Refined sampling methodology and action thresholds for the pepper weevil, Anthonomus eugenii cano (Coleoptera: curculionidae)

MISSING IMAGE

Material Information

Title:
Refined sampling methodology and action thresholds for the pepper weevil, Anthonomus eugenii cano (Coleoptera: curculionidae)
Physical Description:
xii, 179 leaves : ill. ; 29 cm.
Language:
English
Creator:
Riley, David Gerard, 1957-
Publication Date:

Subjects

Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1990.
Bibliography:
Includes bibliographical references (leaves 172-178).
Statement of Responsibility:
by David Gerard Riley.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001572707
notis - AHJ6535
oclc - 22830039
System ID:
AA00003757:00001

Full Text












REFINED SAMPLING METHODOLOGY AND ACTION THRESHOLDS
FOR THE PEPPER WEEVIL, ANTHONOMUS EUGENII CANO
(COLEOPTERA: CURCULIONIDAE)










BY

DAVID GERARD RILEY


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1990














ACKNOWLEDGEMENTS

Many individuals contributed to the research presented in this dissertation. First, I

would like to thank my parents for their care and understanding during the many years of

my formal education. Drs. Dave J. Schuster and Carl S. Barfield provided the fundamental

support and motivation for this research project, and I would like to thank them for making

the research profitable as well as enjoyable. Many of the personnel at the Gulf Coast

Research and Education Center at Bradenton also assisted me in the study and provided

much-valued friendship on a daily basis. I would like to specifically recognize Emily

Vasquez for her kind comments and illustrations; Ken Kiger for assisting in various field

activities; Annette McDonald for assistance in data entry; all of the entomology group for

providing an enjoyable work atmosphere; L. M. McPherson, Mrs. E. G. Jones, and the

rest of the farm crew for their inspiration and good nature; all the personnel in the front

office for their much-appreciated support; Gail Somodi and Nancy West for assistance in

proofreading; and all the staff at the station for periodic assistance in various aspects of the

research. Special mention goes to Kimberly Taylor, who directly assisted in the

experiments conducted in Honduras. Ted Winsberg, Ken Shuler, Ken Porhonezny,

Melonie Kegley, Myrine Hewitt, and Glades Crop Care, Inc., provided the fundamental

cooperation in an action threshold validation trial conducted in Palm Beach County. I

would also like to acknowledge Dr. J.B. Kring for his endless sense of humor, hard

candies, and professional advice concerning insect responses to color. I would like to

recognize the rest of my committee, K. L. Andrews, J. R.Strayer, and M. J. Bassett, for

being readily available and supportive during the program. Finally, I would like to thank

my fiancee, Angie Schmid, for putting up with long work days and nights.
















TABLE OF CONTENTS


page


ACKNOWLEDGEMENTS...............................

LIST OF TABLES ....................................

LIST OF FIGURES ................. .......... ......

ABSTRACT .....................................


. ii


. v


. viii


CHAPTERS


1. INTROD UCTIO N ......................................... 1

2. REVIEW OF LITERATURE ................................... 4
2.1 Pepper Weevil Biology and Bionomics ..................... 4
2.2 Sampling and Spatial Dispersion of the Pepper Weevil and Other
Anthonomii....................................... 10
2.3 Control of the Pepper Weevil ................. ........ 13
2.4 Major Research Needs for the Management of the Pepper Weevil ...... 15

3. SAMPLING AND SPATIAL DISPERSION OF THE PEPPER WEEVIL ...... 17
3.1 Sampling Techniques for the Pepper Weevil, Anthonomus eugenii
Cano (Coleoptera: Curculionidae) ....................... 17
3.2 Spatial Dispersion of the Pepper Weevil, Anthonomus eugenii Cano
(Coleoptera: Curculionidae) ...........................61


4. ACTION THRESHOLDS FOR THE PEPPER WEEVIL ..............
4.1 Threshold Applications of Cypermethrin, Fenvalerate, Methomyl, and
Oxamyl for Control of Anthonomus eugenii and Other Insect Pests
of Peppers in Honduras ............................
4.2 Effect of Diurnal Sample Variation on Insecticidal Action Thresholds
for the Pepper Weevil, Anthonomus eugenii Cano (Coleoptera:
Curculionidae) ..................................
4.3 A Lower Action Threshold for the Pepper Weevil, Anthonomus eugeni
Cano (Coleoptera: Curculionidae) ......................
4.4 Commercial Field Testing of Sampling Plan and Action Threshold for
the Pepper Weevil ..............................

5. SUMMARY AND CONCLUSIONS ...........................


. 82


..82


S. 97

S118

. 135

S142











A PPEND ICES ............................................ 149

A. PEPPER WEEVIL COUNT DATA 1987 ....................... 150

B. PEPPER WEEVIL MARKED-RECAPTURE DATA 1988 .............. 167

C. DISPERSION ESTIMATES FOR FALLEN BUDS ................. 171

REFERENCES ........................................... 172

BIOGRAPHICAL SKETCH .......... ........................ 179















LIST OF TABLES
Table page

2.1 Summary of various biological data for the pepper weevil, Anthonomus
eugenii Cano (Coleoptera: Curculionidae) to 1989 .................. 5

3.1 Effect of time of day on the average numbers of pepper weevil adults per plant
part in exposed versus unexposed positions. ................... ... 35

3.2 Temperature data for three sizes of bell pepper plants for exposed versus
unexposed buds in the top third and bottom two-thirds of the plant .... 36

3.3 Mean numbers of pepper weevil adults on exposed versus unexposed fruiting
buds in the top third and bottom two-thirds of the plant ................ 37

3.4 Comparison of three visual techniques for sampling pepper weevil adults ...... 38

3.5 The diurnal effect on the numbers of pepper weevil adults for terminal bud
inspection, whole plant inspection, and an absolute sample . .... 39

3.6 Comparison of four methods for sampling pepper weevil adults ............ 40

3.7 Comparison of methods for sampling pepper weevil adults on an individual
plant basis . . . .. . .. 4 1

3.8 Numbers of pepper weevil adults captured on cylindrical, colored poster board
traps with a yellow vinyl plastic interior ........................ 42

3.9 The effect of time of day on the numbers of pepper weevil adults captured on
sticky traps ... . . . . ...... 43

3.10 Numbers of pepper weevil adults captured at different trap heights on
cylindrical yellow vinyl sticky traps. ............................ 44

3.11 Numbers of pepper weevil adults captured on five sizes of yellow vinyl sticky
traps. ......... .... ......... .......... ............. 45

3.12 Numbers of pepper weevil adults captured on sticky traps constructed of
colored polyethylene plastic cylinders ..... . ........ 46

3.13 Numbers and sex ratios of pepper weevil adults in bell pepper collected using
terminal bud inspection or captured using yellow sticky traps and white
sticky traps ............................... ...... .... 47









Table page

3.14 Numbers and sex ratios of pepper weevil adults captured in bell pepper on
yellow and white sticky traps ........................... ..48

3.15 Comparison of the numbers and sex of pepper weevil adults observed per
plant in an absolute sample, terminal bud inspection, or captured per yellow
or white sticky traps ..................................... 49

3.16 Impact of time and plant part on spatial dispersion estimates of pepper weevil
ad u lts . . . . . 7 1

3.17 Summary statistics for the spatial dispersion of pepper weevil adults by crop
and bud position ................... .................. 72

3.18 Summary statistics for the spatial dispersion of pepper weevil adults for bell
pepper using three visual sampling methods . . ... 73

3.19 Summary statistics for the spatial dispersion of pepper weevil adults in bell
pepper and Jalapeiio pepper ................................ 74

3.20 Means, variances, and variance/mean ratio for pepper weevil counts by field
block for bell pepper at Green Cay farm and the Baum farm and for
Jalapefio peppers at Hayes farm ................... ........... 75

4.1 Effect of calendar and threshold application regimes for three insecticides on
the numbers of pepper weevil adults and immatures in bell peppers.
Honduras 1988 ................... ................... ..89

4.2 Effect of calendar and threshold application regimes for three insecticides on
arthropod fauna in bell pepper. Honduras 1988 . ... ........ 90

4.3 The effect of application regimes and insecticide treatments for pepper weevil
control on numbers of arthropods in Jalapeiio pepper with treatments
arranged in a split plot design. Honduras 1988 . .. 91

4.4 The effect of application regimes and insecticide treatments for pepper weevil
control on yields of Jalapefio pepper with treatments arranged in a split plot
design. Honduras 1988 ................... ...... ........ 92

4.5 The effect of insecticide treatments on the production of undamaged fruit and
pepper weevil damaged fruit in bell pepper and Jalapefio pepper ......... .107

4.6 The effect of adult pepper weevil threshold treatments on the production of
undamaged fruit and pepper weevil damaged fruit in bell pepper.
Bradenton, Florida 1988 ................... .............. 108

4.7 Effect of a pepper weevil adult action threshold on the yield of undamaged fruit
and pepper weevil damaged fallen fruit for individual harvests of bell
pepper. Bradenton, Florida ............................... 109









Table pag

4.8 The effect of pepper weevil adult thresholds on the production of undamaged
fruit and fruit damaged by pepper weevil in Jalapeio pepper. . ... 110

4.9 Effect of a pepper weevil adult action threshold on the yield of undamaged fruit
and PW damaged fallen fruit for individual Jalapefio pepper harvests.
Bradenton, Florida 1988 ................. ...... ........ 111

4.10 Total numbers of insecticide applications in bell and Jalapefo peppers
managed under different threshold treatments . .... ........ 112

4.11 Budget for bell peppers comparing two morning thresholds and a calendar
spray schedule to a control overall and for permethrin only. ............ 113

4.12 Numbers of fallen pepper buds, pepper weevil immatures, and pepper weevil
parasites, Catolaccus hunter Crawford, in fallen buds for 5 different
sampling dates .............. .... ... ................. 127

4.13 Pepper production and insect damage per plant in bell pepper sprayed with
either oxamyl or permethrin .............................. 128

4.14 The effect of insecticide action thresholds on the production of undamaged
bell pepper fruit and the incidence of abscised fruit due to pepper weevil
dam age ................. ........................... 129

4.15 Harvested fruit with external pepper weevil damage and internal pepper
weevil damage for three harvests on bell pepper.................... 130

4.16 The interaction of insecticide and threshold treatments in bell pepper ........ 131

4.17 Budget for bell pepper production under two action thresholds and a calendar
spray schedule for pepper weevil control overall and in oxamyl and
permethrin whole plots .......... ....................... 132

A. Numbers of PW adults per five distinct plant parts from a two minute whole
plant inspection of individual bell pepper plants taken six times per day,
four days per week, for four weeks. Bradenton, Florida May 1987 ....... 150

B. Marked pepper weevil adult counts in field block B-l at the Gulf Coast
Research and Education Center, Bradenton .................... 167

C. Taylor's Power Law coefficients and variance to mean ratios for terminal bud
inspection and inspection of fallen buds in bell pepper ............... 171















LIST OF FIGURES

Figure page

2.1 The life cycle of the pepper weevil, Anthonomus eugenii Cano (Coleoptera:
Curculionidae), constructed from observations by Elmore et al. (1934) and
Wilson (1986) and illustrated by E. Vasquez ...................... 8

3.1 The numbers of pepper weevil adults observed on five bell pepper plant parts
at six times during the day ................................ 50

3.2 The numbers of pepper weevil adults observed on five plant parts of bell
pepper for morning and afternoon samples by week ................. 51

3.3 The impact of time of day on the numbers of pepper weevil adults observed on
exposed and unexposed fruiting buds .......................... 52

3.4 The numbers of fruiting pepper buds and pepper weevil adults observed in
exposed and unexposed portions of the plant in the top third and bottom
two-thirds of bell pepper plants. Bradenton, Florida spring 1988. ......... 53

3.5 The numbers of fruiting pepper buds and pepper weevil adults observed in
exposed and unexposed portions of the plant in the top third and bottom
two-thirds of Jalapeiio pepper plants. Bradenton, Florida spring 1988. ..... 54

3.6 The proportion of fruiting buds and proportion of pepper weevil adults
between exposed and unexposed portions of the plant in the top third and
bottom two-thirds of bell pepper plants. Bradenton, Florida spring 1988. .... 55

3.7 The proportion of fruiting buds and proportion of pepper weevil adults
between exposed and unexposed portions of the plant in the top third and
bottom two-thirds of Jalapeiio pepper plants. Bradenton, Florida spring
1988...... .. ..... ...... ................ ........ ... 56

3.8 The numbers of pepper weevil adults observed per top exposed fruiting bud
for bell and Jalapefio pepper plants. Bradenton, Florida spring 1988. ...... 57

3.9 The numbers of pepper weevil adult observed by five techniques including
terminal bud inspection, inspection of two exposed terminal buds plus two
unexposed buds per plant, whole plant inspection, an absolute sample, and
an intensive yellow sticky trap sample in bell and Jalapeiio peppers.
Bradenton, Florida spring 1988 ..............................58









Figureag

3.10 Reflectance analysis for materials used to construct sticky traps at Bradenton,
Florida.. ................... .. ............. .. ....... 59

3.11 Field maps indicating blocks sampled for pepper weevil adults at Green Cay
Farm, Hayes Farm, and the Baum Farm in 1989. . ..... 76

3.12 The numbers of pepper weevil adults per sample site in bell pepper at 8 weeks
and 12 weeks after transplant. .............................. 78

3.13 The numbers of pepper weevil adults per sample site at 6 weeks and 9 weeks
after transplant in bell pepper .................... ........ 79

3.14 The mean numbers of pepper weevil adults across samples and across rows at
Green Cay Farm. Palm Beach Co. spring 1989. .................... 80

3.15 The mean numbers of pepper weevil adults across samples and across rows at
Hayes Farm. Palm Beach Co. spring 1989. ...................... 81

4.1 Mean numbers of arthropods over all treatments observed using whole plant
inspection in bell pepper. Honduras 1988 ....................... 93

4.2 Average numbers of pepper weevil adults observed using whole plant
inspection of 14 bell pepper plants under different insecticide application
regimes for pepper weevil control. Honduras 1988 .................. 94

4.3 Average numbers of Lepidoptera larvae observed using whole plant inspection
of 14 bell pepper plants under different insecticide application regimes for
pepper weevil control. Honduras 1988 ........................ 95

4.4 Average numbers of plants infested with aphids and numbers of predators
observed in 14 bell pepper plants under different insecticide applications
regimes for pepper weevil control. Honduras 1988 .................. 96

4.5 Numbers of pepper weevil immatures and pepper weevil parasites in treated
and untreated bell pepper plots. Bradenton, Florida fall 1988. .......... 114

4.6 Numbers of pepper weevil immatures and pepper weevil parasites in treated
and untreated Jalapeiio plots. Bradenton, Florida fall 1988. ............ 115

4.7 Numbers of pepper weevil adults per control, the morning threshold of 1
pepper weevil adult/200 terminal buds, and the calendar insecticide
treatment in bell peppers. Bradenton, Florida fall 1988. .............. 116

4.8 Numbers of pepper weevil adults per control, the morning threshold of 1
pepper weevil adult/100 terminal buds, and the calendar insecticide
treatment in Jalapefio peppers at Bradenton, Florida during the fall of 1988. .. 117









p=ge


4.9 Numbers of pepper weevil adults observed in bell pepper managed using either
oxamyl or permethrin applied weekly or when a threshold of 1 pepper
weevil per 200 or 400 terminal buds were exceeded. ................ 133

4.10 Numbers of pepper weevil adults in an untreated field compared to the
numbers in experimental bell pepper plots treated with either oxamyl or
permethrin.......................................... 134

4.11 Sequential sampling plan for pepper weevil adults based on the assumptions
that 1 adult/100 plants (200 terminals) causes damage and 1 adult/200 plants
(400 terminals) does not cause significant damage. ................. 139










Abstract of the Dissertation Presented to the Graduate
School of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of Doctor of Philosophy


REFINED SAMPLING METHODOLOGY AND ACTION THRESHOLDS
FOR THE PEPPER WEEVIL, ANTHONOMUS EUGENII CANO
(COLEOPTERA: CURCULIONIDAE)

By

David Gerard Riley

May 1990

Chairman: D.J. Schuster
Cochairman: C.S. Barfield
Major Department: Entomology and Nematology


Sampling studies were conducted to assess pepper weevil (PW) adult numbers on

various pepper plant structures, to compare visual sampling methods, and to characterize

adult dispersion on a small- and a large-field basis. More PW adults were observed on

terminal fruiting buds than on other plant structures, but numbers varied diurnally on the

exposed portions of the plant. The number of PW adults on exposed terminal buds was

significantly higher in the morning than the afternoon. A comparison of terminal bud

inspection, whole plant inspection, and an absolute sample for PW adult counts indicated

that terminal bud inspection required the least amount of search time and provided higher

PW counts per unit time than the more labor-intensive whole plant inspection. Initial tests,

using colored sticky traps as an alternative sampling method, indicated that yellow traps

were comparable in efficiency (weevils found per unit search time) to terminal bud

inspection. Also, sticky traps detected proportionally more PW females than were detected

with the plant inspection methods. Further development of traps is needed before the

technique can be adopted. Therefore, terminal bud inspection was determined to be the

more acceptable relative sampling method for PW adults currently available.








Dispersion estimates suggested that PW adults were dispersed randomly in small

pepper fields (0.2 Ha) to slightly clumped in larger, commercial size fields (0.9 to 58 Ha).

A concentration of PW adults in the outer field margins was observed at three distinct farm

sites. Taylor's Power Law coefficients did not vary significantly over time of day, but the

slope of the Log mean by Log variance regression was significantly less for terminal bud

inspection than for whole plant inspection or the absolute adult sample.

An action threshold, based on inspection of two terminal buds per plant, was tested

using oxamyl, cypermethrin, and methomyl at the Escuela Agricola Panamericana in

Honduras in 1988. Results indicated no significant difference between PW adult counts in

plots treated weekly versus those treated by the threshold. In Honduras, oxamyl provided

the best control of PW adults and permitted parasitism of PW immatures by Catolaccus

hunter Crawford (Hymenoptera: Pteromalidae). Subsequent experiments were conducted

in Florida to evaluate the effect of diurnal sample variation on the efficacy of selected

thresholds based upon terminal bud inspection. Pepper production under the action

threshold of 1 adult/200 terminal buds was not significantly different from peppers treated

weekly with permethrin. Scouting in the afternoon rather than the morning did not

significantly decrease the efficacy of the threshold. In the spring of 1989, under a heavier

PW infestation than in the previous season, marketable yield of plots treated with

permethrin, based upon a threshold of 1 adult/400 buds, was not significantly different

from plots treated weekly.

PW adults can be sampled effectively using terminal bud inspection, and effects due

to time of day do not significantly alter the efficacy of a threshold based on this sampling

method. In bell pepper, an action threshold (1 adult/400 terminal buds) lower than

previously established (1 adult/100 terminal buds) was needed to achieve pepper

production similar compared to weekly insecticide treatment under a heavy PW infestation.

















CHAPTER 1
INTRODUCTION


The pepper weevil (PW), Anthonomus eugenii Cano, is an important pest of bell-

type peppers and Jalapeiio, Capsicum spp., peppers in the southern U.S.A., Mexico,

Central America, Hawaii and several Caribbean islands (Burke and Woodruff 1980,

Elmore et al. 1934, O'Brien and Wibmer 1982). The economic importance of PW as a pest

of peppers began to be realized in the USA by Walker (1905), who reported a 33%

commercial crop loss in two consecutive years; by Campbell (1924), who found as much

as 100% infestation of the young pepper fruit in commercial fields; and by Elmore et al.

(1934), who reported yield losses as much as 50%. In Florida, all of the pepper fields

inspected in Manatee Co. in 1935 were infested with PW (Watson 1935) and, more

recently, 100% of the fallen fruit sampled were infested with PW in several commercial

fields (Genung and Ozaki 1972).

Ecological insight into PW dynamics and pest status has been limited by lack of a

robust sampling methodology. Inability to sample any PW life stage with known precision

or accuracy has precluded the development and use of reliable action thresholds. Damage-

infestation relationships in the current literature indicate that a PW population density

exceeding 1 PW/100 terminal pepper buds or 1 PW/10 whole pepper plants can cause

economic loss in high-yielding bell peppers (Andrews et al. 1986, Cartwright et al. 1990,

Segarra-Carmona and Pantoja 1988b). In the absence of a reliable and economical sampling

method to detect these population densities, growers have resorted to the classical

"calendar" spraying regime; thus, PW control programs have, by and large, tracked
1










analogous programs against a close relative, the boll weevil, Anthonomus randis

Boheman. The majority of literature on PW control is associated with but one tactic --

chemical control (see Cartwright et al. 1986, Genung and Ozaki 1972, Schuster 1984,

Schuster and Everett 1982). Very little definitive information exists on alternative

management tactics or on general PW biology and ecology, including natural mortality

factors which ultimately may prove useful.

Obviously, a great deal more information on PW ecology is needed before robust

pest management strategies, which avoid the pitfalls of calendar spraying, can be

developed. Development and evaluation of virtually all management options demand access

to a reliable and economical sampling methodology for PW. Thus, the goal of this study

was to pursue the development of a reliable and economical sampling plan for PW for

incorporation into an action threshold. Present work began with consideration of PW

dispersion pattern; however, the ultimate PW sampling plan must consider dispersion

pattern, sample unit size, number of samples to be taken and the allocation of those samples

in various sized fields.

The present study was conducted under the following two principle objectives and

the sub-objectives indicated for each: TO:

1. Quantify PW dispersion pattern

a. evaluate visual sampling techniques for their accuracy, precision, and

ease of use in PW field scouting programs

b. quantify significant changes in PW dispersion patterns as functions of

field size, pepper type, diurnal (daytime) variation and seasonal variation

c. at the same time, measure the dynamics of PW in-field adults densities,

especially as measurable with color sticky traps











2. Validate a previously developed action threshold for PW

a. evaluate efficacies of sampling methods developed in objective 1

b. validate and/or calibrate the action threshold for Florida peppers, based

on the existing threshold of 1 PW/100 terminal buds (Andrews et al. 1986)

c. based on objective 1; parts a and b, objective 2; the literature; and

extensive field observations, recommend a PW pest management strategy.

Chapter 2 presents a review of PW literature, including historical and modem

documentation of PW bionomics, sampling plans and proposed management strategies.

Chapter 3 deals with present sampling methods and is especially relevant to PW dispersion

pattern and adult sampling with sticky traps. Efforts to validate and/or refine an existing

action threshold for PW is presented in Chapter 4, and an extensive summarization of

present work follows in a SUMMARY AND CONCLUSIONS chapter. To assist future

investigators, three appendices have been added and contain volumes of raw data on PW

field counts (A), mark-recapture data (B), and a dispersion estimate of PW immatures (C).














CHAPTER 2
REVIEW OF LITERATURE

2.1 Pepper Weevil Biology and Bionomics

The PW shares many of the same characteristics of other members of the genus

Anthonomus. Adults are oligophagous and females oviposit on flower buds or fruit.

Immatures complete development in the immature bud or fruit, causing premature bud and

fruit abscision. Like the boll weevil, Anthonomus grandis Boheman, the PW has 3 larval

instars and multiple generations per year (Burke 1976). Unlike A. grandis, diapause has

not been detected in the PW. If facultative diapause does not occur for the PW, then the

northern distribution of PW in the United States should be limited to areas that can support

PW host material throughout the winter (Elmore and Campbell 1954) or areas that are

subject to periodic re-infestation (Boswell et al. 1964, Stacey and Adams 1982). The

similarity of the life histories of PW and the boll weevil suggests that much of the

voluminous literature existing for the boll weevil can be used to guide research on PW

biology and population ecology.

General information on the biology of the PW was reviewed by Wilson (1986).

These data and additional information are summarized in Table 2.1. The largest voids in

PW literature are in the areas of developmental biology and general ecology, particularly

PW behavior and spatial dispersion. Reported generation times for the PW varies widely

among different investigators with perhaps the largest differences occurring between

summer and fall observations. According to Elmore et al. (1934), generation time and the

number of generations per year are determined primarily by host availability and

temperature. Wilson (1986) reviewed the principle literature on PW host plants and











Table 2.1 Summary of various biological data for the pepper weevil, Anthonomus eugenii
Cano (Coleoptera: Curculionidae) to 1989.


Biological Parameter
1. generation time
"1 "
If "


2. generations/year
3. longevity of adults
(without food)
"(without food)
(without food)
4. oviposition period

5. fecundity
t!

6. oviposition rate



7. biotic mortality factors
a. predators


b. parasites





8. abiotic mortality factors
a. temperature

b. physical

9. host plants
(1st references)


10. feeding and
development


Characteristic
ca. 22 to 28 days
16 to 23 days summer
22 to 46 days fall
ca. 13 to 17 days
25 to 40 days
13 to 22+ days
16 to 19 days
ca. 15 to 20 days
11 to 21 days
5 to 8 generations
avg. 78.7 days (to 316)
ca. 90 days
22 to 39 days
avg. 6.8 days
16 to 129 days
avg. 30 days
avg. 341 eggs/adult
(range 28 to 634 eggs)
avg. 198 eggs/adult
avg. 4.7 eggs/day
avg. 6.6 eggs/day
avg. 6 eggs/day
avg. 7.1 eggs/day

Solenopsis geminata
Strunella mana
Tetramorium guineese
Pyometes venticosis
Catolaccus incertus
Pediculoides ventricosus
Bracon mellitor
Habrocytus Diercei
Zatrolis incertus
Catolaccus hunterii

northern distribution
winter mortality of PW
fallen fruit destruction

peppers
Solanum rostratum
Solanum irum
Capsicum and Solanum spp.


Literature Cited
Walker 1905
Elmore et al. 1934

Goff and Wilson 1937
Berry 1959
Toba et al. 1969
Genung and Ozaki 1972
King and Saunders 1984
Wilson 1986
Elmore et al. 1934
Elmore et al. 1934
Goff and Wilson 1937
Elmore et al. 1934
Goff and Wilson 1937
Elmore et al. 1934
Goff and Wilson 1937
Elmore et al. 1934
It* t It
Goff and Wilson 1937
Elmore et al. 1934
Goff and Wilson 1937
Genung and Ozaki 1972
Wilson 1986

Pratt 1907
Genung and Ozaki 1972
Wilson 1986
Meraz 1905
Pierce 1907
Elmore et al. 1934
Chestnut and Cross 1971


Wilson 1986

Elmore and Campbell 1954
Boswell et al. 1964
Pratt 1907
Andrews et al. 1986
Cano 1894
Pierce 1907
Merrill 1929
Elmore et al. 1934


on various host plants


"t 11


Wilson 1986










investigated the suitability of several host plants for PW feeding, oviposition, and

reproduction. He found that the PW was able to complete its larval development in the fruit

and/or flowers of Solanum americanum Mill., S. pseudogracile Heiser, and S. carolinense

L., all of which are found in Florida.

Developmental biology: No specific information currently exists which relates

temperature, in terms of degree days and/or relative humidity, to PW development time.

Temperature data associated with development time have been reported only in three of the

studies listed in Table 2.1. Toba et al. (1969) reported emergence on artificial diet in 17 to

18 days at 25.7-27.7 oC and 70% relative humidity (RH). Genung and Ozaki (1972)

estimated the average generation time of 17.5 days at 23.9-26.7 OC and 60-85% RH.

Wilson (1986) estimated a mean generation time on bell pepper of 14.2 days at 25.7-27.7

oC and 40-100% RH.

In addition to the lack of adequate temperature and humidity data associated with

PW development, there is a need for greater quantification of nutritional effects on adult

PW biomass and fitness. Wilson (1986) demonstrated indirectly that host material may

affect the individual weights of emerging PW adults. Toba et al. (1969) suggested that

extreme ranges in diet quality also may affect the development time as well as biomass of

PW adults. It has long been known for the cotton boll weevil that the quantity and quality

of diet affects the weight of adults (Hunter and Hinds 1905).

The fruiting structure size, the presence of previous PW oviposition or feeding, and

the availability of sufficient host material for the completion of larval development seem to

be important factors in the selection of oviposition sites. The quality of host material was

shown to affect ovipositional preference (Wilson 1986). Bruton et al. (1989) presented data

suggesting that developing fruit (1.31-5.0 cm in diameter) were preferred over smaller fruit

(<1.31 cm in diameter) and larger mature fruit (>5.0 cm in diameter) for oviposition based











on numbers of punctures. A review of the boll weevil literature clearly indicates greater

oviposition activity during the day and that females seemed to avoid substrates previously

used for oviposition (Mitchell and Cross 1969). With these considerations, it would seem

that the frequency of PW ovipositions per fruit is a factor of time, PW population density

and pepper fruit size. Varietal differences in peppers may affect host preference, but these

were thought to be negligible effects under heavy feeding and oviposition pressure

(Genung and Ozaki 1972). For a comparative review of boll weevil literature concerning

developmental biology and ecological fitness (accumulated largely through the development

of mass rearing and sterilization techniques), see Sikorowski et al. (1984).

PW mating behavior and the dynamics of pheromonal sex attraction should be the

next major thrusts in PW research. This is because PW populations, like boll weevil

populations, eventually may be monitored or manipulated by the use of pheromones. In a

preliminary study of mating dynamics, Wilson (1986) found that males attract females over

long distances and that males are attracted to females at close distances. He reported 11-34

minute copulation and that mating occurred at all periods of the day. However, no

observation was made on the frequency of mating.

General ecology: Ecological investigations for the PW at the population and

community levels do not currently exist; however, studies, such as Guitierrez et al. (1979)

for the boll weevil, provide insight into weevil co-adaptation with host plants and suggest

the precursor biological information needed for a thorough examination of PW ecology.

Ecological studies may have a direct impact on PW management by augmenting the

vulnerability of the PW at critical times during its life cycle (Figure 2.1). This is particularly

true with respect to the impact of parasitoids and predators on PW population dynamics.

Although predators have been recorded for PW immatures as early as 1907 (Table 2.1),

this was only for those immatures remaining in the fallen fruit after the fruit wall had been

perforated with an exit hole. Parasitoids, on the other hand, have been found in PW











infested Solanum spp. fruit which were not perforated (Pierce 1907, Wilson 1986).

Therefore, parasitoid attack may be the only biological control tactic available at the early

stages of PW development in the fallen fruit. Two of the reasons why this information has

not been incorporated into a PW management strategy is that critical data on the level of

parasitism possible in commercial settings have not been generated for the PW and the

compatibility of this control tactic with chemical controls for the pepper pest complex has

not been assessed. Therefore, more emphasis is needed in ecological studies if integration

of pest control tactics is to be accomplished.


Figure 2.1 The life cycle of the pepper weevil, Anthonomus eugenii Cano (Coleoptera:
Curculionidae), constructed from observations by Elmore et al. (1934) and Wilson (1986)
and illustrated by E. Vasquez.











Another aspect of PW ecology which merits investigation is dispersal. This is

particularly true since the movement of PW adults on pepper plants may affect sampling

when visual inspection methods are employed (Andrews et al. 1986). Movement to and

from secondary host material and cull sites has been indicated as one of the primary causes

for periodic re-infestation in commercial pepper fields (Elmore et al. 1934, Genung and

Ozaki 1972, Wilson 1986). Specific data regarding within field or between field movement

of PW have not been reported. Until the 1960s, movement and dispersal data were lacking

for the much more intensively studied boll weevil (Beckham and Morgan 1960). Most of

the early observations on boll weevil movement assumed that unmarked weevil populations

could be monitored for dispersal, since the boll weevil primarily reproduced on cotton.

Assuming an absence of stray cotton plants between the source of boll weevil infestation

and the target planting of cotton, boll weevil dispersal to isolated fields could be monitored.

The pepper weevil, on the other hand, has been reported to reproduce on nightshade,

Solanum spp. Solanaceae (Anon. 1928, Genung and Ozaki 1972, Merrill 1929, Wilson

1986). Since nightshade can be ubiquitous in many regions where pepper is grown, the

assumptions on movement of weevils would be largely invalid. The mark-recapture method

described by Johnson et al. (1975) for boll weevil could serve as a valuable tool to future

studies for PW dispersal on a regional basis.

Of course, the foundation of most ecological field studies is a reliable sampling

plan. From a practical standpoint, sampling is also the single most important factor for

implementing an effective integrated pest management program. Dispersion of pest species

has a direct impact on the cost and precision of sampling. In the next section these factors

are reviewed as they pertain to the PW and related species.










2.2 Sampling and Spatial Dispersion of the Pepper
Weevil and Other Anthonomii


There have been relatively few investigations of sampling techniques for the PW.

Early attempts at scouting PW populations were simply presence-absence determinations

and, in many cases, the presence of fallen fruit was the first indication of a PW infestation

(Elmore et al. 1934, Watson 1935). Boswell et al. (1964) recommended the examination of

fallen buds for presence of PW immatures early in the season as a means of determining the

presence of PW in the field. No reference was made to methods for evaluating infestation

levels.

Methods for estimating relative PW population densities recently have been

reported. Andrews et al. (1986) found that counting fallen fruit or using a beat cloth were

unacceptable sampling methods for timing PW control. They suggested that inspecting

terminal buds or bud clusters for PW adults was effective in predicting subsequent

infestations. Segarra-Carmona and Pantoja (1988a) found that detergent-water pans were

ineffective for collecting PW adults and that using a beat cloth or sweep netting were

excessively destructive to the pepper plant. They also found that yellow sticky traps

correlated well with direct adult PW counts, using whole plant inspections, and suggested

the use of traps as an alternative to the more intensive and costly direct count method.

Cartwright et al. (1990) proposed the use of PW feeding or ovipositional damage to

terminal bud clusters as a method for determining the presence of PW in commercial pepper

fields. Wilson (1986) indicated that boll weevil traps baited with PW males compared

favorably with terminal inspections, but no economic analysis was included to support the

use of traps as a viable alternative to visual inspection. In the boll weevil literature, male-

baited traps have been shown to be effective in detecting the time and degree of boll weevil

dispersal (Hardee et al. 1970).











The dispersion of PW has not been investigated sufficiently. The only dispersion

index developed for PW adults indicated that PW adult populations exhibited a moderately

clumped dispersion in the field with a negative binomial k = 2.5 (Segarra-Carmona and

Pantoja 1988b). No aspects of clumping patterns, border effects, or temporal effects were

evaluated.

Sampling plans for the boll weevil have been investigated much more extensively

due to the insect's greater economic impact. Early in the development of boll weevil

sampling methods, the point method (sampling in a grid) was found to provide similar

estimates of infestation levels as a more intensive survey method (a circular systematic

sample) (Gaines 1933). Grossman (1931) reported that in small cotton fields relatively few

plants had to be sampled to obtain an acceptable estimate of the population density (20

plants/0.25 acre). He noted that in large fields the weevil population tended to be clumped

so that small samples would tend to be inaccurate regardless of the sample method

employed. Sequential sampling methods, which are designed specifically to reduce the

number of samples needed for applicable density estimates from large populations, were

not developed for the boll weevil until the 1970s (based on Pieters 1978). The sequential

method has not been accepted widely perhaps because scouts tend to prefer fixed sample

sizes rather than the variable sample size associated with sequential sampling methods

(Winfield Sterling 1989, personal communication). Wolfenbarger (1977) found the grid

system and the X system of sampling (two variations of a systematic sampling plan) were

more effective than the area system (a random sampling plan) for scouting boll weevils

(described by Dupnik et al. 1973). Currently accepted plant inspection methods for

detecting boll weevil infestations rely on the presence of damage and the use of pheromone

traps more than on counts of adults (Bob Cartwright 1989 personal communication,

Winfield Sterling 1989 personal communication).











Boll weevil eggs tend to exhibit a more clumped pattern of dispersion in the field

than do adults (negative binomial k's = 0.16 and 0.58 respectively), which may be a result

of adult mortality and mobility (Pieters and Sterling 1974). In other words, the location of

oviposition sites remains stationary over a certain period, resulting in an accumulation of

oviposition events. Resulting samples of eggs should produce a greater range of counts

than that of the adults, because the adults are more transient. Kuehl and Fye (1971) noted

that boll weevil adult counts often fit the Poisson distribution; however, the index used

(variance/mean) may have erroneously indicated a Poisson distribution when population

levels were low. A more rigorous evaluation of dispersion patterns by Pieters and Sterling

(1973a) indicated that the negative binomial distribution best described adult counts,

feeding damage, and oviposition sampling data. Leggett (1984) found that there was a

greater percent damage to cotton squares on the border of cotton fields than within the field.

Recent studies, dealing specifically with dispersion of the boll weevil, do not

address some possibly important effects on dispersion estimates. The sampling method is

known to affect the dispersion estimate (Elliott 1983). Temporal effects on Taylor's Power

Law coefficients are not generally considered significant (Taylor 1984). However,

temporal behavior patterns, which can affect the apparent clumping of individuals, should

affect the dispersion estimate. The emphasis on boll weevil trapping (Cross 1973), as

opposed to extensive use of plant inspection methods, may explain the lack of more

detailed work on the dispersion of boll weevil adults on cotton plants.

Morris (1960) and Southwood (1978) reviewed much of the previous literature on

sampling insect populations and provided insights into the mechanics of data collection,

determination of population classes, spatial distribution, design of sampling programs, and

the interpretation of sample data. The predominance of information for edaphic or arboreal

arthropods in the early literature could be seen in these reviews. Kogan and Herzog's

Sampling Methods in Soybean Entomology (1980) is a perceptive review which










emphasizes the greater dynamics in sampling an annual crop. Concepts addressed in this

review of particular interest to researchers of vegetable crops include the relationship

between crop phenology and the impact of pest damage, temporal sample variation, and

spatial dispersion of insect pests. Statistical analyses for invertebrate samples and

dispersion data are summarized by Elliott (1983). General considerations on the effect of

bias, stratification, systemizing, and optimal allocation of samples can be found in many

statistical references (Steele and Torrie 1980, Cochran 1953, Hansen et al. 1953). Specific

examples of the applications of sequential sampling are numerous (Pieters 1978, Onsager

1976, Morris 1954, Oakland 1950). A complete review on the development and application

of Taylor's Power Law as a spatial dispersion index is given by Taylor (1984).

Just as sampling studies are the basis for any pest management program, so too is

the availability of effective pest control tactics. The following section deals primarily with

chemical control of the PW, but also reviews related usage of biological and cultural

controls. By and large, the more common form of PW control has been through the use of

insecticides.

2.3 Control of the Pepper Weevil

"Management guides" that are available to growers are often simply intelligent use

guides for pesticide applications (e.g., Cotton Management Guide 1989). Pesticides play

an invaluable role in the control of immediate pest problems, but the limitations of this tactic

can be dramatically seen in the history of cotton production (Walker 1987).

Early in the development of PW control in peppers, the use of pesticides was not

emphasized due to the problem of toxic residues of inorganic insecticides on the fruit

(Elmore et al. 1934, Merrill 1929, Walker 1905). Even so, calcium arsenate was

recommended on peppers before fruit set (Goff 1937, Goff and Wilson 1936, Watson

1935). Intensive dusting with inorganic insecticides, followed by washing of the harvested

fruit, was also recommended (Campbell and Elmore 1942, Elmore 1933). Before the











advent of organic insecticides, crop sanitation was reported as one of the more effective

cultural control methods (Wilson and Watson 1938, Watson and Lobdell 1939). Sanitation

was sometimes recommended in conjunction with inorganic insecticides (Lockwood 1937).

By the 1950s, several commercial organic insecticides were available. These new

insecticides provided good control of PW adults without the problem of excessive residues

on the harvested fruit (Elmore and Campbell 1954). Plant extracts, apparently nontoxic to

mammals and effective in PW control, were reported by Beroza and Bottger (1954). Since

the 1970s, numerous trials on insecticide efficacy have been conducted (Cartwright et al.

1986, Mendoza 1984, Ozaki and Genung 1982, Rolston 1977, Rolston et al. 1977,

Schuster 1983, 1984, 1986, 1987, Schuster and Everett 1982). Some data exists which

suggest that the PW has multifunction oxidases which aid in insecticide detoxification

(synergistic ratio = 5.4, Brattsten and Metcalf 1970).

The availability and low cost of effective insecticides has not been conducive to the

development of action thresholds for the PW. Even so, there has been a recent interest in

the use of thresholds to minimize control costs and to promote a more integrated approach

to insect control in peppers (Andrews et al. 1986, Cartwright et al. 1989, Segarra-

Carmona and Pantoja 1988b). Recommendations on the use of thresholds for other pepper

pests in conjunction with a PW threshold have been proposed (Schuster and Shuler 1987).

Considerations that must now be incorporated into these recommendations are 1) the

relationship between PW and internal mold of bell pepper (Bruton et al. 1989); 2) the

occurrence of new and possibly more devastating pests in the pepper agro-ecosystem

(Pantoja et al. 1988, Oi and Mau 1988); and 3) the use of alternative control tactics such as

parasitism for early season PW control (Chapter 4). The use of crop residue destruction,

transplant sanitation, and nightshade management have been recommended as effective

cultural controls of PW (Boswell et al. 1964, Stacey and Adams 1982, Schuster and Shuler

1987). Even so, caution should be taken with the timing of destruction of host material











since the lack of oviposition sites may trigger the migration of weevils (Fye and Bonham

1970).

PW immatures in the fallen bud or fruit are protected physically from many external

mortality factors by the fruit pericarp (Figure 2.1). Insecticide induced mortality of PW

immatures, even by systemic compounds, has not been reported to date and predation of

immatures has only been noted in fallen fruit with PW exit holes (Pratt 1907, Wilson

1986). Hymenopterous parasitoids, desiccation of fallen buds, and the destruction of fallen

fruit through normal cultural practices are probably the three most important mortality

factors for PW immatures in commercial fields other than insecticides. Mortality due to

fallen fruit desiccation has been well documented for the boll weevil (Curry et al. 1982).

The most important natural mortality factors for the adult PW, other than insecticides,

appears to be related to host quality and availability (Boswell et al. 1964, Elmore et al.

1934, Goff and Wilson 1937, Wilson 1986). This type of biological information has lead

to the cultural practice of eliminating pepper culls and secondary host plants. This has been

cited as the most effective method of reducing PW populations other than the use of

pesticides (Genung and Ozaki 1972).

A strictly biological or cultural approach to PW control generally is not considered

as a viable alternative to the current dependency on insecticides. There has been at least one

case where exclusive use of strictly biological control of pepper insect pests (Bacillus

thuringiensis for noctuid larvae and no sprays for PW) was attempted in Palm Beach Co.,

Florida (1986) without success (Ted Winsberg 1988, personal communication). A gradual

shift to integrated control of the PW, with more emphasis on alternative control tactics for

PW, may receive more general acceptance among growers.

2.4 Major Research Needs for the Management of Pepper Weevil

The available literature specific to PW is insufficient for the development of a robust

management strategy for this pest. In particular, development rates for immatures needs to










be calibrated with a range of temperature and humidity regimes before emergence times can

be accurately predicted. This is information is necessary for the interpretation of

ovipositional sampling data, such as used for damage-based thresholds (Cartwright et al.

1990), and is critical for the timing of insecticide applications based on PW immature

samples in fallen bud and fruit. Also, the dynamics of PW mating behavior, response to

color and pheromone production will need clarification, if effective trapping for adults is to

be realized. Sampling is probable the most crucial area of study needed for the management

of PW populations. This is especially true for PW adults, since it is only the adults that are

immediately controlled by insecticide applications. Intra- and inter-plant dispersion

estimates for PW adults are essential for the development of a cost-effective and reliable

sampling plan for PW adults.

In the present study, I have chosen to focus on the sampling and dispersion studies

for PW adults since this information is critical for the development of adult-responsive

action thresholds. The following Section 3.1 deals primarily with intra-plant dispersion of

adults and investigates PW response to color. Section 3.2 treats inter-plant dispersion of

PW adults in small experimental plots and in commercial-size pepper fields. To further the

investigation of the adult-based sampling plan, action thresholds utilizing terminal pepper

bud inspection for PW adults were validated and refined in Chapter 4.















CHAPTER 3
SAMPLING AND SPATIAL DISPERSION OF THE PEPPER WEEVIL


3.1 Sampling Techniques for the Pepper Weevil.
Anthonomus eugenii Cano (Coleoptera: Curculionidae)


Research on PW has concentrated on biology and control (Elmore et al. 1934,

Wilson 1986), and only recently has sampling for PW been investigated more intensively.

Andrews et al. (1986) proposed a cost-effective PW adult sampling method to be used with

an action threshold. Their method utilized terminal bud inspection; however, this sampling

method was subject to diurnal sample variation. Whole plant inspection for PW adults

(Segarra-Carmona and Pantoja 1988b) and terminal cluster inspection for PW damage

(Cartwright et al. 1990) have also been investigated and applied to action thresholds.

Sequential sampling has been developed for whole plant inspection, but was deemed too

costly for use in commercial pepper production systems (Segarra-Carmona and Pantoja

1988b). The damage-based threshold proposed by Cartwright et al. (1990) may prove

more practical, but dispersion estimates for PW damage and the relationship between bud

cluster damage and yield need to be investigated further before this method can be applied

commercially. Segarra-Carmona and Pantoja (1988a) compared whole plant inspection,

yellow sticky traps, drop cloths, sweep netting, and detergent water (pan) traps for

suitability for sampling PW adults. They determined that only whole plant inspection and

sticky traps were effective PW sampling methods. They also determined that the yellow

sticky trap method at 10/Ha could reduce scouting costs to 80% less than with whole plant

inspection, but no regression analysis was presented for the relationship between these two










sampling methods. A comparison of whole plant versus terminal bud inspection has not

been performed, and the relationship of these relative methods to an absolute has not been

determined. Also, basic behavior studies on the preferred temporal niches of the PW adult

on the pepper plant have not been conducted.

The first objective of this study was to determine where PW adults located diurnally

on the pepper plant. The second objective was to compare terminal bud inspection and

whole plant inspection with an absolute sample for estimating adult PW population density.

The third objective was to evaluate colored sticky traps as an alternative sampling method to

plant inspection. The goal of this study was to provide basic information for the

development of an efficient PW adult sampling method.


3.1.1 Methods and Materials

3.1.1.1 Cultural practices.

All experiments were conducted in the spring of 1987, in the spring and fall of

1988, and the spring of 1989 at the Gulf Coast Research and Education Center in Manatee

County, Florida on 1.52 m x 0.15 m beds of EauGallie fine sand. For all experimental

plots, bell pepper transplants were set in two parallel rows per bed with 30 cm spacing

between plants and rows. The beds were fertilized with a single band of 18-0-25-2 at a rate

of 2.98 Kg/10 linear meters of bed (LMB) mixed with 0-21-0 at 1.12 Kg/10 LMB and

were fumigated with methylbromide: chloropicrin (2:1) at 0.536 Kg/10 LMB prior to

covering with black plastic mulch. Transplants were watered with Nutrileaf (R) 20-20-20

soluble fertilizer at 600 g/100 1 water. Applications of a tank mix of tribasic copper sulphate

at 4.8 g/l, mancozeb (Manzate 200DF (R)) 1.8 g/l, and Bacillus thuringiensis kurstaki (Bt)

(Dipel 2X (R)) 1.2 g/l were made weekly.











In the spring of 1987, bell peppers, var. Jupiter, were transplanted February 20.

The total area planted consisted of two groups of seven 55 m long beds separated by an

irrigation ditch. In the spring of 1988 bell peppers, var. Early Calwonder, and Jalapefio

peppers, were transplanted on March 8 and 10, respectively. Each field of bell and

Jalapeiio peppers consisted of two groups of seven 94 m long beds separated by an

irrigation ditch. Jalapenio peppers were planted in single rows with double banded fertilizer.

In 1988, Plyac (R) spreader-sticker was added to the fungicide-Bt tank mix at 0.3 ml/1.

Each field was partitioned such that partition A = 32 m of one end of each field and

partition B = the remaining 62 m. In the fall of 1988, bell peppers, var. Early Calwonder,

were transplanted on September 7 in a block of six beds 91 m long. In the spring of 1989,

bell peppers, var. Early Calwonder, were transplanted into a block of six beds

approximately 85 m long on March 10. A single application of methomyl (Lannate L (R))

at 3.5 1/Ha was made on April 10, since an armyworm infestation exceeded a threshold

level of one larva per six plants despite weekly applications of Bt.

3.1.1.2 Plant characteristics affecting adult PW counts.

Experiment 1. Ninety eight plants were sampled six times a day (7:00, 9:00, 11:00,

13:00, 15:00, 17:00 hours), four days per week, for four weeks beginning on May 3,

1987. Adult PW were counted on all fruiting buds, young leaves (including terminal leaf

buds), mature leaves, stems and fruit (> 2.5 cm in diameter). In the last week, PW counts

on buds visible from above (exposed) versus buds not visible from above or within the

plant canopy (unexposed) were recorded separately.

Experiment 2, Ninety eight plants were sampled individually from partition B in the

morning and afternoon twice per week for 12 weeks, Spring, 1988. The number of PW

adults were counted on fruiting buds from four locations on each plant: 1) exposed buds in

the top third of the pepper plant, 2) unexposed buds in the top third, 3) exposed buds in

the bottom two-thirds of the pepper plant, and 4) unexposed buds in the bottom two-thirds.










In May of 1989, temperature data was collected for these bud locations for three height

classes of bell pepper plants on an hourly basis from 8:00 to 17:00 hour in order to

characterize diurnal temperature changes. Tele-Thermometer (R) temperature probes were

positioned at 2.6 cm and 8 cm above the soil surface on exposed and unexposed buds in an

8 cm plant; at 6 cm and 18 cm above the soil surface on exposed and unexposed buds in an

18 cm plant; and at 11 cm and 33 cm above the soil surface on exposed and unexposed

buds in an 33 cm plant. Readings were taken over four days.

3.1.1.3 Comparison of plant inspection sampling methods.

Experiment 3. Five contiguous plants in each of three randomly selected sites were

sampled on June 2, 3, and 4, 1987 at 7:00, 13:00, and 17:00 hour for each of three

sampling methods: terminal bud inspection, whole plant inspection, and an absolute

method. Terminal bud inspection consisted of a two-second visual inspection of two

exposed terminal buds per plant. Whole plant inspection represented a direct count of PW

adults from an approximate one-minute inspection of the entire plant. The absolute sample

was based on the "clam trap" method described by Leigh et al. (1970). The clam trap

consisted of two 1.8 m by 1.2 m plastic tarps sealed together by Velcro (R) adhesive sides

and foam-padded board clamps on the top and bottom edges. The two tarps were sealed

around the base of five contiguous pepper plants with the bottom board clamp 24 hours

before sampling. At the time of sampling one person on each end of the tarp raised the top

board clamp up and around the plant canopy, sealing the top and sides together

simultaneously. The trunks of the plants were then clipped, pyrethrin aerosol was sprayed

into both sides of the enclosed trap, and the whole case was taken in for detailed inspection

for PW adults among the plant sample.

Experiment 4. Three separate replicates of five contiguous plants were randomly

sampled from partition A of the 1988 field plot and the average number of PW adults was

estimated for each of the three sample methods described for Experiment 2 plus an








21

intermediate method. This latter method consisted of an inspection of two exposed terminal

buds plus two unexposed buds within the top third of the canopy of each plant. This

method was an attempt to decrease the diurnal sample variation associated with terminal

bud inspection without the excessive sampling time as required for whole plant inspection.

Samples were taken in the morning and afternoon once a week for nine weeks (the absolute

sample was taken for seven weeks).

Experiment 5. Spring (1989) terminal bud inspection, whole plant inspection, and

an absolute method were compared on the same plant so that plant to plant variability could

be estimated. Plants were selected using a systematic design with a random starting point.

For example, a row was randomly selected (from random values between one and six) and

then, a random number of paces between one and ten was selected for the first sample site.

Succeeding samples were taken at ten pace intervals (approximately 7.6 m). A total of 140

individual plants were sampled over a seven-week period. Morning and afternoon samples

were recorded separately. In order to conduct all three sampling methods on the same plant,

the terminal buds were first inspected and PW adults collected into a vial with minimal

disturbance of the plant. Then, a 36 cm diameter x 62 cm height cylinder of clear Lexan (R)

plastic with a 37 mesh gauze top cover with access holes for the hands, was placed firmly

down upon the bed over the plant. An approximate two min. whole plant search was then

conducted, collecting PW adults as they were encountered into a separate vial. Individual

plant parts were clipped and inspected for PW adults. Finally, the plastic mulch and soil

were inspected for fallen PW adults. The sum of the three methods provided the absolute

estimate. Pepper weevil adults from each method were sexed under dissection with

genitalia descriptions by Burke (1958).

3.1.1.4 Sticky trap evaluations for sampling PW adults.

Experiment 6. Sticky traps were evaluated simultaneously with the visual sampling

comparison in partition A of the field plot prepared Spring, 1988. Three yellow sticky











cards (7.6 cm x 25.4 cm) were placed at each of 12 evenly spaced sites in the field at 15

cm, 30 cm, and 91 cm above the soil surface in the center of the beds. Sticky cards were

clean and covered with Stickem Special (R) weekly. Pepper weevil adults were counted and

removed twice weekly.

Experiment 7. Late Spring (1988), colored sticky traps were evaluated in partition

B to evaluate color, height, size, and time of day for captures of PW adults. The traps were

cylindrical, 5 cm width x 20 cm circumference, and constructed of colored poster board

with yellow vinyl backing. Reflectance analysis for all poster board colors and the yellow

vinyl are given in Figure 3.10 A. The color traps were stapled to a wooden stake 60 cm

above the soil surface. Three replicates of the trap colors were placed in bell peppers and

three replicates were placed in Jalapeiio peppers. An evaluation of trap height was made

using cylindrical traps 3 cm width x 20 cm circumference constructed of yellow vinyl

plastic. These cylinders were stapled individually on wooden stakes at 10 cm, 50 cm, 60

cm, 70 cm, 80 cm, and 90 cm, above the soil surface. Three replicates of the trap heights

were placed in bell peppers and three replicates were placed in Jalapeiio peppers. Trap

sizes tested were 1, 3, 5, 7, and 15 cm widths x 20 cm circumference cylinders using the

yellow vinyl plastic. The different trap sizes were all positioned so that the top of the traps

were 60 cm above the soil surface. There were four replicates of the different trap sizes, all

placed in bell peppers. All traps were coated with a thick (1 mm) layer of Stikem Special

(R) because previous tests had shown that weevils were able to free themselves rapidly

from thinner coatings. Even with the thicker coating, weevils were frequently able to

escape from the lower rim of the trap within two hours by hanging and moving back and

forth. For this reason, traps were inspected at two hour intervals (7:00, 9:00, 11:00, 13:00,

15:00, and 17:00 hour) and PW adults removed at each inspection. The order of trap

treatments was randomized within blocks and traps were placed systematically,

approximately every 7.6 m with a random starting point. For example, beginning in row 1










the order of colors in the first block was yellow, white, orange, black, blue, green, and red

followed by a random order of trap heights and trap sizes all within the same block of

approximately two rows. At two hour intervals, all traps were inspected beginning at row 1

and ending with row 14.

Experiment 8. Sticky trap colors yellow, green, white, beige, ivory, colonial blue,

light gray, azure blue, royal blue, gray, maroon, black, red, chocolate brown, or a

combination of half white on top and half yellow on the bottom were evaluated Fall, 1988;

Spring, 1989. Reflectance analyses for all polyethylene sticky trap colors used are given in

Figures 3.10 B and C. Traps were 15 cm width x 25 cm circumference constructed of

cylinders of polyethylene plastic that were stapled onto wooden stakes 25 cm above the soil

surface to maximize captures while minimizing the accumulation of wind-blown debris.

Plastic was used because the poster board used in Experiment 7 deteriorated rapidly in the

elements. There were four replicates per color in the fall of 1988 and 16 replicates per color

in the spring of 1989.

Experiment 9. A paired comparison of yellow versus white traps (15 cm width by

25 cm circumference) was conducted throughout the pepper growing season, Spring

1989. Samples were taken twice weekly at four times of the day (10:00, 12:00, 14:00,

16:00 hours) on one white trap and one yellow trap in each of four blocks. Pepper weevil

adults were collected and sexed under dissection. Concurrently, two terminal bud

inspections per plant on sections of 25 plants were taken at 9:00 hours on the same

sampling dates in the same designated blocks. Pepper weevil sex ratios were also evaluated

for terminal bud inspection samples. On May 24, PW adults were collected on four

samples of five individual plants using terminal bud inspection and the absolute method

described in experiment 5 for comparison to sticky trap samples. Also, PW adults were

collected on four terminal bud inspection samples of 25 contiguous plants each, on four










yellow sticky traps, and on four white sticky traps. Pepper weevil adults from each method

were sexed under dissection consulting genitalia descriptions by Burke (1958).


3.1.2 Results and Discussion

3.1.2.1 Plant characteristics affecting adult PW counts

Experiment 1. More PW adults were found on fruiting buds and, to a lesser extent,

the young leaves surrounding bud clusters than other plant structures (Figure 3.1). This

was true regardless of the week of sampling (Figure 3.2). A significantly greater number of

adults on the exposed terminal leaves at 7:00 hours than at later hours (Table 3.1)

supported the observation that, early in the morning, a significant number of weevils

remained stationary within the curl of newly expanding leaves. The number of PW adults

observed on exposed buds decreased from 7:00 to 17:00 hour while the number observed

on unexposed buds increased (Figure 3.3). Andrews et al. (1986) reported similar results

for PW adults observed on exposed terminal buds. Greater numbers of PW in the

afternoon than in the morning was noted for unexposed terminal leaves, trunk, and stem

(Table 3.1). Based on the overall averages for number of observed PW adults (Table 3.1),

whole plant inspection on mature bell pepper plants seemed to be slightly more efficient in

the afternoon than in the morning. Perhaps this was due to the greater activity of adults in

the afternoon and, therefore, greater visibility of the adults within the canopy. Two field

observations (Spring 1987) that may have affected the PW count data on plant parts were

1) flight activity appeared to be greater in the afternoon and 2) the adult weevils tended to

initiate flight from the older leaves.

Experiment 2. An average of 66% of the fruiting buds were encountered in the

top-third of the bell pepper plants, 39% and 27% in the exposed and unexposed portions,

respectively (percentages represented by divisions of the bars in Figure 3.4 A). Pepper

weevil adult counts on the buds in the top third portion of the bell pepper plant represented










77% of the total PW adult count per plant, 39% and 38% in the exposed and unexposed

portions, respectively (Figure 3.4 B). Thus, PW adult and fruiting bud numbers were

proportional for the top third exposed location in the bell pepper plant. Similarly, in

Jalapeiio pepper an average of 62% and 43% of the fruiting buds were observed in the

top third and top third exposed portion of the plants, respectively (Figure 3.5 A). Pepper

weevil adult numbers on the buds in the top third and top third exposed portion of the

Jalapefio pepper plant represented 65% and 40% of the total PW adult count per plant,

respectively (Figure 3.5 B). These data indicated that both PW adults and fruiting buds

were more concentrated in the top third of the pepper plant. They also indicated that the

relative sampling method, terminal bud inspection, had access to a major portion of the total

bud (within plant sample universe) and adult PW count on the pepper plant. In the first ten

weeks after transplant in both bell and Jalapefio pepper, PW adults were highly correlated

with the number of pepper buds squared (R=0.96; P<0.0001; n=9 and R=0.99; P<0.0001;

n=10, respectively) and the number of pepper buds was significantly correlated with time

in terms of weeks (R=0.94; P<0.0001; n=10 and R=0.93; P<0.0001; n=10, respectively).

As indicated by the range of values between weeks one and ten (Figures 3.4 A and 3.5 A),

the number of fruiting buds increased linearly with time for bell and Jalapeiio pepper (R2

=0.89; slope = 3.84; P<0.0001; intercept = -9.39; P<0.01 and R2 =0.86; slope=9.28;

P<0.001; intercept= -25.5; P<0.01, respectively). Between weeks one and ten, the

increase in PW adults more closely fit a quadratic regression for both bell and Jalapefio

peppers (R2 =0.77; F=11.5; df=2, 7; P<0.01 and R2 =0.79; F=13.3; df=2, 7; P<0.01,

respectively) than a linear regression (R2 =0.43; F=6.1; df=l, 8; P<0.05 and R2 =0.51;

F=8.3; df=l, 8; P<0.05, respectively). After week ten, PW adult numbers and pepper

buds were not significantly correlated; however, PW adults continued to increase with time

for both bell and Jalapefio peppers on all within plant locations (R=0.65; P<0.05; n=12 and











R=0.85; P<0.001; n=12), while bell buds decreased (R= -0.73; P<0.01; n=12). Thus,

there was a significant positive relationship between PW adult numbers and buds from

week one to ten, but later in the season PW numbers continued to increase while the

number of buds decreased.

The ratio of exposed to unexposed fruiting buds decreased after week five for both

bell and Jalapeiio peppers (Figures 3.6 A and 3.7 A, respectively). This phenomenon was

termed canopy closure; i.e., the rapid growth of pepper foliage concealing much of the

plant interior. As a result of canopy closure, the relatively high percentage of exposed

pepper buds early in the season (to five weeks after transplant) decreased significantly in

the last seven weeks of the season (Figures 3.6 A and 3.7 A), possibly affecting PW

counts on exposed buds. However, throughout almost two-thirds of the season, there were

approximately 50% of the PW adults consistently on the top third exposed buds (Figures

3.6 B and 3.7 B) so that, in fact, there were approximately the same to greater numbers of

PW adults per exposed bud after canopy closure than before (Figure 3.8). The number of

adults per bud on the top exposed portion of the plant did not increase greatly until week

ten (Figure 3.8). Regression analysis of adult numbers on exposed buds from the top third

of the plant in relation to adult numbers on all buds (total PW) provided a good fit (bell

pepper adjusted R2=0.91; P<0.0001 and top exposed bud inspection=0.32 x total PW +

0.07; Jalapefio pepper adjusted R2=0.99+; P<0.0001 and top exposed bud

inspection=0.39 x total PW + 0.02). Therefore, inspection of exposed buds in the top third

of the pepper plant was an adequate relative sampling method for estimating adult numbers

on all buds. Also, the number of adults per fruiting bud did not appear to be affected

greatly by seasonal variations in plant phenology and PW population levels until the end of

the season (Figures 3.6 B and 3.7 B).










Temperatures of top exposed buds on small plants (8 cm height) were 1-20C less

than temperatures of bottom unexposed buds, thus indicating an influence of heat

irradiating from the plastic mulch bed (Table 3.2). In the medium and large plants (18 cm

and 33 cm height, respectively), temperatures of exposed buds in the top third of the plant

were 1-20C warmer than temperatures of unexposed buds in the top third of the plant at

9:00 hours. Also, temperatures were 1-30C cooler on the unexposed buds in the top third

of the plant at 14:00 hours (Table 3.2). There was less influence of irradiation from the bed

in the larger plants. One hypothesis for the changes in PW adult counts during the day

(Figure 3.2) is that the adults move within the canopy in response to temperature changes

(i.e., thermo-regulation). When PW counts in similar bud locations are compared over an

eleven-week period, the ratio of adults on exposed to unexposed buds is approximately

1.00 for both morning and afternoon samples early in the season. By week ten, this ratio is

1.59 for the morning sample and 1.10 for the afternoon sample (Table 3.3). Perhaps this

greater difference between morning and afternoon PW counts later in the season is due to a

greater temperature difference between exposed and unexposed buds in a mature pepper

plant. Of course this temperature only applies to spring conditions in Bradenton, Florida

and would have to be reassessed for other locations and weather conditions to make similar

inferences else where.

Differences in crop morphology should not be overlooked in developing sampling

procedures for different pepper crops. In this study, Jalapefio peppers produced over twice

as many fruiting buds as bell pepper var. Early Calwonder (the mean no. buds/plant ten

weeks after transplant was 200 for Jalapefio peppers and 71.3 for bell peppers). Thus, the

probability of encountering PW adults with a fixed bud sample number per plant should be

proportionally less for Jalapefio pepper as for bell pepper, however, through week 10, the

average PW count on the top exposed buds was similar between the two crops (Figure











3.8). By week twelve, more adults were indeed observed per bud on bell than on Jalapefio

pepper. With Jalapefio peppers, the scout's field of vision was disrupted to a greater extent

than for bell pepper due to a greater number of small leaves and increased branching.

Under equal PW population densities, terminal bud inspection in Jalapeiio peppers should

reveal relatively fewer PW adults than for bell peppers due to these phenomena. However,

the greater bud production and diminutive fruit size of Jalapefio peppers may enable this

crop to tolerate greater numbers of adults, compensating for the reduced efficiency in adult

PW counts. Also, damage is a direct result of the ratio of PW adults to fruiting buds so that

the numbers of buds per plant will be, to a certain extent, an irrelevant parameter. The

economic value of a single bud of Jalapefio versus bell pepper is a relevant parameter.

3.1.2.2 Comparison of visual plant inspection techniques

Experiment 3. In the first sampling comparison, more PW adults were observed per

five plants in the absolute sample than the other samples, but the number observed per unit

time for terminal bud inspection and whole plant inspection was greater than the number

observed in the absolute sample (Table 3.4). Furthermore, terminal bud inspection required

the shortest examination time per plant. Overall, there was no significant effect of time of

sampling for whole plant inspection and the absolute sample (Table 3.5); however, the

number of PW adults observed with terminal bud inspection was significantly greater at

7:00 than at 13:00 or 17:00 (Table 3.5). This substantiates the diurnal effect on PW counts

on terminal buds observed previously (Figure 3.2). From these preliminary results,

terminal bud inspection was judged to be more efficient than whole plant inspection in

maximizing the number of PW observed for the least amount of sampling time per plant.

Since it is desirable to cover more sample sites as the degree of clumping increases, then

less time per sample site is preferable to offset the increase sampling effort needed for an

acceptable level of precision.











There were still two questions concerning diurnal sample variation that needed to be

addressed. Was there an alternative sample method with the same efficiency as terminal bud

inspection, but without the diurnal variation? Was the diurnal variation of terminal bud

inspection great enough to cause different responses in an action threshold based on this

method? The latter question will be discussed in Section 4.2.

Experiment 4. Regardless of the pepper type, the intermediate sampling method of

examining two exposed and two unexposed buds per plant required significantly less time

per plant than whole plant inspection without a significant decrease in PW adults observed

per unit search time (Table 3.6). Even so, terminal bud inspection required less search time

per plant than the intermediate sampling method without significant loss in efficiency.

Thus, it should be more profitable to simply take more samples with terminal bud

inspection in the afternoon to compensate for diurnal variation rather than opt for a more

intensive sampling method.

Terminal bud inspection in bell pepper encountered 5.5% and whole plant

inspection 49% of the PW adults found with the absolute sample (calculated from values in

Table 3.6). In Jalapeiio peppers, there was no significant difference between the four

sampling methods in the numbers of PW encountered per unit of search time. Also in

Jalapefio pepper, terminal bud inspection encountered 1.4% and whole plant inspections

33% of the PW adults found with the absolute sample. The lower percentages in Jalapefio

pepper relative to bell pepper were probably related to the greater numbers of buds and the

greater canopy effect previously described for the Jalapeiio pepper growth habit. The

relatively low rate of recovery for whole plant inspection in both crops was due to the time

limit of two minutes per sample. The efficiency of terminal bud inspection as measured in

PW/minute search time was not significantly lower than the efficiency of the other methods

for Jalapefio pepper and was greater than whole plant inspection in bell pepper (Table 3.6).

Since terminal bud inspection was also the simplest method and covered the greatest











number of plants per minute of search time, it would be more acceptable as a sampling

method than the two more intensive methods evaluated in this comparison.

Since the number of PW adults and fruiting buds were shown to be nearly

proportional, that is, adults tend to disperse evenly where fruiting buds are located on the

plant (Figures 3.4 and 3.5), the number of PW adults on all buds was estimated from the

observed adults from terminal bud inspection and an estimate of the average number of

buds per sampling period. For bell pepper, the estimate of PW adults on all buds/plant was

31.8 buds per plant/2 buds per inspection x 0.33 = 5.3, which is reasonably close to the

absolute estimate of 6.1 (Table 3.6). The estimate for Jalapefio peppers (61.2 buds per

plant/2 buds per inspection x 0.11 = 3.4) was not so close. Thus, this observed

relationship of adults to fruiting buds may be too simplistic for accurate estimation of

absolute PW densities from observed adult numbers per bud.

Experiment 5. PW adult counts on exposed terminal buds (individual plant basis)

were again greater in the morning than the afternoon (Table 3.7). Overall, PW adults

observed with terminal bud inspection and with whole plant inspection represented 8.9%

and 41% of the absolute sample, respectively. The percentages of PW accounted for by

terminal bud inspection and whole plant inspection decreased progressively as the plant

increased in size (Table 3.7). However, the decrease in the percent of PW counted by

whole plant inspection was less over time compared to terminal bud inspection. This was

because terminal bud sample size remained constant, two buds/plant, whereas whole plant

inspection sample size increased to the point that the two minute inspection time would

allow. Excluding the first sample date, whole plant inspection averaged 47% of the

absolute sample. For pepper plants less than 14 cm in height and with less than nine

fruiting buds (April 10), the whole plant search was essentially the same as the absolute

PW count. On the last sample period in bell peppers, when plants were at first harvest,

terminal bud inspection represented 6%, and whole plant inspection represented 35% of the










absolute PW count per plant. These estimates were similar to estimates obtained in a bell

pepper crop of similar physiological age in 1987 (terminal bud inspection = 7% and whole

plant inspection = 31% of an absolute sample) (Table 3.4).

The numbers of adults observed by terminal bud inspection and whole plant

inspection had significant linear relationships to the number observed in the absolute

sample (Table 3.7). However, the variation in the absolute estimate accounted for less than

50% of the variation in the terminal bud inspection estimate regardless of sampling in the

morning or afternoon. There was a stronger relationship between whole plant inspection

and the absolute sample (adjusted R2 = 0.88) than terminal bud inspection and the absolute

(adjusted R2 = 0.39) (Table 3.7). This is probably due to the constant sample size with

terminal bud inspection and emphasizes the relative stability of PW counts from terminal

bud inspection over changes in PW population density. The only intercept significantly

different from zero in Table 3.7 was from the regression of whole plant inspection on the

absolute. However, if the same data are fit to a quadratic regression, the R2 is improved

and the intercept is not significantly different from zero (R2 =0.92, slopes X-0.52, X2 =-

0.01; intercept=0.04), indicating that the relationship between the two sampling methods is

curved rather than linear.

3.1.2.3 Sticky trap evaluations for sampling PW adults

Experiment 6. The plant inspection methods of Experiment 4, each consisting of an

approximate sample size of 1% of the plant population (15 plants per treatment/1456 plants

total), did not detect the PW infestation until four to six weeks after transplant (Figure

3.9). However, a high density of sticky traps (12 stations per 0.04 Ha) was able to detect

PW adults about four weeks before any of the other plant inspection sampling methods.

Therefore, a series of experiments was conducted to further explore the possibility of using

sticky traps as an alternative sampling method for PW adults.










Experiment 7. More pepper weevil adults were captured on white traps with a

yellow interior than on other colored traps (Table 3.8). The reflectance analyses of these

colors are presented in Figure 3.10 A. More PW adults were captured between 15:00 and

17:00 hours than earlier in the day (Table 3.9). Trap width and height were evaluated with

light yellow vinyl plastic. More adults were captured at trap heights 10-50 cm for mature

bell peppers (54.8 cm in height) and 50-60 cm for mature Jalapefio pepper plants (83.1 cm

in height)(Table 3.10). The number of adults trapped increased as the trap band width

increased to 15 cm, but the optimum band width was not determined (Table 3.11). Even

so, the number of PW adults captured per cm2 of yellow vinyl did not increase significantly

as band width increased from 5 to 15 cm. Since the colored poster board tended to

deteriorate rapidly in the field, the color trial was re-evaluated with polyethylene plastic in

December 1988 and June 1989.

Experiment 8. In the spring of 1989, yellow captured more PW adults than white,

but yellow did not differ significantly from white in the fall of 1988 (Table 3.12). A

combination of yellow and white trapped more adults than solid white traps in 1989. There

was no significant difference among the red, blues or grays. Spectral reflectance analysis

for all the colored materials tested is presented in Figures 3.10 A and B. From the

reflectance analysis and PW adult count data it can be inferred that there was a significant

difference in response to color by the PW adult and that the greater response was to the

peak reflectance defined by the color yellow (3.1.10 B). Also, since there was a stronger

attraction to white than to blue or red in 1988, then blue or red were apparently not repellant

to the PW adults (Jim Kring 1990, personal communication). This effect was not observed

in 1989.

Experiment 9. Yellow sticky traps tended to capture more PW adults than the white

sticky traps, especially at higher PW population densities (Table 3.13). Yellow traps at a

density of four traps/0.1 Ha were similar to terminal bud inspection of four strips of 25










contiguous plants/0.1 Ha in overall numbers of PW adults detected. However, these

sample sizes were chosen purely on a practical basis (each sample including replicates

required less than five minutes) and inferences concerning this relationship should be

approached with caution. The time involved in the two sampling methods, excluding the

construction of the sticky trap, was less for the trap (50 sec inspection of 50 terminal buds

versus 30 sec inspection of one trap). However, since weevils were able to escape from the

stickem after 1-2 hours. The sex ratio of PW female: male adults observed by terminal bud

inspection (5/22/1989) was significantly lower than the sex ratio of adults captured on

yellow and white sticky traps at high PW population densities (t=3.99; df=6; P<0.01 and

t=3.97; df=6; P<0.01 (see Table 3.13 for means)). Overall, the sex ratios for each of the

two sticky trap colors were higher than the sex ratio of terminal bud inspection. This

indicated that the sticky traps captured proportionally more females than were observed by

terminal bud inspection.

Analyses of factors affecting the sex ratio of PW adult trap catches revealed that,

although more adults were captured on yellow traps than white traps, the sex ratios of the

captured adults were not different (Table 3.14). There was a significant diurnal effect on

sex ratios and numbers of PW females captured per trap. Greater numbers of females were

captured at 14:00 than either 10:00 or 18:00 (Table 3.14), suggesting a mid-afternoon peak

in female activity. Pepper weevil males were not similarly affected.

In May 1989, the sex ratio of adults collected during terminal bud inspection was

not significantly different from that of the absolute sample (Table 3.15). Sex ratios were

significantly higher for both yellow (t=3.13; df=6; P<0.05) and white (t =4.11; df=6;

P<0.01) sticky trap samples than the sex ratio in the absolute sample. The female-biased

response to traps may have been due to greater attraction of the females than males to the

trap colors, female response to males ensnared on the sticky trap, or greater flight activity

of the female such that more females encountered sticky traps.












3.1.3 Summary and Conclusions

Pepper weevil adults were most often encountered on the fruiting buds and on the

young leaves surrounding these buds. There were more buds and proportionally more PW

adults on the top third of the pepper plant. Diurnal variation was validated and quantified

for PW adult counts on the exposed terminal buds; however, the impact of this variation,

when utilizing an action threshold, needs to be evaluated. Terminal bud inspection resulted

in the highest PW counts per minute search time in each of the sampling comparison trials

and, as first described by Andrews et al. (1986), was the more efficient method tested for

the detection of PW adults. Since greater numbers of PW adults were observed on exposed

terminal buds in the morning than the afternoon, this sampling method should be used in

the morning for greater sensitivity to low PW adult populations. This method, used in

conjunction with sampling for immatures in buds and fruit, should provide the best

indicator for the need to spray insecticides against PW adults.

Sticky trap tests results substantiated the utility of yellow sticky traps as a possible

alternative to plant inspection techniques and suggested that the traps may be more sensitive

to the presence of PW females than terminal bud inspection. However, the density of traps

in these tests (approximately 35 traps/Ha) was such that the expense of trap construction

and maintenance could outweigh the benefits. Also, data is lacking on the relationship

between sticky trap capture rates and absolute PW population density or PW damage. The

future utility of PW traps will depend greatly upon the development of pheromones to

enhance trap attraction and upon the commercial field testing of sticky traps in comparison

with commercial scouting practices to determine if the use of sticky traps is economically

desirable. Information is needed on the spatial limits of color attraction and relationship

between trap counts and absolute densities.












Table 3.1 Effect of time of day on the average numbers of PW adults per plant part in
exposed versus unexposed positions, relative to a scout's field of vision without foliage
manipulation, for harvestable bell pepper plants. Bradenton, Florida 1987.



No. adults captured per plant per two hour interval

Plant part 7:00 9:00 11:00 13:00 15:00 17:00

Overall 1.35 a* 1.33 a 1.55 ab 1.50 ab 1.49 ab 1.58 b

Fruiting bud
exposed 0.19 b 0.25 a 0.17 b 0.10 c 0.12 c 0.11 c
unexposed 0.54 a 0.50 a 0.58 ab 0.61 ab 0.61 ab 0.70 b

Terminal leaf
exposed 0.18 a 0.11 ab 0.09 b 0.08 b 0.07 b 0.09 b
unexposed 0.34 a 0.27 a 0.36 a 0.34 a 0.38 a 0.38 a

Mature fruit
exposed 0.00 a 0.01 a 0.01 a 0.01 a 0.01 a 0.00 a
unexposed 0.03 a 0.08 ab 0.10 bc 0.16 c 0.12 bc 0.07 ab

Trunk and stems
exposed 0.00 a 0.00+ a 0.00+ a 0.00 a 0.00 a 0.00 a
unexposed 0.02 a 0.05 ab 0.06 b 0.08 b 0.09 b 0.08 b

Mature leaf
exposed 0.01 a 0.01 a 0.02 a 0.03 a 0.02 a 0.02 a
unexposed 0.05 a 0.06 a 0.15 b 0.11 ab 0.09 ab 0.13 b

Means within rows followed by the same letter are not significantly different (P<0.05) using Duncan's New
Multiple Range test; n-4 replicates; per plant means based on 392 sub-samples.












Table 3.2 Temperature data for three sizes of bell pepper plants for exposed versus
unexposed buds on the top third and bottom two-thirds of the plant at hourly intervals
during the day. Bradenton, Florida 1989.


Temperature at various fruiting bud positions

Top Top Bottom Bottom
Time exposed unexposed exposed unexposed


Small pepper plant (8 cm height)
-**


8:00
9:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
overall


8:00
9:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
overall


8:00
9:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
overall


21.62.0
25.5+2.3
28.32.2
31.32.0
32.7+1.9
33.5+1.5
33.5+2.4
32.3+2.0
30.7+1.5
29.3+1.6
29.8+4.2


21.3+2.3
24.2+0.9
26.6+1.7
28.91.2
29.9+0.9
31.1+1.4
31.4+1.4
31.2+1.0
30.4+0.5
29.0+2.0
28.33.5


20.9+2.3
24.5+1.5
26.8+1.6
29.8+1.6
30.4+1.8
31.10.9
32.31.3
31.9+1.3
30.9+1.5
29.2+2.3
28.6+3.8


21.5+1.9
25.3+1.5
27.91.7
30.62.5
33.52.7
34.1+1.3
34.92.0
33.3+1.0
32.2+2.0
30.72.1
30.3+4.5


22.5+1.8
24.9+1.1
26.3+1.2
28.8+1.4
30.4+1.1
31.1+1.1
31.1+1.4
30.3+1.5
30.3+0.9
29.0+1.6
28.4+3.1


20.5+2.1
22.3+1.1
24.6+1.2
26.2+1.3
27.4+1.2
28.8+0.9
29.5+1.0
29.0+0.7
29.4+0.2
28.2+1.4
26.4+3.3


* Values given in average degrees Celsius std. Measurements were taken in the same field at the
same hour and replicated over 4 days in May 1989. Air temperatures during the entire sample
period averaged by the hours indicated above were 22.2+1.3, 24.7+0.6, 26.30.8, 27.8+1.0,
28.8+0.7, 29.2+1.2, 29.31.1, 28.91.0, 28.20.8, and 27.4 0.8 oC from 8:00 to 17:00 hours,
respectively.
** Missing values in the first five weeks were due to buds not being present.


Medium pepper pla
20.8+2.0
22.9+0.9
25.4+1.1
28.0+1.6
29.1+1.5
30.0+0.8
30.3+1.2
29.7+1.5
29.5+0.5
28.6+2.1
27.33.4

Large pepper plant
20.3+2.1
22.4+1.1
25.0+1.1
26.8+1.5
27.6+1.0
28.30.9
29.31.4
28.7+0.7
29.2+0.6
27.91.5
26.4+3.2


nt (18 cm height)
22.62.5
25.5+1.2
27.01.7
29.0+0.9
29.61.4
30.7+1.7
30.4+1.6
30.8+2.2
30.21.6
28.6+1.5
28.3+3.0

(33 cm height)
21.6+1.6
24.3+1.3
26.00.9
27.8+0.4
28.9+0.8
29.00.8
30.4+1.1
30.2+1.8
30.5+1.5
29.0+1.5
27.6+3.0









37

Table 3.3 Mean numbers of PW adults on exposed versus unexposed fruiting buds in the
top third and bottom two-thirds of randomly selected bell pepper plants. Bradenton, Florida
1988.



Mean no. of adults/2 min. inspection/plant Ratios*

Top Top Bottom Bottom
Week exposed unexposed exposed unexposed Ratio 1 Ratio 2 Ratio 3

Morning sample
1 0.00 -** 0.00 1.00 1.00 1.00
2 0.00 0.00 1.00 1.00 1.00
3 0.01 0.00 1.01 1.01 1.01
4 0.01 0.00 1.01 1.01 1.01
5 0.01 0.00 1.01 1.01 1.01
6 0.01 0.01 0.00 0.01 1.00 0.99 0.99
7 0.02 0.01 0.00 0.01 1.02 1.01 1.01
8 0.08 0.02 0.01 0.00 1.10 1.07 1.06
9 0.16 0.06 0.01 0.02 1.19 1.09 1.07
10 0.86 0.15 0.04 0.04 1.86 1.59 1.50
11 1.68 0.96 0.15 0.43 2.30 1.19 1.06

Afternoon sample
1 0.00 0.00 1.00 1.00 1.00
2 0.00 0.00 1.00 1.00 1.00
3 0.00 0.00 1.00 1.00 1.00
4 0.00 0.00 1.00 1.00 1.00
5 0.00 0.01 1.00 1.00 1.00
6 0.01 0.01 0.00 0.01 1.01 1.00 1.00
7 0.01 0.01 0.00 0.02 1.00 0.99 0.99
8 0.08 0.05 0.00 0.02 1.11 1.01 1.01
9 0.13 0.05 0.00 0.03 1.15 1.05 1.05
10 0.53 0.35 0.05 0.09 1.64 1.10 1.03
11 1.41 1.46 0.07 0.43 2.58 0.86 0.81

Ratio 1 = top third versus bottom two thirds PW count, ratio 2 = exposed versus unexposed PW
count, ratio 3 = top third exposed versus all others combined (ratios calculated after the addition of 1
to the mean of each value).
** Missing values in the first five weeks were due to buds not being present.











Table 3.4 Comparison of three visual techniques for sampling pepper weevil adults
on fully mature bell pepper plants, var. Jupiter. Bradenton Florida 1987.


Number Mean search Adults per
of adults time (min.) minute
Type of sample per 5 plants per 5 plants search time


terminal bud inspection 2.63 a* 0.67 a 3.93 a
whole plant inspection 11.9 b 3.44 b 3.58 a
absolute plant sample 38.3 c 90.0 c 0.43 b

* Means within columns followed by the same letter are not significantly different
(P<0.05) using Duncan's New Multiple Range test (n=9).










Table 3.5 The diurnal effect on numbers of pepper weevil adults for terminal bud
inspection, whole plant inspection, and an absolute sample on fully mature bell pepper
plants, var. Jupiter. Bradenton, Florida 1987.


No. adults observed/5 plants

Terminal bud Whole plant Absolute
Time of day inspection inspection sample


7:00 hour 4.33 a* 9.3 a 29.0 a
13:00 hour 1.00 b 10.7 a 47.3 a
17:00 hour 0.67 b 8.7 a 48.7 a


* Means within columns followed by the same letter are not significantly different
(P<0.05) using Duncan's New Multiple Range test (n=3). Sample unit = 5 contiguous
pepper plants.











Table 3.6 Comparison of four methods for sampling pepper weevil adults in bell and
Jalapeiio peppers averaged over the entire growing season. The methods were examination
of 2 exposed fruiting buds per plant, examination of 2 exposed plus 2 unexposed fruiting
buds per plant, whole plant inspection including buds, stems and leaves, and a destructive,
entire plant absolute sample. Bradenton, Florida 1988.


No. adults Search time No. adults
Sampling method n /5 plants /5 plants (min.) /min.


Bell peppers

2 exposed buds 36 0.33 a* 0.2 a 1.53 a
2 exposed + 2 unexposed 36 0.44 a 0.4 b 0.91 ab
whole plant inspection 36 3.00 b 3.0 c 0.45 bc
absolute 24 6.11 c 30.0 d 0.20 c

Jalapefio peppers

2 exposed buds 36 0.11 a 0.2 a 0.41 a
2 exposed + 2 unexposed 36 0.28 ab 0.4 b 0.52 a
whole plant inspection 36 2.69 b 2.5 c 0.47 a
absolute 24 8.11 c 30.0 d 0.27 a

* Sample unit = 5 contiguous plants. Means within columns within pepper type followed
by the same letter are not significantly different (P<0.05) using Duncan's New Multiple
Range test.












Table 3.7 Comparison of methods for sampling pepper weevil adults in bell pepper on an
individual plant basis for morning and afternoon samples and over five sampling dates.
Sampling methods include terminal bud inspection, whole plant inspection, and an entire
plant absolute sample. Comparison includes regression analyses. Bradenton, Florida 1989.
No. adults per
No. adults per plant std (Ratio to absolute) plantstd
Terminal bud Whole plant
Sampling period n inspection inspection Absolute

overall 140 0.210.48 (0.09) 0.991.85 (0.41) 2.4+4.9
morning 50 0.300.58 (0.10) 1.181.94 (0.39) 3.05.7
afternoon 50 0.2290.47 (0.06) 1.462.24 (0.41) 3.65.5
April 10 40 0.100.30 (0.67) 0.150.36 (1.00) 0.2+0.4
April 25 40 0.08+0.35 (0.20) 0.200.72 (0.53) 0.4+1.2
May 3 30 0.03+0.18 (0.14) 0.130.43 (0.57) 0.2+0.7
May 24 20 0.65+0.59 (0.08) 3.501.61 (0.42) 8.4+3.8
June 2 10 0.90+0.74 (0.06) 5.001.94 (0.35) 14.2+6.8

Linear regressions: *
1. TBI on Absolute (overall): TBI = 0.06(Absolute) + 0.07,
Adjusted R2 = 0.39, P<0.0001

2. TBI on Absolute (morning): TBI = 0.07(Absolute) + 0.08,
Adjusted R2 = 0.49, P<0.0001

3. TBI on Absolute (afternoon): TBI = 0.05(Absolute) + 0.04,
Adjusted R2 = 0.33, P<0.0001

4. TBI on WPI (overall): TBI = 0.18(WPI) + 0.04,
Adjusted R2 = 0.46, P<0.0001

5. WPI on Absolute (overall): WPI = 0.35(Absolute) + 0.14,
Adjusted R2 = 0.88, P<0.0001

Abbreviations for linear regression equations: TBI = terminal bud inspection, WPI = whole plant inspection
(equations given as Y = slope(X) + intercept).










Table 3.8 Numbers of pepper weevil adults captured on cylindrical (7 cm band width by
25 cm circumference), colored poster board traps with a yellow vinyl plastic interior placed
in mature bell and Jalapeiio pepper. See Figure 3.10 A for reflectance analyses.
Bradenton, Florida June 1988.

Color No. adults/trap (n=102)


white 0.49 a*
yellow 0.25 b
green 0.18 b
blue 0.16 b
black 0.15 b
red 0.12 b
orange 0.09 b

* Means within the column followed by the same letter are not significantly different
(P<0.05) using Duncan's New Multiple Range test.








43

Table 3.9 The effect of time of day on the numbers of pepper weevil adults captured on
sticky traps in mature bell and Jalapefio pepper. Bradenton, Florida June 1988.

Time of day No. adults/trap (n=156)


7:00- 9:00 0.20 a*
9:00 11:00 0.22 a
11:00-13:00 0.17 a
13:00- 15:00 0.15 a
15:00- 17:00 0.56 b

* Data averaged over all colors in Table 3.8. Means within the column followed by the
same letter are not significantly different (P<0.05) using Duncan's New Multiple Range
test.










Table 3.10 Numbers of pepper weevil adults captured at different trap heights on
cylindrical (3 cm width by 25 cm circumference) yellow vinyl sticky traps placed in mature
bell and Jalapefio peppers. Bradenton, Florida June 1988.

Height of trap No. adults/trap (n=36)
above the bed
Bell pepper* Jalapeiio pepper


10 cm 0.28 a** 0.69 bc
50 cm 0.19 ab 1.25 a
60 cm 0.06 bc 1.22 ab
70 cm 0.00 c 0.53 cd
80 cm 0.00 c 0.67 c
90cm 0.00 c 0.11 d

* the average bell pepper plant height was 54.8+8.1 cm and the average Jalapefio pepper
plant height was 83.1+7.8 cm.
** Means within columns followed by the same letter are not significantly different
(P<0.05) using Duncan's New Multiple Range test.








45

Table 3.11 Numbers of pepper weevil adults captured on five sizes of yellow vinyl sticky
traps placed in mature bell pepper. Bradenton, Florida June 1988.

No. adults/cm2
Trap band width No. adults/trap (n=48) of trap surface


1 cm 0.02 a* 0.02 a
3 cm 0.08 a 0.03 a
5 cm 0.25 a 0.05 ab
7 cm 0.56 b 0.08 b
15 cm 1.21 c 0.08 b

* Means within columns followed by the same letter are not significantly different (P<0.05)
using Duncan's New Multiple Range test.










Table 3.12 Numbers of pepper weevil adults captured on sticky traps constructed of
colored polyethylene plastic cylinders 15 cm width by 25 cm circumference and covered
with Stickem Special (R) and placed in mature bell pepper. See Figures 3.10 B and C for
reflectance analyses. Bradenton, Florida 1988 and 1989.


Color of trap No. adults/trap 1988 (n=4) No. adults/trap 1989 (n=16)


yellow 1.54 a* 2.88 a
yellow+white -2.25 ab
green 1.37 a 1.88 bc
white 1.04 ab 0.81 de
beige 0.72 bc 1.06 cde
ivory 0.52 bcd 1.25 cd
colonial blue 0.52 bcd
light gray 0.35 cd 1.00 de
azure blue 0.35 cd 0.63 de
gray 0.35 cd 0.63 de
royal blue 0.35 cd
maroon 0.35 cd 0.44 de
black 0.17 cd 0.31 e
red 0.00 d 0.50 de
brown 0.00 d

* Means within columns followed by the same letter are not significantly different
(P<0.05) using Duncan's New Multiple Range test.












Table 3.13 Numbers and sex ratios of pepper weevil adults in bell pepper collected using
terminal bud inspection or captured using yellow sticky traps and white sticky traps.
Bradenton, Florida 1989.

Terminal bud inspection Yellow sticky traps White sticky traps
Date No. adults Sex ratio No. adults Sex ratio No. adults Sex ratio
/50 buds /trap/8 hours /trap/8 hours

3/24 0.0-0.0 1.00+0.00 0.00.0 1.000.00 0.000.00 1.00+0.00
3/27 0.5+0.6 1.130.63 0.3+0.5 1.250.50 0.500.58 1.230.63
3/30 0.3+0.5 0.880.25 0.3+0.5 1.250.50 0.000.00 1.00+0.00
4/02 0.3+0.5 1.250.50 0.50.6 1.13+0.63 0.250.50 0.88+0.25
4/03 0.81.0 1.250.50 0.30.5 1.250.50 0.000.00 1.00+0.00
4/06 0.3+0.5 1.250.50 0.00.0 1.000.00 0.750.96 1.500.58
4/10 0.3+0.5 1.250.50 0.3+0.5 1.25+0.50 0.250.50 1.25+0.50
4/13 0.00.0 1.000.00 0.00.0 1.000.00 0.000.00 1.00+0.00
4/17 0.00.0 1.00+0.00 0.0+0.0 1.000.00 0.00+0.00 1.000.00
4/20 0.00.0 1.00+0.00 0.00.0 1.000.00 0.00+0.00 1.00+0.00
4/24 0.8+1.0 1.25+0.50 1.02.0 1.250.50 0.25+0.50 1.250.50
4/27 0.3+0.5 0.880.25 1.81.3 1.130.25 0.50+1.00 1.000.00
5/01 0.5+0.6 0.75+0.29 2.5+2.7 0.49+0.37 0.500.58 0.750.29
5/04 0.3+0.5 0.880.25 0.8+1.5 0.810.38 0.250.50 1.250.50
5/08 0.8+1.0 1.25+0.50 2.0+1.2 0.38+0.14 1.25+1.50 0.75+0.32
5/11 4.0+1.8 0.900.34 0.81.0 1.38+1.11 0.000.00 1.000.00
5/16 7.3+4.8 0.760.54 6.8+1.9 4.633.04 2.752.06 3.75+2.06
5/18 6.5+3.0 0.410.39 6.84.0 1.50+0.00 3.75+2.87 2.290.00
5/22 12.8+2.9 0.23+0.08 6.5+0.6 2.421.28 2.75+2.99 1.560.97
5/24 10.8+3.6 0.330.20 8.5+2.7 1.84+1.14 2.75+1.71 1.46+0.42
5/31 23.5+10.1 0.670.31 18.8+1.7 1.17+0.58 7.25+3.40 2.03+1.20
6/02 22.5+8.58 0.46+0.18 20.5+6.3 1.630.31 9.50+2.89 2.04+1.02
overall 4.2+7.12 0.920.34 3.65.8 1.350.84 1.51+2.51 1.360.68
Values reported as meanstd, n=4. Ratio=number females+l/number males+1. Overall mean averaged
over means of each date (n=22).











Table 3.14 Numbers and sex ratios of pepper weevil adults captured in bell pepper on
yellow and white sticky traps with seasonal averages given for different times of the day.
Bradenton, Florida 1989.


No. of pepper weevil adults/trap

Treatment n All adults Females Males Sex ratio*


Trap color:
yellow trap 384 0.81 a** 0.48 a 0.32 a 1.20 a
white trap 384 0.35 b 0.24 b 0.11 b 1.14 a

Time (hour): ***
1600 160 0.70 a 0.40 ab 0.26 a 1.14 b
1400 160 0.67 ab 0.47 a 0.21 a 1.27 a
1200 160 0.58 abc 0.37 ab 0.23 a 1.16 ab
1000 160 0.48 bc 0.29 b 0.20 a 1.14 b
1800 128 0.43 c 0.25 b 0.18 a 1.10 b

* Sex ratio = (no. females + 1)/(no. males + 1).
** Means within columns within treatment category followed by the same letter not
significantly different (P<0.05) using Duncan's New Multiple Range test.
*** Combined values for both yellow and white traps.










Table 3.15 Comparison of the numbers and sex of pepper weevil adults observed per plant
in an absolute sample, or terminal bud inspection, or captured per yellow or white sticky
traps in bell pepper. Bradenton, Florida 1989.


No. pepper weevil adults/sample

Type of sample Total Females Males Sex ratio**

Absolute plant sample 7.85+1.66*** 2.90+0.74 4.95+1.08 0.60+0.11
Terminal bud inspection 0.65+0.59 0.15+0.10 0.11+0.38 0.67+0.41
Yellow sticky trap 2.13+1.50 1.31+1.08 0.81+0.83 1.53+0.91
White sticky trap 0.69+0.79 0.44+0.63 0.25+0.45 1.25+0.66

* Absolute sample taken on the same plants as terminal bud inspection (sample unit=5
plants). The terminal bud inspection sample consisted of 2 buds per plant on 5 individual
plants. Yellow sticky traps and white sticky traps were inspected over an eight hour
period (sample unit=l trap).
** Sex ratio = (no. females + 1)/(no. males + 1).
*** Mean+std (n=4).




















0.3 -


0.2 -


0.1 -


-I.


7:00


9:00 11:00 13:00 15:00 17:00


Time of day








Figure 3.1 The numbers of pepper weevil adults observed per bell pepper plant parts
(fruiting buds, young leaves, old or fully mature leaves, and fruit greater than 2.5 cm in
diameter) at six times during the day averaged over a four-week sampling period.
Bradenton, Florida 1987. Parts in legend followed by the same letter are not significantly
different (P<0.05) using Duncan's New Multiple Range test on means averaged over all
hours of the day with sample dates as replicates.


S-- on buds a
S on young leaves b
i-- on old leaves c
on stems c
-- on fruit c






















on old leaves

on stems
on fruit

on young leaves
on buds


1 2 3 4 1 2 3 4


Week


Week


Figure 3.2 The numbers of pepper weevil adults observed on five plant parts of bell
pepper (fruiting buds, young leaves, old or fully mature leaves, and fruit greater than 2.5
cm in diameter) for morning and afternoon samples by week. Bradenton, Florida 1987.


































7:00 9:00


11:00


13:00


15:00 17:00


Time of day









Figure 3.3 The impact of time of day on the numbers of pepper weevil adults observed on
exposed and unexposed fruiting buds in bell pepper. Bradenton, Florida May 1987. Data
points represented with standard error bars (n=4).


0.6 8-1



0.6-1 T L --4 --


0.4-



0.2-


.1 *














exposed top A
exposed top
exposed bottom
exposed bottom


e
u
e
E u









1 2


10 11 12


exposed top
unexposed top
exposed bottom
unexposed bottom


1 2 3 4 5 6 7 8 9
123456789Weeks after transplant
Weeks after transplant


10 11 12


Figure 3.4 The numbers of fruiting pepper buds (A) and pepper weevil adults (B)
observed in exposed and unexposed portions of the plant in the top third and bottom two-
thirds of bell pepper plants. Bradenton, Florida spring 1988.


60-


40-


20-


3 4 5 6 7 8 9
Weeks after transplant


0.02
0.02 U1


0.01


I


/


80 -














300


200




100-


OF exposed top A
unexposed top
exposed bottom
unexposed bottom










1 2 3 4 5 6 7 8 9 10
Weeks after transplant


m
m



0.05


0.02


exposed top
unexposed top
exposed bottom
unexposed bottom


12


1 i 3 3 0 / l0

1 2 3 4 5 6 7 8 9 10 11 12
Weeks after transplant


Figure 3.5 The numbers of fruiting pepper buds (A) and pepper weevil adults (B)
observed in exposed and unexposed portions of the plant in the top third and bottom two-
thirds of Jalapefio pepper plants. Bradenton, Florida spring 1988.


8


6


4


2


I






























1 2 3 4 5
Weeks


6 7 8 9 10 11 12
after transplant


SI I I I I
---- top third/bottom 2 thirds B
exposed/unexposed
top third exposed/all










0 2 4 6 8 10 12 14


Weeks after


transplant


Figure 3.6 The proportion of fruiting buds (A) and proportion of pepper weevil (PW)
adults (B) between exposed and unexposed portions of the plant in the top third and
bottom two-thirds of bell pepper plants. Bradenton, Florida spring 1988.


* top third/bottom 2 thirds
B exposed/unexposed
* top third exposed/all


























01o


123


A

top third/bottom 2 thirds
exposed/unexposed
H top third exposed/all









4 5 6 7 8 9 10 12


Weeks after transplant


Weeks after transplant


Figure 3.7 The proportion of fruiting buds (A) and proportion of pepper weevil (PW)
adults (B) between exposed and unexposed portions of the plant in the top third and
bottom two-thirds of Jalapefio pepper plants. Bradenton, Florida spring 1988.









57



3- A


0.03
S2-
0.02

0.01
S1-

1 2 3 4 5 6 7 8 9 10


1 2 3 4 5 6 7 8 9 10 11 12
Weeks after transplant





0.08 B


= 0.06- o.oo00


V 0.04


z 0.02- o.oo0
3 4 5 6 7 8 9 10

0.00-
3 4 5 6 7 8 9 10 12
Weeks after transplant




Figure 3.8 The numbers of pepper weevil adults observed per top exposed fruiting bud for
bell (A) and Jalapeiio (B) pepper plants. Bradenton, Florida spring 1988.


























0 2 4 6 8 10
Weeks after transplant


0 2 4 6 8 10
Weeks after transplant






Figure 3.9 The numbers of pepper weevil adults observed by five techniques including
terminal bud inspection (TBI), inspection of 2 exposed terminal buds plus 2 unexposed
buds per plant (2+2), whole plant inspection (WPI), an absolute sample, and an intensive
yellow sticky trap sample (YST) in bell (A) and Jalapefio (B) peppers. Bradenton, Florida
spring 1988.


















A
300 400 500 600 700 800





SIrI "
------ 'y**- -~' Ly ~' '


4"white .- -., gr e
so M O/ ", / .- s /


80

S ]I I / i
'\
SV 1,i
I, I I ,I


SB'. IC I
I4
/- ". I/ i /
20 .- / /A / / / /
i *-!./ / / / /
"_.-, ,,\ // .. I-I i /!
/- ".. blue ."

black
300 400 500 600 700 800









Figure 3.10 Reflectance analyses for the materials used to construct sticky traps at
Bradenton, Florida. In the summer of 1988 yellow vinyl and colored poster board were
used (A). In the fall of 1988 and the spring of 1989 colored polyethylene plastic was used
(B and C).


















500


600


500 O00
Light Wave Length (Nm)


Figure 3.10 (continued)


300


400


700


800


u


40


0


S o

40





20









3.2 Spatial Dispersion of the Pepper Weevil.
Anthonomus eugenii Cano (Coleoptera: Curculionidae)


Sampling for the PW has recently received more research attention due to the

increased interest in action thresholds for the control of this pest in bell peppers. Currently,

sampling information specifically for the PW includes an evaluation of terminal bud

inspection (Andrews et al. 1986), investigations in the use of male-baited traps (Wilson

1986), sampling for PW damage (Cartwright et al. 1990), a comparison of relative

sampling techniques (Segarra-Carmona and Pantoja 1988a) and a preliminary dispersion

analysis fitting sample data to the negative binomial distribution (Segarra-Carmona and

Pantoja 1988b). The latter study was based on whole plant inspection and point samples

spaced at ten-plant intervals. Information is lacking on the spatial dispersion of PW adults

in commercial size fields.

The purpose of this study was to characterize in greater detail the physical

dispersion of PW adults in bell and Jalapefio pepper fields. Specifically, the objectives

were: 1) to investigate temporal effects on PW adult dispersion indices, 2) to evaluate the

impact of bud orientation on the dispersion estimates for PW adults, 3) to quantify PW

adult dispersion estimates for various visual sampling techniques, 4) to investigate

differences in the dispersion estimate due to sample quadrat size, and 5) to characterize

adult PW dispersion in large (>1 Ha) pepper fields.


3.2.1 Method and Materials

3.2.1.1 Small plot dispersion studies

Bell peppers, var. Jupiter, were transplanted February 20, 1987 into double rows

with 30 cm spacing between plants and between rows. The total area planted consisted of

two adjacent lands each with seven 1.52 m beds x 55 m length. The black plastic mulch










beds were fertilized with a single band of 18-0-25-2 at a rate of 2.98 Kg/10 linear meters of

bed (LMB) mixed with 0-21-0 at 1.12 Kg/10 LMB. The beds were fumigated with

methylbromide: chloropicrin (2:1) at 0.536 Kg/10 LMB and transplants were watered with

Nutrileaf 20-20-20 (R) at 600 g/100 1 water. Applications of a tank mix of tribasic copper

sulphate at 4.8 g/l, mancozeb (Manzate 200DF (R)) 1.8 g/l, and Bacillus thuringiensis

kurstaki (Dipel 2X (R)) 1.2 g/1 were made on a weekly basis from April 10 to May 29.

Sampling began on May, 4 1987, approximately six weeks after transplant.

Samples of 98 plants were taken six times per day (7:00, 9:00, 11:00, 13:00, 15:00, 17:00

hours) four days per week for four weeks for a total of 9,408 whole plant inspections.

Seven, single plant samples were selected systematically at intervals of 25 plants

(approximately 7.6 m) with a random starting point in each of 14 rows. The order of rows

sampled also was randomized. Adult PW were counted on all fruiting buds, young leaves

near terminal buds, mature leaves, stems, and fruit greater than two cm in diameter. In the

last week, the numbers of PW adults on buds exposed to a upper view and buds not

exposed (concealed within the plant canopy) were recorded separately. Taylor's Power

Law (TPL) coefficients were estimated for six time periods during the day, for each week,

and for certain plant parts (Taylor 1961). Differences in dispersion estimates were tested

using comparisons of slope and intercepts with Student t tests for independent regressions

and F tests of homogeneity of regression slopes (Steele and Torrie 1980, Gomez and

Gomez 1984). Variance to mean ratios and means also were compared using Duncan's

New Multiple Range test. The raw data for the whole plant sums by plant part are presented

in Appendix A.

In the spring of 1988 in the same location, bell peppers, var. Early Calwonder, and

Jalapefio peppers were transplanted on March 8 and 10 respectively. Each field of bell and

Jalapeiio peppers consisted of two adjacent lands of seven 1.52 m beds x 62 m length.










Cultural practices were similar to those for 1987 with the exception that Jalapefio peppers

were planted in single rows with double banded fertilizer and Plyac (R) spreader-sticker

was added to the tank mixes at 0.3 ml/1. Samples of 98 individual plants were examined in

the morning and afternoon twice per week for nine weeks. Samples consisted of seven

individual plants systematically selected at approximately 8.5 m intervals within each of 14

rows. A random starting point was selected at the beginning of each row and the order of

rows randomized. Pepper weevil adults and fruiting buds were enumerated for four

categories per plant: 1) exposed in the top third of the pepper plant, 2) unexposed in the top

third, 3) exposed in the bottom two-thirds of the pepper plant, and 4) unexposed in the

bottom two-thirds. TPL coefficients were estimated for morning and afternoon samples and

by bud position. Differences in dispersion estimates were evaluated using Student t tests

for differences between two independent regression slopes and Duncan's New Multiple

Range analysis on means and variance to mean ratios (Steele and Torrie 1980, Gomez and

Gomez 1984).

In 1989, bell peppers, var. Early Calwonder, were transplanted on March 10 in a

single land of six 1.8 m beds x 91 m length. Cultural practices were similar to those for

1987 with the exception of a single application of methomyl (Lannate L (R)) at 3.5 1/Ha

made on April 10 for an armyworm infestation exceeding a threshold level of one larva per

six plants. A sampling comparison was done for terminal bud inspection, whole plant

inspection, and an absolute method, on an individual plant basis so that plant to plant

variability could be estimated for the relationship between these sampling methods. Rows

were randomly selected out of 12 possible rows (two rows per bed). Then, individual

plants were selected using a systematic design (samples at 7.6 m intervals) with a random

starting point within each row. Nine data sets (40, 20, 20, 15, 15, 10, 10, 5 and 5

individual plant samples) were taken over a seven-week period. The absolute method was

modified so that all three methods could be performed on the same pepper plant. The










terminal buds were inspected first and PW adults collected into a vial with minimal

disturbance of the plant. Then, a 36 cm diameter x 62 cm height Lexan (R) clear plastic

cylinder with a 36 mesh gauze top cover (with access openings for the hands) was placed

over the plant, and pressed firmly down upon the bed. An approximate two min. whole

plant search was then conducted collecting PW adults into a separate vial as they were

encountered. Individual plant parts were then clipped and inspected for PW adults. Finally,

the plastic mulch and soil was inspected for fallen PW adults. The sum of the three

methods provided the absolute estimate. TPL coefficients were estimated for each of the

sampling methods and comparisons were made using Student t tests and Duncan's multiple

range analysis (Steele and Torrie 1980, Gomez and Gomez 1984).

3.2.1.2 Large plot dispersion studies

Whole plant inspection of individual pepper plants in three commercial fields was

employed to evaluate the dispersion of PW adults on a large acreage basis in 1988. A

systematic sample with whole plant inspection of individual plants was employed. Since

only one PW adult was encountered over the entire course of these trials, the description is

omitted. However, due to the excessive time spent on individual plants with whole plant

inspection, terminal bud inspection was chosen over whole plant inspection in 1989 so that

more plants could be sampled per unit search time.

Extensive sampling was conducted in three commercial pepper fields in Manatee

and Palm Beach counties in 1989. In Palm Beach County, Green Cay farm encompassed

143.6 acres of bell pepper, primarily var. Jupiter, and approximately 13.9 acres were in

designated sample areas (block 1 was at first harvest maturity and block 2 was at post

harvest maturity) (Figure 3.11 A). Also in Palm Beach County, Hayes farm consisted of

2.2 acres of Jalapefio peppers and, of this, 2.0 acres were included in the sample area

(Figure 3.11 B). In Manatee County, the Baum farm encompassed 112.5 acres of bell

pepper, var. Early Calwonder. Of this, ten sample areas in the field totaling approximately










12.9 acres were sampled (Figure 3.11 C). At the Green Cay and the Baum farms,

contiguous strips of 25 plants spaced at approximately 15 meter intervals were sampled

systematically with random starting points within rows. At Hayes farm, contiguous strips

of 25 plants were sampled at approximately nine meter intervals to accommodate more

samples in the smaller acreage. The number of PW adults was recorded for each five-plant

interval in the 25 plant sample so that quadrat size could be evaluated in terms of TPL

coefficients. Samples were taken during the middle of the day on most sample dates. The

total numbers of sampling dates in 1989 for Hayes, Green Cay and the Baum farms were

1, 3, and 10, respectively.


3.2.2 Results and Discussion

3.2.2.1 Small plot dispersion studies

Time of day did not significantly affect spatial dispersion of whole plant counts of

PW adults based on a comparison of TPL coefficients or variance to mean ratios utilizing a

simple t test for each time period (Table 3.16). Variance to mean ratios increased with PW

population density as expected (week 9 versus weeks 6-8), but TPL coefficients remained

stable over time. Taylor (1984) discusses in detail the reasons for this stability in the

estimate of "b" over changes in population density.

The estimate of slope "b" and the variance to mean ratio were significantly lower

for the whole bud sample than for the whole plant sample (t=2.54, df=96, P<0.05) (Table

3.16). This may have been related to the greater range of PW numbers on a whole plant

basis than on a whole bud basis (0-9 and 0-6 respectively). Although more adults were

observed in the morning than in the afternoon on exposed terminal buds in 1987, the

variance to mean ratios and the estimates for "b" were not similarly affected (Table 3.16).

Bud position did not affect variance to mean ratios, but did affect the estimates of

"b" in 1988 (Table 3.17). The estimate of "b" for the top exposed bud position was










significantly lower than that for the other bud positions in bell pepper. In Jalapeiio pepper,

the estimate of "b" for the top exposed bud position was significantly higher than for other

bud positions (Table 3.17). The greater variability of PW counts in Jalapeiio peppers than

in bell peppers (discussed in Section 3.1) may have contributed to the difference in these

trends.

The adult PW counts in the 0.2 Ha bell pepper plots in 1987 and 1988 were slightly

clumped ("b">l) for all of the data criteria in both years (Tables 3.16 and 3.17). However,

the total numbers PW adults observed per sample site late in the season in 1987 (Figure

3.12) and early in the season in 1988 (Figure 3.13) gave no indication of strong clumping

patterns on a small field basis. Three factors may have contributed to the apparent low

degree of clumping of adults in these small experimental plots. First, the single plant

inspections restricted the number of PW adults to the observed maximum of nine PW

adults per sample, reducing the numerical impact of sample count frequencies. This

restriction was even greater for terminal bud inspection where counts rarely exceeded one

or two weevils per bud. The apparent low degree of clumping of PW adults may have been

partly an artifact of the sampling technique. Secondly, the PW adult population found in the

small experimental plots could represent the population in margins or borders of larger

fields. Thirdly, the small field size and the observed movement of adults could result in

multiple counts of the same adults. It was noted that pepper plants observed to have PW

adults present in the morning, flagged, and rechecked in the afternoon, were consistently

lacking in weevils in the afternoon (Riley, unpublished data). This indicated that there was

substantial plant-to-plant movement of PW adults during the day.

As expected, the absolute sample provided a significantly higher mean adult count

than either relative sampling method (Table 3.18). However, the absolute sample also

provided a higher estimate of "b" and a higher variance to mean ratio than terminal bud

inspection (P<0.05), which was not expected. Elliott (1983) states that the sampling








67

method affects the estimate of "a" and that quadrat size may affect the estimate of "b". The

methods of sampling, which in this case consisted of increasing portions of the plant being

sampled, may be considered as increasing quadrat sizes within the plant structure. If we

consider these methods as different "quadrats" of the within-plant sample universe, then the

quadrat size approaching clump size would produce the greater sample variation (Elliott

1983) and therefore the higher estimate of "b". The whole pepper plant is a natural clump

size for PW adults due to its spatial separation from other plants. Therefore, the absolute

sample should produce greater sample variation and a higher estimate of "b" than sampling

two terminal buds (essentially a small, within-plant quadrat). This effect on the estimate of

"b" may also partly explain the differences observed between estimates for "b" for distinct

bud positions (Table 3.17). The reason the estimates for "b" for terminal bud inspection

were lower than other estimates in this section was attributed to biased selection of plant

samples with PW adults which caused to the dispersion to appear more even.

3.2.2.2 Large plot dispersion studies

In a bell pepper planting at Green Cay farm, a higher concentration of PW adults

was observed along the outer field margins than in the inner field of block 1 (Figure 3.14

A). Forty percent of all the adults were observed in the outer 73 m of both ends of the 730

m length of block 1. Twenty four percent of the adults were observed in the outer 37 m,

which comprised only ten percent of the total area. Block 2, on the west side of the farm,

had a greater density of PW adults than block 1 (Figure 3.14 A). This was because

insecticide applications had not been made for several weeks in block 2, whereas an

application of permethrin (Ambush (R)) had been made within a week of the sample date in

block 1 (see Figure 3.15 A for orientation of blocks on farm). The concentration of PW

adults in the outer field margins was not as distinct in block 2. This may have been related

to the stage of maturity of the pepper planting and duration of PW infestation before the

time of sampling. There was no significant difference in the means across rows in either










block (Figure 3.14 B). Nightshade plants infested with PW adults were observed along the

outer margins of block 1 at the end of the season, after the crop had been harrowed. It was

speculated that these sources of PW could have contributed to the border effect previously

observed.

In a small acreage of Jalapefio peppers, there was a higher concentration of PW

adults on the east end of block 2 (samples 1-5 in Figure 3.15 A). Also, there tended to be

greater numbers of PW adults observed on the north end of block 2 (rows 14-17 in Figure

3.15 B). The mean number of adults increased from blocks 1 to 3 (0.03+0.19, 0.47+0.83,

and 0.56+0.73 respectively) indicating a progression of infestation across the field (Figure

3.11 B).

The contiguous plant samples at Green Cay farm and Hayes farm (5, 10, 15, 20,

and 25 plants) represented linear quadrat sizes along the row lengths. The variance to mean

ratios for 20 and 25 contiguous plants was higher than for 5 and 10 contiguous plant

samples (P<0.05) for bell pepper (Table 3.19). This was expected since variance to mean

ratios tend to increase with the mean. The smallest linear quadrat size of five contiguous

plants in bell pepper (Table 3.19) resulted in an estimate of "b" (1.16) that was not

significantly different from the overall estimate of "b" (1.12) in 1987 for individual whole

plant inspection (t=0.59, df=104).The estimate of "b" for 25 contiguous plants was

significantly higher than the estimate of "b" for 5 plants (P<0.05). This occurred despite

the observation by Taylor (1984) that the estimate of "b" should remain stable over a range

of means. The effect of quadrat size on dispersion indices may explain this effect. Elliott

(1983) stated that maximum variance occurs when quadrat size and clump size are

approximately equal. It follows that "b" would also be greater when estimated from the

quadrat size approaching the clump size. Thus, the PW clump size along row length might

be on the order of 25 plant strips with 15 m between strips or greater (Table 3.19). In the

smaller Jalapefio pepper field (Figure 3.11 B), no increase in "b" with quadrat size was










detected (Table 3.19). This may have been due to the shorter distances between sample

strips (9 m) in the Jalapeiio pepper or clumping may not have been apparent in the smaller

acreage in which these Jalapelio peppers were planted.

At the Baum farm, there were significant numbers of PW adults observed in only

two field blocks (block 2: t=2.73; df=500; P<0.01and block 7: t=2.25; df=1815; P<0.05)

indicating a very low population density (Table 3.20). Also at the Baum farm, 71% of the

adults encountered were located within 15 m of the field margins indicating a concentration

of adults in the outer field margins. Blocks 2 and 7 were the only locations where PW

adults occurred on more than one sampling date. Higher concentrations of PW adults

seemed to occur in block 2 in the previous year (Agricultural Pest Management Inc. 1989,

personal communication). This re-occurrence of PW may have been associated with

nightshade plants sites near the field margins in block 2.

The observations on the physical dispersion of PW adults in commercial fields

suggests that scouting plans may benefit from concentrating on field margins, so as to bias

their estimates to be as sensitive as possible for PW detection Adequate randomization of

samples in all field margins should be conducted to identify these "hot spots" of PW

activity and to provide greater sensitivity in the PW detection. However, sample locations

based solely on the previous season's concentrations of PW adult populations may be

misleading. The effect of linear quadrat size suggests that a given number of bell pepper

plants sampled with the smaller sample size of 5 contiguous plants may incur less sample

variation than with the larger sample size of 25 plants.



3.2.3 Summary and Conclusion

The Taylor's Power Law dispersion estimate "b" was affected by the plant part

sampled, the type of sample, and quadrat size. Estimates of "b" were not affected by

temporal variations in PW population density which concurred with the prediction by


I










Taylor (1984). Quadrat and field size can affect spatial dispersion estimates (Elliott 1983),

but the type of sample is generally thought to affect only the estimate "a" or the regression

intercept. The difference in the dispersion index "b" between plant parts and sampling

methods may have been related to the quadrat effect on dispersion estimates discussed by

Elliott (1983). Also, the movement of PW adults within the plant canopy and plant

morphology may have contributed to the apparent differences in adult clumping observed

between plant structures.

The sampling in commercial fields revealed two phenomena associated with spatial

dispersion of PW adults. First, the number of PW adults observed on the smaller strips of

five contiguous plants displayed less sample variation than the longer strips of contiguous

plants. Therefore, taking smaller individual strip samples for a given number of plants may

reduce sample variation. Secondly, PW adults tended to concentrate on the field margins in

the larger commercial acreage. This second observation coincided with the occurrence of

nightshade plants along the outer field margins, which may have contributed to the apparent

border effect. Thus, allocating more samples in the field margins and taking some samples

in nightshade plants along field margins could increase the probability of detection of PW

adults. The practice of concentrating samples along field margins was observed on several

occasions while watching scouts from Glades Crop Care Inc., and Agricultural Pest

Management Inc. Also, scouting of nightshade plants around pepper fields before the

growing season was done by Glades Crop Care Inc. to detect early infestations of PW

adults (Charles Mellenger 1989, personal communication). As previously mentioned, we

have found PW adults and immatures along old pepper field margins between growing

seasons on various occasions. Thus, these practices should be incorporated into

commercial pepper field scouting plans that include PW as an important pest.










Table 3.16 Impact of time and plant part on spatial dispersion estimates of pepper weevil
adults in a 0.2 Ha bell pepper plot. Bradenton, Florida 1987.


Taylor's Power Law coefficients

Data criteria Slope ("b") Intercept ("a") s2/


Time of day (n=64)
07:00 1.10 a -0.01 a 0.32 a 1.02 a
09:00 1.06 a 0.00 a 0.34 a 1.02 a
11:00 1.16 a 0.02 a 0.37 a 1.03 a
13:00 1.15 a -0.02 a 0.37 a 1.02 a
15:00 1.13 a -0.00 a 0.36 a 1.03 a
17:00 1.02 a 0.01 a 0.36 a 1.02 a

Weeks (n=96)
after transplant
6 weeks 1.15 a -0.01 a 0.09 a 1.06 a
7 weeks 1.07 a -0.00 a 0.14 a 1.06 a
8 weeks 1.08 a -0.00 a 0.25 b 1.08 a
9 weeks 1.11 a -0.01 a 0.92 c 1.14 b

Plant part (n=96)
whole plant 1.12 a -0.01 a 0.56 a 1.04 a
whole bud 1.05 b 0.00 a 0.29 b 1.01 b

Exposed bud (n=12)
morning 1.36 a -0.05 a 0.20 a 1.07 a
afternoon 1.22 a -0.02 a 0.11 b 1.02 a

* Taylor's Power Law slope and intercept values within data criteria and within columns
followed by the same letter not significant at P<0.05 using Student t test. Mean (per
plant) and variance/mean values within data criteria and within columns followed by same
letter not significant at P<0.05 using Duncan's New Multiple Range test.










Table 3.17 Summary statistics for the spatial dispersion of pepper weevil adults by crop
(bell and Jalapefio pepper) and bud position. Bradenton, Florida 1988.


Taylor's Power Law coefficients

Bud position (n=21) ** Slope ("b") Intercept ("a") 7 s2/


Bell pepper
top exposed 1.09 a 0.01 a 0.35 a 1.03 a
top unexposed 1.19 b -0.00 a 0.29 ab 1.03 a
bottom exposed 1.29 bc -0.00 a 0.03 b 1.01 a
bottom unexposed 1.53 c -0.02 a 0.12 ab 1.04 a

Jalapefio pepper
top exposed 1.33 a -0.02 a 0.29 a 1.02 a
top unexposed 1.19 b -0.00 a 0.18 ab 1.01 a
bottom exposed 1.09 be 0.00 a 0.09 b 1.01 a
bottom unexposed 1.08 c 0.01 a 0.14 ab 1.01 a

* Taylor's Power Law slope and intercept values within data criteria within columns
followed by the same letter not significant at P<0.05 using Student t test, mean (per plant
per position) and variance/mean values within data criteria within columns followed by
same letter not significant at P<0.05 using Duncan's New Multiple Range test.
** Terminal fruiting buds visible from above (exposed) or concealed within the plant
canopy (unexposed) for the top third and bottom two-thirds of the plant.










Table 3.18 Summary statistics for the spatial dispersion of pepper weevil adults for bell
pepper using three visual sampling methods. Bradenton, Florida 1989.


Taylor's Power Law coefficients

Sample method Slope ("b") Intercept ("a") x s2/


Terminal bud inspection 0.58 a 0.04 a 0.38 a 0.80 a
Whole plant inspection 0.79 a 0.12 a 1.98 b 1.15 ab
Absolute sample 1.29 b 0.16 a 5.16 c 2.50 b

* Taylor's Power Law slope and intercept values within columns followed by the same
letter not significant at P<0.10 using Student t test, mean (per plant) and variance/mean
values within columns followed by same letter not significant at P<0.05 using Duncan's
New Multiple Range test.










Table 3.19 Means, variances, and variance/mean ratio for pepper weevil (PW) counts by
field block for bell pepper at Green Cay farm (Palm Beach Co.) and the Baum farm
(Manatee Co.) and for Jalapeiio pepper at Hayes farm (Palm Beach Co.) with Taylor's
Power Law coefficients calculated over all 3 farms.


Field Block No. plants PW 7 PW s2 s2/


Green Cay farm 1 8000 0.294 0.456 1.55
S2 2000 0.121 0.146 1.21
Hayes farm 1 2700 0.032 0.034 1.09
2 3650 0.473 0.681 1.44
3 250 0.560 0.537 0.96
Baum farm 1 4000 0.000 0.000
2 9075 0.003 0.003 1.00
3 1375 0.000 0.000
4 2475 0.000 0.000
5 4400 0.000 0.000
6 1300 0.000 0.000
7 2500 0.018 0.022 1.21
8 1050 0.000 0.000
9 850 0.000 0.000

Log Variance = 1.17 Log Mean + 0.003; R2=0.96; P<0.001.

* Refer to Figure 3.11 for location of individual blocks, means (per plant) calculated over
the entire sample period for each location. In all locations terminal bud on 5 contiguous
plants (2 buds/plant) were systematically sampled with a random starting point in each
row. # plants = total numbers of pepper plants sampled per location.










Table 3.20 Summary statistics for the spatial dispersion of pepper weevil adults in bell
pepper (Green Cay Farm) and Jalapeiio pepper (Hayes farm) using five linear quadrat sizes
(5, 10, 15, 20, 25 contiguous pepper plants).


Taylor's Power Law coefficients

Quadrat size (n=12) Slope ("b") Intercept ("a") 7 s2/7


Bell pepper
5 plants 1.16 a -0.01 a 0.49 a 1.15 a
10 plants 1.33 ab -0.02 a 1.02 ab 1.34 a
15 plants 1.44 ab -0.07 a 1.66 bc 1.51 ab
20 plants 1.42 ab -0.04 a 2.24 cd 1.75 b
25 plants 1.47 b -0.07 a 2.82 d 1.84 b

Jalapeiio pepper**
5 plants 1.33 a -0.00 a 0.32 a 1.41 a
10 plants 1.24 a -0.01 a 0.59 b 1.35 a
15 plants 1.25 a -0.01 a 0.89 c 1.31 a
20 plants 1.17 a -0.01 a 1.24 d 1.34 a
25 plants 1.21 a -0.01 a 1.67 e 1.45 a

* Taylor's Power Law slope and intercept values within data criteria within columns
followed by the same letter not significantly different (P<0.05) using Student t test,
mean and variance/mean values within data criteria within columns followed by same
letter not significantly different (P<0.05) using Duncan's New Multiple Range.
** Distance between samples in Jalapefio pepper (approx. 9 m) was smaller than the
distance between samples in bell peppers (approx. 15 m) to accommodate more samples
in the smaller field.
























*
U


cu
C3

ctt


6
r.
Fi

U
E

CQ


s
ge
B~e3sc


CI~
' ~ ~ ~ ~ ~ ~ ~ ~


Figure 3.11 Field maps indicating blocks sampled for pepper weevil adults at Green Cay
Farm (A), Hayes Farm (B), and the Baum Farm (C) in 1989.


2400 ft.










S I
0 0
I-a


I--i-ili-.-i-i..
II!III












Hayes Farm, Palm Beach Co.

(2.2 acres total, approx. 2.0 acres sampled) B.





100 ft.
fallow









Saum Farm, Manatee Co.
112.5 acres total, approx. 12.9 acres sampled) I.


'3U IL.=


Figure 3.11 (continued)


B
(


"















Mean no. adultl/plant


1 2 3 4 5 6 7 8 9 10 11 12 13 14
Row number


Mean no. adults/plant


12 3 4 56 7 8 9 10 11 12 13 14
Row number


Figure 3.12 The numbers of pepper weevil adults per sample site in bell pepper at 8 weeks
(A) and 12 weeks (B) after transplant. Bradenton, Florida spring 1987. Each sample site
consisted of a single plant selected at 8.5 m intervals in each of 14 rows.














Mean no. adults/plant


1 2 3 4 5 6 7 8 9 10 11 12 13 14
Row number


1 2 3 4 5 6 7 8 9 10 11 12 13 14
Row number


Figure 3.13 The numbers of pepper weevil adults per sample site at 6 weeks (A) and 9
weeks (B) after transplant in bell pepper. Bradenton, Florida spring 1988. Each sample site
consisted of a single plant selected at 8.5 m intervals in each of 14 rows.















Mean no. adults/plant
3
block 1 5 block 2
.5
A
2

.5


".5


1 6 10


15 20 25 30 35 40
Sample number


1 block 1 M block 2


B


1 2 3 4 5 8 7 8 9 10
Row number


Figure 3.14 The mean numbers of pepper weevil adults across samples (A) and across
rows (B) at Green Cay Farm. Palm Beach Co. spring 1989. See Figure 3.11 for locations
of blocks, samples and rows.


1










81




Mean no. adults/plant

SI block 1 block 2 O block 3


A
0.8 A


0.6


0.4


0.2

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Sample number







Mean no. adults/plant

block 1 block 2 0 block 3

B


0.8-

0.6-

0.4-

0.2 -

0-
1 2 3 4 5 6 7 8 9 1 1 13 14 15 16 17
Row number












Figure 3.15 The mean numbers of pepper weevil adults across samples (A) and across
rows (B) at Hayes Farm. Palm Beach Co. spring 1989. See Figure 3.11 for locations of
blocks, samples and rows.















CHAPTER 4
ACTION THRESHOLDS FOR THE PEPPER WEEVIL


4.1 Threshold Applications of Cypermethrin. Fenvalerate.
Methomvl and Oxamvl for Control of Anthonomus eugenii
and Other Insect Pests of Peppers in Honduras.

The PW is a common pest of Jalapeiio and bell peppers in Central America,

Mexico, the southern United States and on various Caribbean islands (Wilson 1986). Its

status as a pest in many areas of Honduras has stimulated interest in control efforts which

have been based primarily on the use of insecticides. Recent efficacy trials have identified

insecticides for controlling the PW (Cartwright et al. 1986, Ozaki and Genung 1982,

Rolston 1977, Schuster and Everett 1982, Schuster 1983, 1984, 1986, 1987). Action

thresholds using visual sampling techniques for PW adults have been developed for timing

insecticidal applications (Andrews et al. 1986, Segarra-Carmona and Pantoja 1988b).

However, these thresholds have not been validated for different insecticides with

potentially different residual effects, efficacies and/or modes of action.The objective of

these studies was to compare the efficacy of four insecticides for PW control under

calendar and threshold applications in Honduras.


4.1.1 Method and Materials

Experiment 1. Bell peppers, var. Agron6mico, were transplanted on June 8, 1988

at the Escuela Agricola Panamericana in Honduras. Plant spacing was 30 cm and row

spacing was 150 cm. Treatment plots consisted of three rows by five meters (22.5 m2) and

were arranged in four randomized complete blocks. The total experimental plot dimension











was six rows by 70 m. One untreated buffer row of Jalapeiio peppers bordered one entire

side of the experimental plot and a partial row of untreated pepper, var. Agron6mico,

bordered the other side. These untreated rows provided a reservoir of PW for re-infestation

to ensure heavy feeding and ovipositional pressure.

Cultural practices in the first experiment consisted of three hand weedings on June

28, July 12, and July 26. Mancozeb (Manzate (R)) was applied at 1.2 Kg/Ha on June 27

and July 1; and mancozeb (Dithane (R)) was applied at 1.2 Kg/Ha on July 8 and July 27.

Bacillus thuringensis kurstaki (Bt) (Dipel 2X (R)), was applied on July 5 and July 12 to

suppress an early infestation of lepidopterous larvae. Ditch irrigation was employed as

needed. The average rainfall was 4.46 mm/day. The average high and low temperatures

were 29.10C and 19.40C, respectively.

Treatments for PW control began on June 29. Calendar applications were made

weekly and threshold applications were made when an action threshold of 1 adult/100

terminal buds (Andrews et al. 1986) was equalled or exceeded. Application rates for each

of the insecticides were: cypermethrin (Arrivo 200 CE (R)) 60 g AI/Ha, oxamyl (Vydate

240 CE (R)) 750 g AI/Ha, and fenvalerate (Belmark 100 CE (R)) 150 g AI/Ha. The rate of

fenvalerate was reduced to 100 g AI/Ha after the first application (June 29) because

phytotoxicity in the form of severe yellowing of the terminal buds was evident one day

following the application. Treatments were made with a CO2 pressurized backpack sprayer

using a single TGL solid cone nozzle (Spraying Systems Co.) on a single boom. Spraying

pressure was maintained at 2.46 Kg/cm2 with an application ground speed of 0.56 m/sec

and a boom height of 25-35 cm (delivery approx. 200 1/Ha).

Twice weekly sampling began on June 27 and lasted until a premature termination

of the field trial due to an accidental insecticide application on July 19. Counts were taken

on 14 plants within the central and outside row of each treatment plot with a one meter











buffer between plots. Using the simple visual inspection technique of Andrews et al.

(1986), PW adults were counted on two terminal buds per plant at 07:00-09:00 hours

without disturbing the plant. After completing the terminal bud inspection, a more rigorous

whole plant inspection was conducted on the same plants. The following arthropods were

counted in the whole plant inspection: noctuid larvae (Spodoptera sunia Guen., S. exigua

(Hubn.) and others), pyralid larvae, curculionid adults (PW and others), chrysomelid

adults (mainly Diabrotica spp. and Colaspis sp.), aphids, and various predators including

forficulids (D1ru sp.) coccinelid adults and immatures (Hippodamia convergens Gudrin,

Coleomegilla maculata Deg. and others), chrysopid larvae, staphylinid adults, carabid

adults, and spiders. Aphid counts were based on the number of plants visibly infested with

aphids whereas all other arthropods were counted individually. Arthropods were identified

using the keys and pictorial information in King and Saunders (1984). Fallen pepper buds

were collected from five meters of row on July 21 and the number of PW immatures within

the buds was recorded.

Experiment 2. Instead of bell pepper, Jalapefio pepper seedlings were transplanted

in the first week of July in a single row per bed with 23 cm between plants. Treatments

were the same as those in experiment 1 except that methomyl (Lannate L (R)) at 1 Kg/Ha

was included. Treatment plots each consisted of four rows (80 cm apart) by seven meters

for a total of 22.4 m2. Treatments were arranged in a split plot design with schedules of

treatments as whole plot and insecticides as subplots. The sampling period lasted from

August 12 to September 21 and only PW, insect predators, and Spodoptera spp. were

monitored. Due to inclement weather only one harvest was taken on September 5, and data

were taken only on the following harvest categories: number and weight of marketable

fruit, number of PW infested fruit, number of fallen fruit, and weight of non-marketable

fruit due to causes other than PW damage. Due to the diminutive yield, an economic

analysis was not included for the Jalapefio pepper harvest.











For both experiments homogeneity of variances and a non-additivity test were

employed to indicate the need for transformations prior to the analysis of variance.

Multiple comparisons for treatment means were made using orthogonal contrasts and

Duncan's New Multiple Range test Simple correlations were calculated for all dependent

variables, but only significant correlations are reported. All means were reported in the

original scale on a per treatment plot basis. Analysis reported as "overall" signifies that

insect counts from individual sampling dates were summed by plot and the analysis was

preformed on the sums. Means are reported on a per plot basis.


4.1.2 Results and Discussion

Experiment 1. Pepper weevil adults were detected on the first sampling date, June

26, just three weeks after transplanting (Figure 4.1). By the second week of sampling,

lepidopterous larval numbers increased, followed in the third week by predator numbers

and the numbers of plants infested with aphids. The impact of insecticide treatments on

these insect groups was evaluated in addition to treatment effects on PW numbers.

The number of PW adults on terminal buds reached the threshold level for all

insecticides on June 28 and July 18. The treatment effects on PW numbers were evaluated

in terms of whole plant samples to exploit the higher per plant counts associated with this

sampling method. Overall, the number of PW adults per plant, based on whole plant

inspection, was significantly lower in the threshold (1.7) and calendar (2.2) treatments than

in the control (4.5), both inclusively (F=5.32; df=l, 18; P<0.05) and separately (F=5.60;

df=l, 18; P<0.05 and F=3.80; df=l, 18; P<0.07, respectively). Overall, there was no

significant difference between the calendar and the threshold treatments in numbers of PW

adults per plant (F=0.35; df=l, 18). After June 28, PW numbers were consistently higher

in the control than in the calendar or threshold plots, but significantly higher only on July 4

and July 11 (F=15.6; df=l,18; P<0.01 and F=4.84; df=1,18; P<0.05, respectively)











(Figure 4.2). There was no significant difference between threshold and calendar

treatments of any insecticide in the number of PW adults per plant (Table 4.1).

Furthermore, PW adult numbers from whole plant inspection were lowest for threshold

applications of oxamyl and fenvalerate. Thus, no advantage in PW adult control resulted

from the greater number of insecticide applications in the calendar treatment compared to

the threshold treatment. This further supported the use of action thresholds over the use of

weekly insecticide applications.

No significant difference was noted between treatments in terms of PW immature

numbers in fallen buds. The oxamyl calendar treatment resulted in the fewest PW

immatures numerically, but the difference between oxamyl and other insecticides was not

statistically significant (Table 4.1). The mean+std (n) number of fallen buds with PW

immatures collected on July 21 for each insecticide averaged over both calendar and

threshold plots were: control = 6.50+11.03 (4), fenvalerate = 4.50+10.39 (8),

cypermethrin = 4.25+6.67 (8), and oxamyl = 1.75+4.17 (8). Hymenopterous PW

parasitoids (probably Catolaccus hunter Crawford) were found only in the oxamyl-treated

plots. These parasitoid specimens were collected only from fallen pepper buds less than 2

cm in diameter.

Noctuid larvae were the most abundant large arthropods encountered in the

experimental plots. Spodoptera exigua and S. sunia caused considerable defoliation in

localized areas of the pepper field early in the season, but were partially controlled by the Bt

applications on July 5 and 12 (Figure 4.3). Overall, the only significant difference in

numbers of noctuid larvae was between fenvalerate and oxamyl calendar treatments (Table

4.2). On July 4 (Figure 4.3) there were significantly greater numbers of larvae in the

control than in both calendar and threshold treatments (F=15.4; df=1,18; P<0.01). Also on

this date, the control and calendar oxamyl treatments had significantly greater numbers of

noctuid larvae than both fenvalerate treatments and the threshold cypermethrin treatment











(Duncan's New Multiple Range test; P<0.05). These results indicate that the fenvalerate

and cypermethrin treatments provided control of noctuid larvae in addition to PW control.

The numbers of plants infested with aphids were significantly lower in plots treated

with fenvalerate or oxamyl compared to plots treated with cypermethrin regardless of the

treatment schedule (Table 4.2). The frequency of aphid infestations increased dramatically

after July 13 (Figure 4.4 A). Overall, greater numbers of predators were observed in the

threshold plots (6.0) and in the control plots (6.0) than the calendar plots (2.0) (F=22.0;

df=l,18; P<0.01 and F=11.01; df=1,18; P<0.01, respectively). There was no significant

difference in the numbers of predators in the control versus the threshold plots. The

number of predators, consisting primarily of Coccinelid larvae and adults, increased

dramatically with the increase in aphid numbers (Figure 4.4 B). Predators and aphids were

significantly correlated in the control plots (Correlation coefficient R = 0.68; P<0.001);

however, there were no significant correlations between PW adults or lepidopterous larvae

and predators. Although there were greater numbers of predators in the threshold

treatments, this probably would have no effect on PW numbers since few instances of PW

predation have been observed. Predators were observed only attacking aphids and

lepidopterous larvae (Riley, unpublished data). In threshold plots, greater numbers of

predators were found in cypermethrin treatments than in fenvalerate or oxamyl treatments

(each consisted of two applications), suggesting that cypermethrin was less detrimental to

predators (Table 4.2). Observed numbers of chrysomelids and weevils other than PW were

not affected by the insecticide treatments.

Experiment 2. The only significant difference observed in the number of arthropods

between the insecticide schedules was that coccinelids were reduced by calendar insecticide

applications (Table 4.3). All insecticides resulted in fewer PW adults relative to the control.

Cypermethrin significantly reduced the number of spiders by 70% compared to the control,

but no insecticide significantly affected the number of coccinelids. There were more











Noctuid (Spodoptera spp.) larvae observed in plots sprayed with fenvalerate than the

control plots (Table 4.3). Therefore, fenvalerate induced a noctuid infestation, possibly

through a reduction of predator numbers.

Fewer fruit damaged by PW were harvested from plots treated weekly by

insecticide than plots treated according to a threshold (Table 4.4). However, the numbers

of fallen and undamaged fruit were not affected. Threshold plots received four fewer

applications of insecticide than the calendar plots. Though not statistically significant,

threshold plots produced 16% more marketable fruit weight than the calendar plot (Table

4.4). Since there were no significant differences in marketable yield and thresholds save

insecticide applications, using the threshold resulted in a net savings in insecticide cost.

None of the insecticides significantly affected the numbers of PW damaged fruit or

undamaged fruit. Oxamyl appeared to stimulate growth in some plots whereas fenvalerate

exhibited phytotoxicity at the higher rate. Nevertheless, no significant differences in

marketable yield among insecticide treatments were observed.


4.1.3 Summary and Conclusions

In bell peppers, early season PW infestation was adequately controlled by oxamyl,

fenvalerate and cypermethrin when applied according to the action threshold of 1PW/100

terminal buds. Season long effects could not be evaluated since the experiment was

terminated prematurely. Both fenvalerate and oxamyl provided better aphid control than

cypermethrin. Only oxamyl allowed parasitism of PW immatures. Calendar applications of

all insecticides depressed predator numbers significantly compared to the control; however,

threshold applications had no such effect.

For Jalapeio peppers, all of the compounds tested performed equally well in PW

control. However, cypermethrin may have been more detrimental to spiders. Also, calendar

applications significantly reduced coccinelid numbers. Threshold applications based on