricultural
practices or erosion, earthworm abundance declines. By contrast, when organic matter is
added to soil, such as in manure applications, earthworm abundance increases. The fact
that the birds had no effect on organic matter indicates that earthworms should not have
been influenced positively by their presence. However, there was also no evidence of
negative influence due to chicken predation or vegetation alterations. Although the
chickens were observed to feed on earthworms (see chapters 3 and 4)), their effect on
earthworm populations, at least at these stocking densities, appears to be absent or
relatively minor. '

The relationship between earthworm abundance and distance from the old field was
an interesting and surprising finding. It suggests an explanation for why corn field #2, a
field with relatively little organic matter, had such high earthworm abundance. It is
probable that there is a substantial amount of surface activity by the earthworms and that
there is a net movement out of areas with higher densities into areas with lower densities

(Mather and Christensen, 1988). This com field, which is annually tilled, apparently

101

benefits by its close proximity to undisturbed land with high earthworm densities. This
combined with its small size, makes it easily colonizable.

In summary, free-ranging chickens and geese resulted in some reduction in spider
and harvestrnan activity. The mechanism for this reduction may have been direct predation
by chickens or habitat disturbance and alteration by both bird species. Chickens have been
found to feed on the two common carabid species in this landscape, Pterostichus
melanarius (Illiger) and Cyclotrachelus sodalis (LeConte), which represented 93% of all
carabids collected during the sampling period. However, no reductions in the abundance
of carabids or staphylinids were detected. A plausible explanation for this lack of effect is
that these predators are generally active at night when the birds were in coops (chickens are
inactive at night regardless of whether they are in a coop or not). Furthermore, the
presence of the birds did not result in any detectable changes in soil organic matter or any
other chemical or physical variable measured. Thus, no positive effects on earthworm
abundance would be expected. At the same time, no negative effects on earthworm

abundance were found.

102

Table 5-1. Management histories of the agricultural habitats sampled for earthworm

 

abundance estimates.

Haunt W Karim 1111.48;

Apple Orchard intercropped with potatoes 1993-1995 none minimum
(in alleys)

Old Field never cropped, mowed once per year none none

Corn Field #1 rotated with wheat herbicides conventional

Corn Field #2 rotated with wheat, rye, and oats herbicides conventional

Alfalfa re-planted in 1994 none conventional

( 1994)

 

103

Table 5-2. Pearson correlations between earthworm abundance and soil texture, organic

matter, nutrient levels, cation exchange capacity, and distance from an old field.

 

% clay 0.67 0.018
% silt 0.40 0.197
% sand -0.55 0.065
% organic matter 0.78 0.003
nitrate (ppm) 0.09 0.772
phosphorus (kg/ha) -0.37 0.242
potassium (kg/ha) 0.13 0.698
calcium (kg/ha) 0.40 0.199
magnesium (kg/ha) -0.29 0.359
cation exchange capacity (me! 100g) 0.71 0.009
distance from old field (m) -0.77 0.003

 

104

 

 

Distance (meters)
150 100 50
l
l I Road j I

 

 

Corn Field #1

i

*

 

 

 

 

Apple Orchard

**

**

**

**

**

**

it

**

 

/

Corn Field V

ass Paths

Old Field

*

*

 

 

Figure 5-1. Landscape map showing the experimental apple orchard and surrounding

fields and the earthworm sampling locations (*).

 

105

Tree Unmulched Tree Mulched Tree

 

Row Potatoes Row Potatoes Row
¥3¥¥ $353!
¥*¥ *¥*
*9¥¥ 3333*
¥#* ***
*¥¥ ¥¥*
*$¥ **¥
*** ¥3¥3¥
**¥ *¥*
¥3¥¥ ¥¥¥
¥¥* ¥**
¥ ¥ * ¥
¥*¥ ¥¥*
¥** ***
**¥ ¥¥¥
**¥ 33*?!
¥¥¥ ¥*¥

 

 

 

Figure 5-2. Plot diagram showing the general layout and the locations of the pitfall traps
(shown with arrows). Plots measured 16.8m X 9.9m and included three rows of three
trees and two inter-row alleys which were intercropped with potatoes (mulched and
unmulched).

106

Araneae
12.6 %
Staphylinidae
Opiliones 35.5 %
18.9 %

 

Carabidae
32. l %

Figure 5-3. The relative abundance of predator taxa collected in pitfall traps in the orchard
from 2 July until 29 July, 1993.

107

 

Z Unmulched Potatoes
Staphylinidae Tm Row

main treatments; p: 3.1 I Mulched Potatoes
subtreatments; p: .004

interaction; p: 9.2
‘° Ali
0

Control Chicken Goose

 

 

 

8

Mean / Trap
N
e

Opiliones
main treatments; p: .05

 

a subtreatments; p: .007
g 20 Interaction; p= .
E-t
\ ab
= 10 ab ab ab
8 b b b b
2 V741
0 l I l
Control Chicken Goose
Carabidae

20- main treatments; p: .91
subtreatments; p: .50
. interaction; p: .8

Mean / Trap
H
°.

 

 

 

0 l
Control Chicken Goose
Araneae main treatments; p: .01
15 subtreatments; p: .0001

interaction; 1): .63

    

p-I
6

Mean / Trap
0|

0

Control Chicken Goose
Figure 5-4. Mean abundance (:11 SEM) of predatory arthropods collected in pitfall traps in
main treatments (control, chicken, and goose) and subtreatments (unmulched potatoes,

108

mulched potatoes, and tree row). Different letters indicate significant differences according
to Tukey’s multiple range test (p s 0.10).

as

Earthworms/sq. m
5. 5,

     

109

 

'5' Orchard

‘ Old Field

‘ Corn Field 1
° Corn Field 2
Alfalfa

 

 

 

 

 

300‘

200‘

100‘

Earthwormslsq. m

 

 
 

100 200
Distance from Old Field (In)

 

 

A
' I ' I i I

% Organic Matter

Figure 5-5. The relationship between earthworm abundance (number/m2) and distance

from old field and percent organic matter, based upon samples taken from the apple orchard

and nearby fields. Earthworm abundance was negatively correlated with distance from old

field (F077, p=0.003) and positively correlated with organic matter (r=0.78, p=0.003).

CHAPTER 6

FARMER PERCEPTIONS OF FREE-RANGE DOMESTIC BIRDS
AS BIOLOGICAL CONTROL AGENTS ON SMALL FARMS

INTRODUCTION

In recent years the use of domestic birds as insect pest and weed biological control
agents has received increased attention by both farmers and researchers. Although some
published historical information on the use of domestic birds for biological control does
exist, it is mostly anecdotal and lacks detail on management practices and measurements of
the birds’ effectiveness (Quaintance and Jenne, 1912; Hoare, 1928; Mayton et al., 1945;
Johnson, 1960). Thus, it has not adequately served the purposes of farmers or researchers
seeking to learn more on the use of birds as alternatives to pesticides.

Recently published case studies, on-farm trials, and research experiments have
reported on the use of geese for weed control in fruits, vegetables, and nursery plantings
(Cramer, 1992; Bachman, 1993; Ware, 1995; Wurtz, 1995), ducks for housefly control in
dairy operations (Glofcheskie and Surgeoner, 1990; 1993), and chickens for insect and
weed control in potato and apple systems (Hough-Goldstein et al., 1993). Nevertheless,
there are still questions regarding the potential for widespread applicability of domestic

birds as biological control agents on farms.

In order to address these questions, cooperation between farmers and researchers is
clearly needed. Discussion on the appropriate roles of farmers, researchers, and land-grant
uni versities in generating agricultural knowledge and technologies has been an important
blit controversial topic in the sustainable agriculture movement (see for example: American

Journal of Alternative Agriculture, 1990, volume 5(4); Agriculture and Human Values,

110

111

1994, volume 11(2&3). Although it has been stated that when farmers do not accept
technologies it is simply because they are either unable or unwilling to do so (N owak,
1992), the reasons for such inability or unwillingness are varied and complex. Some have
claimed that the experiences, values, perceptions, and goals of researchers and farmers
have diverged over the years to such an extent that fiuitful communication between these
groups is difficult and rare. Such a view is expressed by Soule and Piper (1992, p. 51)
who stated that “Farming has moved from a cultural art, handed down from generation to
generation, to an industry taught as a profession by expert specialists who may never have
farmed a day in their lives.” Regardless of whether this view is correct, it seems reasonable
to assume that if sustainable technologies are to be developed for use on farms, farmer
involvement in research and researcher experience is crucial (Gerber, 1992; Ikerd, 1993).
This chapter reports the results of an effort to gather information from small
farmers, through the use of structured interviews, on the applicability of using domestic
birds as biological control agents for insect pests and weeds. The purpose is to build upon
the controlled studies of the previous chapters to gain a greater understanding of the

potential roles of domestic birds on small farms and help direct future research efforts.

MATERIALS AND METHODS

Six farmers were selected for interviews based upon their experience with raising
free-range domestic birds. The farmers were identified through their expressed interest in
research conducted at Kellogg Biological Station in southwestern Michigan, to evaluate the
potential of domestic chickens and geese as biological control agents. Five of the six
farmers were from southern Michigan, while the sixth was from Ontario (Table 6-1). All

of the farmers were practicing certified organic, nonchemical, or low-chemical farming
methods and used some means of direct marketing to sell their products. While all of the
fanns were relatively small, there was considerable diversity in the farm enterprises which

ineluded fruits, vegetables, herbs, poultry, and larger livestock. Only two of the six

112

farmers made a full-time income from their farm. The other four stated that their farms
provided only supplemental income although some worked the equivalent hours of a full-
time job on the farm.

The interview questions were based upon interests and concerns stimulated by
observations during research or from discussions with farmers, either at their farms or at
the experimental site. The purposes of the interviews were to identify why farmers had
free-range domestic birds on their farms; to find out what, if any, biological control
benefits they thought the birds provided; and to determine what aspects of bird management
they considered to be excessively difficult or problematic (Table 6-2). Five of the six
interviews were conducted in person; the sixth (Tony McQuail) was mailed. The interview
questions were designed to be objectively analyzed while also allowing growers to describe
their unique situations, attitudes, and problems. Thus, the growers were asked to answer
nine of the 11 questions with one of five choices, then to elaborate on the answer in their
own words. The choices provided a range of answers between “yes” and “no” (yes,
probably, maybe/not sure, probably not, and no).

For analysis, the multiple choice answers were ranked from 0 to 4 with “no”
corresponding to O and “yes” to 4 (see bottom of Table 6-2). Means and standard errors of
the means were calculated based on the rankings from the six interviews (N =6) for most of
the questions. Questions related to geese were not answered by one farmer (Tony
McQuail) because he had no experience with them. Thus, those statistics are based upon
five interviews (N=5). The interview responses were interpreted based upon these

descriptive statistics and on the more elaborate answers provided by each of the farmers.

RESULTS AND DISCUSSION
The farmers were in general agreement that domestic birds, including chickens,
geese, and ducks, played important roles on their farms (Fig. 6-1). This was not

Particularly surprising considering these farmers were selected based of their interest in

113

raising and using domestic birds. However, a wide variety of reasons were given for
keeping birds on the farm, including weed management, insect pest management, egg
production, food and waste recycling, aesthetics, meat production, income, manure
fertilizer, and general enjoyment.

Most of the farmers felt certain that chickens helped in controlling insect pests
around their farm. However, only two farmers mentioned specific pests which they had
observed the chickens to eat. These included scarab beetle larvae and Colorado potato
beetles. Identification of chicken diets through observation is difficult (see chapters 3 and
4) and it is therefore not surprising that the farmers mentioned few specific incidents of pest
control by chickens. One farmer did mention the differences in chicken breeds, noting that
birds bred for meat production were poor insect predators.

There was considerably less agreement on how effective chickens were as weed
control agents, as indicated by the “maybe/not sure” mean rating and the larger standard
error (Fig. 6-1, question 4). While three of the growers stated that chickens had potential
as weed control agents, the other three were quite doubtful of that prospect. Two reasons
were provided to justify the negative responses. First, in comparison to geese, chickens
have very little impact on vegetation, particularly when free-ranged at low densities.
Second, chickens can do more damage than good by pecking and feeding upon crop plants
as well as weeds. In contrast, one of the farmers who responded positively to this question
maintained relatively high chicken densities within a small block of fruit trees (several birds
per tree) and was quite satisfied with the resulting weed control.

All of the farmers considered geese to be effective weed control agents (Fig. 6-1,
question 5), however not all felt that geese were compatible with their operation. Two of
the five respondents thought geese, or at least some breeds, were too aggressive to keep on
the farm (Fig. 6-1, question 8). One farmer noted that while the geese were good-natured
as immatures, they became too aggressive as adults, particularly around children. Other

farmers also expressed some concern about geese around young children.

114

Most of the farmers had experienced at least some problem with bird loss to
predators. Two of the growers felt that predator problems on their farms were minimal
because of a high level of human activity on the farm and in the landscape around the farm.
Those farmers with the most pressure from predators mentioned putting the birds into
coops during the night to protect them from dogs, opposums, and raccoons. However,
even with such management losses still occurred occasionally. Geese were considered to
be less vulnerable than chickens although one grower did mention loosing geese to
wandering dogs.

Although all of the farmers stated that they would definitely or probably continue to
raise domestic birds on their farm (Fig. 6—1, question 10), there was less agreement on the
potential profitability of small-scale production systems. Three of the six growers were
actively marketing poultry and considered their enterprises to be relatively successful. One
the growers was keeping a chicken flock for egg production and was planning to direct
market in the future, though he doubted that it could be profitable based on the high cost of
feed and the labor required in maintaining the birds during the winter. Similarly, another
farmer, who attempted to use geese in an orchard for vegetation management, stated that
labor requirements were too greater because the birds had to be moved frequently to
prevent them from damaging the groundcover and digging holes.

Overall, the interviews indicated that these farmers kept domestic birds on their
farm for a variety of reasons and although insect pest and weed biological control were
commonly considered benefits, the birds were not managed primarily for these purposes.
Rather, insect pest and weed control were considered more as fringe benefits than as
primary reasons for having the birds. Most of the farmers expressed some interest in
finding more efficient and more effective means of managing the birds for biological
control. In addition, they were interested in knowing; 1) which crops were compatible
with the birds,; 2) which pests the birds were most effective in controlling; 3) which breeds

of birds were the best performers; and 4) how many birds would be needed per unit area

115

of crop production for controlling weeds and insect pests. Future research directed at these
questions would likely yield the most useful information for addressing small farmer
interests and concerns related to the on-farm use of domestic birds as biological control

agents.

116

Table 6-1. General descriptions of the farmers interviewed in this study.

 

EanLmaticn Eannfiize
Quinn & Shelly Cumberworth 16 ha
Dirnondale, Michigan

Markus Held 1 ha
Mason, Michigan
Jane Bush 7 ha

Charlotte, Michigan

Bruce Schultz 2 ha

Kalamazoo, Michigan

Pauline Lee lha
Williarnston, Michigan

Tony McQuail 26 ha

Lucknow, Ontario

Bum.

part—time, nonchemical, poultry,

pork, beef, potatoes, and squash

part-time, low-chemical,

apple orchard

full-time, certified organic,

apple orchard, herbs

part-time, certified organic,

fruit and vegetable

part-time,

fruit and vegetable

full-time

apples, sheep, poultry

 

117

Table 6-2. Questions used in interviews with farmers on role of domestic birds on small

farms as biological control agents.

 

1) Do you feel that free-range domestic birds have a role on your farm?*

2) What is/are the role(s)?

3) Do you think that free-range chickens have potential as insect pest biological control
agents?*

4) Do you think that free-range chickens have potential as weed biological control agents?*

5) Do you think that free-range geese have potential as weed biological control agents?*

6) Do you feel that predators are a serious problem for free-ranging birds on your farrn?*

7) Do you feel that the labor required in free-ranging domestic birds is too much?*

8) Do you think that geese are too aggressive to use as weeders?*

9) Do you think that raising free-range domestic birds on a small scale can be a profitable
enterprise?“

10) Will you (continue to) raise free-range domestic birds on your farm?*

11) Why? Why not?

 

*the following multiple choice responses were used with the corresponding ranks:

no, (0); probably not (1); maybe/not sure, (2); probably, (3); yes, (4).

118

1) domestic birds have a role

3) chickens for insect pest control
4) chickens for weed control

5) geese for weed control

6) predators are a serious problem
7) labor requirements are too high
8) geese are too aggressive

9) birds on small scale/ profitable

10) continue to raise domestic birds

 

no prob. not maybe/ probably yes
not sure

Figure 6—1. Mean (+SEM) responses by farmers to interview questions 1 and 3 -10. The
original questions are summarized into statements (see entire questions in Table l)

and the statistics were calculated based on ranks (see text).

CHAPTER 7

SUMMARY & CONCLUSIONS

The overall objective of this research was to evaluate the potential for using free-
range chickens and geese as biological control agents for insect pests and weeds. In 1993,
the compatibility of the birds was assessed in a nonchemical apple orchard with
intercropped potatoes, located at the Kellogg Biological Station, Hickory Comers, MI.
The compatibility assessment was based primarily on visual observations of the birds’
behavior. This included their diurnal activity patterns, their tendency to disperse from the
coop and water source, and most importantly, their feeding habits.

Based upon structured observations, chickens were found to be omnivorous,
feeding on insect pests, beneficial insects, weeds, and in some instances, the crops. They
were found to feed on two important insect pests, Japanese beetle and Colorado potato
beetle, however there was noticable inter-bird variability in the willingness to feed on the
latter pest. Although chickens fed upon vegetation, their effects on weed biomass were
small in comparison to the geese. The tendency of the chickens to feed on unearthed potato
tubers was generally not a problem during this study because the potatoes were hilled,
providing protection from the scatching and pecking which typify chicken foraging.
However in later studies in which the potatoes were not hilled, chicken damage to the
tubers was a common occurrence. One other problem during this study was the loss of
chickens to predators. Loss to predators is probably the most obvious drawback to free-
ranging chickens. Because the birds were put into coops at night, the most serious losses

to predators were due to hawks, which hunt in the day.

119

120

The geese were found to be solely herbivorous. They were observed to feed less
often than the chickens but consumed large amounts of vegetation while foraging. The
most preferred plants included common dandelion, common plantain, and grasses, but a
variety of other species were also consumed. Geese required large quantities of water and
spent more time than chickens close to the coop and water source. Geese occasionally fed
upon unearthed potato tubers but did not feed on potato or apple foliage. Weed biomass
comparisons clearly demonstrated that geese had potential as weed control agents.
Furthermore, the geese, which grew faster and larger than the chickens, were less
susceptible to hawk predation.

The compatibility study suggested that chickens might have an effect on at least two
pests (Japanese beetle and Colorado potato beetle), and possibly others, which spend part
of their life cycle on or near the soil surface. By contrast, the geese showed clear promise
as weed control agents. Therefore, in 1994 and 1995, the effectiveness of free-range
chickens and geese as biological control agents of insect pests and weeds was studied. The
ultimate goal was to determine if the presence of free-range chickens or geese provided any
crop protection benefits. The effects of the birds on the abundance of four insect pests
(plum curculio, apple maggot, Japanese beetle, and Colorado potato beetle ), weed growth,
potato growth and yield, and apple damage and yield were measured. The direct and
indirect ecological effects of the birds, as determined in this study, and the compatibility
study, are presented in Figure 7-1.

The feeding activities of the geese resulted in clear reductions in weed growth
which led to increased potato plant growth. In 1994, potato tuber yields in plots weeded
by geese were substantially higher than those of control plots. However, in 1995, no
increase in potato yields were found due to a disease infestation which affected all
ueatments. Interestingly, weeding by the geese also resulted in greater Colorado potato

beetle abundance. Presumably, the presence of large potato plants in a relatively weed-free

121

environment encouraged the colonization and survival of this pest species. Other published
research supports this explanation.

A surprising, but consistent, finding was the reduction in plum curculio damage
due to the presence of geese. In 1994 and 1995, the percentage of fruit with plum curculio
damage in plots with geese was less than in control plots. This resulted in a corresponding
increase in the percentage of undamaged or pest-free fruit. The compatability study
suggested that the effects of the geese in reducing weeds could result in higher apple fruit
yields, however, harvest weights in the plots with geese were not statistically greater than
those of the controls in 1994 or 1995.

The results of the compatibility study indicated that the chickens could have some
effect on weed growth but that densities would have to be increased substantially to achieve
adequate control. The chickens showed more potential, however, as insect control agents.
Insect pest abundance patterns in 1994 and 1995 indicated that the chickens had an effect
on Japanese beetle. Furthermore, analyses of chicken gut contents showed that Japanese
beetle was a common prey. However, fruit damage assessments showed no differences
between the plots with chickens and those without. There are several possible explanations
for these contradictory findings: 1) the small plot size allowed for rapid colonization during
the period of time between chicken removal from the plots and apple harvest; 2) severely
damaged fruit tended to drop from trees resulting in an underestimation of damage in the
plots without chickens; and/or 3) wasp damage was mistaken for Japanese beetle damage,
causing an overestimation of Japanese beetle damage in some plots.

In 1994, plum curculio damage in the plots with chickens was intermediate between
the control and goose treatment but did not differ statistically from either of them. If
chickens did feed on plum curculio it was probably a relatively rare occurrence and
chickens could not be counted on to provide control for this pest. Unfortunately, the effect
of chickens on Colorado potato beetle could not be adequately assessed in this study
because of the confounding effects of weeds on the abundance of this pest. Further

122

research is needed in this area, particularly to address the effects of intra- and inter-breed
variability on the tendency of chickens to feed on this pest.

The effects of free-range chickens and geese on beneficial soil invertebrates was
assessed with two studies. In 1993, the effects on epigeic predators, including ground
beetles, rove beetles, spiders, and harvestmen, were measured in a pitfall trapping study.
Neither bird species was found to have an effect on the activity of ground beetles or rove
beetles, however, the presence of chickens and geese resulted in a reduction in spider and
harvestmen activity. Although the mechanism for these predator reductions were not
determined, these finding suggest that diumally-active predators were negatively influenced
by the birds while noctumally-active predators were not. Presumably, the predator
reductions in the chicken plots were due to predation because chickens were found to feed
on a variety of beneficial arthropods including ground beetles, rove beetles, and spiders.
The effect of geese was apparently indirect, likely resulting from the reduction in vegetation
or from habitat disturbance. Although the presence of the birds did result in a reduction in
predator activity, their influences were relatively minor in comparison to the effects of
intercropping potatoes, which resulted in greater reductions in epigeic predator activity.

In 1995, soil sampling in areas with and without birds showed no differences in
earthworm abundance. Soil analyses showed that earthworm abundance was most highly
correlated with soil organic matter and the distance from an untilled, old field located 25m
east of the orchard. After three summers of free-ranging chickens and geese there were no
detectable increases in soil organic matter. Therefore, increases in earthworm abundance
should not have been expected. However, the results also suggest that the activities of the
birds, particularly chicken predation, did not have an important negative influence on
earthworm abundance, at least at the stocking densities used in this study.

In 1995, six small farmers, with experience in free-ranging domestic birds, were
interviewed to gain some insight into the feasibility of using domestic birds for biological

control and to provide direction for future research. Nearly all of the farmers believed that

123

chickens and geese had potential as insect pest and weed biological control agents,
however, none kept birds solely or primarily for that purpose. Most of the farmers
expressed interest in finding efficient and effective management systems for small scale
bird management which would reduce losses to predators, purchased feed inputs, and labor
requirements. Suggested future research included: assessing the compatibility of different
crops with chickens and geese, determining which insect pests can be managed with
chickens or other domestic fowl, comparing the effectiveness of different bird species and
breeds for biological control, and estimating bird stocking rates for particular management
needs.

Overall, the findings of this study indicate that there are possibilities for integrating
domestic birds into ecologically-based agricultural systems for biological pest management.
The geese, in particular, demonstrated clear potential as weed control agents. Although not
suited for all types of systems, geese show the greatest potential in diversified horticultural
operations, especially when agrochemicals, such as herbicides, are not used. Production
systems comprised of perennial shrubs and trees are most compatible with geese because
they provide shade for the birds and are relatively immune to feeding and trampling
damage. However, annual vegetables and herbs can also be compatible with weeder geese,
but may require additional care, such as trellising or staking, to prevent damage. The
indirect effects of geese on insect pests may be positive, negative, or absent. Further
research is needed to evaluate how disturbance by foraging geese affects insect pests and
their damage.

One factor which may limit the use of weeder geese is the perception that they are
too aggressive. Although the threat by domestic geese is commonly overstated, farmers
and gardeners who are unfamiliar with these birds may be intimidated by their defensive
displays. Education and experience with geese may be needed to counter this perception.

This study provides relatively little support for integrating free-range chickens into

agroecosystems for insect pest or weed control. Although chickens did consume insects

124

and weeds in this study, their impact did not result in any real detectible crop protection
benefits. The results did suggest that chickens may have some potential in controlling
Japanese beetles, especially if used in conjunction with lures. However, achieving any
measurable level of insect pest or weed control with free-range chickens would require
much higher stocking rates than were used in this study. Another factor which can limit the
applicability of free-range chickens in pest management is loss to predators. Although
coops and electric fencing can provide some protection, losses can still be high in rural
areas with little human presence. In situations in which high stocking rates can be achieved
and predator pressure is minimal, free-range chickens may have some potential to play a

role in biological pest management.

125

 

l GEESE

 

 

feeding, trampling

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I WEED I
GROWTH . _
reduced I (-) I habitat alteration ?
weed l disturbance?
competition reduced
weed
competition
POTATO APPLE PLUM
GROWTH GROWTH CURCULIO
(+) (+) H
larger potato plants
fewer weeds
OLORADO
TATO
EETLE (+)
ICHICKENS _ _ _
I _l
. | .
feedIng predation ?
scratching predation | scratching ?
l
WEED JAPANESE OLORADO
GROWTH BEETLE TATO
f) (-) EETLE (-)

 

 

 

 

 

Figure 7-1. Diagram summarizing the detected (solid arrows) and suggested (dashed
arrows) ecological effects of geese and chickens in the experimental agroecosystem. The

effects are indicated as positive (+) or negative (-) for each of the measured variables.

LIST OF REFERENCES

126

LIST OF REFERENCES

Agnello, A. M., J. Kovach, J. P. Nyrop, W. H. Reissig, D. I. Breth, and W. F. Wilcox.
1994. Extension and evaluation of a sim‘ lified monitoring program in New York
apples. American Entomologist 40: 37-49.

 

 

 

 

 

Allen, W. R. and E. A. C. Hagley. 1990. Epigeal arthropods as predators of mature larvae
and pupae of the apple maggot (Diptera: Tephritidae). Environmental Entomology
19: 309-312.

Altieri, M. A. 1987. Agroecology: The Scientific Basis of Alternative Agriculture.
Westview Press, Boulder, CO. 227p.

Altieri, M. A. 1994. Biodiversity and Pest Management in Agroecosystems. Food
Products Press, New York. 185p.

Andow, D. A. 1991 . Vegetational diversity and arthropod population response. Annual
Review of Entomology 36: 561-586.

Appleby, M. C. and B. 0. Hughes. 1991. Welfare of laying hens in cages and alternative
systems: environmental, physical and behavioural aspects. World’s Poultry Science

Journal 47: 109-128.

Axtell, R. C. and J. J. Arends. 1990. Ecology and management of arthropod pests of
poultry. Annual Review of Entomology 35: 101-126.

Bachmann, J. 1993. Weeder geese: A practical alternative to herbicides. Kerr Center for
Sustainable Agriculture Newsletter 19: 1-2.

Bellinder, R. R. and R. W. Wallace. 1991. An integrated production management
approach to weed control in potatoes. In D. Pimentel and A. A. Hanson (eds.).
CRC handbook of pest management in agriculture, CRC Press, Boca Raton, pp.
677-688.

Bender, J. 1994. Future Harvest: Pesticide-Free Farming. University of Nebraska Press,
Lincoln, Nebraska. 159p.

Berry, E. C. 1994. Earthworms and other fauna in the soil. In: J. L. Hatfield and B. A.
Stewart (Editors), Soil Biology: Effects on Soil Quality. CRC Press, Boca Raton,
pp. 61-90.

Biever, K. D. & R. L. Chauvin. 1992. Suppression of Colorado potato beetle (Coleoptera:
Chrysomelidae) with augmentative releases of predaceous stinkbugs (Herniptera:
Pentatomidae). Journal of Economic Entomology 85: 720-726.

Brown, L. R. and H. Kane. 1994. Full House: Reassessing the Earth’s Carrying Capacity.
W. W. Norton, New York. 250p.

127

Brunner, J. F. and A. J. Howitt. 1981. Tree Fruit Insects. Michigan State University
Cooperative Extension Service. 60p.

Brust, G. E. 1994. Natural enemies in straw-mulch reduce Colorado potato beetle
populations and damage in potato. Biological Control: 4: 163-169.

Brust, G. E., B. R. Stinner and D. A. McCartney. 1986. Predator activity and predation in
com agroecosystems. Environmental Entomology 15: 1017-1021.

Bugg, R. L. and C. Waddington. 1994. Using cover crops to manage arthropod pests of
orchards: A review. Agriculture Ecosystems and Environment: 50: 11-28.

Bush, G. L. 1992. Host race formation and sympatric speciation in Rhagoletis fruit flies
(Diptera: Tephritidae). Psyche 99: 335-357.

Casagrande, R. A. 1987. The Colorado potato beetle: 125 years of mismanagement.
Bulletin of the Entomological Society of America 33: 142-150.

Chouinard, G., S. B. Hill, and C. Vincent. 1993. Spring behavior of the plum curculio
(Coleoptera: Curculionidae) within caged dwarf apple trees. Annals of the
Entomological Society of America 86: 333-340.

Coli, W. M. and R, Ciurlino. 1990. Interaction of weeds and apple pests. Proceedings of
the New England Fruit Meetings 96: 52-58.

Cramer, C. 1992. Weeder geese boost berry profits $222 per acre. The New Farm 14(4):
38-40.

Crews, T. E. and S. R. Gliessman. 1991. Raised field agriculture in Tlaxcala, Mexico: An
ecosystem perspective on maintenance of soil fertility. American Journal of
Alternative Agriculture 6: 9-16.

Curtis, J ., T. Profeta, and L. Mott. 1993. After Silent Spring: The Unsolved Problems of
Pesticide Use in the United States. Natural Resources Defense Council, Inc.,
Washington DC. 56p.

Daly, H. E. 1993. Sustainable growth: an impossibility theorm. In H. E. Daly and K. N.
Townsend [eds]. Valuing the earth: economics. ecology, and ethics. pp. 267-273.
The MIT Press, Cambridge, Massachusetts.

Davidson, R. H. and W. F. Lyon. 1987. Insect Pests of Farm, Garden, and Orchard. John
Wiley and Sons, New York. 644p.

Dazhong, W. and D. Pimentel. 1986. Seventeenth century organic agriculture in China:
cropping systems in Jiaxing region. Human Ecology 14: 1-14.

Dazhong, W., T. Yingxin, Z. Xunhua, and H. Yungzhen. 1992. Sustainable and
productive agricultural development in China. Agriculture Ecosystems Environment
39: 55-70.

Duan, J. J. and R. J. Prokopy. 1995. Control of apple maggot flies (Diptera: Tephritidae)
with pesticide-treated red spheres. Journal of Economic Entomology 88: 700-707.

128

Edens T. C. and D. L. Haynes. 1982. Closed system agriculture: resource constraints,
management options, and design alternatives. Annual Review of Phytopathology
20: 363-395.

Edwards, C. A. 1989. The importance of integration in sustainable agricultural systems,
Agriculture Ecosystems Environment 27: 35-35.

Edwards, C. A. 1993. The impact of pesticides on the environment. In D. Pimentel and H.
Lehman (Eds), The pesticide question: environment, economics and ethics.
Chapman and Hall, New York. pp. 13-46.

Edwards, C. A., P. J. Bohlen, D. R. Linden, and S. Subler. 1995. Earthworms in
agroecosystems. In P. F. Hendrix (ed.), Earthworm Ecology and Biogeography in
North America. CRC Press, Boca Raton. pp. 185-213.

Edwards, C. A., T. L. Grove, R. R. Harwood, and C. J. P. Colfer. 1993. The role of
agroecology and integrated farming systems in agricultural sustainability.
Agriculture, Ecosystems and Environment 46: 99-121.

Feder, J. L. 1995. The effects of parasitoids on sympatric host races of Rhagoletis
pomonella (Diptera: Tephritidae). Ecology 76: 801-813.

Fuan, L. 1985. Integration of duck production and rice culture in south China. In D. J.
Farrell and P. Sarnpleton (eds). Duck Production Science and World Practice.
University of New England, Armidale, Australia, pp. 385-392.

Gerber, J. M. 1993. Farmer participation in research: A model for adaptive research and
education. American Journal of Alternative Agriculture 7: 118-121.

Giampietro, M. and D. Pimentel. 1992. Energy efficiency and nutrition in societies based
on human labor. Ecology of Food and Nutrition 28: 11-32.

Gillen, R. L. and C. J. Scifres. 1991. Selective grazing as a weed control method. In D.
Pimentel and A. A. Hanson (eds.). CRC Handbook of Pest Management in
Agriculture, CRC Press, Boca Raton, pp. 369-376.

Glass, E. H. and S. E. Lienk. 1971. Apple insect and mite populations developing after
discontinuance of insecticides: lO-year record. Journal of Economic Entomology
64: 23-26.

Glofcheskie, B. D. and G. A. Surgeoner. 1990. Muscovy ducks as an adjunct for the
control of the house fly (Diptera: Muscidae). Journal of Economic Entomology 83:
788-79 1 .

Glofcheskie, B. D. and G. A. Surgeoner. 1993. Efficacy of muscovy ducks as an adjunct
for house fly (Diptera: Muscidae) control in swine and dairy operations. Journal of
Economic Entomology 86: 1686-1692.

Gould, F. 1991. The evolutionary potential of crop pests. American Scientist 79: 496-507.

Groden, E., F. A. Drummond, R. A. Casagrande, and D. L. Haynes. 1990. Coleomegilla
maculata (Coleoptera: Coccinellidae): Its predation upon the Colorado potato beetle
(Coleoptera: Chrysomelidae) and its incidence in potatoes and surrounding crops.
Journal of Economic Entomology 83: 1306-1315.

129

Hagley, E. A. C. and W. R. Allen. 1988. Ground beetles (Coleoptera: Carabidae) as
predators of the codling moth, Cydia pomonella (L.) (Lepidoptera: Tortricidae).
Can. Entomol. 120: 917-925.

Hall, C. A. S. 1990. Sanctioning resource depletion: Economic development and neo-
classical econornics. The Ecologist 20: 61-66.

Hall, C. A. S., C. J. Cleveland, and R. Kaufmann. 1992. Energy and Resource Quality.
University Press of Colorado, Niwot, CO. 577p.

Hardin, G. 1968. The tragedy of the commons. Science 162: 1243-1248.

Hare, J. D. 1990. Ecology and management of Colorado potato beetle. Annual Review of
Entomology 35: 81-100.

Hazzard, R. V., D. N. Ferro, R. G. Van Driesche, and A. F. Tuttle. 1991. Mortality of
eggs of Colorado potato beetle (Coleoptera: Chrysomelidae) from predation by
Coleomegilla maculata (Coleoptera: Coccinellidae). Environmental Entomology 20:
841-848.

Hendrix, P. F., B. R. Mueller, R. R. Bruce, G. W. Langdale, and R. W. Parmelee. 1992.
Abundance and distribution of earthworms in relation to landscape factors on the
Georgia piedmont, U.S.A. Soil Biology and Biochemistry 24: 1357-1361.

Hill, S. B. and R. J. MacRae. 1992. Organic farming in Canada. Agriculture Ecosystems
and Environment 39: 71-84.

Hoare, A. H. 1928. The English Grass Orchard and the Principles of Fruit Growing. The
Riverside Press, Great Britain, 227p.

Holderread, D. 1981. The Book of Geese. Hen House Publications, Corvallis, OR.

Horton, D. R. and J. L. Capinera. 1987. Effects of plant diversity, host density, and host
size on population ecology of the Colorado potato beetle. Environmental
Entomology 16: 1019-1026.

Hough-Goldstein, J. A., G. E. Heimpel, H. E. Bechmann, and C. E. Mason. 1993.
Arthropod natural enemies of the Colorado potato beetle. Crop Protection 12: 324-
334.

Hough-Goldstein, J. A., J. Geiger, D. Chang, and W. Saylor. 1993b. Palatability and
toxicity of the Colorado potato beetle (Coleoptera: Chrysomelidae) to domestic
chickens. Annals of the Entomological Society of America 86: 158-164.

Hough-Goldstein, J. and J. Whalen. 1993. Inundative release of predatory stink bugs for
control of Colorado potato beetle. Biological Control 3: 343-347.

Howarth, F. G. 1991. Environmental impacts of classical biological control. Annual
Review of Entomology 36: 485-509.

Howitt, A. J. 1993. Common Tree Fruit Pests. Michigan State University Extension NRC
63.

130

Hoy, M. A. 1992. Biological control of arthropods: genetic engineering and environmental
risks. Biological Control 2: 166-170.

Ikerd, J. E. 1993. The question of good science. American Journal of Alternative
Agriculture 8: 91-93.

Johnson, C. 1960. Management of weeder geese in commercial fields. California
Agriculture 14(8): 5.

Johnson, J. W. and M. Herr. 1995. Apple Insect Scouting Manual. Great Lakes
Publishing. Sparta, Michigan.

Johnson, W. T. and H. H. Lyon. 1991. Insects that Feed on Trees and Shrubs. Cornell
University Press, Ithaca. 560p.

Kennel, W. 1990. The role of the earthworm Lumbricus terrestris in integrated film
production. Acta Hort. 285: 149-156.

Kiley-Worthington, M. 1993. Eco-Agriculture: Food First Farming. Souvenir Press,
London. 276p.

Kim, K. C. 1993. Insect pests and evolution. In: K. C. Kim and B. A. McPheron (eds.),
Evolution of Insect Pests. John Wiley and Sons, New York, pp. 3-25.

King, L. D. 1990. Soil nutrient management in the United States. In C. A. Edwards, R.
La], P. Madden, R. H. Miller, and G. House (eds.). Sustainable agricultural
systems. Soil and Water Conservation Society, Ankeny, Iowa. pp. 89-106.

Lacy, W. B. 1994. Biodiversity, cultural diversity, and food equity. Agriculture and
Human Values 11(1): 3-9.

Lafleur, G. and S. B. Hill. 1987. Spring migration, within-orchard dispersal, and apple-
tree preference of plum curculio (Coleoptera: Curculionidae) in southern Quebec.
Journal of Economic Entomology 80: 1173-1187.

Lafleur, G., S. B. Hill, and C. Vincent. 1987. Fall migration, hibernation site selection,
and associated winter mortality of plum curculio (Coleoptera: Orrculionidae) in a
Quebec apple orchard. Journal of Economic Entomology 80: 1152-1172.

Landis, D. A. 1994. Arthropod sampling in agricultural landscapes: ecological
considerations. In L. P. Pedigo and G. D. Buntin [Eds]. Handbook of sampling
methods for arthropods in agriculture. CRC Press. Boca Raton. pp. 15-31.

Lappe, F. M. 1991. Diet for a Small Planet. Ballantine Books, NY. 479p.

Leius, K. 1967. Influence of wild flowers on parasitism of tent caterpillar and codling
moth. Canadian Entomologist 99: 444-446.

Letoumeau, D. K. 1987. The enemies hypothesis: tritrophic interactions and vegetational
diversity in tropical agroecosystems. Ecology 68: 1616-1622.

Lockeretz, W., G. G. Shearer, and D. H. Kohl. 1981. Organic farming in the corn belt.
Science 211: 540—547.

131

Loomis, R. S. 1984. Traditional agriculture in America. Annual Review of Ecology and
Systematics 15: 449-478.

Loomis, R. S. and D. J. Connor. 1992. Crop Ecology: Productivity and Management in
Agricultural Systems. Cambridge University Press, Great Britain. 538p.

Luna, J. M. and G. J. House. 1990. Pest management in sustainable agricultural systems.
pp. 157-173. In: C. A. Edwards, R. Lal, P. Madden, R. H. Miller, and G. House
(eds.). Sustainable Agricultural Systems. Soil and Water Conservation Society,
Ankeny, Iowa.

Maddox, J. V. 1994. Insect pathogens as biological control agents. pp 199-244. In: R. L
Metcalf and W. H. Luckmann (eds.). Introduction to insect pest management. John
Wiley and Sons, New York.

Madsen, H. F. & B. J. Madsen. 1982. Populations of beneficial and pest arthropods in an
organic and a pesticide treated apple orchard in British Columbia. Canadian
Entomologist 114: 1083-1088.

Mather, J. G. and O. Christensen. 1988. Surface movements of earthworms in agricultural
land. Pedobiologia 32: 399-405.

Mayton, E. L. E. V. Smith and D. King. 1945. Nutgrass eradication studies. IV. Use of
chickens and geese in the control of nutgrass, Cyperus rotundus L. Journal of the
American Society of Agronomy 37: 785-791.

Meadows, D. H., D. L. Meadows, and J. Randers. 1992. Beyond the Limits: Confronting
Global Collapse, Envisioning a Sustainable Future. Chelsea Green Publishers, Post
Mills, Vermont.

Merwin, I. A. and M. P. Pritts. 1993. Are modem fruit production systems sustainable?
HortTechnology 3: 128-136.

Meyer, W. B. and B. L. Turner 11. 1992. Human population growth and global land-use/
cover change. Annual Review of Ecology and Systematics 23: 39-61.

Mollison, B. 1988. Permaculture: A Designers’ Manual. Tagari Publications, Tyalgum,
NSW 2484, Australia. 576p.

National Research Council. 1989. Alternative agriculture. National Academy Press,
Washington, DC. 448p.

National Research Council. 1991. Microlivestock. National Academy Press, Washington,
D. C. 449p.

Nelson, R. H. 1995. Sustainability, efficiency, and God: Economic values and the
sustainability debate. Annual Review of Ecology and Systematics 26: 135-154.

Niggli, U., F. P. Weibel, and W. Gut. 1990. Wed control with organic mulch materials
in orchards: Results from 8 year field experiments. Acta Horticulturae 285: 97 -102.

Nordlund, D. A., D. C. Vacek, and D. N. Ferro. 1991. Predation of Colorado potato
beetle (Coleoptera: Chrysomelidae) eggs and larvae by Chrysoperla rufilabris

132

(Neuroptera: Chrysopidae) larvae in the laboratory and field cages. Journal of
Entomological Science 26: 443-449.

Nowak, P. 1992. Why farmers adopt production technology. Journal of Soil and Water
Conservation 47: 14-16.

Ojeniyi, A. A. 1985. Comparative bacterial drug resistance in modern battery and free-
range poultry in a tropical environment. Veterinary Record 117: 11-12.

Olkowski, W., S. Daar, and H. Olkowski. 1991. Common-Sense Pest Control. Taunton
Press, Newtown, Connecticut. 715p.

Osteen, C. 1993. Pesticide use trends and issues in the United States, pp. 307-336. In D.
Pimentel & H. Lehman [eds], The Pesticide Question. Chapman & Hall, New
York.

Parker, C. F. 1990. Role of animals in sustainable agriculture. In C. A. Edwards, R. Lal,
P. Madden, R. H. Miller, and G. House (eds.). Sustainable agricultural systems.
Soil and Water Conservation Society, Ankeny, Iowa. pp. 238-245.

Peacock, L. and G. A. Norton. 1990. A critical analysis of organic vegetable crop
production in the UK. Agriculture, Ecosystems and Environment: 31 :187-197.

Pfeiffer, D. G., W. Kaakeh, J. C. Killian, M. W. Lachance & P. Kirsch. 1993. Mating
disruption for control of damage by codling moth in Virginia apple orchards.
Entomol. exp. appl. 67: 57-64.

Pimentel, D. 1984. Energy flows in the food system. pp. 1-24. In D. Pimentel and C. W.
Hall [eds], Food and Energy Resources. Academic Press, Orlando.

Pimentel, D. 1991. Introduction, pp. 3-11. In D. D. Pimentel [ed.], CRC Handbook of
Pest Management In Agriculture. CRC Press, Boca Raton.

Pimentel, D. 1993. Economics and energetics of organic and conventional farming. Journal
of Agricultural and Environmental Ethics 6: 53-60.

Pimentel, D. and W. Dazhong. 1990. Technology changes in energy use in US.
agricultural production. In C. R. Carroll, J. H. Vandermeer, and P. Rosset [eds.],
Agroecology. McGraw-Hill, New York. p. 147-164.

Pimentel, D. C. Kirby, and A. Shroff. 1993c. The relationship between “cosmetic
standar ” for foods and pesticide use.pp. 85-105. In D. Pimentel & H. Lehman
[eds], The pesticide question. Chapman & Hall, New York.

Pimentel, D., C. Harvey, P. Resosudarrno, K. Sinclair, D. Kurz, M. McNair, S. Crist, L.
Shpritz, L. Fitton, R. Saffouri, and R. Blair. 1995. Environmental and economic
costs of soil erosion and conservation benefits. Science 267: 1117-1123.

Pimentel, D., G. Berardi & 8. Fast. 1983. Energy efficiency of farming systems: organic
and conventional agriculture. Agriculture Ecosystems and Environment 9: 359-372.

Pimentel, D., H. Acquay, M. Biltonen, P. Rice, M. Silva, J. Nelson, V. Lipner, S.
Giordano, A. Horowitz, & M. D’Amore. 1993a. Assessment of environmental and

133

economic impacts of pesticide use, pp. 47-84 . In D. Pimentel & H. Lehman
[eds.], The Pesticide Question. Chapman & Hall, New York.

Pimentel, D., L. McLaughlin, A. Zepp, B. Lakitan, T. Kraus, P. Kleinman, F. Vancini,
W. J. Roach, E. Graap, W. S. Keeton, & G. Selig. 1993b. Environmental and
economic impacts of reducing agricultural pesticide use, pp. 223-278. In D.
Pimentel & H. Lehman [eds.], The Pesticide Question. Chapman & Hall, New
York.

Pope, C. W. 1991. Poultry production’s environmental impact on water quality. Poultry
Science 70: 1123-1125.

Pradgitt, S. and P. Petrzelka. 1994. Making sustainable agriculture the new conventional
agriculture: social change and sustainability. p. 261-285. In J. L. Hatfield and D.
L. Karlen [eds.]. Sustainable Agricultural Systems. CRC Press, Boca Raton.

Prokopy, R. J. 1991. A small low-input commercial apple orchard in eastern North
America: management and economics. Agriculture Ecosystems and Environment
33: 353-362.

Prokopy, R. J. 1994. Integration in orchard pest and habitat management: A review.
Agriculture Ecosystems and Environment: 50: 1-10.

Prokopy, R. J. and B. A. Croft. 1994. Apple insect pest management. pp. 543-585. In: R.
L Metcalf and W. H. Luckmann (eds.), Introduction to insect pest management.
John Wiley and Sons, New York.

Racette, G., G. Chouinard, C. Vincent, and S. B. Hill. 1992. Ecology and management of
the plum curculio, Conotrachelus nenuphar (Coleoptera: Curculionidae), in apple
orchards. Phytoprotection 73: 85-100.

Radcliffe, E. B., K. L. Flanders, D. W. Ragsdale, & D. M. Noetzel. 1991. Pest
management systems for potato insects, pp. 587-621. In D. Pimentel [ed.], CRC
handbook of pest management in agriculture. CRC Press, Boca Raton.

Raese, J. T. 1990. The importance of weed control and nitrogen fertilizer on growth and
yield of young bearing apple and pear trees. Journal of Sustainable Agriculture
1(2): 7- 1 8.

Raw, F. 1962. Studies of earthworm populations in orchards 1: Leaf burial in apple
orchards. Annals of Applied Biology 50: 389-404.

Reganold, J. P., A. S. Palmer, J. C. Lockhart. and A. N. Macgregor. 1993. Soil quality
and financial performance of biodynamic and conventional farms in New Zealand.
Science 260: 344-349.

Reinken, G. 1986. Six years of biodynamic growing of vegetables and apples in
comparison with the conventional management. In H. Vogtmann, E. Boehncke,
and I. Fricke (eds.). The importance of biological agriculture in a world of
diminishing resources: Proceedings of the 5th IFOAM international scientific
conference, University of Kassel, Germany. pp. 161-174.

134

Reissig, W. H., R. W. Weires, G. C. Forshey, W. L. Roelofs, R. C. Lamb & H. S.
Aldwinckle. 1984. Insect management in disease-resistant dwarf and semi-dwarf
apple trees. Environmental Entomology 13: 1201-1207.

Rice, R. A. and J. Vandermeer. 1990. Climate and the geography of agriculture. In
Agroecology. C. R. Carroll, J. H. Vandermeer, and P. Rosset [eds.]. MacGraw
Hill, New York, pp. 21-64.

Richards, J. F. 1990. Land transformation. In B. L. Turner 11, W. C. Clark, R. W. Kates,
J. F. Richards, J. T. Mathews, and W. B. Meyer [Eds]. The Earth as A
Transformed by Human Action. Cambridge University Press, Cambridge, pp. 163-
178.

Riddick, E. W. and N. J. Mills. 1994. Potential of adult carabids (Coleoptera: Carabidae)
as predators of fifth-instar codling moth (Lepidoptera: Tortricidae) in apple orchards
in California. Environmental Entomology 23: 1338-1345.

Riechert , S. E. & L. Bishop. 1990. Prey control by an assemblage of generalist predators:
spiders in garden test systems. Ecology 71: 1441-1450.

Rifltin, J. 1992. Beyond Beef. Penguin Books, NY. 353p.

Risch, S. J ., D. Andow. and M. A. Altieri. 1983. Agroecosystem diversity and pest
control: data, tentative conclusions, and new research directions. Environmental
Entomology 12: 525-629.

Robbins, J. 1987. Diet for a New America. Stillpoint Publishing, Walpole, NH. 423p.
Rodale Institute. 1991. On-farrn research: summary of 1990 results. Emmaus, PA. 65p.

Roltsch, W. J. and S. G. Gage. 1990. Influence of bean-tomato intercropping on
population dynamics of the potato leafhopper (HomopterazCicadellidae).
Environmental Entomology 19: 534-543.

Ruddle, K. and G. Zhong. 1988. Integrated Agriculture-Aquaculture in South China: The
Dike-Pond System on the Zhujiang Delta. Cambridge University Press,
Cambridge. 173p.

Rupp, L. A. and J. L. Anderson. 1985. Grth and fruiting responses of young apple and
tart cherry trees to weed control. HortScience 20: 727-729.

Russell, E. P. 1989. Enemies hypothesis: a review of the effect of vegetational diversity on
predatory insects and parasitoids. Environmental Entomology 18: 590—599.

Salatin, J. 1991. Profit by appointment only. The New Farm 13(6): 8-12.

Schuman, S. 1993. Risk of pesticide-related health effects: An epidemiologic approach.
pp. 106-125. In D. Pimentel & H. Lehman [eds.], The Pesticide Question.
Chapman & Hall, New York.

Sharpley, A. and M. Meyer. 1993. Minimizing agricultural nonpoint-source impacts: a
symposium overview. Journal of Environmental Quality 23: 1-3.

Shirley, C. Put stock in orchards and woodlots. The New Farm 14(4): 35-37.

135

Silversides, F. G., R. D. Crawford, and H. C. Wang. 1988. The cytogenetics of domestic
geese. Journal of Heredity 79: 6-8.

Sims, J. T. and D. C. Wolf. 1994. Poultry waste management: agricultural and
environmental issues. Advances in Agronomy 52: 1-83.

Smolik, J. D., T. L. Dobbs, T. L., D. H. Rickeri, L. J. Wrage, G. W. Buchenau, and T.
A. Machacek. 1993. Agronomic, economic, and ecological relationships in
alternative (organic), conventional, and reduced-till farming systems. Bulletin of the
Agricultural Experiment Station, South Dakota State University. 57p.

Soule, J. K. and J. D. Piper. 1992. Farnring in Nature’s Image: An Ecological Approach
to Agriculture. Island Press, Wash. DC. 286p.

Soule, J ., D. Carre, and W. Jackson. 1990. Ecological impact of modern agriculture. In C.
R.’ Carroll, J. H. Vandermeer, and P. Rosset [eds.], Agroecology. McGraw-Hill,
New York. p. 165-188.

Spalding, R. F. and M. E. Exner. 1993. Occurrence of nitrate in groundwater-a review.
Journal of Environmental Quality 22: 392-402.

Stanhill, G. 1990. The comparative productivity of organic agriculture. Agriculture
Ecosystems and Environment 30: 1-26.

Stoner, K. A. 1993. Effects of straw and leaf mulches and trickle irrigation on the
abundance of Colorado potato beetles (Coleoptera: Chrysomelidae) on potato in
Connecticut. Journal of Entomological Science 28: 393-403.

Stout, B. A. 1993. Handbook of Energy for World Agriculture. Elsevier Science
Publishers, New York.

Subinprasert, S. 1987. Natural enemies and their impact on overwintering codling moth
populations (Laspeyresia pomonella L.) (Lep., Tortricidae) in South Sweden.
Journal of Applied Entomology 103: 46-55.

Terry, L. A., D. A. Potter, and P. G. Spicer. 1993. Insecticides affect predatory
arthropods and predation on Japanese beetle (Coleoptera: Scarabaeidae) eggs and
fall arrnyworrn (Lepidoptera: N octuidae) pupae in turfgrass. Journal of Economic
Entomology: 86: 871-878.

Traupman, M. 1990. Profitable poultry on pasture. The New Farm 12(4): 20-23.

US. Department of Agriculture. 1990. Management Guide for Low-Input Sustainable
Apple Production. 83p.

VABF. 1993. Proceedings of the Virginia Association of Biological Farming New River
chapter farmer to farmer conference. Floyd, VA.

van Ravenswaay, E. 1994. The consumer’s perspective on pesticide use. The IPM Report:
Integrated Pest Management in Michigan 3: 5-8.

Vandermeer, J. 1989. The Ecology of Intercropping. Cambridge University Press,
Cambridge. 237p.

136

Vandermeer, J. 1995. The ecological basis of alternative agriculture. Annual Review of
Ecology and Systematics 26: 201-224.

Vereijken, P. 1989. From integrated control to integrated farming, an experimental
approach. Agriculture Ecosystems and Environment 26: 37-43.

Vereijken, P. and C. D. Van Loon. 1991. A strategy for integrated low-input potato
production. Potato Research 34: 57-66.

Vincent, C. and M. Roy. 1992. Entomological limits to biological control programs in
Quebec apple orchards. Acta Phytopathol. Entomol. Hung. 27: 649-657.

Vitousek, P. M. 1992. Global environmental change: An introduction. Annual Review of
Ecology and Systematics 23: 1-14.

Vitousek, P. M., P. R. Ehrlich, A. H. Ehrlich, and P. A. Matson. 1986. Human
appropriation of the products of photosynthesis. Bioscience 36: 368-373.

Wagstaff, H. 1987. Husbandry methods and farm systems in industrialized countries
which use lower levels of external inputs: a review. Agriculture Ecosystems and
Environment. 19: 1-27.

Ware, A. 1995. Geese in the strawberry patch. Small Farm Today 12(3): 34-36.

Whalon, M. E. & B. A. Croft. 1984. Apple IPM implementation in North America.
Annual Review of Entomology 29: 435-470.

Whalon, M. E., D. L. Miller, R. M. Hollingworth, E. J. Grafius, and J. R. Miller. 1993.
Selection of a Colorado potato beetle (Coleoptera: Chrysomelidae) strain resistant to
Bacillus thuringiensis. Journal of Economic Entomology 86: 226-233.

White, D. C., J. B. Braden, and R. H. Hombaker. 1994. Economics of sustainable
agriculture. p. 229-260. In J. L. Hatfield and D. L. Karlen [eds.]. Sustainable
Agricultural Systems. CRC Press, Boca Raton.

Wurtz, T. L. 1995. Domestic geese: biological weed control in an agricultural setting.
Ecological Applications 5: 570-578.

York,, E. T. 1991. Agricultural sustainability and its implications to the horticulture
profession and the ability to meet global food needs. HortScience 26: 252-256.

Zar, J. H. 1984. Biostatistical Analysis. Prentice Hall, Englewood Cliffs, NJ. p. 142.
Zehnder, G. W. and G. K. Evanylo. 1989. Influence of extent and timing of Colorado

potato beetle defoliation on potato tuber production in eastern Virginia. J. Economic
Entomology 82: 948-953.

 

HICHI IES

lllrrrrlrllrll