INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy subm itted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.
Photographs, included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
Bell & Howell Information and Leaming 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 800-521-0600 UMI“
BIOLOGY, ECOLOGY, AND DECISION RULES FOR CARROT WEEVIL, LISTRONOTUS OREGONENSIS (LECONTE) (COLEOPTERA: CURCULIONIDAE) MANAGEMENT IN PARSLEY IN THE GREAT LAKES REGION
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
The Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By
Angel Nolberto Torres, M.S.
*****
The Ohio State University 2001
Dissertation Committee: Approved by Casey Hoy, Adviser
David Horn
Celeste Welty Adviî Department of Entomolog Mark Bennett UMI Number: 9999449
UMI
UMI Microform 9999449 Copyright 2001 by Bell & Howell Information and Leaming Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.
Bell & Howell Information and Leaming Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 Copyright by Angel Nolberto Torres 2001 ABSTRACT
Carrot weevil adults infest early-planted parsley fields in spring, between mid
May and the end of June, but most activity is during the first half of June. Adults start laying eggs after parsley plants reach the 4‘*'-true-leaf stage, on average, from about mid
May until the end of June, with a peak of oviposition activity in June. Weevil larvae hatch, approximately 5 days after the eggs are laid and feed in parsley roots for about 20 to 30 days, mainly in July. Carrot weevil damage and yield loss are observed mostly at the second cutting, by the end of July in parsley planted before the 4*** week of April.
Sampling for oviposition scars is easier than finding adults in parsley fields. Carrot weevil infestations ranged from 0.1% to 20.8% plants infested with oviposition scars in commercial parsley fields and these observed populations fit the Poisson distribution when the infestation was below 0.047 oviposition scars per plant and the negative binomial distribution when the infestation was above 0.06 oviposition scars per plant.
One oviposition scar per plant causes significant damage either by reducing yield or killing the plant no matter the plant size as a ftmction of planting date. The relationship between carrot weevil infestations and parsley yield is best described as a linear ftmction;
Yield=l64.9-103*INF%, and based on this relationship the economic injury level (EEL) was found to be about 1% plants having larval feeding in the roots or oviposition scars in the petioles. An economic threshold (ET) of 1% plants infested with carrot weevil oviposition scars is recommended for control decision making. Because the ET is so low, binomial or presence/absence sampling is the most convenient way to sample oviposition scars in parsley fields. A decision rule for carrot weevil management is: if 148 plants can be inspected without finding any oviposition scars, there is a 95% chance that the percentage of infested plants is less than 2% and no treatment is needed. On the contrary, if one infested plant is found before all the 148 plants are examined, then the scout can stop and decide to treat. An X-shaped transect was found to be the best path for sampling carrot weevil oviposition scars in parsley fields. This transect results in the most correct decisions and the fewest incorrect decisions compared to other transect patterns.
Ill Dedicated to my parents, Angel Onistes and Maria Ricarda; my children, Mariangel and Jorge Andrés; and my wife, Genova Maria
IV ACKNOWLEDGMENTS
In the first place I want to thank my adviser, Casey Hoy, for teaching me
how to do sound research, for helping me to understand statistics, for supporting me
intellectually, for his patience in correcting my English style and grammar, and for many
other things.
I would like to express my gratitude to the members of my graduate committee:
David Horn, Celeste Welty, and Mark Bennett for their guidance in my career as a graduate student, as well as to the entire Entomology faculty for inspiring me with their teaching and fiiendly attitude.
I am very thankful to Mike Dunlap for all the wonderful help provided in my lab, greenhouse, and field research, as well as to the summer workers: Deanna Skelton,
Jennifer Bicker, Nick Fioritto, T. J. Tausch, Joe Dunlap, and Penny Howman; their help in my field research was absolutely invaluable. I thank also Tim Miklasiewicz, for his great help in my greenhouse experiments, and Bill Evans and his crew at the Muck Crops
Branch of OARDC for their cooperation in my field research.
I wish to give special thanks to Bruce Buurma, Rich Danhoff, Dick Zellers, Cecil
Kene, and Tom and John Brenckle for allowing me to sample their parsley fields.
I would like to thank Liyang Zhou for helping me with my simulation studies. I appreciate the spontaneous help given to me by post docs, visiting scientists, technicians. janitors, and support staff at OARDC. I thank also my fellow students who shared with me all or part of this doctorate program.
Special thanks to Mabel Kirchner and Shirley Holmes for their kindness, to Ken
Chamberlain and Margaret Latta for the pictures, Bonnie Bing for the posters, and the library staff: Connie Britton, Della Bardall, and Pat Sword, for their assistance in the library, and to many people who made Ohio feel like home to me.
I’d like to express my appreciation to the Ohio Vegetable and Small Fruit
Research and Development Program for the funds provided to do this research, and my eternal gratitude to INIA (FONAIAP), Venezuela, for paying my tuition, fees, and living expenses through the PRODETEC H Program. Many thanks to Richard Lindquist and
Casey Hoy for supporting my stay during the 2001 winter quarter of my program.
Finally, I am indebted to my wife, Maria for accompanying me in this mission, and for encouraging me all this time, as well as to my children, Mariangel and Jorge, who make my life wonderful.
VI VITA
February 28, 1955 ...... Bom - Târiba, Tâchira, Venezuela
198 0 ...... Ingeniero Agrônomo, Universidad del Tâchira, San Cristobal, Venezuela
1981 — 1987 ...... Researcher, National Fund for Agricultural Research (FONAIAP), Valle de la Pascua, Guarico, Venezuela
1987 - 1990 ...... M.S. Entomology, Auburn University, Auburn, Alabama, USA
1990 — 1993...... Researcher, FONAIAP, Valle de la Pascua, Guarico, Venezuela
1994 — 1997...... Researcher, FONAIAP, Bramon, Tâchira, Venezuela
1997 - 2000 ...... Graduate Research Assistant, The Ohio State University
2001 — present Graduate Research Associate, The Ohio State University
PUBLICATIONS
Research Publication
1. Torres, A. and C. Yanez. Evaluacion de técnicas de control de babosas en firesas y hortalizas en zonas altas del Estado Tâchira. Agronomia Trop. 48(3): 291-303. 1998
2. Torres, A. and C. Mufioz. Insectos plagas del sorgo granlfero. FONAIAP. Estaciôn Experimental Guarico, Sub-Estaciôn Valle de la Pascua. Serie B: no. 2-07. Valle de la Pascua, (1986)
vii 3. Torres, A. and C. Mufioz. Principales insectos plagas del sorgo granifero. FONAIAP-DIVULGA, Caracas 2(16):21-25, (1984)
FIELDS OF STUDY
Agronomy
Major Field: Entomology
vm TABLE OF CONTENTS
Page
A bstract ...... ü
Dedication ...... iv
Acknowledgments ...... v
V ita...... vii
List o f Tables ...... xii
List of Figures ...... xiv
Chapters:
1. Literature Review ...... 1
2. Preliminary studies in carrot weevil biology and research methods ...... 23
2.1 Carrot weevil sex identification ...... 24 2.2 Carrot weevil infestation techniques ...... 24 2.3 Average number of eggs per oviposition scar ...... 29 2.4 Carrot weevil rearing technique in parsley ...... 29 2.5 Egg stage duration ...... 31 2.6 Carrot weevil oviposition preference ...... 33 2.7 Oviposition activity ...... 33 2.8 Trench study ...... 35 2.9 Parsley growth study ...... 36 2.10 Parsley growth stage and oviposition activity ...... 39 2.11 Relationship among carrot weevil infestations, planting dates of parsley, and nematode control ...... 39 2.11.1 Materials and methods ...... 41 2.11.1.1 Plant size ...... 41 2.11.1.2 Pots, potting medium, and transplanting ...... 42 2.11.1.3 Plot layout and experimental design ...... 42 2.11.1.4 Greenhouse settings ...... 42 2.11.1.5 Oviposition scars and infestation technique 42 ix 2.11.1.6 Nematodes ...... 43 2.11.1.7 Yield ...... 44 2.11.1.8 Adult weevil collection ...... 44 2.11.1.9 Root damage evaluation ...... 44 2.11.2 Results...... 45 2.11.2.1 Fresh yield ...... 45 2.11.2.2 Root damage ...... 45 2.11.2.3 Adult Emergence ...... 46 2.11.3 Discussion ...... 47
3. Determining the relationship between carrot weevil infestations, damage and planting dates of parsley ...... 49 3.1 Materials and methods ...... 50 3.1.1 Parsley Plantings and Experimental D esign ...... 50 3.1.2 Weevil infestation ...... 51 3.1.3 Parsley plots and agronomic practices ...... 51 3.1.4 Adult counts ...... 52 3.1.5 Egg and larval counts ...... 52 3.1.6 Parsley cuttings ...... 53 3.1.7 Statistical analysis ...... 53 3.2 Results...... 54 3.2.1 Carrot weevil infestation and planting dates of parsley ...... 54 3.2.2 Eggs removal at harvest time ...... 61 3.3 Discussion ...... 62
4. Estimating economic injury levels and action thresholds for carrot weevil in parsley ...... 68 4.1 Materials and methods ...... 69 4.1.1 Observations and measurements in research plots ...... 69 4.1.2 Observations and measurements in commercial fields ...... 72 4.2 Results...... 73 4.2.1 Research plots ...... 73 4.2.2 Commercial parsley fields ...... 80 4.3 Discussion ...... 80
5. Developing a sampling scheme for carrot weevil in parsley ...... 88 5.1 Materials and methods ...... 89 5.1.1 Commercial parsley fields ...... 89 5.1.2 Oviposition scar counts ...... 92 5.1.3 Population distribution ...... 92 5.1.4 Sample size ...... 94 5.1.5 Simulation studies ...... 94 5.2 Results and discussion ...... 97 5.2.1 Spatial distribution of oviposition scars ...... 97 5.2.2 Spatial pattern of oviposition scars counts in parsley fields 99
X 5.2.3 Sample size ...... 102 5.2.4 Best transect for sampling parsley fields ...... 102 5.3 Conclusions ...... 106
An integrated pest management program for carrot weevil in parsley 108 6.1 Life history ...... 109 6.1.1 Adult stage...... 109 6.1.2 Egg stage...... 113 6.1.3 Larval stage ...... 113 6.1.4 Action thresholds ...... 120 6.1.5 Sampling scheme ...... 120 6.1.6 How to walk the field ...... 121 6.1.7 Control strategies ...... 121
Bibliography ...... 124
XI LIST OF TABLES
Table Page
2.1 Analysis of variance for the effect of plant size, nematodes, and oviposition scars on parsley yield arranged in a 2x3x5 factorial experiment with 10 replicates ...... 46
2.2 Analysis of variance for the effect of plant size, nematodes, and oviposition scars on parsley root damage arranged in a 2x3x5 factorial experiment with 10 replicates ...... 47
3.1 Analysis of variance results for seasonal means of carrot weevil population activity among planting dates of parsley. Muck Crops Branch, Celeryville, Ohio ...... 57
3.2 Analysis of variance results for rate of population increase (slope) and pattern of population change (quadratic) of carrot weevil adults, eggs, larvae, and root feeding among planting dates of parsley. Muck Crops Branch, Celeryville, Ohio ...... 58
3.3 Total number of carrot weevil eggs present above and below the cutting point, and percentage of eggs removed in five planting dates o f parsley ...... 61
3.4 Polynomial regression analysis to test the effect of five levels of oviposition scars on firesh weight per plant of parsley and orthogonal contrasts to test for differences among the oviposition levels, at two different cuttings in 1999 ...... 75
4.2 Analysis of variance to test the effect of five levels of oviposition scars on fi%sh weight per plant (g) and plant mortally of parsley and the interaction between planting date and oviposition scars at two different cuttings in 2000 ...... 78
4.3 Cost of control ($) per hectare of carrot weevil in parsley during the growing season and market value ($) of parsley in O hio ...... 82
5.1 Transect patterns compared by simulation to select the best for sampling parsley fields infested with carrot weevil oviposition scars ...... 95 xii 5.2 Population distribution fitted to counts of carrot weevil oviposition scars in 16 different population densities. NA - not applicable ...... 98
X lll LIST OF FIGURES
Figure Page
1.1 Carrot weevil adult, eggs, and larva ...... 3
1.2 Forrester diagram of carrot weevil-parsley interactions ...... 22
2.1 Carrot weevil sex identification. From left to right and from top to bottom; female abdomen, male abdomen, female elytra, and male elytra 25
2.2 Petiole sections bearing at least one oviposition scar fastened to the host parsley petioles ...... 28
2.3 Histogram of number of carrot weevil eggs per oviposition scar in parsley ...... 30
2.4 Carrot weevil egg stage duration based on percentage of eclosed larvae after oviposition ...... 32
2.5 Carrot weevil oviposition preference in parsley leaf petioles ...... 34
2.6 Parsley plant at the four-true-leaf stage showing the 4* leaf fully expanded. Carrot weevils seldom lay eggs in smaller plants ...... 37
2.7 Histogram showing the number and proportion of parsley plant having different numbers of true leaves at two sampling dates in two parsley fields when carrot weevil oviposition scars were present. Field 1 was in Celeryville, Ohio and field 2 was in Hartville, Ohio, 2000 ...... 40
3.1 Carrot weevil seasonal activity in five planting dates of parsley. Celeryville, 1998 ...... 55
3.2 Carrot weevil seasonal activity in three planting dates of parsley. Celeryville, 1999...... 56
3.3 Daily maximum and minimum temperatures during the sampling period at the Muck Crops Branch ...... 64
4.1 Average fresh weight per plant and percentage plant mortality in parsley infested with different levels of carrot weevil oviposition scars, 1999 ...... 74
XIV 4.2 Average fresh weight per plant and percentage plant mortality in parsley infested with different levels of carrot weevil oviposition scars, 2000 ...... 77
4.3 Proportion of plants infested with carrot weevil oviposition scars that had root feeding (dead and live plants) ...... 79
4.4 Regression line for parsley yield per 30 cm-row section vs. the proportion of plants with larval feeding in the roots ...... 81
4.5 Economic injmy level for carrot weevil oviposition scars in parsley ...... 83
5.1 The progression of carrot weevil infestation over four weeks in four commercial parsley fields ...... 100
5.2 An X-shape is the best transect to sample parsley fields infested with carrot weevil oviposition scars. The starting point can be any comer, and 7 randomly selected plants per cell must be sampled along one diagonal axis, and 8 plants, along the other diagonal axis ...... 103
5.3 Transect patterns showing the mistake gap and the good attributes that help to deciding the best one for sampling parsley fields infested with carrot weevil oviposition scars ...... 104
5.4 Average sample number curve for 6 different transect patterns. The economic threshold is approximately equal to 0.01 ...... 105
6.1 Carrot weevil adult ...... 110
6.2 Overwintering parsley field with few surviving plants as it looks by early April ...... I l l
6.3 Seasonal activity for three different carrot weevil stages; avg. number of adults per 3 m of row, avg. number of eggs per plant, and avg. number of larvae per root ...... 112
6.4 Parsley plant at the four-true-leaf stage showing the 4* leaf fully expanded. Carrot weevils seldom lay eggs in smaller plants ...... 114
6.5 Carrot weevil oviposition scar in a parsley petiole ...... 115
6.6 Carrot weevil eggs ...... 116
6.7 Parsley root showing severe internal damage caused by the carrot weevil larva ...... 117
XV 6.8 Dead and damaged parsley plants due to carrot weevil larva ...... 118
6.9 Carrot weevil larva ...... 119
6.10 An X-shaped transect is the best pathway to sample parsley fields infested with carrot weevil oviposition scars. The starting point can be any comer, and each diagonal must be divided in 10 sections approximately, so that 7 plants per section must be sampled along one axis and 8 plants along the other axis ...... 122
XVI CHAPTER 1
LITERATURE REVIEW
Taxonomy
The carrot weevil, Listronotus oregonensis LeConte (Coleoptera: Curculionidae),
belongs to the Superfamily Curculionoidea Latreille. 1802 Subfamily Brachycerinae
Billberg 1820 (EIS 1996). This species was included within the tribe Phytonomini by
LeConte (1876), and within the tribe Hyperini by Blatchley and Leng (1916). The genus
Listronotus was described by Jekel in 1865, vAûiMacrops Kirby 1837 referenced as a junior homonym, and Hyperodes Jekel 1865, Mascarauxia Desbrochers 1898, and
Aulametopiellus Bretches 1826 referenced as junior synonyms (EIS 1996). LeConte first
described it as Listroderes oregonensis in 1857, then in 1876 revised the name to
Listronotus oregonensis, (Henderson 1939, cited in Whitcomb 1965, EIS 1996). Since
that time, carrot weevil has been identified with the synonyms of L. impressifrons
LeConte 1876, L. tessellatus Casey 1895, and L. rudipennis Blatchley 1916 (EIS 1996).
There are 78 species described in the genus Listronotus (EIS 1996), and early literature is
somewhat confusing regarding those species that feed on Umbelliferae in the Midwestern
and Northeastern states. The carrot weevil, as recognized by the Association of American
Entomologists, has been named in the past according to its host plant, e.g. Sagittaria curculio, parsley stalk weevil, carrot curculio (Blatchley and Leng 1916, Pepper 1942,
Wright and Decker 1958). Pepper (1942) published an extensive article on carrot weevil
as L. latiusculus (Bohe), which seems to be the same as L. oregonensis.
Morphology
Carrot weevil eggs are ovoid or elliptical, 0.75 mm to 0.89 mm long and 0.48 mm to 0.54
mm wide. They are light yellow when first oviposited then turn brown and finally become
almost black before the larva hatches (Fig. 1.1). The larvae range firom 5 to 9 mm long
and from 1 to 3 mm in thickness just before pupation (Fig. 1.1). The grubs are legless,
slightly curved, and white with amber heads. Each segment of the larval body bears a series of setae, which in the case of the head are spread in an irregular pattern. The spiracles are conspicuous. The pupae are 5 to 8 mm long and are creamy white. Strong spines can be observed on each abdominal segment, as well as on the head, thorax, and femora. The adults are elongate-oblong, dome shaped weevils that measure 3.5 to 7 mm long (Fig. 1.1), the males being smaller than the females. They are light reddish brown when newly emerged, and then turn dark brown or nearly black with very distinctive cuprous tinged scales covering the thorax and elytra. The head bears short cuprous setae and has scattered scales. The beak is slender, and the antennae and tarsi are reddish brown. Female weevils have the first and second abdominal segment swollen while in males these segments are rather depressed or concave (Blatchley and Leng 1916, Pepper
1942, Whitcomb 1965, Perron 1971, Martel et al. 1976, Cress and Wells 1977, Grafius et aL 1983, Ghidiu 1987). Adult
Eggs
Larva
Figure 1.1: Carrot weevil adult, eggs, and larva The following quotation is the description given by LeConte (1876) to L. oregonensis, ’’One abraded $ firom Oregon. The last ventral segment is not at all impressed. The beak is feebly carinate, and obsoletely sulcate, the firontal fovea not deep.
The prothorax is a little wider than long, much rounded on the sides, and feebly channeled, the sculpture is peculiar, consisting of small granules, separated by deep rugosities, and mixed with large, scattered punctures; on each granule is a small puncture; the few scales which remain are small, and indicate three stripes. The elytra are one-third wider than the prothorax; the humeri are rounded, and the base slightly emarginate; striae strongly punctured, interspaces slightly convex. Length 6 mm.; .23 inch.”
Geographical Distribution
The carrot weevil, L. oregonensis, originated in the northern United States and
Canada (Pepper 1942). This insect has been found in Ontario, Quebec, and Nova Scotia
Canada (Perron 1971, Roberts and Stevenson 1974, Martel et aL 1975, Stevenson 1976,
Le Blanc and Boivin 1993). In the United States, the carrot weevil has been recorded in
Pennsylvania, New Jersey, Ohio, Michigan, Illinois, Indiana, Missouri, Massachusetts, and many other states (Blatchley and Leng 1916, Pepper 1942, Wright and Decker 1958,
Whitcomb 1965, Cress and Wells 1977). In Ohio this species is noticed principally in the muck vegetable production area in Celeryville, Huron County and Hartville, Stark County
(Simonet 1981). Life History
Life Stages
The carrot weevil has four developmental stages; egg, larva, pupa and adult In laboratory rearing studies Wright and Decker (1958), at room temperature, and using carrots as a host, the life cycle o f L. oregonensis, firom egg to adult ranged firom 21 to 53 days, and adult females lived fi-om 103 to 428 days. The preoviposition period ranged firom 9 to 43 days. The egg stage lasted firom 4 to 10 days with means of 5.5 and 6.1 days in two different experiments. The larva passed through 5 stadia of approximately equal duration, the first stadium lasted 2-6 days, the second stadium, 2—5, the third stadium, 2—
7, the fourth stadium, 2 to 7, and the fifth stadium, 1—10 days. The prepupa lasted 2—6 days, and the pupa lasted 5—20 days. According to Simonet (1981), carrot weevil develops firom egg to adult in approximately 34 days at 23.9 °C, and two generations may be produced each year in Ohio. Roberts and Stevenson (1974) reported an average of 19 days for the preoviposition period, 32 days for the egg-larval stage, 11 days for the prepupal stage, and 12 days for the pupal stage at 19 °C, 70% RH, and 16 h photoperiod.
Martel et al. (1975) recorded a range of 14 to 27 days after oviposition for larvae to reach maturity at 27 °C, and firom 20 to 33 days, at 21 °C. Adult emergence ranged firom 27 to
42 days after oviposition at 27 °C, and 39 to 49 days, at 21 °C. The photoperiod was 16 h for all the recordings. Life history records reported by Pepper (1942) under insectaiy conditions indicate that eggs required an average of 8.2 to 10.8 days to complete development. Larvae required 13.3 to 14.5 days, and prepupae and pupa averaged 3 to 3.1 days and 8.1 to 9.1 days respectively. Habits
The female oviposits in small cavities or oviposition scars in petioles, stems, crowns, or exposed roots of carrot, parsley, and celery plants (Pepper 1942, Wright and
Decker 1958, Whitcomb 1965, Stevenson 1981, Grafius et aL 1983). According to Boivin
(1985), oviposition punctures are small, deep apertures containing 1 or 2 eggs in contrast with feeding punctures or scars that are shallow and rather large. The number of eggs per oviposition scar varied from 1.72 to 3.52 for individual females, and the maximum number of eggs laid by individual females was 870, on carrots under laboratory conditions (Wright and Decker 1958). Martel et al. (1976) reported 1.6±0.24 eggs per oviposition scar on carrot slices, and 1.59±0.24 eggs per oviposition scar on carrot petioles, at 21 °C, 60% RH, and 16 h photoperiod. Pepper (1942) observed that the number of eggs per oviposition scar varied from 1 to 10, depending on the size of the petiole, the average was only 3 or 4. Pepper (1942) recorded up to 23 oviposition punctures per plant in celery and 16 in carrot, with a total of 93 and 51 eggs respectively.
Once the eggs are laid, the oviposition puncture is immediately sealed with a black exudate emitted from the caudal end of the female abdomen (Pepper 1942, Whitcomb
1965, Stevenson 1981). The majority of oviposition punctures are found on the concave part of the petiole. The newly emerged larvae feed inside the petiole, then either make a tunnel downward toward the root or drop to the ground, and enter the root from the soil surface. When the eggs are deposited in very young carrot or parsley plants, the newly hatched larvae go inunediately to the roots. The larvae feed either in the crown, stem or roots of the host plant. Larvae that exhaust the food before reaching maturity may migrate to neighboring plants, to continue feeding. When their development is complete, larvae leave the plant and build an earthen cell in which to pupate about 5 to 7.6 cm deep in the soil. After the pupal cell is formed, the larva slows its metabolism and pupates within 1 to
5 days. This period is called the prepupal stage. The pupa remains inactive in its earthen cell until the adult stage is reached and the weevil is ready to emerge. Adults remain in the pupal cells for 1 to 5 days and then tunnel their way to the surface. Adults feed for a few days, then mate, and females start laying eggs. Adult weevils have the habit of dropping to the ground when disturbed and remain motionless for several minutes.
(Pepper 1942, Wright and Decker 1958, Cress and Wells 1977, Stevenson 1981, Simonet
1981, Grafius et aL 1983).
Phenology
Carrot weevil females oviposit at different threshold photoperiods depending on temperature. When the temperature was 16 °C, oviposition occurred between 16 and 18 hours of daylight, at 20 °C, oviposition occurred between 14 and 16 h, at 24—25 °C, oviposition occurred about 14.5 h, and at 28—30 °C, oviposition occurred between 10 and
12 h (Stevenson and Boivin 1990). Observations made by Stevenson (1985), using carrot root sections, in Ontario, Canada indicate that peak of oviposition may occur as early as mid May as long as the air temperature exceeds 24 °C for 5 or more days. A delayed peak of oviposition was observed in mid-June when the temperature did not reach 24 °C during the second half of May. Stevenson (1976) observed that females began to lay eggs in celery transplants as early as mid May and the average date of peak oviposition was June 6th. Carrot fields planted in early May, from 1973 to 1976, began to show damage due to
carrot weevil larvae in early June and reached a peak in early July. Full-grown larvae
emerged from about mid-July until the end of September. First generation adults appeared
from late July until November. Pepper (1942) states that the adult weevils are in the fields
when carrots and parsley are 2.54 cm high or when celery is transplanted to the field, in
Southern New Jersey. Carrot weevil adults become active and seek host plants in late
April and early May in Ohio (Simonet 1981).
Reproductive diapause in the carrot weevil, studied by Stevenson and Boivin
(1990) indicated that carrot weevil females that began to reproduce at 25 °C and 16:8
(L:D) were able to mate when transferred to cages containing males at 25 °C and 12:12
(L:D), but the females eventually entered diapause after about 2 weeks exposure to the short photoperiod.
In Quebec, Boivin (1988) found that carrot weevils started to lay eggs in carrots placed on the ground after 147 ± 9 degree-days (DD) (base 7 °C) and by 456 ± 47 DD,
90% of the eggs were laid. However, females did not start to oviposit in carrot seedlings until the plant reached the four-true-leaf stage, and the majority of the eggs were foxmd on carrot leaves, petioles and crowns. According to these findings carrots planted after 400—
450 DD may escape carrot weevil infestation. Stevenson and Boivin (1990) established that the carrot weevil is less likely to attack carrots planted late in the season.
Cumulative degree-days can predict insect phenology for an integrated pest management program (Higley and Wintersteen 1987, Pope 1998). The accumulated degree-days, base 7 ®C for carrot weevil, after January can be used to predict the egg laying period in spring (Boivin 1988). Growers can use degree-day accumulation to reduce unnecessary scouting efforts outside the egg laying period.
Dispersal
The carrot weevil overwinters in the adult stage along margins o f previously infested fields, in windbreaks, in grass, sod along ditches, hedgerows, woods, and under debris and crop residues. (Pepper 1942, Simonet 1981). When temperature declines, or after the host plants are harvested, the adults begin to seek overwintering sites, probably in the nearest suitable place (Pepper 1942). Weevils were never found overwintering in celery or carrot fields; however some overwintering weevils were found in overwintering parsley rows (Pepper, 1942). Grafius and Collins (1986) indicated that adult weevils might overwinter extensively in previously infested fields, and that they do not move out of the field unless it is plowed early enough in the year for active weevils to walk to other overwintering sites. Although the weevils have functional wings, they rarely fly. Pepper
(1942) reported about 10 adults captured while flying, over a 5-year period. Perron (1971) captured five adults in a light trap suggesting that this insect can fly at night. In spring, as the adults emerge firom the overwintering sites they seek carrot, parsley or celery fields.
Once the weevils are in the host crop they apparently do not walk far. Adult weevils can live for a relatively long time with sufficient food and moisture (Pepper 1942). Natural Mortality Factors
Both adverse environmental conditions and natural enemies reduce carrot weevil populations. Survival rate of overwintering weevils in a field was 70% through February, similar to that of a population kept in laboratory at ca. 20 °C, despite temperatures in the field that were firequently below 0 °C (Grafius and Collins 1986). By March, survival in the field decreased markedly due probably to the duration of cold temperature. According to Whitcomb (1965), natural enemies did not significantly affect carrot weevil populations in experimental plantings. Several species of egg parasitoids have been reported for carrot weevil, Anaphes cote, A. listronoti, A. victiis, and A. sordidatus
(Hymenoptera: Mymaridae) (Collins and Grafius 1986, Cormier et aL 1996, Hooper et aL
1996, Huber et aL 1997). In Ontario commercial fields, Cormier et aL (1996) estimated a maximum of 33% of carrot weevil eggs parasitized by A naphes spp. Baines et aL (1990), studied the consumption of carrot weevil by five species of carabids that are abundant in
Quebec (Canada) carrot fields. Because these carabids are polyphagous, they may exert some natural control of the carrot weevil population.
Host Plants
Carrot weevil is a pest of economic importance to growers of several crop plants belonging to the family Umbelliferae: parsley, Petroselinum crispum (Mill.) Nym.; carrot,
Daucus carota L., and celery, Apium graveolens L. Also wild carrot; parsnip, Pastinaca sativa L.; dill, Anethum graveolens L.; broad leaved plantain, Plantago major L.; patience dock, Rumex patiena L.; sour grass or garden sorrel, Rumex acetosa (Pepper 1942).
10 Parsley
According to the botanical classification, parsley belongs to the dicotyledoneae, family Apiaceae (Umbelliferae), genus Petroselinum, species crispum var. , neapolitanum
(Italian flat-leaf parsley), and var. crispum (curly leaf parsley) (Gleason 1952, Rubatzlqr and Yamaguchi 1997). Parsley cultivars can be further grouped into five types: plain leaf, celery or neapolitan leaf, curled leaf, fern leaf, and turnip root type. The plant has an umbel-type inflorescence (Chapman et al. 1976). The foliage is the edible part, so the crop is generally categorized as “greens” based on its primary use (Peirce 1987). Parsley is a biennial species but is grown mostly as an annual. This plant requires vernalization prior to flowering and is categorized as a hardy or cool season crop because it tolerates firost, germinates in cold soil, plants are able to obtain water firom cold soil, and the harvested product is stored at cold temperature (Maynard and Hochmuth, 1997). Parsley roots are shallow, 45.7-61 cm depth, pivoting or taproot type, and are tolerant of pH in the range 6.S-5.5 (Maynard and Hochmuth 1997, Rubatzky and Yamaguchi 1997). The growing degree-day base for parsley has not been measured, but that of carrots, 3.3 °C, can be used (Sanders et al. 1980, cited in Maynard and Hochmuth, 1997).
Parsley is grown as an herb for flavoring and garnish, either firesh or dry. It has a high content of vitamin A, 5,200 lU per 100 g of edible portion (Haytowitz and Mattews
1984, and Gebhardt et al. 1982, cited in Maynard and Hochmuth 1997). In Ohio, parsley is planted by direct sowing as early as February using spun polyester row cover to protect the crop or seeds firom fireezing temperatures and to get early firesh produce. In early spring, parsley is planted every 1-3 weeks in order to supply the firesh market. Three-four
11 cuttings are usually made during the growing season in each planting. Seeds require 7 to
20 or more days to germinate, depending on the temperature, and the germination percentage can be low due to immature embryos and inhibitory substances located within the seed coat (Rubatzky and Yamaguchi 1997, Hassel 1993).
Parsley is grown mainly on muck soils in Ohio. Fields are usually 1 to 15 acres.
The soil is plowed and disked after fertilizers are broadcast. Flat or raised beds are the usual way to plant parsley, with 5 and 2 rows, respectively and 25 to 33 cm between rows. A contact herbicide is usually applied at planting if weeds are present, or after planting when the plant has at least 2 true leaves. Hand weeding is a common practice also. Barley, rye, or oats are sometimes interplanted to protect the seedlings from the wind, then the cover crop is killed with selective herbicides in early May. Most parsley growers spray insecticides for carrot weevil control (Grower’s survey; Ohio Vegetable
Production Guide, 2000).
Damage and Economic Importance
The majority of the damage in carrot, celery, and parsley is caused by the larval stage in the root, stem, and crown. Adult feeding injury can be grooves, channels, and scars or punctures formed by chewing the epidermis of the petiole with the mandibles.
Adults also feed on leaves and the females cause minor injures due to deep cavities made in the petioles for oviposition, but the resulting injury is not economically important
(Pepper 1942, Whitcomb 1965, Ghidiu 1987). Whitcomb (1965) cites two types of damage that may depend on the oviposition site in the carrot plant; if eggs are oviposited
12 in the axils or base of the petiole, then larvae tunnel directly to the root firom the crown and usually kill young plants; if eggs are laid elsewhere in the petiole, the larvae drop to the soil and enter the carrot root fi*om the ground surface. According to Whitcomb (1965) this second of damage is the most destructive; the larvae make shallow tunnels by feeding around the carrot circumference rendering the carrot unmarketable. The larvae tend to feed mainly in the upper one-third of the root (Stevenson 1981). Infested carrots break in the crown at harvest, and in celery the plant may be stunted or killed when the larvae feed on the apical meristem (Pepper 1942). The potential for damage in celery is high because the feeding injury in the heart makes the plant unmarketable (Cress and
Wells 1977, Simonet 1981). The damage to parsley is not visible early in the season, when damage has not yet accumulated. As larval feeding accumulates it weakens the plant, reducing the stand, and causing yellowing in the foliage (Pepper 1942, Simonet
1981, Ghidiu 1987). Damage on celery has been reported to vary firom 0% to 90%, on carrots, 25% damage has been reported, and the stand of parsley may be seriously reduced under heavy infestations (Pepper 1942). According to Perron (1971), damage to carrots ranged fi"om 12.5% for early-planted carrots to 4% for late-planted carrots in Quebec,
Canada, during 1970. Boivin (1988) mentions that 0.14 eggs per plant on average can produce 1% damage to carrots at harvest. Parsley growers in Ohio have observed increased carrot weevil infestation since carrot production has declined, probably because parsley has become the most available host for this insect.
13 Management
Predictive Models
Boivin (1988) investigated the relationship between planting dates of carrots
based on DD 7 0 C and weevil damage to the crop, and found that carrot damage can be
predicted by number of eggs and planting date. In laboratory studies at 16 h photoperiod,
Stevenson (1986) determined development rates from eggs through adults as a frmction of
temperature within the range 17.5-30 °C. Development times decreased with increasing
temperature, and a cubic polynomial function was fit to describe this relationship.
Stevenson (1986) suggested that this equation for carrot weevil egg to adult development
may be the most appropriate for a predictive model. Zhao (1991) developed a simulation
model for predicting the population dynamics of L. oregonensis, in field carrots.
Mortality rates of eggs, larvae, and pupae for different planting dates were included in the
model along with density of the overwintered adults; temperature dependent development
rates for each life stage; age-, temperature-, and phenology-dependent oviposition rates;
and the effect of an egg parasitoid. The population densities predicted by this model were
very close to those observed in the field.
Economic Thresholds
Economic thresholds are not available for carrot weevil in parsley. However, 5 eggs per 100 carrots in early carrots and a range of 3.8 to 5.3 eggs/100 carrots in mid
season plantings are estimated economic thresholds for carrot weevil (Zhao 1990).
14 Stevenson (1985) suggested 0.3 oviposition cavities per carrot section per day (CSD) or
0.5 CSD, if less than 50% of carrot root sections (CRS) placed on the ground are infested, as a tentative action threshold for carrot weevil.
Monitoring
Carrot weevil populations can be monitored by counting adults, eggs, oviposition scars, or feeding punctures. Different monitoring methods have been tested in order to determine population levels either for research or for control decision-making purposes.
Different carrot weevil traps have been tested in the United States and Canada. Some of these are: 1) Boivin traps (Boivin 1985) consisting of 22 wooden plates separated by metal washers and placed over a carrot used as bait; 2) pitfall traps, consisting of a 1.25-L plastic-jar, with an aperture of 9 cm in diameter, which contains a solution of trisodium orthophosphate as a preservative; 3) jar traps with a funnel baited with a carrot slice
(Ryser 1975); 4) carrot root sections (CRS) consisting on counting the number of egg cavities per carrot root section per day (CSD) in a 3 to 4 day period (Stevenson, 1985);
5) carrot trap consisting on pieces of fresh carrots placed vertically in the ground to record the number of adults (Stevenson and Barszcz, 1997).
Boivin found his trap to be the most efficient for capturing carrot weevil adults at the beginning of the crop season. Its effectiveness decreases after the carrot plants have two true leaves. The adults of these insects are more attracted to the host plant than to the trap after the two-true-leaf stage. The Boivin trap has the additional advantage of being selective enough to avoid problems of identification for inexperienced scouts (Boivin
15 1985). Stevenson and Barszcz (1997) pointed out that the method of counting egg cavities
in carrot root sections may be useful for research purposes but not for routine monitoring
because it requires too much labor. Boivin (1985) found that the CSD method can be
used during the entire growing season, but trap captures were not consistent with damage
observed in carrot fields among years. Stevenson and Barszcz (1997) pointed out that the
method of counting egg cavities per carrot section per day (CSD) has the advantage of
detecting oviposition activity, and hence would be useful for pest management specialists
to study seasonal activity, even if counting egg cavities is impractical for parsley growers.
The use of carrot root sections to detect adult occurrence may be acceptable for growers
who prefer to do their own monitoring. A good method must provide reliable estimates of
weevil population with a minimum effort and the least expenditure of money in order to
be accepted by parsley growers. According to Boivin (1985), pitfall traps captured more
individuals when the adults disperse after the second true-leaf stage of the carrot plant.
According to Stevenson and Barszcz (1997), any of the methods tested were
effective only when infestation ranged fi-om moderate to heavy in carrots; the same may
be true for other members of the Umbelliferae like parsley and celery. Parsley growers
require a monitoring technique for the weevil population at any time or stage of the crop to make treatment decisions. The adult weevils are difGcult to locate on the host plant
because they are difBcult to see and drop fiom the plant and remain motionless for
several minutes when disturbed. Stevenson (1985) also indicates that due to the high
16 correlatioa between carrot sections per day (CSD) and carrot root sections (CRS), determining the proportion of infested CRS rather than the total oviposition cavities may provide adequate data for decision-making.
Control
When dieldrin was banned for use in food crops in 1972, carrot weevil damage increased and became severe in 1979-80. Weekly applications of azinphosmethyl
(Guthion®) produced plants free of carrot weevil damage (Simonet 1981). Toxicity studies made by Martel et aL (1975) indicated that Guthion® killed 100% of the weevil population at 0.1% concentration. Stevenson (1981) recorded no injured carrots when
Guthion® was used at 1.1 kg a.i./ha Applications of Fenvalerate® and methyl parathion were found to be effective for controlling L. texanus in Texas, and a control program of 6 or 12 applications was recommended throughout the season (Woodson et al. 1989).
Assays conducted by Hoy et al. (1997) in Ohio indicated that the organophosphate insecticides, azinphos-methyl (Guthion®) and malathion, resulted in the most effective weevil control. Most pyrethroids tested in laboratory by Free et al. (1996) were more toxic than azinphosmethyl; however permethrin toxicity decreased at temperatures above
22 °C.
The only insecticide labeled for carrot weevil control in parsley in Ohio is azinphos methyl (Guthion®), which may not be applied within 21 days of harvest (Ohio Vegetable
Production Guide 2000).
17 In addition to monitoring and timing insecticide applications, cultural practices help reduce carrot weevil numbers. Crop rotation using non-host plants and planting susceptible host crops away firom infested areas, eliminating volunteer or overwintering host crops and proper disposal of culled celery and carrots, as well as controlling weed hosts such as wild carrot constitute good control tactics to reduce infestations of this pest
(Simonet 1981, Grafius et al. 1983). Pepper (1942), however, found that cultural practices did not significantly reduce the population of carrot weevil in experimental plantings.
The use of barriers or trenches may stop walking weevils firom. reaching new host crops. Whitcomb (1965) accidentally found that an early-planted lettuce strip between the hibernating crop and the carrot plots, which was sprayed with insecticide (DDT and malathion), fieed the carrot plots firom weevil infestation. Trenches lined with plastic located between the overwintering crop and the new crop may prevent adult weevils firom reaching the other side of the trench, as was demonstrated for Colorado potato beetle
(Boiteau et al. 1994).
Biological control of the carrot weevil can be achieved to some extent by predators, parasitoids, parasitic nematodes, and entomopathogenic bacteria and fungus. A study conducted by Saadé et al. (1996), revealed that different strains o f Bacillus thuringiensis differed in their toxicity to the carrot weevil, the LC 50 values for the three top strains ranged firom 102 to 143 pg protein/ml.
Four species of entomophagous nematodes; Steinemema feltiae, S. carpocapsae,
S. bibionis, and Heterorhabditis heliothidis, have been tested under controlled conditions
18 for control of carrot weevil larvae, pupae, and adults. The results reported by Bélair and
Boivin (1985, 1995), indicate that these species have good potential for carrot weevil
control in the field.
Several species of parasitoids have been identified as egg parasitoids that have potential for integrated management of carrot weevil. The species described are: Anaphes cote, A. listronoti, A. victus, and A. sordidatus (Collins and Grafius, 1986, Cormier etaL
1996, Hooper et aL 1996, Huber et aL 1997). Field surveys conducted by Cormier et aL
(1996) in Ontario (Canada) indicate up to 33% of egg parasitism in commercial fields, and up to 94% in untreated plots.
Conceptual Model
The hypothetical model illustrated by the Forrester diagram (Fig. 1.2) shows the interactions that take place within the parsley crop system in muck vegetable production areas of Ohio. This model is based on a current understanding of all important variables in this system as described in the previous literature review. Most of the research published to date describes the carrot weevil in carrots. Therefore a large part of the carrot weevil-parsley model system consists of new information gathered during the last three years. Carrot weevil adults, eggs, and larvae and parsley plant populations (healthy and damaged) are state variables. Auxiliary variables that influence the change in state variables include temperature, insecticides, photoperiod, etc. The adults immigrate to the parsley fields in spring at a rate (IM) that is influenced by temperature, distance to overwintering sites, and insecticides. The closer the overwintering sites are to new parsley
19 plantings, the faster and denser the colonization by the weevils. However, if the overwintering site contains overwintered parsley from the previous season, the emerging weevils may not leave in search of food, but if the overwintering parsley is plowed at the beginning of the season, then the adult weevils may be forced to walk toward new parsley fields.
The mortality rate (DH) of adult weevils is accelerated by insecticide sprays. Eggs are laid at a rate (B) that is influenced by the weevil population size and the presence of parsley fields. Eggs and larvae develop at a known development rate (D). The parsley crop decreases according to a damaging rate (DG) influenced by the larval population density present in the field. Damaged parsley plants die at a death rate (DH) given by the population of larvae and the planting date. The emigration rate (EM) is influenced by photoperiod, temperature, and the presence of the host crop.
Thesis
According to growers and the results obtained in my field study on weevil infestations, damage, and planting dates of parsley, most carrot weevil damage occurs in early-planted parsley. The overwintering weevils enter parsley fields in early spring, dispersing from nearby overwintering sites. One way to prevent early infestations is by spraying insecticides to kill the adult weevils in parsley fields from the previous season.
Carrot weevil oviposition activity starts when the plants have four fully expanded true leaves. My research will support decision-making by specifying the critical period for sampling parsley fields infested with carrot weevil. Sampling for oviposition scars in
20 parsley is preferable to sampling for adults. Up to this moment the only reliable control method available is to spray the adults with insecticide. Therefore, adult treatment must be done before female weevils lay a significant number of eggs capable of producing yield losses in parsley. From intensive sampling in commercial parsley fields, I determined the relationship between parsley yield and carrot weevil infestation in order to define action thresholds for this pest in parsley. A practical decision rule was formulated firom knowledge of weevil distribution in the field and action thresholds.
21 Egg Parasitoids
0 0 -~-~\ y z*P>CëZH' x
Larvae /\ ------^ Adul
...
. O /" .. Nematodes < A Damaged Parsley
..- "••••► P 's □ State Variable o Auxiliary Variables Material Flow Rate Equation Information Flow Source / Sink ow Overwintering Sites T Temperature 1 Insecticide PH Photoperiod c Parsley Cutting H Harvest Rate PD Planting Date
Figure 1.2: Forrester diagram o f carrot weevil-parsley interactions CHAPTER 2
PRELIMINARY STUDIES IN CARROT WEEVIL BIOLOGY AND RESEARCH
METHODS
INTRODUCTION
Numerous aspects of carrot weevil biology must be understood to design sound research. For example, carrot weevil adult sex identification is important for manipulating plant infestation under controlled environments. An estimate of the average number of eggs per oviposition scar is needed to design economic threshold studies. Measurement of egg stage duration might be useful for sampling and control decision-making. Knowledge of carrot weevil preference for oviposition in different parts of the host plant could focus sampling efforts. Adult weevil longevity is relevant in the design of control and sampling strategies. Understanding parsley growth patterns is prerequisite for planning sampling schedules and pest control. In preparation for other research, various aspects of carrot weevil biology and control were studied and the results are discussed in this chapter.
23 Carrot Weevil Sex Identification
Two references describe carrot weevil sex differentiation by comparing the ventral segments' appearance. None of the references mention the tip of the elytra as a means for sex differentiation. An easy way to tell between males and females is to look at the ventral segments and at the tip or apex of the elytra. The tip of the elytra in females is lobulated beyond the point where they meet, whereas in males, the elytra meet at the tip without lobules (Fig. 2.1). The 1^ and 2"** ventral segments in males are depressed or concave, whereas in females these ventral segments are rather bulky or distended with a small depression in the center at their junction (Fig. 2.1). Blatchley and Leng (1916) described the first two abdominal segments of a male of Listronotus Jekel, as flattened or concave, which is similar to my observations of L. oregonensis LeConte. Whitcomb
(1965) illustrated the characteristic depression in the first abdominal segment of males and the swelling appearance in females.
Carrot Weevil Infestation Techniques
Five different procedures were tested to determine the best method of infesting parsley plants with larvae firom a known number of carrot weevil oviposition scars.
Parsley plants were exposed to 20 or more weevils for 24 hours, and then placed in a growth chamber with 85% RH, and 25 °C for 5 days. These infested plants were used as a source of oviposition scars for all the techniques described below. All the techniques were tested under greenhouse conditions at ca. 25 °C.
24 V
$
?
Figure 2.1: Carrot weevil sex identification. From left to right and from top to bottom; female abdomen, male abdomen, female elytra, and male elytra
25 Petiole Section Attached to the Host Plant at the Base of the Petioles
Petiole sections ca. 1 cm long containing at least 1 oviposition scar were cut ârom an infested plant 5 days after the eggs were laid and just before the larvae were expected to hatch. The petiole section was inserted at the base of the petioles of the host plant.
Keeping the humidity high increases the success of this procedure. Therefore, one variation of this procedure was to place the entire potted plant with the attached petiole section, inside a plastic bag with the top open. The bag was removed after 3 days and excess water at the bottom of the bag was emptied to prevent damage to the plant roots.
Egg Implanting
Punctures that simulated oviposition scars were made with a micro syringe in host petioles, then eggs laid 5 days before were removed carefully from the donor plant and transplanted to the host plant with an artist's brush. Once the eggs were transplanted, the puncture opening was sealed with a small layer of plant tissue fastened with parafrlm.
Keeping the humidity high can also enhance the success of this method, for example by encasing the plant in a plastic bag.
Grafting Petiole Sections
Petiole sections containing oviposition scars were grafted in the host plants using parafrlm to fasten the graft in place. The host petiole was cut transversely at the upper
26 third and the top discarded, then a "v" shape incision was made longitudinally to receive the infested petiole with the tip shaped as an inverted "v". Plants were kept inside plastic bags under greenhouse conditions for 3 days after grafting.
Petiole Section Insertions
A small portion of the petiole, just enough to bear 1 ovipuncture with eggs inside was inserted into the host petiole through a longitudinal incision. Parafîlm was used to hold the insert in place.
Petiole Sections Fastened to the Host Petiole
Petiole sections containing at least 1 ovipuncture were fastened to the host petioles well above the base of the plant. Parafîlm was used to wrap the section tightly to the host petiole so that the convex part of the petiole section made contact with the concave part of the host petiole (Fig. 2.2).
Results and Discussion
The techniques were evaluated by inspecting all plants; the criterion for success was larval feeding in the crown. Egg implanting was the best technique followed by petiole section insertions, and petiole sections attached to the host plant at the base of the petioles. These techniques were very effective especially at high humidity. The method of grafting was not effective at all, because the grafted section dried quickly and the eggs did not survive. The technique of wrapping a petiole section onto an uninfested petiole was
27 Figure 2.2: Petiole sections bearing at least one oviposition scar fastened to the Ihost parsley petioles
28 not effective; a large percentage of dead larvae was observed inside the sections. This technique can be improved by selecting more succulent sections that do not dry quickly.
When wrapping with parafîlm, newly hatched larvae can become trapped. Although egg implanting was very effective, it also was the most time-consuming method. The sections inserted at the base o f the petioles were the most time efficient and good survival is possible by selecting more succulent petiole sections.
Average Number of Eggs per Oviposition Scar
Caged parsley plants were exposed to carrot weevil adults for 5 days. Sixty-four randomly selected oviposition scars were dissected carefully under the stereoscope and eggs were counted. The results revealed an average of 3.1 ±1.3 eggs per oviposition scar
(Fig. 2.3), which is not signicantly different fi-om the results reported by Boivin (1985).
Carrot Weevil Rearing Technique in Parsley
Parsley plants grown in 10 cm pots with potting medium at 25 °C and 16 h light were caged with mated carrot weevil females when plants were at least 1 month old.
Cages were prepared by placing a clear plastic container, with window screen top and sides, inverted into a black plastic pot with the drain holes sealed with adhesive tape.
Plants were checked for oviposition scars 24 h later and weevils were removed firom infested plants. Uninfested plants were left in the cage with weevils until they become infested. The infested plants were kept in the greenhouse for approximately 25 d, and
29 0.3 "5
0.2 ^
Figure 2.3: Histogram of number of carrot weevil eggs per oviposition scar in parsley
30 then placed inside a cage with window screen sides. Plants were watered every day; fertilizer and fiingicide were added to the water once per week. Adult weevils were collected weekly or every 3 d beginning 30 days after oviposition.
Egg Stage Duration
Parsley petioles infested with carrot weevil eggs laid during the previous 24 hours were cut in sections immediately after the weevils were removed from the plant. About
60 petiole sections were distributed in 8 petri dishes and placed along with 5 carrot slices as food source for larvae, on moist filter paper. The dishes were kept in a growth chamber set for 16:8 (L:D) and 15 °C on average, ranging from 13 to 18 °C, and 74% RH on average, ranging from 60% RH to 88% RH. Five days after the eggs were laid and every day thereafter, carrot slices were dissected and larvae were counted. Larvae and carrot slices were discarded daily after counting. Time of eclosion ranged from 7 to 26 days after oviposition (DAO) (Fig. 2.4). In this experiment, the carrot weevil egg stage duration averaged 9.6 days and the median was estimated as 8.43 days. Martel et al.
(1983), determined an average of 8.3 days, and 4.5 ±0.9 days for egg development at 21
°C, and 27 °C, respectively, both at 60% RH, and 16 h photoperiod. Whitcomb (1965) reported an average of 6.5 days at 24 °C.
31 Carrot Weevil Egg Span
40
35
30 I.
►J 2 0 'S & s 15 I. 1 1 1 X 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
Days After Oviposition
Figure 2.4: Carrot weevil egg stage duration based on percentage of eclosed larvae after oviposition
32 Carrot Weevil Oviposition Preference
Parsley plants grown in 10 cm pots in a greenhouse at 25 °C and 16:8 (L:D) were caged with carrot weevil females to determine their preferred oviposition sites. The plants chosen for this trial were young plants that had all the first leaves still in place.
Results (Fig. 2.5) indicate that most of the eggs are deposited in the second trifoliate leaf (fourth true leaf) followed by the first trifoliate (third true leaf). More than
80% o f all oviposition scars were in the 2"** unifoliate, 1^ and 2"** trifoliate, and 1®^ and 2"** multifoliate. Eggs were inserted either on the concave or the convex part of the petiole; however the concave part of the petiole is preferred. Field observations of oviposition scar distributions matched the greenhouse observations.
Oviposition Activity
First generation adult weevils collected in August 1998 were kept in two cages, one containing 81 adults and the other containing 64 adults. Adults were provided with potted parsley plants, carrot pieces and water in an environmental growth chamber at 15
°C, 70% RH and 16:8 (L:D). The oviposition scars made by the female weevils in potted parsley plants were counted weekly most of the time, otherwise biweekly. Parsley plants were replaced after counting oviposition scars, as were water and carrot pieces. Female weevils oviposited for 305 days after they were caged. The number of weekly oviposition scars fluctuated between 100 and 500 during the period firom October 13^*, 1998 until
March 29*, 1999. After this date, the number of oviposition scars steadily declined until the end of June. The reduction in egg laying was due to adult mortality, primarily.
33 25
O 20
u_
Unifoliate Trifoliate Multifoliate Multifoliate 1st 1st 1st 3rd 5th LEAVES
Figure 2.5. Carrot weevil oviposition preference in parsley leaf petioles
34 Trench Study
In 1999, at the Muck Crops Branch of OARDC in Celeryvile, a trench was
installed to test its effectiveness in preventing carrot weevil adults from entering new parsley fields. The trench surrounded a 1.5 m^ area, had a depth of 30 cm, a width o f 30 cm, and was lined with black polyethylene. In cross section, one of the trench walls was vertical and the other had a 45° angle. Twelve adult weevils were released at the bottom of the trench. Two weevils were observed climbing the vertical wall very quickly and one of them was able to leave the trench. At the time these observations were made the plastic had few dust or muck soil particles on it. These fine particles were dry and easy to remove from the plastic surface by blowing. According to Boiteau et al. (1994), the
Colorado potato beetle cannot climb plastic-lined trenches when the plastic is coated with fine soil particles that adhere to it after a rain.
A few days later, after a rain, 12 new weevils were released at the bottom of the trench. The plastic wall was coated thoroughly with fine soil particles well adhered to the plastic surface after the rain. This time the number of attempts made by each weevil was recorded for one hour. Eleven weevils were active, they made 102 total attempts to climb the sides of the trench, and only one succeeded. The one success was by a weevil that climbed the wall along a fold in the plastic that was at a 45° angle. Fine soil particles interfere with Colorado potato beetle adhesive tarsal pads making it difficult for the beetles to walk up the plastic wall (Boiteau et al. 1994).
35 Parsley Growth Study
Parsley growth stages were described to know oviposition preferences for plants of different sizes. Ten parsley seedlings selected at random were transplanted 11 days after seeding in a jiffy pot tray filled with potting medium (Pro-Mix). Plants were transplanted to 11 cm round pots (G-450) filled with Pro-Mix. The j i ^ pot was removed before transplanting and the seedlings were carefully removed without breaking their roots. This study was conducted under greenhouse conditions at 25 °C and 16:8 (L:D).
Every other day for 74 days after seeding the leaves were counted. Leaves were counted in the following categories: cotyledons, unifoliate, trifoliate, and multifoliate leaves (Fig.
2.6). Cotyledons are the first pair of leaves, unifoliate leaves constitute the second pair, or first pair of true leaves with three-lobes. The first leaf of this pair does not have branched petioles or petioules, but the second leaf may have petioules somewhat resembling a trifoliate leaf. The trifoliate leaves constitute the second pair of true leaves and have 3 leaflets. All the leaves after the second pair of true leaves are multifoliate leaves having 5 or more leaflets. The rules for classifying unifoliate, trifoliate, or multifoliate leaves were the following: The first unifoliate leaf was recorded when the leaf surface was completely expanded, the second unifoliate leaf was recorded when it reached or surpassed the first unifoliate leaf in height. The same rules applied for trifoliate and multifoliate leaves as for unifoliate leaves.
Parsley seedlings were transplanted using procedures described above to pure muck soil and muck soil mixes with vermiculite and perlite, in two different proportions for each combination, 1:1 and 3:1 muck: perlite and muck: vermiculite. This trial was
36 Figure 2.6: Parsley plant at the four-true-leaf stage showing the 4*** leaf fully expanded. Carrot weevils seldom lay eggs in smaller plants
37 conducted to compare muck with other potting soils under greenhouse conditions. The two cotyledon leaves were expanded 12 days after planting (DAP) and they both remained on the plant for 19 days when planted in the potting medium “pro-mix”. The first unifoliate leaf was fully expanded at 19 DAP on average, and the second one was fully expanded at 25 DAP on average. The unifoliate leaves began to fall off the plant 48
DAP. The first trifoliate leaf was fully expanded 29 DAP on average, amd the second one was expanded 38 DAP on average. The trifoliate leaves began to fall oiff 63 DAP. The first multifoliate leaf, that is actually a pentafoliate, was fully expanded at ca. 40 DAP on average, then a new set of leaves appeared fully expanded, every 5 days on average. By
71 DAP, 7 multifoliate leaves were present. Parsley grown in pure m uck soil and in different muck soil mixes with perlite and vermiculite did not differ significantly in growth pattern (F=0.479; df=4,130; P=0.751), except for the unifoliate; stage (F=2.5; df=4,130; p=0.046).
In general terms, the growth of parsley in potting soil and all thie muck mixes was similar; the cotyledon leaves began to fall off between 34 and 44 DAP_, the 2"** unifoliate leaf was fully expanded between 25 and 28 DAP, the 2"*^ unifoliate lealf started to fall off between 48 and 65 DAP, the 1®^ trifoliate leaf was fully expanded betw^een 29 and 34
DAP, the 2"** trifoliate leaf was fully expanded between 38 and 44 DAJP, and the 1^ multifoliate leaf was fully expanded between 40 and 50 DAP.
38 Parsley Growth Stage and Oviposition Activity
Parsley growth stage was recorded when carrot weevil oviposition activity was just beginning in two commercial parsley fields seeded on February 28*** and March 7***,
2000. In the first field, 61 plants were sampled on May 15*, and 27 were sampled on May
25*; in the second field, 10 plants were sampled on May 24*, and 24 were sampled on
June 1^. The number of true leaves was recorded regardless of oviposition scars in the
samples. However, the fields were recently infested with carrot weevils. The infestations, based on 1000 samples (chapter 5), were 1% and 1.8% in field 1 and 2 when the first sampling was done. The results (Fig. 2.7) indicate that most plants had 4 or more true leaves by the time the oviposition activity started. Seven to ten days later, most plants had more than 6 true leaves (Fig. 2.7).
Relationship among Carrot Weevil Infestations, Planting Dates of Parsley, and Nematode Control
Introduction
The number of oviposition scars per infested parsley plant can vary firom 1 to 10 or more depending on the population density of carrot weevil, L. oregonensis (LeConte), but usually these numbers are low, no more than 3 oviposition scars per plant. One ovipuncture per plant causes economic damage and may kill young or small plants. Older or larger plants may withstand carrot weevil feeding, depending on the number of oviposition scars per plant. Carrot weevil control is achieved mainly by spraying insecticides. There is a potential for biological control, however, by using insect
39 F ie ld I F ie ld 2
24-May 15-May
0.7 3 0.8 à
O 30 0.5 S
0.4 0) 00
'2 3456789 4 5 6 7 No. of True Leaves No. of True Leaves
1-June 25-May 20 0.7 0.6 15 -0 0.5 . | #3 0 0.4 t- o 10 Ü 10 3 s 0.3 % 1 03 0.2 % 0.1 0.0 '2 3 4 5 6 7 8 3 4 5 6 7 8 9 No. of True Leaves No. of True Leaves
Figure 2.7: Histogram showing the number and proportion of parsley plants having different numbers of true leaves at two sampling dates in two parsley fields when carrot weevil oviposition scars were present. Field 1 was in Celeryville, Ohio and field 2 was in Hartville, Ohio, 2000
40 parasitoîds and entomopathogenic nematodes (Bélair and Boivin 1985,1995). The use of nematodes to control carrot weevil adults and pupae seems feasible in parsley fields because these stages are in contact with the soil. On the other hand, the use of nematodes to control larvae may be restricted to the brief period of time when the newly hatched larvae make contact with the soil, just before entering the plant root, perhaps 1 hour or less. There is a possibility, however, that nematodes enter the root.
The objective of this research was to compare yield effects of 5 levels of oviposition at 2 plant sizes treated with 2 species of nematodes. A check treatment with no nematodes was used for comparison. Two cuttings were made to determine firesh yields for treatment comparison.
Materials and Methods
Plant Size
Two plant sizes were produced by planting at 2 different planting dates, August
11*, 1999 and September 10*, 1999. Carrot weevil infestation started 75 days after the first planting and 45 days after the second. Plants were large enough for female weevils to lay eggs 45 days after planting. These plants had 5-6 true leaves (2 unifoliate leaves, 2 trifoliate, and 1 or 2 leaves with 5 or more leaflets). I expected the older, larger plants to tolerate more weevil damage than the younger, smaller plants.
41 Pots, Potting Medium, and Transplanting
Round plastic (15 cm diameter) pots were filled with “Pro-Mix” potting soil.
Parsley seedlings were transplanted 16 days after seeding, one plant at the center of each pot. Plants were watered every day or as needed, and fertilizer was added to the water once per week.
Plot Layout and Experimental Design
Ten replicates of a 2x3x5 factorial experiment consisting of 2 plant sizes, 3 nematodes treatments (2 different species and no nematodes), 5 levels of oviposition scars per plant (0, 1,2,3, and 4) were conducted under greenhouse conditions during
August — December, 1999.
Greenhouse Settings
The temperature ranged between 15.8 and 33 °C on average with a minimum of
10.3 °C and a maximum of 45.4 °C. Lights (HID, 60 HZ, 9.50 A, HPF, 1000 W) were on from 6:00 to 22:00. Lights were kept on during the daytime to compensate for clouds and reduced light intensity during September through January.
Oviposition Scars and Infestation Technique
Five levels of oviposition scars per plant were considered in this experiment.
Weevil infestation was done manually by attaching petiole sections containing 1, 2, 3, or
4 oviposition scars to the host petiole. Petiole sections bearing oviposition scars were cut
42 from plants 5 days after oviposition, just before larvae were expected to hatch. A check plant with no petiole sections attached (no oviposition scars) also was included. The petiole section was attached about 2.5 cm above the base of the host petiole in trifoliate leaves if available (Fig. 2.2). Parafrlm was used to wrap and hold the petiole section to the host petiole. The paraftlm wrapping was cut 3 days later to prevent neonates from being trapped. Individual plants were kept inside plastic bags for 5 days after nematode infestation to ensure high humidity for eggs and larvae. One replicate was infested per day for 10 consecutive days because the manual infestation was time consuming.
Nematodes
Nematode treatments were applied in a water solution containing either
Steinemema carpocapsae (Weiser) ‘All’ strain (Rhabditida: Steinemematidae),
Heterorhaditis bacteriophora Poinar ‘Lewiston’ strain (Rhabditida: Heterorhabditidae), or water alone as the control treatment. These species and strains caused >90% mortality to carrot weevil larvae in other laboratory experiments (Miklasiewicz, unpublished data).
Cultures were obtained from Integrated Biological Control Systems (Greendale, IN,
USA) and were maintained in the Ohio State University TurfiNematology Laboratory. All nematodes were cultured in vivo in the laboratory at 20-23 °C by infesting fourth instars of greater wax moth. Galleria mellonella L. (Dutky et al. 1964), then recovering infective juvenile stage nematodes (IJ’s) using white traps (White 1927). The IJ’s used for tests emerged from wax moth cadavers 2-10 days before application. Nematodes were applied to the surface of the potting medium within 48 h of attaching petiole sections, by dripping
43 5 ml of solution containing approximately 12,385 IJ’s, equivalent to approximately
750,000 IJ’s per m^ of surface. Solutions were agitated until application, and the solution was dripped evenly over the potting medium. Pots were watered immediately after nematode applications and as needed thereafter to maintain moist soil conditions.
Yield
Yield of parsley foliage was measured by cutting plants 5 cm above the soil surface. Yield of the first planting was measured on November 10* and December 21®*
1999; the second planting on December 10*, 1999 and January 20*, 2000. Fresh weight was determined for each plant at each cutting.
Adult Weevil Collection
To collect carrot weevil adults emerging firom the soil in the pots, the upper part of each pot was covered with nylon fabric tied tightly to the parsley stem just below the crown, or base of the petioles so that the soil was covered but not the plant. Pots were covered with this material before adult emergence, and for approximately 30 days.
Carrots were placed in each pot as a food supply for the emerging adult weevils. Adults were counted on December 27*, 1999 or approximately 60 days after the last infestation.
Root Damage Evaluation
Root damage was evaluated immediately after the last yield sample. Damage in the external part of the root was rated firom 1 to 7 depending on the number of root
44 sections showing feeding scars. Seven 1-cm sections were examined firom the crown to the tip of the root; section number 7 corresponded to all root beyond 6 cm. The same sections were rated for the internal damage. A plant with a score of 14 would be a plant with damage in all 7 root sections, both externally and internally.
Results
Fresh Yield
Fresh yields obtained in the first cutting were analyzed by analysis of variance using MINITAB™ (Minitab, 1999) and Systat (SPSS, 1999). The data comply with the
ANOVA assumption for equal variances (Bartlett’s test: 1®* cutting P=0.299,2"** cutting
P=0.707), so transformation was not necessary. The number of oviposition scars per plant had a significant effect on firesh yield in both cuttings (table 2.1). Plant size also had a significant effect on yield in the second cutting. The nematode treatment did not have any significant effect on firesh yield. None of the interactions terms were significant in either cutting.
Root Damage
There were significant differences in root damage associated with the number of oviposition scars per plant (table 2.2). Plant size and nematodes did not have any significant effect on root damage due to larval feeding. None of the interactions terms had any significant effect on root damage in parsley.
45 Analysis o f Variance
Source 1®* Cutting 2"“ Cutting D f F P d f F P Rep 9 3.332 0.001 9 1.645 0.103 Size 1 0.006 0.937 1 51.059 0.000 Oviposition scars 1 7.760 0.006 1 3.968 0.047 Nematodes 2 1.222 0.296 2 0.370 0.691 Size*Ovip 1 0.610 0.435 1 0.739 0.391 Size*Nem 2 0.106 0.899 2 0.034 0.967 Ovip*Nem 2 0.757 0.470 2 0.241 0.786 Size*Ovip*Nem 2 0.271 0.763 2 0.384 0.681 Error 279 278
Table 2.1 ; Analysis of variance for the effect of plant size, nematodes, and oviposition scars on parsley yield arranged in a 2x3x5 factorial experiment with 10 replicates
Adult Emergence
No adult weevils were collected firom the potted parsley. No adult feeding scars were found on carrots supplied as food, and no adults were found outside the screened pots in the greenhouse. Only 1 adult was found, dead, inside the root of 1 plant.
Discussion
Larval mortally appeared to be very high in the experiment, because no adult weevils were collected firom the infested plants, even in those plants without nematodes.
46 Analysis o f Variance
Source df F P Rep 9 2.162 0.025 Size 1 0.072 0.788 Oviposition scars 1 42.606 0.000 Nematodes 2 0.062 0.940 Size*Ovip 1 0.538 0.464 Size*Nem 2 0.074 0.929 Ovip*Nem 2 0.260 0.771 Size*Ovip*Nem 2 0.464 0.629 Error 276
Table 2.2: Analysis of variance for the effect of plant size, nematodes, and oviposition scars on parsley root damage arranged in a 2x3x5 factorial experiment with 10 replicates Adult Emergence
Little damage was observed in the roots. I checked the larval eclosion in all 10 replicates
and found the chorion in most of the attached petioles, suggesting a high eclosion
percentage. High greenhouse temperatures may have caused larval mortality. Maximum
daily temperatures ranged from 27.5 to 45.4 °C. I did not sample for entomopathogens but they cannot be ruled out as the cause of larval mortality, especially Beauveria spp.
High temperature and humidity favors this fungus.
The number of oviposition scars per plant had a significant effect on parsley yield both in the first and second cutting, and also on root damage. These differences are
47 difficult to explain because the larval population died with little observable damage being
done. The diffierence in mean yield, in the second cutting, between the first and second
planting (plant size) was significant. However, because little weevil damage occurred, the
difference does not appear to be due to weevil larvae. Mean yield was higher in the older,
larger plants than in the younger, smaller plants. Potted parsley is difficult to manage because it grows quickly, the root system fills the inside of the pot, and water retention becomes poor.
48 CHAPTERS
DETERMINING THE RELATIONSHIP BETWEEN CARROT WEEVIL
INFESTATIONS, DAMAGE AND PLANTING DATES OF PARSLEY
INTRODUCTION
The carrot weevil, Listronotus oregonensis (LeConte) is the most damaging pest of Ohio parsley (Simonet 1981). Adults feed on and lay eggs in carrot, celery and parsley; larvae tunnel in the roots and cause most of the damage. Larvae often kill the plant, especially when it is small and feeding is in the upper part of the root, blocking the flow of water and nutrients. Adults walk rather than fly from their overwintering sites to new parsley fields in early spring, or they stay to feed and reproduce if they emerge in the overwintered parsley.
Carrot weevil adults become active in early spring as soon as the temperature and photoperiod are favorable. Stevenson (1985) observed that oviposition peaks as early as mid May when the temperature exceeds 24 °C. According to Boivin (1988), carrot weevil started to lay eggs in carrots placed on the ground by 147 ± 9 degree-days (DD) (base 7
°C) and 90% of the eggs were deposited by 456 ± 47 DD. Stevenson and Boivin (1990) established that the carrot weevil is less likely to attack carrots planted late in the season.
49 The life cycle of L. oregonensis, from egg to adult lasts from 24 to 47 days within the temperature range of 27 to 21 °C (Wright and Decker 1958, Martel et aL 1975, Simonet
1981). Adult weevils may colonize parsley fields anytime during crop growth, but females will not lay eggs before the plant reaches the four-true-leaf stage. Parsley is planted from late February until May so that supply is maintained for the fresh market.
Three or four cuttings are usually made; the first one in mid June, and the following cuttings are made every four weeks, approximately. According to parsley growers, significant crop damage and losses are seen in the second cutting of early-planted parsley.
Based on their observations, two hypotheses were considered; 1. carrot weevils lay eggs principally in the larger early-planted parsley, 2. weevils lay eggs in both early and late plantings, but the plants are less susceptible to damage in late plantings. To design an efficient control strategy, I needed to distinguish between these two possibilities and determine the relationship between carrot weevil phenology and planting dates of parsley.
Therefore, I monitored carrot weevil activity by counting eggs, larvae, adults and feeding scars in parsley planted at difierent times.
MATERIALS AND METHODS
Parsley Plantings and Experimental Design
Five parsley planting dates in 1998 and three planting dates in 1999 were arranged in a randomized complete block design with 4 replicates, each replicate containing one planting on each planting date. Parsley was planted on March 30*, April 13*, April 24*,
50 April 30*, and May 5* 1998, and on April 2"“, April 22"“, and May 10*, 1999, at the
Muck Crops Branch of OARDC, in Celeryville, Huron County, Ohio.
Weevil Infestation
In 1998, in addition to the naturally occurring infestation, 58 adult weevils collected by using Boivin traps baited with carrots from commercial fields in early spring were released broadcasting them by hand. The weevils were distributed evenly throughout the plots, between May 13* and 29*. In 1999,200 adults were collected fix)m an overwintering parsley field then released in the experimental plots on two different dates, 100 weevils on May 4*, and 100, on May 6*, again distributed evenly throughout the plots. The difference in releasing dates between 1998 and 1999 was due to the collecting method used; Boivin traps took longer than manually collecting adults in overwintering parsley fields.
Parsley Plots and Agronomic Practices
Soil was plowed to 20 cm, then fertilized with 16-8-26 at 784 kg/ha, and disked.
Parsley was seeded at a rate of 49 seeds/m in raised beds using a Stanhay 870 belt-type planter. The bed dimension was 1.83 m x 45.72 m with 3 rows/bed, and 41.9 cm between rows, so the plot dimension was 36.6 m x 45.72 m in 1998 and 22 m x 45.72 m in 1999.
The cultivar Forest Green was planted using seeds obtained from local parsley growers.
Lorox® was used for weed control when plants had two or more true leaves. Hand weeding was also done for weed control during the growing season. Fertilizer was
51 sprayed after each cutting to aid parsley regrowth. No insecticide was sprayed during the
monitoring period. Two cuttings were taken from each planting, each when the parsley
was suitable for commercial use (Ohio Vegetable Production Guide 2000).
Adult Counts
Adult weevils were counted weekly, for 17 weeks in 10 randomly selected sections of row, 3 m long, within each plot. Parsley plants were shaken and then adult weevils were counted on the soil surface or under plant residues. The adult of this insect tends to drop from the plant and remain motionless when disturbed, but moves quickly after approximately 30 seconds. Adults were left in the field after counting except in the last two sampling weeks when they were removed.
Egg and Larval Counts
Every week, for 17 weeks, eggs were counted on petioles of 20 randomly selected plants in each plot. Each plant petiole was cut from the plant and inspected careftilly.
Oviposition scars were recognized as round holes covered with dark exudates.
Oviposition scars were dissected and eggs were inspected carefully to make sure that they were viable and less than 5-6 days old. Plants sampled for eggs were also inspected for larvae and feeding on the roots. Larvae were counted by carefully dissecting the stem and root if it showed feeding damage externally. Roots and stem with no apparent damage were cut longitudinally to inspect for larva feeding inside. Feeding on roots or stems or both was recorded, regardless of the presence of larvae.
52 Parsley Cuttings
Two parsley cuttings were made in each planting in both years. Cuttings were made sequentially at weekly intervals beginning with the first planting. Cuttings were made to simulate commercial practice, taking each plant at approximately 5 cm above the crown. In 1998, just before the first cutting in all five parsley plantings, 40 plants per plot were taken at random and the number of oviposition scars was counted both above and below the cut. These data were used to estimate the proportion of eggs removed during harvest.
Statistical Analysis
Multivariate repeated measures analysis of variance was used to compare the pattern of population change over time for the different planting dates of parsley (von
Ende 1993); seasonal means, slope coefficients and quadratic coefficients were calculated for each plot. Seasonal means were calculated for all 17 weeks of sampling, while slope and quadratic coefficients were calculated for specific periods of time or particular weeks during which the pattern of population change was important for comparison or useful for decision making. The slope coefficient was calculated for overwintering adults firom
May 15* to June 5*, 98 and firom May 14* to June 18*, 99; for first generation of weevils fi"om July 10* to August 28*, 98 and firom July 16* to August 27*, 99; for eggs firom
May 22"** to June 26*, 98 and fi"om May 28* to June 18*, 99; for larval population fi"om
June 5* to July 10*, 98 and firom June 11* to July 16*, 98; and root feeding firom June
12* to July 10*, 98 and firom June 4* to August 20*, 99. The quadratic coefficient was
53 calculated for adults from May 15* to July 13*, 98 and from June 14* to July 16*, 99; for eggs from May 15* to July 10*, 98 and from May 28* to July 16*, 99; and for larval population from June 19* to August 7*, 98 and from June 11* to August 20*, 99.
Seasonal means, linear rates of increase (slopes) o f carrot weevil adults and larvae, and proportion of plants with root feeding, and quadratic coefBcients describing the shape of population change over time were compared among planting dates by analysis of variance. Systat® 9.0 (SPSS, Chicago, Illinois) was used for all the statistical analyses.
RESULTS
Carrot Weevil Infestation and Planting Dates of Parsley
The initial population of overwintering adults, one complete generation, and at least a partial second generation were observed in parsley between May 14* (week 1) and
September 4* (week 17), in 1998 (Fig. 3.1) and during the same period in 1999 (Fig. 3.2).
The majority of eggs were deposited between May 22"** and July 3* in 1998 and between
June 4* and July 16* in 1999 (Figs. 3.1 and 3.2). Most of the larvae were counted from
June 19* through July 31^ in 1998 (Fig. 3.1) and from June 18* through August 6* in
1999 (Fig. 3.2). The overall seasonal means of adults, eggs, larvae, and root feeding in both 1998 and 1999 were significantly different among plantings (table 3.1). Root feeding and all carrot weevil life stages sampled steadily decreased from the earliest to the latest planting in both 1998 and 1999 (Figs. 3.1 and 3.2).
The rate of increase of the population of overwintering adult weevils was not significantly different among the first four plantings between May 15* and June 5*, 1998
54 CARROT WEEVIL ACTIVITY 1998
P lan tin g I P lan tin g 2 Planting 3 Planting 4 Planting S
Z 0 .5
0.7 5 0.6 Ï 05 I 0.4 o*. 0 3 ^ 0 3 s? < 0.1
■5 0.» c5 0.7 # 0 .6 a Oj iH £I 0
Figure 3.1 : Carrot weevil seasonal activity in five planting dates of parsley. Celeryville, 1998
55 CARROT WEEVIL ACTIVITY 1999
-Plant mg 1 -Planting 2 A Planting 3
1.2
g 0.8 ui 0.6 à 0 .4 ^ 0.2
0 .2 5
0 .0 5
0 .6 0 .5 0 .4 0 .3 0 .2
5/14 5/28 6/11 6/257/9 7/23 8/6 8/20 9/3
Figure 3.2; Carrot weevil seasonal activity in three planting dates of parsley. Celeryville, 1999
56 1998
df F P-Value
Adults 4 18.320 <0.0001 Eggs 4 25.138 <0.0001 Larvae 4 27.342 <0.0001 Root Feeding 4 97.447 <0.0001 Error 12
1999
df F P-Value
Adults 2 7.708 0.022 Eggs 2 23.903 0.001 Larvae 2 10.029 0.012 Root Feeding 2 63.602 <0.0001 Error 6
Table 3.1: Analysis of variance results for seasonal means of carrot weevil population activity among planting dates of parsley. Muck Crops Branch, Celeryville, Ohio
57 (table 3.2). The fifth planting had a low infestation for the entire sampling period and
therefore a much lower rate of arrival of overwintering adults. Based on quadratic terms,
the pattern of population change over time in overwintering adults was significantly
different among the first four plantings in 1998 (table 3.2). In 1999, the overwintering
adults started to increase on May 14* and peaked on June 18* (Fig. 3.2). The rate of
population increase during this time was significantly different among the three plantings
Slope Quadratic
Year df d f
Overwintering Adults 1998 3,9 3.226 0.075 3,9 7.797 0.007 Overwintering Adults 1999 2,5 28.254 0.002 2,6 64.544 <0.0001
1^ Generation Adults 1998 4,12 6.735 0.004 ------
1^ Generation Adults 1999 2,6 3.786 0.086 -™ ------Eggs 1998 2,6 27.243 0.001 3,9 40.125 <0.0001 Eggs 1999 1,3 9.216 0.056 2,6 26.167 0.001 Larvae 1998 3,9 29.944 <0.0001 4/12 29.278 <0.0001 Larvae 1999 1,3 1.342 0.331 1,3 3.627 0.153
Root Feeding 1998 4,12 25.756 <0.0001 — — — —— - — ——
Root Feeding 1999 2,6 13.113 0.006 ----- ————
Table 3.2: Analysis of variance results for rate of population increase (slope) and pattern of population change (quadratic) of carrot weevil adults, eggs, larvae, and root feeding among planting dates of parsley. Muck Crops Branch, Celeryville, Ohio
58 (table 3.2) as was the pattern of change over time (table 3.2). The first generation of adults began to increase on July 10^ and peaked on August 28*, 1998 (Fig. 3.1). The rate of population increase during this period was significantly different among the five planting dates (table 3.2). In 1999, the first generation of adults started to increase on July
16* and peaked on August 27*, and there were no significant differences in the rate of increase among the plantings (table 3.2).
Carrot weevil oviposition activity occurred mainly between May 15* and July 10* in 1998, and between May 28* and July 16* in 1999 (Figs. 3.1 and 3.2). Adult females started to lay eggs at different times, however, depending on the planting date, so the oviposition peak tended to be progressively later in successive plantings (Figs. 3.1 and
3.2). In 1998, peaks of oviposition occurred on June 12* and June 26* and the rate of increase since the oviposition started was significantly different among the first three plantings, as well as the pattern of change over time in eggs among the first four plantings during the period of greatest oviposition activity (table 3.2). The fourth and fifth planting in 1998 had no eggs on most sample dates; therefore, they were not included in the statistical analysis for the rate of population increase comparison, and the fifth planting was not included in the statistical analysis for the pattern of change comparison (table
3.2). In 1999, the greatest oviposition increase occurred from May 28* to June 18* in the first planting and from June 4* to July 9* in the second planting at the same rate for both plantings. The pattern of change over time in egg population was significantly different among the three plantings during the greatest oviposition activity in 1999 (table 3.2).
59 In 1998, the larval population started to increase on June 5* and peaked on July
10* in the first two plantings, on July 17* in the third plating, and on July 31^ and
August 7* in the fifth and fourth planting respectively (Fig. 3.1). The rate of larval increase between June 5* and July 10* was significantly different among the first four planting dates (table 3.2). The greatest larval activity occurred between June 19* and
August 7* (Fig. 3.1). The pattern of larval population change over time was significantly different among the 5 plantings during that period (table 3.2); however, this pattern was the same between the first two plantings (F=2.417; df=l,3; P=0.218). In 1999, the larval population started to increase on June 4* and peaked on July 16* in the first planting
(Fig. 3.2). Larvae in the second planting started to increase one week later and peaked two weeks earlier than the first planting. The third planting had very little infestation. The rate of larval population increase between June 11* and July 16* was not significantly different between the first and second planting (table 3.2), although the time of the increase differed. Most larvae were present between June 11* and August 20* (Fig. 3.2) and the pattern of population change over time was not significantly different between the first two plantings (table 3.2).
In 1998, root feeding started to increase on June 12* and peaked in July 10* (Fig.
3.1). The rate of increase in proportion of plants with feeding during this period was significantly different among plantings (table 3.2). In 1999, root feeding started to increase on June 4*, June 11*, and July 2"** in the 1^, 2"**, and 3"* plantings, respectively.
60 and steadily increased until August 20* in all plantings (Fig. 3.2). The rate of increase in proportion of plants with feeding during this period was significantly different among plantings (table 3.2).
Eggs Removal at Harvest Time
The number of eggs removed from the parsley plants during the first cutting was significantly different among planting dates (F=5.212; df=4,12; P=0.011) (table 3.3). The percentage of eggs removed was significantly greater in the first two plantings compared to the last thr-?e plantings (F=12.17; df=4,12; P<0.0001). The effect of planting date on the percentage of eggs removed can be described by a linear contrast (F=43.008; dfr=l,12;
P<0.0001).
Planting Eggs Present Eggs Present Eggs Removed Date Above Cutting Below Cutting %
1 58 92 38.7 2 75 77 49.3 3 7 45 13.5 4 3 22 12.0 5 1 23 4.2
Table 3.3: Total number of carrot weevil eggs present above and below the cutting point, and percentage of eggs removed in five planting dates of parsley
61 DISCUSSION
As noted in the introduction, parsley growers claim that most carrot weevil damage is seen in the second cutting of early-planted parsley. Based on results reported herein, carrot weevils lay eggs principally in the larger early-planted parsley. Weevils deposit eggs in all plantings but more in early than later plantings. Likewise, the larval population is present in all plantings but at higher population density in early plantings.
Root feeding also is seen in all plantings but stand loss is significantly greater in the first two plantings, which means that parsley plantings sown firom late February until the third week of April are the most susceptible to damage by the carrot weevil in Ohio. A critical amount of damage, caused by a critical number of larvae, apparently is required to reduce parsley stand or yield.
In this study, parsley fields planted after the third week of April had low infestations of overwintering adults, and corresponding low numbers of eggs, larvae, and roots with feeding, which is comparable to the findings of Stevenson and Boivin (1990) who established that the carrot weevil is less likely to attack carrots planted late in the season in Ontario and Quebec (Canada). However, the number of adults corresponding to the first generation increased at the end of the season, especially in the last planting in
1999. Because the parsley plots were planted close to each other, the emerging weevils may have redistributed in all the plantings at the end of the season.
The rate of increase of the population of overwintering weevils was the same among the first four plantings in 1998, so the increase took place at different times, but about the same rate once started. In contrast, the pattern of population change over time
62 was different, so the shapes of the peaks were different; early plantings had high sharp peaks and later plantings had low broad peaks. In 1999, the rate of increase of the population of overwintering adults was high in the first planting, intermediate in the second, and low in the third and last planting. The maximum peak o f overwintering adults was reached two weeks later than in 1998, due possibly to weather conditions (Fig.
3.3). The period of adult population increase is probably the most critical with respect to carrot weevil sampling and control decisions. The first generation of weevils declined by the end of the sampling period in both years. Because all weevils found were collected and removed firom the field, sampling may have contributed to the decline. Stevenson
(1976) reported that the first generation of adults appeared fi"om about mid-July until
November in Ontario (Canada), which is similar to what I found in this study, except that
I did not sample beyond the first week of September. A small population of a second generation of weevils may have emerged because I observed a little increase in egg and larval populations at the end of the sampling period. According to Simonet (1981), carrot weevil may develop two generations each year in Ohio.
The oviposition activity took place between mid May and mid July in 1998 and between the end of May and mid July in 1999 with the greatest peaks in June in both years. Stevenson (1976) reported similar results in Ontario (Canada), observing that females started to lay eggs in celery as early as mid May and the average date of peak oviposition was June 6*. After this period egg counts were very low; however, by the end of July these counts began to increase a little, both in 1998 and 1999. Plantings made on
May 5*, 1998 and May 10*, 1999 had almost no eggs in June. The rate of increase in egg
63 TEMPERATURE 1998
Min. Temp. ■Max. Temp. 40 35 I 25 I
% % % % % 'O
TEMPERATURE 1999
40 35 30 25 2
% ^
Figure 3.3: Daily maximum and minimum temperatures during the sampling period at the Muck Crops Branch
64 population was difTerent among the first three plantings; the egg population in thel^
planting increased faster than the 2"** and 3^** plantings in 1998. In contrast, the rate of
population increase was the same between the 1** and 2"** plantings in 1999, the increase
started at different times but at the same rate. The pattern of population change over time
in egg population was different among plantings in both years, so the shape of the peaks
was different; the 1®‘ planting in 1998 and the 2"** planting in 1999 had high and broad
peaks, the 2"*^ and 3"* plantings in 1998 and the 1^ planting in 1999 had high and sharp peaks, and the last two plantings in 1998 and the last planting in 1999 had low and broad peaks.
Carrot weevil larvae were present in the parsley plots fi’om the first week of June until the first week of September in 1998 and 1999, but the greatest numbers occurred during July. By the time of the 2"** cutting in commercial parsley fields, about mid July in the earliest planting and late July for the following planting (parsley growers’ survey), a great amount of damage may have occurred, explaining why growers notice yield loss at this cutting. In 1998, the population of larvae began to rise again, apparently in all the plantings, after the third week of August, and in 1999, after the second week of August in the second planting. This small increase may have been the partial second generation of weevils that was previously detected in the egg population. In 1998, the highest rate of larval population increase occurred in the 1®* and 2"** planting and steadily decreased until the 4* planting. In 1999, the rate of larval population increase was similar for the first two plantings, but at different times. In 1998, the pattern of change over time in larval population was different among the first four plantings, the 1^ planting had a delayed
65 high broad peak, the 2"** had a delayed high sharp peak, the 3^** had a low broad peak, and the 4^** planting had a low increase without peaks. In 1999, plantings 1 and 2 had successive high low peaks of similar shape.
Parsley roots with larval feeding steadily increased week after week in early plantings until July 10*, July 17*, and July 24* in the first, second, and third plantings, respectively in 1998. The number of plants with feeding scars in the roots declined steadily until August 7* (plantings 2 and 3), and August 21®* (planting 1), and then increased again until August 14* (plantings 2 and 3), and August 28* (planting 1). In
1999, the number of plants with root feeding increased until August 20*, decreased until
August 27*, and then increased again in all three plantings. The brief decline in number of roots with feeding may be due to plant death caused by the carrot weevil. As damaged plants died and were no longer counted, the proportion of plants with root feeding declined. The second increase in the proportion of plants with larval feeding in the roots may be due to the partial second generation of weevil larvae that started after the second week of August. Stevenson (1976) recorded that carrot fields planted in early May began to show damage due to carrot weevil in early June and reached a peak in early July.
Even though the number of carrot weevil eggs removed during the first cutting was high in the first two plantings, the impact of this removal may not be significant in infestation reduction. Only one cutting is made during the period of oviposition activity, and therefore, only a small proportion of the eggs laid throughout this period is removed.
66 The number of eggs removed firom each planting was associated with the peak of oviposition activity in each planting, as well as the steady decline in oviposition activity observed firom the 1®* to the 5^ planting.
67 CHAPTER 4
ESTIMATING ECONOMIC INJURY LEVELS AND ACTION THRESHOLDS FOR
CARROT WEEVIL IN PARSLEY
INTRODUCTION
Carrot weevil economic thresholds can be defined either by number of adult weevils per unit area or proportion of plants with oviposition scars. Either measure could identify when the pest population must be reduced in order to prevent the economic injury level firom being reached (Stem et al. 1959).
Most carrot weevil damage in parsley is caused by the larval stage, but insecticide sprays target the adult stage. Newly hatched larvae usually go down to the roots by walking on the external surface of the petiole, or by dropping to the ground and entering the roots by tunneling through the crown or the roots. However, some larvae tunnel through the petioles all the way down to the crown and roots. Because of these behaviors, larvae are exposed and susceptible to insecticides only for a very short period. Therefore, larval control currently is not feasible, and this insect must be detected before a significant number of eggs are laid.
68 I plan to define the EIL based on the percentage of plants with root damage, but the sampling protocol will be based on percentage of plants with oviposition scars as a predictor of damage in the roots. Therefore, I need to know the relationship between different numbers of oviposition scars per plant and root damage to the plant.
Early plantings of parsley are more afiected by carrot weevil than late plantings due to greater infestations in early-planted parsley (chapter 3). However, it is not known if different plant sizes, as a function of date of planting, will have the same yield reduction when the infestation for early and late plantings is the same. In this chapter I want to test the hypothesis that there is no significant difference in yield loss between two different plant sizes infested with the same number of carrot weevil oviposition scars.
I know of no references for carrot weevil economic thresholds in parsley. Zhao
(1990), suggests 5 eggs/100 carrots as the economic threshold for carrot weevil in early carrots, and a range of 3.8 to 5.3 eggs for every 100 carrots in a mid season carrot crop.
This research is intended to define the EIL for carrot weevil in parsley based on the percentage of plants with oviposition scars, or infested plants. If thresholds can be defined in terms of percentage of plants with oviposition scars, then this simple measure can be used for pest management decisions.
MATERIALS AND METHODS
Observations and Measurements in Research Plots
In 1999, parsley plants that were naturally infested with different numbers of oviposition scars and planted at each of two different planting dates were observed at the
69 Muck Crops Branch of OARDC in Celeryville, Huron County, Ohio. Using a completely randomized design, thirty plants were marked with ribbons for each of the following categories: 0,1,2, 3, and 4 or more oviposition scars. Plant size was varied by planting at each o f two different dates, so parsley was planted in two plots, one on April 2"** and the other on April 22"**. Yield per plant was measured in one cutting on July 8* for the 1®* planting, and in two cuttings, one on August 3^** and the other on September 8* for the 2"** planting. Dead plants with carrot weevil damage were also counted at each cutting.
In 2000, parsley plants that were planted at each of two different dates and infested with different numbers of oviposition scars were observed at the Muck Crops
Branch of OARDC. Using a 4x2 factorial experiment with 10 replicates arranged in a randomized complete block design 10 plants were marked with ribbons for each of the following categories: 0,1,2, and 3 or more oviposition scars. Parsley was planted in two plots, one on April 14'*’ and the other on April 28***. Yield per plant was measured in two cuttings: on July 11* and August 16'*’ for the 1®' planting, and on July 25'*’ and August 16'*' for the 2"** planting, respectively. Dead plants with carrot weevil damage were also counted at each cutting.
Selected plants were laced securely and tightly with long ribbons at the base of the plants under the leaves. Ribbons of different patterns and colors were used to identify each oviposition category. Flags of different colors were also used to help localize each infested plant in all the plots at the research farm. In 2000, the entire plant was taken in the second cutting of both plantings to correlate the number of oviposition scars with root damage.
70 In 1999, parsley was seeded at a rate of 49 seeds/m in raised beds using a Stanhay
870 belt-type planter. The bed dimension was 1.83 m x 45.72 m with 3 rows/bed, and
41.9 cm between rows. No insecticide was sprayed during the monitoring period. Two cuttings were done in each planting. In 2000, plots consisted of two sets of 3 rows, 41.9 cm apart between rows, and 1.02 m between sets. Plots were 6.1 m long arranged in blocks, 4 m apart Parsley seed at a rate of 11.2 kg/ha was sown in flat soil using a cone planter. Insecticide (Guthion®) was sprayed right after the experiment set up was completed in order to kill adult weevils and stop females firom laying more eggs. In both years, the cultivar Forest Green® was planted. Soil preparation, fertilization, and weed control were done according to the Ohio Vegetable Production Guide (2000).
In 1999, polynomial regression analyses were performed to estimate the relationship between parsley firesh weight and number of oviposition scars. Chi-square analyses were performed to test for significant differences in plant mortality as a fimction of number of oviposition scars. Data were transformed to meet the ANOVA assumptions according to the ladder of powers cited in Fry (1996). The reciprocal of the square root was used to transform the data firom cutting 1 of planting 1, and square root, to transform the data firom both cuttings of planting 2. In 2000, analysis of variance was used to test the effect of number of oviposition scars and planting date of parsley, and their interaction, on average firesh yield per plant and percentage of plant mortality.
71 Observations and Measurements in Commercial Fields
During 2000, the relationship between percentage of infested parsley plants and yield was investigated in three commercial parsley fields located in Celeryville and
Hartville, Ohio. FLfty-30-cm lengths of parsley row were randomly selected for observations. The sections of row contained a variable number of plants and percentages of infested plants. All plants were removed with a trowel right before the second cutting and taken to the lab in plastic bags for yield determination. Each plant was cut about 5 cm above the crown, the top was weighed, and the health status of the root was recorded as having no carrot weevil damage, few feeding scars present, extended feeding injury on the surface of the root, or having extensive feeding damage inside the root. Dead plants were also recorded. Infestation was based on proportion of plants that had root damage in each 30-cm row section.
Commercial plantings consisted of either raised beds with two rows, 25-28 cm apart, 20.3 cm height, and 91.4 cm from bed center to bed center, or flat beds with five rows, 33 cm apart, and 1.83 m from bed center to bed center. The soil was plowed and disked; fertilizer (NPK) was broadcasted before and after plowing and the amount varied between 560-840 kg/ha. Lorox® was used to control weeds when parsley had 2 true leaves, or at planting. Parsley was planted using a Stanhay planter for raised beds or a
Planet Junior for flat beds with a seeding rate that varied between 11.2-and 16.8 kg/ha.
Guthion®, Malathion®, Pounce®, and Ambush® were the insecticides used to control carrot weevil adults. The commercial fields sampled in this experiment ranged from 1.3 ha to 3 ha, and were planted between February 28* and March 8* (Grower’s survey).
72 The proportion of infested plants per 30 cm-row section was estimated and
regressed against the total yield (grams firesh weight) of row in commercial fields.
Regression analyses were performed on the pooled data firom all the three commercial
fields because no significant interaction between field and the relationship between firesh yield and carrot weevil infestation was observed. The number of plants per row section was considered as a covariate, but it was not used because it accounted for little variation in yield (F^.O; df=l,3-5; P>0.075).
The EIL was estimated, by using the mathematical expression given by
Southwood and Norton (1973), as the infestation percentage at which the cost of control
(C) is exactly equal to the difference between the value of the parsley crop when control is used and the value of the parsley crop when control is not used (YcP-YncP), or when
C/(YcP-YncP)=l. The cost of control per hectare and the market value of the produce used in these calculations were obtained firom parsley growers in a survey.
RESULTS
Research Plots
In 1999, no significant efiect of oviposition scars on firesh weight was observed in the first cutting of the first planting (Fig. 4.1, table 4.1). The two cuttings firom the second planting show a clear pattern of yield loss (Fig. 4.1, table 4.1). Yield decreased firom the control with no infestation to those plants with the greatest number of oviposition scars
(Fig. 4.1). In cutting 2 the relationship between yield and number of oviposition scars was nonlinear (table 4.1). In planting 2 ,1 observed significant differences in yield between 0
73 PLANTING 1 CUTTING I
I Avg. Weight per Plant (g) • Dead Plants (99 •SB I 3 r 70 CL. 3 60 & 2 50 40 2 I I 30 1 20 1 I 1 IP < 0 0 I 2 3 4 Oviposition Scars
PLANTING 2 CUTTING I •SB 16 70 i 14 - 60 & 12 50 ID - 40 8 i 6 - 30 4 - 20 1 i 2 op < 0 0 1 2 3 4 Oviposition Scars
PLANTING 2 CUTTING 2 « 16 100 14 I 12 & & a 10 8 CU Î 6 - 40 I 4 - 20 I 2 I 0 0 1 2 3 4 Oviposition Scars
Figure 4.1 : Average firesh weight per plant and percentage plant mortality in parsley infested with different levels of carrot weevil oviposition scars, 1999
74 Polynomial Regression Analysis
Planting Cutting Polynomial d f F P Model
1 1 Quartic 4,142 0.987 0.417
2 1 Linear 1,140 42.735 <0.0001
2 2 Linear 1,144 16.545 <0.0001
2 2 Quartic 4,141 6.009 <0.0001
Orthogonal Contrasts
Planting Cutting Contrast df F P
2 1 0 and 1 Ovip. 1,140 5.449 0.021
2 2 0 and 1 Ovip. 1,141 6.487 0.012
2 1 3 and 4 Ovip. 1,140 0.018 0.893
2 2 3 and 4 Ovip. 1,141 38 0.538
Table 4.1 : Polynomial regression analysis to test the effect of five levels of oviposition scars on fiesh weight per plant of parsley and orthogonal contrasts to test for differences among the oviposition levels, at two different cuttings in 1999
75 and 1 oviposition scars, but not between 3 and 4 oviposition scars (Fig. 4.1, table 4.1).
Percentage of dead plants increased significantly as the number of oviposition scars
increased in all cuttings in both plantings (%^=13.7, df=4, P=0.008; x^=31.8, df=4,
P<0.0001; x^=25.9, df=4, P<0.0001, respectively) (Fig. 4.1).
In 2000, the interaction between planting dates and oviposition scars in their
effect on parsley yield was not significant for either cutting in either planting, nor was the
effect of planting date on parsley yield for both cuttings (Fig. 4.2, table 4.2). The number
of oviposition scars did not affect the yield significantly in either cutting for either
planting (Fig. 4.2, table 4.2). In the first cutting in both plantings; however the proportion
of dead plants was significantly correlated with number of oviposition scars (Fig. 4.2,
table 4.2). Planting date and the interaction between planting date and oviposition scars
had no significant effect on proportion of dead plants in the first cutting. The interaction
between planting date and oviposition scars did not have a significant efiect on percentage of dead plants in the second cutting of either planting (table 4.2). Planting date
and number of oviposition scars both had a significant effect, however, on the proportion of dead plants in the second cutting (Fig. 4.2, table 4.2). In cutting 2 of both plantings there were significant differences in plant mortality between 0 and 1 oviposition scars
(F=5.402; df=l,21; P=0.03), but not between 1 and 2, and 2 and 3 oviposition scars
(F ^ .l 11; df=l,21; P>0.743). About 70% (SE±0.052) of the plants infested with 1 oviposition scar had root feeding (Fig. 4.3).
76 PLANTING 1 CUTTING I
Avg. Weight per Plant (g) - Dead Plants (%)
§ S.
Oviposition Scars PLANTING 2 CUTTING I
1 2 Oviposition Scars PLANTING 1 CUTTING 2 so
5 #0 a. 4 40 3 2 1 0 0 1 2 3 Oviposition Scars PLANTING 2 CUTTING 2 a 7 J 6 a. 5 & 4 20 Z 3 2 Î 1 0 I 01 2 3 Oviposition Scars
Figure 4.2: Average &esh weight per plant and percentage plant mortality in parsley infested with different levels of carrot weevil oviposition scars, 2000
77 Analysis of Variance
Cutting 1
df FP F P
FreshWeight DeadPlants
Planting Date 1 0.497 0.487 0.094 0.761 Oviposition Scars 1 0.612 0.441 9.991 0.004 Rep 3 7.447 0.001 1.636 0.206 PD*Ovip 1 0.530 0.473 0.743 0.397 Error 25
Cutting 2
Planting Date 1 4.182 0.052 21.578 <0.0001 Oviposition Scars 1 0.498 0.487 7.804 0.010 Rep 3 2.768 0.063 0.662 0.583 PD*Ovip 1 0.216 0.646 1.603 0.217 Error 25
Table 4.2: Analysis of variance to test the effect of five levels of oviposition scars on firesh weight per plant (g) and plant mortality of parsley and the interaction between planting date and oviposition scars at two different cuttings in 2000
78 i I
§
1
Oviposition Scars per Plant
Figure 4.3. Proportion of plants infested with carrot weevil oviposition scars that had root feeding (dead and live plants)
79 Commercial Parsley Fields
The relationship between fresh weight/30 cm row and carrot weevil infestation was best described as a linear function of the proportion of plants with root feeding
(linear: F=6.91; df=l,146; P=0.009, quadratic: F=0.011; df=l,146; P=0.915) (Fig. 4.4).
Although the percentage of plants infested varied among commercial fields (F=4.688; dfr=2,143; P=0.011) no signifrcant differences among fields in the relationship between yield and percent plants infested with carrot weevil were observed (field by % infestation interaction: F=1.51; df=2,143; P=0.224). The slope coefficients estimated separately for the three fields had overlapping 95% confidence intervals; therefore the linear relationship is robust. The estimated function (and the standard errors for the regression coefticients for the pooled data) is.
Yield = 164.9 - 103 (INF%) (8.96) (38.8)
Based on this relationship, and taking into account the cost of control per hectare and market value of the produce provided by parsley growers (table 4.3), the Economic
Injury level (EIL) was found to be approximately 1% plants infested with ovipiosition scars (Fig. 4.5).
DISCUSSION
Parsley plants infested with oviposition scars were selected in two different planting dates to guarantee two different plant sizes. Plant size varied a lot among the selected samples even within planting dates in both 1999 and 2000. Nevertheless, I found
80 Regression Plot Yield = 164.584 - 103.510 Inf S = 56.9881 R-Sq = 4.6 % R-Sq(adj) = 4.0 %
300
i 200
& £ I 100 8
0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Infestation
Figure 4.4: Regression line for parsley yield per 30 cm-row section vs. the proportion of plants with larval feeding in the roots
81 Items Field 1 Field 2 Field 3
Cost of labor ($)/ha 52 52 123.5
Insecticide cost ($)/ha 247 247 123.5
Equipment ($)/ha 37 37 123.5
Insecticide sprays (#)/ha 3 3 3
Control cost ($)/ha 336 336 370.5
Price ($)/kg o f parsley 2.56 2.56 1.87
Table 4.3: Cost of control ($) per hectare of carrot weevil in parsley during the growing season and market value ($) o f parsley in Ohio
82 ECONOMC INJURY LEVEL
2
1.8
1.6
1.4
1.2
i ' C 0.8 ^0.6
0.4
0.2 25 0 0 1 2 3 S 6 7 8 8 10 15 20 304 40 SO Infestation (%) EIL
Figure 4.5: Economic injury level for carrot weevil oviposition scars in parsley interesting results regarding the effect of different numbers o f oviposition scars on plant yield and mortality at two different cutting times, as well as the effect of plant size as a function of planting date on these relationships.
In 1999, the effect of different numbers of oviposition scars on fresh weight per plant and on plant mortality was significant for either cutting in the second planting. The average fresh weight per plant significantly declined from uninfested plants to plants infested with 4 or more oviposition scars. In the same way, plant mortality steadily increased from uninfested plants to plants with 3 oviposition scars. An exception to this trend was the effect of oviposition scars on fresh weight per plant in the first cutting of the first planting. But, damage to the plants caused by carrot weevil larvae usually is perceived at the second cutting (chapter 3), as long as the first cutting is made before mid
July, and cutting in this experiment was on July 8*. The first cutting in the second planting was made on August 3^**, when most of the larval feeding activity has already been done.
I found significant differences in fresh weight per plant between non-infested plants and plants with 1 oviposition scar m either cutting of the second planting. One oviposition scar contains 3 eggs on average (chapter 2), which may result in 3 larvae feeding parsley roots, and based on my results, one oviposition scar per plant is enough to reduce parsley yield significantly either through plant mortality or reduction in plant size, or both. On the other hand I did not find any significant difference in fresh weight per plant between 3 and 4 oviposition scars per plant, which may indicate that any infestation above 3 oviposition scars per plant produce the same damage. Larval mortality may
84 increase in crowded conditions, and in many cases the host is killed as was seen in this
experiment. Plant mortality steadily increased from 0 to 3 oviposition scars, then leveled
off.
In those plants with 4 oviposition scars in the 1^ cutting of the 1^ planting, 60% of plants were dead but living plants had no significant reduction in fresh weight per plant, which may be explained by compensation. The surviving plants may grow larger in the absence of competition. In addition, plant size and stand was quite variable in these samples, so it is possible that some of the surviving plants were quite large.
In 2000,1 did not find any significant effect of oviposition scars on fresh weight per plant in either cutting for both plantings, rather I found some discrepancies. For example, plants with 3 or more oviposition scars, in the 2"** cutting of the 1^ planting had no significant reduction in yield but 60% mortality. Again, compensatory growth as well as large surviving plants may explain these results. The 1** cutting in the 1^ planting was made on July 11*, when larval feeding may be incipient, (chapter 3), so fresh weight per plant among the different levels of oviposition scars was the same. Plant stand and plant size were also quite variable in these samples. In addition, I observed about 50% mortality in plants without oviposition scars in the 2"** cutting of the 1^ planting. It is possible that larvae from neighboring infested plants moved to plants not initially infested with oviposition scars. Pepper (1942) reported that carrot weevil larva can migrate from root to root, thus being capable of destroying several plants. Furthermore, some plants may have been infested afrer marking and before the insecticide was sprayed. Plant
85 mortality in the 1®* planting was over 50% for all the oviposition scar categories, including non infested plants, while this percentage was less than 40% in the two oviposition scars category in the 2"** planting.
In conclusion, planting date did not have any significant influence on the relationship between parsley yield and oviposition scars, or on the relationship between plant mortality and oviposition scars, except for plant mortality in the second cutting. One oviposition scar per plant is sufGcient to cause significant reduction in firesh weight per plant. The weevil infestation occurred earlier in the first planting than in the second planting. Therefore, plant size in both plantings was about the same at the time weevil females laid eggs, or after plants reached the 4 true leaf stage (chapter 5). Plant damage in the field is explained by a combination of greater numbers of eggs and larvae in earlier plantings and a critical amount of root damage resulting in yield loss.
The EIL was estimated for each individual commercial field and firom the combined data of all the three commercial fields. The results were similar in all cases, between 0% and 2% plants with carrot weevil damage, depending on the cost of control per hectare and the linear relationship between yield and proportion of plants infested in each individual field as well as the combined fields. It is important to mention that even though the fields sampled were treated with insecticide, the results of this experiment are valid because the row sections sampled had a variable number of plants already having carrot weevil larvae established regardless of subsequent insecticide treatment. The EIL appropriately was estimated based on plants with root damage, because approximately
70% of plants infested with 1 oviposition scar had root feeding or died or both.
86 The EIL was defined by Stem et al. (1959) as the lowest pest population densi^
that will cause economic damage. So approximately 1% plants infested with carrot weevil
oviposition scars is the lowest population density that will cause economic damage to
parsley. This EIL appears to be low and perhaps conservative. However, because
sampling is to be based on oviposition scars, treating parsley fields at early infestations is
pmdent. Furthermore, I did not take into account the indirect yield losses manifested as
yellow leaves due to larvae feeding in the roots. Growers prefer to abandon long rows of
parsley in the field rather than sort green leaves firom yellow leaves at harvest. Because
parsley is a cash crop, or a crop of high value in net profits, a low threshold is affordable
as long as a good sampling scheme is carried out for timing insecticide treatments.
The economic threshold (ET) is the population density according to the definition
given by Stem et al. (1959), at which a control measure must be applied to prevent an
increasing pest population firom reaching the EIL. The ET is difficult to estimate (Pedigo
1986, 1996). Besides the EIL already estimated for carrot weevil on parsley, one needs to
know the phenology of this insect and the parsley host, the rate of population and injury
increase, and the control tactic to be applied. Because the EIL is so low and the ET cannot
accurately be calculated (Pedigo 1986, 1996), I recommend an action threshold for parsley fields of 1% plants infested with oviposition scars. This action threshold is the population density of infestation level proposed as a decision rule for managing carrot weevils. It will serve as the basis of a sampling protocol (chapter 5).
87 CHAPTERS
DEVELOPING A SAMPLING SCHEME FOR CARROT WEEVIL IN PARSLEY
INTRODUCTION
Carrot weevil activity in parsley can be detected by monitoring adults, oviposition scars in the petioles, or feeding scars in the roots. Sampling for adults requires shaking plants along randomly selected sections of row and inspecting for adults that stay motionless on the soil surface and under litter. Detecting adults is difGcult because their color can blend easily with the muck soil. Sampling for feeding scars in the roots is not convenient, because parsley growers need to treat before this damage occurs. However, detecting oviposition scars is much easier than finding adults. It takes less effort and time to randomly select plants and examine the petioles one by one looking for round holes covered with dark exudates, than to search the soil surface for adults.
Sampling protocols require models that describe sampling data, and probability density functions such as the negative binomial. Poisson, or Adès distribution, are useful models for these data (Binns and Nyrop 1992). The spatial pattern of insect counts, oviposition scar counts in this case, may also constitute a useful tool for devising sampling strategies (Binns and Nyrop 1992). Different sampling schemes or procedures
88 vary in cost and accuracy. One of these procedures is binomial or presence/absence sampling, which does not require that we further examine a plant when the first infestation sign, an oviposition scar in this case, is found.
The ultimate objective of this research is designing a sound sampling scheme that requires the least number of samples and the least walking to decide whether a parsley field needs treatment. In this chapter, I fit the parameters for probability distributions to counts of carrot weevil oviposition scars, describe the spatial pattern of carrot weevil infestation based on oviposition scars in commercial parsley fields, estimate the sample size required for treatment decisions, and determine the best transect for sampling carrot weevil oviposition scars in parsley fields.
MATERIALS AND METHODS
Commercial Parsley Fields
Four commercial parsley fields were sampled intensively at least four times on a
10x10 grid pattern during May and June 2000 to determine the spatial distribution of carrot weevil oviposition scars. Fields were divided in 10 rows of equal size and 10 columns of equal size using flags along the edges as markers. The intersection of rows and columns gave each of the 100 cells used for sampling. Within each cell, 10 plants were sampled at random for a total of 1000 plants per field per week. All fields sampled were planted by March 8*, which is relatively early and, therefore, most likely to be at risk of carrot weevil damage (chapter 3). Two of the plantings were in Celeryville, Huron
County, Ohio, and the other two were in Hartville, Stark Coimty, Ohio.
89 Field 1
Parsley was seeded on March 8*^, 2000 at the muck soil vegetable production area of Hartville. The soil was plowed in the fall and fit with disk and roller before planting.
Lime was applied at 225 kg/ha, and 336 kg/ha o f0-0-60,225 kg/ha of 46-0-0, and 225 kg/ha of 18-46-0 fertilizer were broadcasted before planting. Weeds were controlled with
Lorox® at 0.6 kg/ha, 4 weeks after planting and weeded with a cultivator when necessary. Parsley was planted using a Planet Junior planter on fiat beds, each containing
5 rows, 27.9 cm apart. The distance from bed center to bed center was 1.83 m. Field dimensions were 130 m x 27 m. Guthion® was sprayed on May 26*. Parsley had not been planted in this field since 1995.
Field 2
Curly parsley, cv Forest Green was planted on March 7*, 2000 at the muck vegetable production area of Celeryville. The soil was plowed and disked, and 560 kg/ha of 0-0-60 was broadcasted before plowing and 840 kg/ha o f 17-17-17, after plowing.
Weeds were controlled with Lorox® when plants had 2 true leaves. Weeding by hand or cultivator was done when necessary. Parsley was planted in raised beds, 20 cm high, each containing two rows, 25.4 cm apart. The distance from bed center to bed center was 91.4 cm. The crop was seeded using a Stanhay planter. Guthion® was sprayed on June 3^**. The sampling area was 1.08 ha on a 295 m x 36.6 m rectangle. This field was close to parsley fields planted the previous year.
90 Field 3
This field was planted on March 8***, 2000 at the muck vegetable production area o f Celeryville. The soil was plowed and fit with a disk and roller, and 560 kg/ha o f 17-17-
17 fertilizer was broadcasted and plowed down before fitting the soil, while 560 kg/ha of the same fertilizer was spread on top of the beds. Weeds were controlled with Lorox® when they were about 2.5 cm in height. Weeding also was done by hand or cultivator when necessary. Parsley beds were 12.7 — 15.2 cm high and consisted of two rows 28 cm apart. The distance firom bed center to bed center was 91.4 cm. The sampling area was
1.08 ha on a 113 m X 96 m rectangle. This field was located about 1 km firom the nearest parsley planted in 1999.
Field 4
This field was seeded with cv Forest Green at 11.2 kg/ha on February 28*, 2000 at the muck soil vegetable production area of Hartville. The field was covered with spun polyester firom the date of planting until early May. The soil was plowed and fit with disk and roller, and 739 kg/ha of 16-8-38 was broadcasted before fitting the soil. Weeds were controlled with Lorox® at planting and by hand or cultivator when necessary. Parsley was planted on fiat beds, each containing 5 rows, 33 cm apart. The distance (width) fi"om bed center to bed center was 1.83 m. Guthion® at 1.1 kg/ha was sprayed two times, on
May 13*, and June 10*. The sampling area was 1.62 ha on a 209 m x 77.5 m rectangle.
This field was located about 0.8 km away firom previous parsley fields.
91 Oviposition Scar Counts
Ten randomly selected plants were pulled in each grid cell, petioles were removed one by one, inspected carefully on both the concave and convex side, and the number of oviposition scars present was recorded for each plant. Each oviposition scar was dissected in the field to count the number of eggs and to determine whether the eggs were still viable or the larvae had already hatched. Oviposition scars were recognized as small round holes, ca. 1 mm in diameter, covered with dark exudates. Each field was sampled four times weekly starting in mid May in the earliest field and throughout the period of maximum oviposition activity (chapter 3). Plots were generated using Systat (Wilkinson
1988).
Population Distribution
Carrot weevil oviposition scar counts firom commercial fields of parsley were used to determine a probability density function that best describes the observed population distribution in the field. Parameters of the negative binomial and poisson distributions were fit to the field data and then the expected and observed distributions of counts were compared using the Chi Square (%^) test of goodness of fit. A condition for this test establishes that the expected value for any fi'equency class must be equal or higher than 5, otherwise those classes below 5 must be pooled with an adjacent class above 5 if necessary. The number of degrees of freedom for Poisson is equal to the number of
92 classes (a) minus 1 (a-1), and for the NBD is equal to the number of classes minus 3 (a-3)
(Anscombe 1949, Bliss and Fisher 1953, Southwood 1966, Horn 1988, Binns and Nyrop
1992, Madden and Hughes 1995, Sokal and Rohlf 1995).
If the negative binomial distribution is used to describe counts of oviposition scars
then the following expression gives the probability of any given number of oviposition
scars per plant:
p{n oviposition scars}= n!(k-l)!
where R=p/q, and p=q-l, p and k are the parameters of the negative binomial
distribution. For the poisson distribution, the expression is,
p{n oviposition scars}= ^ n! where x is the mean number of oviposition scars per plant. By substituting zero for n, either expression gives the expected proportion of plants that are free of oviposition scars.
Binomial sampling or presence/absence is a convenient method for sampling many
insects, because is easier, faster, and less expensive than counting all the individuals present on each sample unit. The proportion of plants containing no oviposition scars may, therefore, be used to estimate the extent of infestation (Anscombe 1948, Pielou
1960, Binns and Nyrop 1992).
93 Sample Size
Based on the binomial distribution and taking into account the economic threshold for carrot weevil oviposition scars in parsley (chapter 4), the number of samples with no infestation needed for a decision not to treat was estimated. An equation used by Couey and Chew (1986) is the primary formula to estimate the sample size when the number of infested plants with oviposition scars is zero. The equation:
x=i ni S ------PuTl-pJ- =,_c x !(n -x )! where x = 0,1,2, 3, i, simplifies when the infestation (i) is zero (i=0) to:
n=[log( 1 -C]/log( 1 -pu), where n = the sample size, C = the confidence level, and p„ = the upper bound. This function can be used to calculate the number of parsley plants with no infestation that must be sampled to achieve a given confidence level that the infestation is less than a given upper bound. If the upper bound is set at the economic threshold (chapter 4), then the function provides a useful decision rule for carrot weevil in parsley.
Simulation Studies
A program written in Basic (PowerBASIC, 1996) was used to simulate sampling along 6 different transects in 4 directions (table 5.1). The simulation was performed for each of the 17 data sets describing fields with different spatial arrangements of carrot weevil infestation. The program was run 10 times for each combination of transect.
94 TRANSECT TRANSECT STARTING DIRECTION SHAPE NAME POINT n Perimeter NW Clockwise n Perimeter NW Counterclockwise □ Perimeter SE Clockwise Perimeter SE Counterclockwise 1g X-Shape NW/SW Eastward K X-Shape NE/SE Westward X X-Shape SW/SE Northward X X-Shape NW/NE Southward Z-Shape NW Southward
Z-Shape SE Northward
Z-Shape SW Northward
Z-Shape NE Southward El V-Shape SW SW-N-SE 0 V-Shape SE SE-N-SW E V-Shape NW NW-S-NE V-Shape NE NE-S-NW S Cross 5N/5E 5NS-5EW Cross 6N/6E 6NS-6EW 0 Diagonal NE NE-SW
Table 5.1: Transect patterns compared by simulation to select the best for sampling parsley fields infested with carrot weevil oviposition scars
95 direction, field, and date, and the combined results were used to select the best transect
Each data set was structured in a long column of 1000 samples arranged by row and column, so the simulated sample consisted of a sequence of steps that randomly select a given number of samples firom a series of row-column combinations that represent a given path. Examples of paths simulated include Z-shape, V-shape, X-shape, perimeter, cross, and diagonal (table 5.1). The program stopped when one infested plant was found and printed the number of simulated samples up to that point; otherwise it stopped at the simulated end of the transect and reported the maximum number of samples. The simulated sample results for each field data set were compared among the 6 transects and between the simulated transects and the full 1000-plant sample in each field. The transect that most often found infested plants when the infestation was above the ET, taking the least number of samples to stop, and most often completed the transect when the infestation was below the ET, was considered the best. Each simulated sample fell into one o f 4 categories: 1-the field required insecticide spray (infestation above the EEL, based on 1000-plant sample) and the simulated sample indicated that insecticide spray was needed (SS), 2-the field required spray but the simulated sample indicated no spray was needed (SNS), 3-the field did not require spray and the simulated sample indicated spray (NSS), and 4-the field did not require spray and the simulated sample indicated no spray (NSNS). The transect leading to the most correct decisions (SS and NSNS) and the least incorrect decisions (SNS and NSS) was selected as the best for sampling parsley fields infested with carrot weevil oviposition scars.
96 The average sample number (ASN) (Wald 1947, Binns and Nyrop 1992, Legg and
Chen 2000) is another criteria for comparing the simulated transects. The ASN is in
essence the average number of samples required to categorize an infestation above or below a given threshold, and the average sample number curve that is used for
comparison consists of a plot of average sample number as a function of the true
infestation. In this study, the ideal ASN curve for a transect would be the maximum sample size when the true infestation is below the ET, and minimum number of samples
(close to 1), when the true infestation is above the ET.
RESULTS AND DISCUSSION
Spatial Distribution of Oviposition Scars
The sample distribution of oviposition scar counts was best fit by the Poisson distribution when the mean number of carrot weevil oviposition scars per plant was less than or equal to 0.047. When the mean was greater than or equal to 0.06 oviposition scars per plant, the distribution of oviposition scars counts was best fit by the negative binomial distribution (NBD) (table 5.2). The NBD is an extension of the Poisson distribution (Bliss and Fisher 1953). As the variance approaches the mean, the aggregation parameter of the
NBD, k, tends toward infinity. As the variance increases beyond the mean, k tends toward zero (Fisher et al. 1943). In field 4-week 3 (F4-W3), the mean and variance are very close, and k tends to infinite (table 5.2). In some of the samples I could not perform the
Chi Square (%^) test of goodness of fit for the NBD, because the number of degrees of fireedom was equal to zero (df = the number of expected firequency classes minus 3), or
97 FIELDMEAN VARPOISSONNBD
X^=2.5E-07 Fl-Wl 0.001 0.001 NA P>0.995 xW .224 F1-W3 0.016 0.020 dfi=0 P>0.5 X^=0.0003 F1-W4 0.011 0.011 NA P>0.975 x W .o o l F2-W1 0.018 0.018 NA P>0.900 X^=l 0.020 X^=0.141 F2-W2 0.064 0.082 P<0.005 P>0.75, k=0.28 X^=0.084 F2-W3 0.047 0.051 dfi=0 P>0.75 x"=8.246 X^=0.076 F2-W4 0.152 0.183 P<0.025 P>0.75, k=0.81 X^=6.78E-06 F3-W1 0.003 0.003 NA P>0.995 X^=27.078 X^=0.400 F3-W2 0.060 0.090 P<0.005 P>0.50, k=0.15 X^=37.280 X^=3.394 F3-W3 0.281 0.400 P<0.005 P>0.10, k=0.64 xW O.262 X ^ .5 3 8 F3-W4 0.189 0.264 P<0.005 P>0.70, k=0.52 X^=0.315 F4-W1 0.012 0.016 df=0 P>0.500 X^=0.025 F4-W2 0.007 0.007 NA P>0.975 X^=0.082 NA F4-W3 0.011 0.013 P>0.750 k—^ oo X^=l 9.069 X^=0.369 F4-W4 0.068 0.099 P<0.005 P>0.50, k=0.18 x"=23.147 x W .5 6 4 F4-W5 0.077 0.109 P<0.005 P>0.25, k=0.20
Table 5.2: Population distribution fitted to counts of carrot weevil oviposition scars in 16 different population densities. NA - not applicable
98 because the mean and variance had the same value (NBD was not applicable), nevertheless the data fit the Poisson distribution in those samples (table 5.2).
Spatial Pattern of Oviposition Scars Counts in Parsley Fields
Seventeen different spatial patterns for parsley fields infested with carrot weevil oviposition scars were obtained firom sampling 3 commercial fields every week during 4 weeks and 1 field during 5 weeks on a 10x10 grid pattern (Fig. 5.1). These plots are very useful for detectrug possible infestation patterns, like edge effects and the infestation
progression during the oviposition period. In these plots, carrot weevil infestation measured by the number of oviposition scars per grid cell, steadily increased in most commercial fields over the 4-5 week period. Field 1 had a relatively low infestation because it was planted in a relatively isolated area that had been fi-ee of parsley for 5 years. Field 2 was planted very close to the previous year’s parsley fields, which may explain the high population density almost Grom the first week on. Field 3 also was planted near a previous year’s parsley field, and a high population density of carrot weevil was observed. Field 4 was sprayed about 3 days before the first sampling, which may explain the relatively light infestation during the first three weeks. By the 4* and 5^ week
(the 5* week is not shown in figure 5.1), the weevil infestation increased abruptly in field
4 (table 5.2, Fig. 5.1). Adult weevils can continue entering parsley fields for several weeks (chapter 3). Based on the spatial pattern of oviposition scar counts in the four parsley fields sampled, there is no apparent concentration of activity of the adult weevils in any particular part of the field.
99 Field 1 Field 2
Week 1 0.0 I I ■ ■ T ■" 1 ---- 1
2.4
4.8 -- OMP 7.2 __ CMP
9.6
BIII 0 .5 □ 0.0 12.0 ____ 1_____ 1_____ 1_____ 1_____ L_ a IQ 12 0 2 4 6 8 10 12 II 0 Week 2
3
I 6
9
12 12 0 2 4 6 8 10 12 - |------1------1 I ' ~r~ Weeks 0
3
6
8
12 12, 2 4 S 8 10 12 0 2 4 6 8 10 12
I I ----- I I Week 4
_l I L. 12 4 6 8 10 12 2 4 6 8 10 12 COLUMN COLUMN
Figure 5.1 : The progression of carrot weevil infestation over four weeks in four commercial parsley fields
100 Field 3 Field 4 0 | ------1------1 I I ~ r Week 1 A m i OVIP CMP
BB 12 10 12 0 2 4 6 8 10 12 ii 0 0 Week 2
3 3 I 6 6
9 9
12 12 0 2 4 6 8 10 12 0 2 4 6 8 10 12
Weeks I I I I — r i 1 Ü •ÎF
12 b 2 4 6 8 10 12 Week 4
EH
2 4 6 8 C0LIM4 COLUMN
Figure 5.1: Continued
101 Sample Size
Using a 95% confidence level (CL), 148 parsley plants having no oviposition
scars would place the upper bound of the proportion infested at 0.02 (2% plants infested).
If the CL is 99% and the same upper bound is used, the sample size increase to 227 plants
with no oviposition scars. If the upper bound, is set to 0.01, the sample sizes become 298
and 458 plants for 95% and 99% CL, respectively. For decision-making purposes I use a
threshold of 1% plants infested (chapter 4) meaning that we want to be sure that the
percentage of infested plants is less than 2%. Therefore a decision rule is; if 148 plants
can be inspected without finding any oviposition scars, there is a 95% chance that the
percentage of infested plants is less than 2% and no treatment is needed. On the contrary,
if one infested plant is found before 148 plants have been examined, then the scout can
stop and decide to treat. I estimate that it takes an average of 1 hour to inspect 148 plants.
Best Transect for Sampling Parsley Fields
Sampling a commercial parsley field in an X-shaped pattern in any direction was
found to be the best transect for sampling carrot weevil oviposition scars in parsley fields
(Fig. 5.2). This transect results in the most correct decisions and the fewest incorrect
decisions overall (Fig. 5.3), and has one of the highest ASN when the infestation is below the ET, and one of the smallest ASN when the infestation is above the ET (ET=0.001 proportion of plants infested with carrot weevil oviposition scars) (Fig. 5.4).
We can incur two kinds of errors or incorrect decisions when sampling; spray when there is no need and not spraying when there is a need. The worst error is the latter
102 0
2
4
I 6
8
10
12 0 2 4 6 8 10 12 COLUMN
Figure 5.2: An X-shape is the best transect to sample parsley fields infested with carrot weevil oviposition scars. The starting point can be any comer, and 7 randomly selected plants per cell must be sampled along one diagonal axis, and 8 plants, along the other diagonal axis
103 a SS a SNS □ NSS ■nsns 1 I
Q Q n B a a B B a
Perimeter X-Shape Z-Shape V-Shape Cross Diagonal Transects
Figure 5.3. Transect patterns showing the mistake gap and the good attributes that help to deciding the best one for sampling parsley fields infested with carrot weevil oviposition scars 2 0 0
Figure 5.4: Average sample number curve for 6 different transect patterns. The economic threshold is approximately equal to 0.01 105 because an infestation above the ET that is not detected by a sampling scheme and is not treated may increase well above the EIL and cause serious economic losses. The X- shaped transect incurred this mistake in only 2% of the simulated samples (Fig. 5.3). The other mistake, treating when there is no need, wastes the cost of spray, which tends to be small compared with the value of the crop. The simulated samples using the X-shaped transect incurred this error in 16% of the cases approximately (Fig. 5.3). CONCLUSIONS Because the economic injury level is so low, the decision rule was based on the binomial distribution, absence or presence of oviposition scars in parsley plants. In addition, most infested plants have only 1 oviposition scar. Parsley fields should be sampled every week once the plants reach the 4^ true leaf stage on average (chapter 2), Adult weevils tend to arrive in fields during mid May and early June (chapter 3), so this is the critical period for sampling early-planted parsley fields. One hour is the estimated time for one person to sample 148 parsley plants at random in a commercial field following an X-shaped transect. At infestations below 1% plants with oviposition scars, the X-shape transect will result in a decision that the field does not need to be sprayed more reliably than the other transect patterns simulated in this study. On the other hand, when the infestation is above 1%, the X-shape transect will result in fewer samples and more reliably lead to a decision to treat than the other transects. I observed that some of 106 the parsley growers sprayed when there was no infestation, while others sprayed well above the EIL. The decision rules developed in this chapter can assist parsley growers with better timing for pesticide applications or biological controls. 107 CHAPTER 6 AN INTEGRATED PEST MANAGEMENT PROGRAM FOR CARROT WEEVIL DST PARSLEY INTRODUCTION Parsley growers in the Great Lakes Region of the United States and Canada face yield reduction in parsley from carrot weevil infestations. Carrot weevil larvae feed in the roots, killing plants and lowering the yield and market quality of the produce. According to growers in Ohio, most of the damage is seen at the second parsley cutting but the only way to control weevils is the use of insecticide against adults many weeks before damage is evident. Questions growers must consider include when to spray, at what action thresholds, how to monitor this pest population with accuracy, and what transect or path through the field is the best for sampling. The answer to these questions will be addressed in this chapter along with useful information on carrot weevil biology and seasonal activity. The control decision rules or protocols presented here constitute a valuable tool for parsley growers to keep the carrot weevil population low by timing pesticide applications, thereby reducing economic losses. This chapter will form the basis of an extension publication that integrates my research into a control strategy. 108 Life History Adult Stage Carrot weevil adults have functional wings but rarely fly (Fig. 6.1). They walk from their overwintering sites to new parsley fields, or any other host like carrot and celery. The adult weevils overwinter in or near parsley fields, particularly those with parsley left standing during the winter season (Fig. 6.2). They spend the cold weather under plant residues or about 2.5 cm below the soil surface around parsley plants. On the other hand, if the parsley field is plowed at the end of the crop season, and before the weevils have entered the resting or diapause stage, then the adults may be forced to walk away to overwinter in other sites nearby, under plant debris, sod, wood pieces, hedgerows, etc. In early spring, depending on the weather, the surviving overwintering parsley plants resume vegetative growth at the same time that the overwintering adults become active, seeking host plants to feed and reproduce. If the overwintering parsley field can provide food and places to lay eggs in early spring, then the adult weevils do not need to walk away; if the overwintering parsley does not provide food, or if it is plowed in spring, then the weevils have to move away and any parsley field planted nearby could become infested. Adult weevils tend to be very active in new parsley fields from mid May until the end of June approximately, but the most activity occurs from late May until mid June. At the end of August, a new population of adult weevils is present in the field, but the majority of this weevil population will overwinter and reproduce in the following crop season (Fig. 6.3). 109 Figure 6.1 : Carrot weevil adult 110 Figure 6.2: Overwintering parsley field with few surviving plants as it looks by early April 111 Carrot Weevil Adult, Oviposition, and Larva Activity Periods # ^ Date Figure 6.3: Seasonal activity for three different carrot weevil stages; avg. number of adults per 3 m of row, avg. number of eggs per plant, and avg. number of larvae per root 112 Egg Stage When adult weevils infest a parsley field in which plants have four true leaves on average (Fig. 6.4), egg laying starts. Weevil females make punctures in the plant petioles and deposit 3 eggs on average inside. They then cover the oviposition scars with a dark tar-like substance. The oviposition scar looks like a round dark spot of approximately 1 mm in diameter (Fig. 6.5), and may be present anywhere in the petiole, but is most often in the concave part To check a suspected oviposition scar, growers should rub off the dark spot and check for a puncture beneath. If a puncture is found, a weevil has deposited eggs in it. The eggs are white or yellow during the first day, then gradually darken to black when the larvae are ready to hatch at about 5 days after the eggs are laid (Fig. 6,6). The oviposition period occurs from mid May until the end of June, with the highest numbers of eggs per plant occurring in June (Fig. 6.3). Larval Stage The newly hatched larvae move down to the crown and root to feed by walking down the petiole or dropping to the ground. Sometimes the larvae tunnel inside the petiole when the oviposition scar is close to the base of the petioles. Larval feeding may be internal, which is usually severe (Fig. 6.7), or close to the outer layer of the root. Infested young plants generally die (Fig. 6.8), but large plants may withstand even severe internal damage. Carrot weevil larvae (Fig. 6.9) feed in the roots for approximately 20-30 days (Fig. 6.3), then leave the root and pupate in the surroimding soil. 113 Figure 6.4: Parsley plant at the four-true-leaf stage showing the 4* leaf fully expanded. Carrot weevils seldom lay eggs in smaller plants 114 Figure 6.5: Carrot weevil ovipositioa scar in a parsley petiole 115 Figure 6.6. Garrot weevil eggs 1 1 6 Figure 6.7. Parsley root showing severe internal damage caused by the carrot weevil larva 117 Figure 6.8: Dead and damaged parsley plants due to carrot weevil larva 118 Figure 6.9: Canot weevil larva 119 Action Thresholds Loss firom carrot weevil infestations can be predicted firom the relationship between percentage of parsley plants infested with carrot weevil and parsley yield. Taking into account predicted loss, the average cost of control, and average market value of parsley provided by parsley growers, more than 1% plants infested with carrot weevil oviposition scars would produce enough damage to justify the cost of control. Sampling Scheme Carrot weevil infestations can be detected by examining plants for oviposition scars, which is easier than sampling for adults. Sampling should start when parsley plants have 4 true leaves (Fig. 6.4), or about mid May for the earliest plantings. The action threshold is one plant infested out of 100 plants. Because the action threshold is so low, a grower should be able to inspect many plants without finding oviposition scars if treatment is not necessary. Based on statistical formulae, if 148 plants can be inspected without finding any oviposition scars, there is a 95% chance that the percentage of infested plants is less than 2% and no treatment is needed. If the decision is not to treat, the field should be sampled again the following week. If an infested plant is found before all 148 plants are examined; however, then the chance that the field is above the action threshold is high enough that the grower can stop sampling and decide to treat. 120 How to Walk the Field I compared some of the paths a grower could take through parsley fields to determine which would give the most accurate sample with the least walking. An X- shaped path that goes fi*om comer to comer, in any direction was found to be the best (Fig. 6.10). This path led to the most correct decisions and the fewest incorrect decisions regarding the need for treatment of any of the patterns tested. Control Strategies The first generation and a partial second generation of adult weevils spend the winter weather in or near the parsley fields where they developed as larvae. These adults could be sprayed at the end of the crop season the previous fall, but if some of the weevils had already taken refuge for the winter under plant residues or below the soil surface, they might escape the control treatment. Growers can more reliably spray and kill these weevils in May, when all the adults have become active. Leaving the overwintering parsley fields as a trap crop without spraying could be a good cultural practice to keep adult weevils away firom new parsley fields, but more research must be conducted on overwintering and first generation adult activity after the overwintering parsley stops the vegetative growth to flower and reproduce (seeds). The early generation of adult weevils may walk away seeking new parsley hosts. The use of entomopathogenic nematodes might prove to be a good biological control practice in overwintering fields if nematodes are sprayed before the adult weevils become active. 121 Figure 6.10: An X-shaped transect is the best pathway to sample parsley fields infested with carrot weevil oviposition scars. The starting point can be any comer, and each diagonal must be divided in 10 sections approximately, so that 7 plants per section must be sampled along one axis and 8 plants along the other axis 1 2 2 Sampling in new parsley plantings should begin as soon as plants reach the 4-true-leaf stage. One person can sample 148 parsley plants in a commercial field in 1 hour; each plant takes very little time. If parsley growers sample their fields following the rules provided here, then they would be able to time their sprays carefully to avoid economic loss, either fi"om spraying too soon or too late, too much or not enough. My results provide an easy-to-use decision rule to assist parsley growers with timing their pesticide applications. 123 BIBLIOGRAPHY Anscombe, F. J. 1948. On estimating the population of aphids in a potato field. Ann. Appl. Biol. 35:567-71. Anscombe, F. J. 1949. The statistical analysis of insect counts based on the negative binomial distribution. Biometrics 5, 165-73. Baines, D., R. Stewart, and G. Boivin. 1990. Consumption of carrot weevil (Coleoptera: Curculionidae) by five species of carabids (Coleoptera: Carabidae) abundant in carrot fields in Southwestern Quebec. Environ. Entomol. 19(4): 1146-1149. Bélair, G., and G. Boivin. 1985. Susceptibility of the carrot weevil (Coleoptera: Curculionidae) to Steinememafeltiae, S. bibionis, and Heterorhabditis heliothidis. Journal ofNematology 17(3):363-366. Bélair, G., and G. Boivin. 1995. Evaluation o f Steinemema carpocapsae Weiser control of carrot weevil adults, Listronotus oregonensis (LeConte) (Coleoptera: Curculionidae), in organically grown carrots. Biocontrol Science and Technology 5:225-231. Berenbaum, M. R. 1990. Evolution of specialization in insect-umbellifer associations. Annu. Rev. Entomol. 35:319—343. Binns, M. R., and J. P. Nyrop. 1992. Sampling insect populations for the purpose of IPM decision making. Annu. Rev. Entomol. 37:427-453. Blatchley, W. S., and C. W. Leng. 1916. Rhyncophora or weevils o f North Eastern America. Indianapolis: The Nature Publishing Company. Bliss, C. I., and R. A. Fisher. 1953. Fitting the negative binomial distribution to biological data. Biometrics 9:176-200. Boiteau, G., Y. Pelletier, G. C. Misener, and G. Bernard. 1994. Development and evaluation of a plastic trench barrier for protection of potato firom walking adult Colorado potato beetles (Coleoptera: Chrysomelidae). J. Econ. Entomol. 87(5):1325-1331. 124 Boivin, G. 1985. Evaluation of monitoring techniques for the carrot weevil, Listronotus oregonensis (Coleoptera: Curculionidae). Can. Ent. 1117:927—933. Boivin, G. 1988. Effects of carrot developmental stages on feeding of carrot weevil, Listronotus oregonensis (Le Conte) (Coleoptera: Curculionidae). Environ. Entomol. 17(2):330-336. Collins, R. D., and Grafius, E. 1986. Biology and life cycle a£Anaphes sordidatus (Hymenoptera: Mymaridae), an egg parasitoid of the carrot weevil (Coleoptera: Curculionidae). Environ. Entomol. 15:100-105. Cormier, D., A. B. Stevenson, and G. Boivin. 1996. Seasonal ecology and geographical distribution o f Anaphes listronoti and A victus (Hymesnoptera: Mymaridae), eggs parasitoids of the carrot weevil (Coleoptera: Curculiortidae) in Central Ontario. Environ. Entomol. 25(6): 1376-1382. Couey, H. M., and V. Chew. 1986. Confidence limits and sanaple size in quarantine research. J. Econ. Entomol. 79:887-890. Cress, D., and A. Wells. 1977. Celery and carrot insect pests. Mich. State Univ. ExL Bull. E-970. 2 p. Dutky, S. R-, J. V. Thompson, and G. E. Cartwell. 1964. A technique for the mass propagation of the DD-136 nematode. J. Insect Pathol- 6:417-422. Edelson, J. V. 1985. Biology of a carrot weevil, Listronotus texanus (Coleoptera: Curculioniodae): Range and seasonality of infestations. J. Econ. Entomol. 78:895-897. Edelson, J. V. 1986. Biology and control of a carrot weevil in the Lower Rio Grande Valley, Texas. J. Rio Grande Valley Horticultural Society 39:79-82. [EIS] Entomological Information Services. 1996. Nomina inspecta neartica, a check list o f the insects of North America. Vol. 1 : Coleoptera, Strepsiptera. Rockville, MD: Robert W. Poole and Patricia Gentili, editors. Fry, J. C. 1996. One-way analysis of variance. In Fry, J. C. 1996. Biological data analysis: A practical approach. Oxford; IRL Oxford University Press, p 1-39. Gebhardt, S. E., R Cutrufelli, and R.H. Mattews. 1982. Composition o f foods, firuits, and finit juices-raw, processed, prepared, USD A Agricultural handbook 8-9, 1982). Cited in Maynard and Hochmuth 1997. 125 Ghîdiu, G. M. 1987. Carrot weevil. Rutgers Coop. Ext. New Jersey Agr. Exp. Sta. FS250. Ghidiu, G. M., and R. W. VanVranken. 1995. A modified carrot weevil (Coleoptera: Curculionidae) monitoring trap. Florida Entomologist 78(4):627-630. Gleason, H. 1952. The new Britton and Brown illustrated flora of the Northeastern United States and adjacent Canada. Vol. 2. Lancaster, PA: Lancaster Press, Inc. 655 p. Grafius, E., M. Otto, and B Collins. 1983. Detection and control of carrot weevil. Mich. State Univ. Ext. Bull. E-890. Grafius, E., and R. D. Collins. 1986. Overwintering sites and survival of the carrot weevil, Listronotus oregonensis (Coleoptera: Curculionidae). Environ. Entomol. 15:113-117. Hance, T., and G. Boivin. 1993. Effect of parasitism by Anaphes sp. (Hymenoptera: Mymaridae) on the cold hardiness of Listronotus oregonensis (Coleoptera: Curculionidae) eggs. Can. J. Zool. 71:759-764. Hassel, R. L. 1993. Studies of the inhibition complex in seed of Petroselinum crispum L., and the development of methods to improve field stand establishment (parsley). [Dissertation]. Columbus, OH: The Ohio State University. 171 p. Haytowitz, D.B., and R.H. Mattews. 1984. Composition of foods, vegetables and vegetable products-raw, processed, prepared, USD A Agricultural handbook 8-11. Cited in Maynard and Hochmuth 1997. Henderson, L. S. 1939. Revision o f genus Listronotus (Curculionidae-Coleoptera) Kan. Univ. Sci. Bull. 26:215-337 (Cited in Whitcomb 1965). Higley, L. G., and W. Wintersteen. 1987. Using degree days in an integrated pest management program. Iowa State University Extension [electronic version, 1997], URL: http://www.extension.iastate.edu/publications/PM1296. Iowa Sate University, Ames, lA. Hooper, L. R. E., J-P. R. Le Blanc, and G. Boivin. 1996. The detection o f Anaphes sp. Nov. (Hymenoptera: Mymaridae), an egg parasitoid of the carrot weevil in Nova Scotia. Phytoprotection 77:79-82. Horn, D. 1988. Ecological approach to pest management. New York: The Guilford Press. 285 p. 126 Hoy, C. W., C. Jennison, A. M. Shelton, and J. T. Andaloro. 1983. Variable intensity Sampling: A new technique for decision making in cabbage pest management. J. Econ. Entomol. 76:139-143. Hoy, C. W., C. Welty, M. J. Dunlap, and W. B. Evans. 1997. Insect control evaluation summary for Ohio vegetable crops. The Ohio State University, OARDC. Huber, J. T., S. Coté, and G. Boivin. 1997. Description of three new species species (Hymenoptera: Mymaridae), egg parasitoids of the carrot weevil, Listronotus oregonensis (LeConte) (Coleoptera: Curculionidae), and redescription of Anaphes sordidatus Girault. The Canadian Entomologist 129:959-977. Le Blanc, J-P R., and G. Boivin. 1993. A note on the detection of the carrot weevil in Nova Scotia. Phytoprotection 74:113-115. LeConte, J. L. 1876. Rhyncophora of America north of Mexico. Proc. American Philosophical Soc. 15(96) 455 p. Legg, D. E., and B. M. Chen. 2000. Toward fair comparisons of binomial sequential sampling plans. American Entomologist 46(4):250-260. Madden, L. V., and G. Hughes. 1995. Plant disease incidence: Distributions heterogeneity, and temporal analysis. Annu. Rev. Phytopathol. 33:529-564. Martel, P., H. J. Svec, and C. R. Harris. 1975. Mass rearing of the carrot weevil, Listronotus oregonensis (Coleoptera: Curculionidae), under controlled environmental conditions. Can. Ent. 107:95-98. Martel, P., H. J. Svec, and C. R. Harris. 1976. The life history of the carrot weevil, Listronotus oregonensis (Coleptera: Curculionidae) under controlled conditions. Can. Ent. 108:931-934. Maynard, D. N, and G. J. Hochmuth, X. X. 1997. Knott’s Handbook for vegetable growers. Fourth Edition. New York: J Wiley & Sons. Minitab. 2000. MINITAB, Release 13 for Windows [Statistical software]. State College, PA: Minitab, Inc. Ohio Vegetable Production Guide. 2000. The Ohio State University Extension. 256 p. Pepper, B. B. 1942. The carrot weevil, Listronotus latiusculus (Bohe), in New Jersey and its control. N. J. Agricultural Experiment Station Bulletin 693:1-20. 127 Peirce, L. C. 1987. Vegetables: characteristics, production, and marketing. New York: J Wiley & Sons. Perron, J. P. 1971. Insect pests of carrots in organic soils of Soutwestem Quebec with special reference to the carrot weevil, Listronotus oregonensis (Coleoptera: Curculionidae). Can Ent. 103:1441-1448. Pedigo, L. P., S. H. Hutchins, and L. G. Higley. 1986. Economic injury levels in theory and practice. Annu. Rev. Entomol. 31:341-368. Pedigo, L. P. 1996. Economic thresholds and economic injury levels. In E. B. Radcliffe and W. D. Hutchison [eds.], Radcliffe's IPM World Textbook, URL: http://ipmworld.umn.edu. University of Minnesota, St. Paul, MN. Pielou, D. P. 1960. Contagious distribution in the European red mite, Panonychus ulmi (Koch), and method of grading population densities from a count of mite-free leaves. Can. J. Zool. 38:645-653. Pope, R. 1998. Degree days and crop management. Integrated Crop Management [electronic version]. URL: http://www.ipm.iastate.edu/ipm/icm/1998/4-6- 1998/dd.html. Department of Entomology, Iowa State University, Ames, LA. PowerBASIC. 1996. FirstBasic© compiler for DOS [computer program]. Version 1. Carmel, CA: Power Basic, Inc. Pree, D. J., A. B. Stevenson, and E. S. Barszcz. 1996. Toxicity of pyrethroid insecticides to carrot weevils: Enhancement by synergists and oils. J. Econ. Entomol. 89(5):1254-1261. Roberts, M-D,, and A. B. Stevenson. 1974. Carrot weevil: a laboratory-rearing method. J. Econ. Entomol. 67(1): 140 Ryser, B. W. 1975. Investigations regarding the biology and control o f the carrot weevil, Listronotus oregonensis (LeConte), in New Jersey. [Thesis]. New Brunswick, NJ: Rutgers University. 138 p. Rubatzky, V. E., and M. Yamaguchi. 1997. World vegetables: principles, production, and nutritive values. New York: Chapman & Hall. 843 p. Saadé, F. E., G. B. Dunphy, and R. L. Bemier. 1996. Response of the carrot weevil, Listronotus oregonensis (Coleoptera: Curculionidae), to strains o f Bacillus thuringiensis. Biological Control 7:293-298. 128 Sanders, D. C., H. J. Kirk, and C. van den Brink. 1980. Growing degree days in North Carolina. N. C. Agr. Ext. Serv. AG-236, p. 90. Cited in Mayüjrd and Hochmuth 1997 Scheiner, S. M., and J. Gurevitch. 1993. Design and analysis of ecological experiments. New York: Chapman & Hall. 445 p. Semel, M. 1957. Control of the carrot weevil attacking parsley. J. Econ. Entomol. 50(2):183-184. Shelton, A. M., J. Theunissen, and C. W. Hoy. 1994. Efficiency of variable-intensity and sequential sampling for insect control decisions in cole crops in the Netherlands. Entomol. Exp. Appl. 70:209-215. Simonet, D. E. 1981. Carrot weevil management in Ohio vegetables. Ohio Report 66: 83-85. Sokal, R. R., and F. J. Rohlf. 1995. Biometry: The principles and practice o f statistics in biological research. 3”* Ed. New York: Freeman. Southwood, T. R. E. 1966. Introduction to the study of animal populations. In: Ecological methods with particular reference to the study of insect populations. London: Methuen and Co LTD. p 1-56. Southwood, T. R. E., and G. A. Norton. 1973. Economic Aspects of pest management strategies and decisions. Ecol. Soc. Aust. Mem. 1:168-184. SPSS. 1999. Systat 9.0 for Windows 98, 95, and NT. Chicago, 111: SPSS Inc. Starnes, O. 1955. Insects injurious to carrots and celery in New Jersey, the carrot weevil, L. latiusculus (Bohe). 2 p. Steel, R. G., and J. H. Torrie. 1960. Principles and procedures of statistics. New York: McGraw-Hill Co. 481 p. Stem, V. M., R. F. Smith, R. van den Bosch, and K. S. Hagen. 1959. The integrated control concept. Hilgardia 29:81-101. Stem, V. M. 1973. Economic thresholds. Annual Review of Entomology 18:259-280. Stevenson, A. B. 1976. Seasonal history of the carrot weevil, Listronotus oregonensis (Coleoptera: Curculionidae) in the Holland Marsh, Ontario. Proc. Ent. Soc. Ont. 107:71-78. 129 Stevenson, A. B. 1981. Carrot insects. Min. Agric. Food Ont. Factsheet 81-007, Agdex 258/605. 4 p. Stevenson, A. B. 1983. Chemical control o f carrot weevil, Listronotus oregonensis (Coleoptera: Curculionidae), and damage to carrots in the Holland Marsh, Ontario. Proc. Ent. Soc. Ont. 114:101-103. Stevenson, A. B. 1985. Early warning system of the carrot weevil, Listronotus oregonensis (Coleoptera: Curculionidae) and its evaluation in commercial carrots in Ontario. J. Econ. Entomol. 78:704-708. Stevenson, A. B. 1986. A comparison of three methods of monitoring the carrot weevil, Listronotus oregonensis (LeConte) (Coleoptera: Curculionidae). The Canadian Entomologist 129:187-194. Stevenson, A. B., and G. Boivin. 1990. Interaction of temperature and photoperiod in control of reproductive diapause in the carrot weevil (Coleoptera: Curculionidae). Environ. Entomol. 19(4):836-841. Stevenson, A. B., and E. S. Barszcz. 1997. A comparison of three methods of monitoring the carrot weevil, Listronotus oregonensis (LeConte) (Coleoptera: Curculionidae). The Canadian Entomologist 129:87-194. Von Ende, C. N. 1993. Repeated-measures analysis; Growth and other time-dependent measures. In: Scheiner, S. M., and J. Gurevitch. 1993. Design and analysis of ecological experiments. New York: Chapman & Hall, p 113-137. Wald, A. 1947. Sequential analysis. New York: J Wiley. Whitcomb, W. 1965. The carrot weevil in Massachusetts, biology and control. Univ. Mass. Agric. Exp. Sta. Bull. 550. 30 p. White, G. F. 1927. A method for obtaining infective juvenile nematode larvae from cultures. Science 66:302-303. Wilkinson, L. 1988. SYGRAPH. Evanston, IL: Systat, Inc. Woodson, W. D., J. V. Edelson, and T. A. Royer. 1989. Control of a carrot weevil, Listronotus texanus (Coleoptera: Curculioniodae) : Timing pesticide applications and response to selected pesticides. J. Econ. Entomol. 82(1):209-212. Wright, J. M., and G. C. Decker. 1958. Laboratory studies o f the life cycle of the carrot weevil. J. Econ. Entomol. 51(l):37-39. 130 Zhao, D. X. 1990. Mathematicai models for the population dynamics and management of the carrot weevil, Listronotus oregonensis (Leconte) (Coleoptera: Curculionidae) [Dissertation]. Sainte-Anne-de-Belleuve, Quebec: McGill University. 254 p. Zhao, D. X., G. Boivin, and R. K. Stewart. 1991. Simulation model for the population dynamics of the carrot weevil, Listronotus oregonensis (LeConte) (Coleoptera: Curculionidae). Can. Ent. 123:63-76. 131