DAMAGE ASSESSMENT AND HOST PLANT RESISTANCE
FOR TWO MAJOR PESTS OF TEXAS HIGH PLAINS CROPS
by
LEONARDO DE AZEVEDO CAMELO, B.S.
A THESIS
IN
ENTOMOLOGY
Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of
MASTER OF SCIENCE
Appfpved
May, 2003 ACKNOWLEDGMENTS
I would like to show my appreciation to Dr. Scott Armstrong who worked with
me in all aspects of this project and made sure that I obtained a strong and solid
education, based on hard work and on the simple philosophy of giving all you can to
achieve success. Dr. Harlan Thorvilson and Dr. Leslie Thompson, thank you for all
suggestions and for always being there for when I needed help. Dr. Carlos Blanco and his
wife Suzana Blanco, thank you for having an enormous heart and making Graduate
School possible for me. Without you, I would not be here. Thank you for having such an
open mind.
Thank you my colleagues and close friends that made everything easier and a
little more fun: Amber Basinger, Igino Teolis, Michael Fanghaenel, and Ana Abarca.
Distances some times bring people close together especially when they play such
a big role in our life. Mariana Wongtschowski and Paulo Cesar Debenest, thank you for
being my best friends and for having given me so much inspiration. To be so far from
home and all my family makes everything harder in many different ways. I would like to
show my appreciation for my parents Evelin de Azevedo Camelo and Severino Firmino
Camelo, without your guidance and support I would never have conquered what I have so
far. Thank you for being the best parents a son could wish for. My brother: Marcelo de
Azevedo Camelo, thank you for being my hero and my role model and for always being there for me, Elizangela Camelo thank you for looking out for him and for being part of this extraordinary family.
11 TABLE OF CONTENTS
ACKNOWLEDGEMENTS ii
LIST OF TABLES v
LIST OF FIGURES vii
CHAPTER
I. INTRODUCTION 1
II. COMPARISON OF Lygus hesperus AND Lygus elisus
DAMAGE TO COTTON ANTHERS 4
Abstract 4
Introduction 5
Literature Review 8
Objectives 13
Materials and Methodology 14
Results 17
Field Experiments 17
Laboratory Trials 18
Discussions 19
Conclusions 22
Field Experiments 22
Laboratory Trials 24 References 34
111 III. DIETARY EVALUATION OF CYSTEINE PROTEINASE INHIBITORS scN AND E-64 AGAINST SOUTHERN CORN ROOTWORM {Diabrotica undecimpunctata Howardi Barber) INFESTING PEANUT 38
Abstract 38
Introduction 39
Literature review 42
Insect bioassays 42
Rearing southern corn rootworm 46
Objectives 48
Materials and Methodology 48
Purification of recombinant soyacystatin N,
and E-64 acquisition 48
Southern corn rootworm feeding bioassay 49
Results 50
Discussions 52
Conclusions 54
References 62
APPENDIX
A: CHAPTER II SAS PROGRAMS AND OUTPUTS FOR MEAN SEPARATION TESTS AND RAW DATA TABLES, 2001 AND 2002 SEASONS 65
B: CHAPTER III SAS PROGRAMS AND OUTPUTS FOR MEAN SEPARATION TESTS, LINEAR REGRESSION ANALYSES AND RAW DATA TABLES FOR scN AND E-64 TRIAL 122
IV LIST OF TABLES
-) Mean abscission, yield, and square injury comparison of Z. hesperus and L. elisus from field and laboratory trials from artificially infesting one-third grown, 6"^ node, first position, cotton squares, 2001 and 2002. 25
2.2. Mean abscission, yield, and square injury comparison of I. hesperus and L. elisus from field and laboratory trials from artificially infesting one-third grown, 9"^ node, first position, cotton squares, 2001 and 2002. 28
2.3. Mean abscission, yield, and square injury comparison of Z. hesperus and L. elisus from field and laboratory trials from artificially infesting one-third grown, 11"^ node, first position, cotton squares, 2001 and 2002. 31
3.1. Effect of scN and E-64 on mortality of one-week-old southem com rootworm larvae reared on artificial diet. 55
A. 1. Lint and seed weight combined (gms) for cotton bolls, first position 6' node infested for 24h with Lygus in the field during the 2001 season. 81
A.2. Percent damage estimates from artificially infesting cotton squares from first position, 6* node with Lygus for 24h (2001). 82
A.3. Lint and seed weight combined (gms) for cotton bolls, first position 6"^ node infested for 24h with Lygus in the field during the 2002 season. 83
A.4. Percent damage estimates from artificially infesting cotton squares from first position, 6"^ node with Lygus for 24h (2002). 84
A.5. Lint and seed weight combined (gms) for cotton bolls, first position 9' node infested for 24h with Lygus in the field during the 2001 season. 100
A.6. Percent damage estimates from artificially infesting cotton squares from first position, 9'*" node with Lygus for 24h (2001). 101
A.7. Lint and seed weight combined (gms) for cotton bolls, first position 9' node infested for 24h with Lygus in the field during the 2002 season. 102
A.8. Percent damage estimates from artificially infesting cotton squares from first posifion, 9* node with Lygus for 24h (2002). 103 A.9. Lint and seed weight combined (gms) for cotton bolls, first position 11"^ node infested for 24h with Lygus in the field during the 2001 season. 118
A. 10. Percent damage estimates from artificially infesting cotton squares from first position, 11"' node with Lygus for 24h (2001). 119
A. 11. Lint and seed weight combined (gms) for cotton bolls, first position 11"^ node infested for 24h with Lygus in the field during the 2002 season. 120
A. 12. Percent damage estimates from artificially infesting cotton squares
from first position, 11"' node with Lygus for 24h (2002). 121
B.l. Number of larvae dead in each replication for CPI (scN) test number 01. 135
B.2. Larval weight (mg) of living larvae from each replication for CPI (scN)
test number 01. 136
B.3. Number of larvae dead in each replication for CPI (scN) test number 02. 137
B.4. Larval weight (mg) of living larvae in each replication for CPI (scN)
test number 02. 138
B.5. Number of larvae dead in each replication for E-64 test number 01. 139
B.6. Larval weight (mg) of living larvae in each replication for E-64
test number 01. 140
B.7 Number of larvae dead in each replication for E-64 test number 02. 141
B.8. Larval weight (mg) of living larvae in each replication for E-64 test number 02. 142
VI LIST OF FIGURES
2.1. First position, 6'^ node, cotton squares abscised after 24hr infestation of Lygus spp. at 3, 7, 14, 21, 28, 35 DAT, and at harvest for 2001. 26
2.2. First position, 6"^ node, cotton squares abscised after 24hr infestation of Lygus spp. at 3, 7, 14, 21, 28, 35 DAT, and at harvest for 2002. 27
2.3. First position, 9"^ node, cotton squares abscised after 24hr infestation of Lygus spp. at 3, 7, 14, 21, 28, 35 DAT, and at harvest for 2001. 29
2.4. First position, 9"^ node, cotton squares abscised after 24hr infestation of Lygus spp. at 3, 7, 14, 21, 28, 35 DAT, and at harvest for 2002. 30
2.5. First position, 11^"^ node, cotton squares abscised after 24hr infestation of Lygus spp. at 3, 7, 14, 21, 28, 35 DAT, and at harvest for 2001. 32
2.6. First position, 11' node, cotton squares abscised after 24hr infestation of Lygus spp. at 3, 7, 14, 21, 28, 35 DAT, and at harvest for 2002. 33
3.1. Weight gain (mg + SEM) of one-week-old southern coon rootworm larvae after 4 days of feeding on artificial diet containing scN. 57
3.2. Weight gain (mg + SEM) of one-week-old southern coon rootworm larvae after 7 days of feeding on artificial diet containing scN 58
3.3. Weight gain (mg + SEM) of one-week-old southern coon rootworm larvae after 4 days of feeding on artificial diet containing E-64. 59
3.4. Weight gain (mg + SEM) of one-week-old southern coon rootworm larvae after 7 days of feeding on artificial diet containing E-64 60
3.5. Linear relationship of SCRW larval weight gain from two individual trials after seven days of assays on scN. 61
3.6. Linear relationship of SCRW larval weight gain from two individual trials after seven days of assays on E-64. 62
Vll CHAPTER I
INTRODUCTION
Integrated pest management can be defined as a strategy that focuses on long-term prevention or suppression of pest problems through a combination of techniques such as monitoring for pest presence, establishing treatment thresholds, using non-chemical practices, improving sanitation, and employing mechanical, physical, chemical and cultural controls strategies. According to Pedigo (2002), integrated pest management deals with a wide variety of pests including arthropods, plant-parasitic nematodes, microbial and viral plant pathogens, weeds, and vertebrates. The primary objective of the approach is to reduce losses from pests in ways that are effective, economically sound, and ecologically compatible. Pest management usually is equated with plant protection, focusing on value rather than the pest. It can be characterized as the way several techniques are employed simultaneously to solve specific pest problems. Therefore, a single pest management program may involve the use of pesticides, host-plant resistance, biological control, and damage assessment as a way to better understand and control pest complexes in different agro-systems.
The first topic of this thesis (Chapter II) focuses on damage assessment of an emerging pest complex on Texas High Plains cotton and concerns two different species from the same genera, the western tamished plant bug {Lygus hesperus Knight) and the pale legume bug {Lygus elisus Van Duzee). North American cotton fields are most likely to be infested by these insects from May until September when host plants other than cotton become less attractive or unavailable. Economic losses to cotton caused by plant
bug feeding can occur from the time cotton emerges until early bloom. A recent increase
in Lygus problems on the Texas High Plains is possibly a result of several mild winters,
that have increase the over-wintering of adults and reproductive densities for the
following spring. Precipitation in early spring will generate acceptable weed hosts, but
Lygus will move to cotton and other crops when precipitation ceases. The acreage of
other crop species such as alfalfa, potato, and canola has increased on the Texas High
Plains, which might help generate higher Lygus numbers to threaten cotton. All of these
factors combined together are the best possible explanation for an increase in densities of
L. hesperus and L. elisus and increased damage over the last four years. Lygus reduced
cotton production by an estimated 115,781 bales in 1999, 4,570 in 2000 and 13,913 in
2001 on the Texas High Plains (Williams, 1999, 2000, 2001). Recent research on Lygus
species and distributions on the Texas High Plains (Armstrong and Camelo, in press) has documented that in 1999-2001, L. elisus is significantly more prevalent than L. lineolaris, and in some cases, more predominant than L. hesperus. Management decisions for Lygus control on the Texas High Plains are based on L. hesperus densities. A comparative assessment of the damage is needed to determine if thresholds should be adjusted for both species.
The second topic (Chapter III) is the development of laboratory techniques that will improve host-plant resistance development for peanuts against southern com rootworm (SCRW, Diabrotica undecimpunctata hoM'ardi Barber) and possibly other insect pests, that can cause significant damage by feeding on roots and pods of major crops, causing significant economic losses to corn, sweet potato, and peanuts. Peanuts have become a significant crop and a major economic commodity grown in the state of
Texas. Texas currently ranks second to Georgia in peanuts, producing 320,000 to 370,000 acres annually. In 1999, 457,407 tons of peanuts were harvested (Lemon, 2001) generating 560 million dollars for farmers and 1.8 billion dollars of general revenue.
Over 70% of all the peanuts grown in Texas come from the Western Texas growing region. Research developed by Koiwa et al. (2000) has proved that a soybean cysteine proteinase inhibitor N (soyacystatin N, scN) cultured in laboratory was highly active against western corn rootworm {Diabrotica virgifera virgifera), substantially inhibiting growth and development of the larvae by attenuating digestive proteolysis. This inhibitor was named scN and is a derivative of soyacystatin N. Similar biology of Western com rootworm and SCRW and research by Edmonds et al. (1996) and Orr et al. (1994) implied that the southern corn rootworm would also be affected by scN ingestion. The
Effects of scN on SCRW were assessed and reported in chapter III. Screening for CPI
(scN) in different peanut lines is in progress as a separated phase of this research and is being conducted at the Norman Bourlag Center for Southern Crop improvement, Texas
A&M University, College Station, TX. In case of significant inhibition of larval development and growth of SCRW by scN in laboratory bioassays, peanut lines containing CPI (scN) will be used in a traditional breeding programs. E-64 (trans- epoxysuccinyl-leucyl-agmatine) is as powerful and specific as cysteine proteinase inhibitor on coleopterans (Kitch and Murdock, 1986). E-64 will be used as a calibrator factor to the methodology on scN studies. CHAPTER II
COMPARISON OF Lygus hesperus AND Lygus elisus DAMAGE
TO COTTON ANTHERS
Abstract
The pest status of Lygus in the Texas High Plains is becoming more important.
The increase in Bt cotton acreage, boll weevil eradication, and new insecticide technologies are some of the factors that may contribute to increased densities. Lygus species infesting cotton on the Texas High Plains was previously believed to be Lygus hesperus. but recent distribution surveys proved different (Armstrong and Camelo, in press), that Lygus elisus is also present in and often confused with Lygus hesperus. Field and laboratory experiments were conducted to assess injury to cotton anthers and compare L. elisus damage potential to L. hesperus. Both species were compared in the field by artificially infesting 6, 9 and 11"' node, first position cotton squares for 24 hours.
Each infested square was followed to yield to determine if it abscised or resulted in lint cotton. Laboratory experiments were developed to compare feeding to cotton anthers.
Results from field and laboratory studies demonstrated that L. elisus can cause similar or greater damage when compared to L. hesperus and visual damage from the digestive enzymes of both species were identical. Cotton will compensate for Lygus feeding injury, partially masking yield losses, and square abscission. Lygus elisus caused similar yield losses and square shading when compared to L. hesperus and proved to have a similar damage potential. Introduction
Western Tarnished Plant Bug {Lygus hesperus Knight) is a major pest of cotton, especially in the mid-south and southeast cotton production regions (Layton, 2000).
Lygus hesperus and the pale legume bug {Lygus elisus Van Duzee) are shown as western
species that are not found east of Texas, while the tarnished plant bug, Lygus lineolaris
(Palisot de Beauvois), is considered an eastern United States pest, threatening cotton production from the mid-south to southeast regions of the United States (Schwartz and
Foottit, 1998). North American cotton fields are most likely to be infested by these
insects from May until September when host plants other than cotton become less
attractive or unavailable (Stern, 1975). According to Black (1973 )and Tugwell (1976),
economic losses to cotton caused by plant bug feeding can occur from the time cotton
emerges until early bloom. Infestation levels of Lygus spp. will mostly be regulated by
over-wintering conditions, host plant availability, and weather.
The Pale Legume Bug, L. elisus, is not as common in cotton as L. hesperus and L.
lineolaris even though the two species overlap in distribution within the United States
(Schwartz and Footit, 1998). Inl927, McGregor documented L. elisus infesting cotton in
Arizona and California and described the shedding of squares, blooms, and young bolls.
He also recognized that the enzymatic activity of feeding from L. elisus was the end
result of the damage. McGregor (1927) provided the only literature that documented L.
elisus as a pest of cotton for the Western United States, until an extension publication from the University of California (Anonymous, 1996) specified that there are four species of Lygus that occur in the western United States, all of which are similar in appearance and habits, and identifying them to species when monitoring cotton would not be
necessary. Lygus hesperus would be the most common specie throughout the region,
being the only specie that causes economic damage to cotton. Lygus elisus would appear
sporadically, infesting cotton for only a few days. Less common species for the western
cotton growing regions would be L. lineolaris and L. desertinus. Diehl et al. (1998)
suggested that Lygus bugs infesting cotton in Arizona were a complex of Z. hesperus, L.
lineolaris and L. elisus, and that management decisions would not require identification
to species.
Other cotton growing states of the southeast such as Georgia, Alabama,
Tennessee, and Mississippi have documented that Lygus spp. are causing significantly
more damage than has been previously recorded (Layton, 2000). Boll weevil eradication
and the increase in Bt cotton acreage in the southeast and mid-south regions have reduced
the number of foliar insecticide applications that may have reduced Lygus populations.
Insecticides developed for specific lepidopteran pests that do not kill Lygus, and the
development of resistance to organophosphate and pyrethroid insecticides have
contributed to plant bugs becoming more of a key pest to cotton in the southeast (Layton,
2000). Cotton growing states of the western part of the United States have also documented an increase on Lygus densities. In recent years, Lygus control efforts in
Arizona have become more frequent and costs have increased due to more insecticide applications according to Ellsworth (2000). Factors such as the use of more specific insecticides, an increase in Bt cotton acreage, and more available alternate host plants in the State of Arizona helped plant bug problems increase. Arizona lost 2.9 million dollars of cotton in 1990, jumping to 7.6 million in 1999 from Lygus damage. Yield reductions
went from 0.95% to 3,32% in the same years. Bailey (1986) observed a significant yield
reduction on eight cotton cultivars after artificially infesting terminals with nymphs of Z.
lineolaris. According to the author, 25% reduction of the total yield across cotton
vaiieties was observed.
An increase in Lygus problems on the Texas High Plains is possibly a result of
several mild winters, which increases over-wintering of adults and reproductive densities
for the following spring. Precipitation in early spring will generate weed hosts, but Lygus
will move to cotton and other crops when precipitation ceases, or drought conditions
persist. The acreage of other crops species such as alfalfa, potato, and canola has
increased on the Texas High Plains, which might help generate higher numbers to
threaten cotton. All of these factors combined together have been responsible for an
increase in densities of Z. hesperus and L. elisus and increased damage over the last four
years. Lygus reduced cotton production by an estimated 115,781 bales in 1999, 4,570 in
2000, and 13,913 in 2001 on the Texas High Plains (Williams, 1999, 2000, 2001). Recent
research on Lygus species and distributions on the Texas High Plains (Armstrong and
Camelo, 2002) has documented that in 1999-2001, Z. elisus is significantiy more
prevalent than Z. lineolaris, and in some cases, more predominant than Z. hesperus.
Management decisions for Lygus control on the Texas High Plains are based on Z. hesperus densities per row foot in relation to square set (Muegge et al., 2002). The threshold does not consider combinations of other Lygus species, and assumes that they all cause the same amount of damage. According to Armstrong and Camelo (in press). the Texas High Plains is the only cotton production region in the United States where Z. hesperus and Z. elisus can be similar in densities. A comparative assessment of the damage needs to be conducted to determine if thresholds should be adjusted for both species.
Literature Review
Several authors have described western tarnished plant bug and tarnished plant bug damage to cotton anthers and its relation with injury to squares, blooms and young bolls, however pale legume bug, Lygus elisus, damage has not been thoroughly studied and its injury to cotton anthers and fruiting organs compared to other species of Lygus in unknown. Gutierrez et al. (1977) caged Z, hesperus on cotton terminals and observed the damage caused by adults (male and female) and nymphs. He concluded that western tarnished plant bug had a significant impact on square shed, but was not able to quantify a reduction in yield. Leigh (1988) correlated the numbers of Z, hesperus found in 50 net sweeps of cotton with damaged and abscised squares, and made two contributions two the understanding of Z. hesperus damage to cotton in California. His first observation was that dissecting small cotton squares was a very effective way of determining injury was caused by Z. hesperus feeding. The feeding injury was distinct and could be differentiated from other insect feeding injury. The second contribution was that Z. hesperus densities from 50 sweeps samples were correlated with square abscission and cotton yield, and concluded that Z. hesperus caused a significant increase in square abscission, although cotton has a natural compensation factor that masked the effects of Lygus damage to cotton squares. Russel (1999) observed the impact of tarnished plant bug (Z. lineolaris) feeding on cotton boll abscission and yield by caging the insects on one-day-old bolls for 72 hours. Russel concluded that plant bugs caused significant yield losses through abscised or injured cotton bolls. Jubb and Carruth (1971) infested individual cotton plants with one Z. hesperus during the first six weeks of flowering and observed the impacts of feeding to cotton growth and yield. Although nymph and adult infestations altered cotton plant development, the data indicated that artificially infesting did not caused significant reductions on yield. Conversely, Mauney and Henneberry
(1984) claimed the major cause of square abscission in Arizona before July 15 is from
Lygus hesperus feeding.
Literature that explains when cotton is most susceptible to plant bug injury is based on tarnished plant bug research. Littie research is available on this subject for other species, and the study of Lygus elisus damage to cotton is only found in McGregor
(1927). According to Scales and Furr (1968), cotton that has not flowered through early squaring is the most susceptible to yield loss from tarnished plant bug injury, and the developmental stage where most research has been conducted. Cotton is in these early developmental stages in May or June, when plant bugs are migrating from native hosts and searching for other suitable hosts. Tarnished plant bugs can feed on cotton plants during the entire season, however economic damage will occur from first-square to first bloom (Layton, 1995; Craig and Luttrell, 1997). Studies examining the effects of plant bug infestation on flowering and later developmental stages are fewer in number. The affects of plant bug injury on flowers and bolls have been considered relatively unimportant (Tugwell et al., 1976). Plant bugs have always been considered an early season pest, and little damage was thought to occur on young to mature bolls, especially when that plant bug densities are low during the late season or after the majority of bolls, that will produce lint, are one day in age (Russel, 1999).
Tarnished plant bug feeding results in a characteristic necrosis of the anthers and atrophy of the pollen sacs, leading to the abscission of the square if sufficient injury occur
(Pack and Tugwell, 1976). Lygus feeding on bolls appears as a dull, dark necrotic area about 1 to 2 mm in width with a glossy spot in the center of the larger spot (Pack and
Tugwell, 1976). Tugwell et al. (1976) observed from one to 10% yield reduction by plant bugs feeding on bolls. However, the age of the bolls, or stage of plant development were not considered factors in that study. Pack and Tugwell (1976) reported that tarnished plant bugs caged on two-day-old bolls significantly reduced seed cotton yield and seed quality, and increased boll abscission due to plant bug injury. Williams and Tugwell
(1999) described histological damage of tamished plant bug feeding on small cotton floral buds. Staminal columns, developing anthers and corollas appeared to be most damaged from insect feeding. All of these tissues demonstrated different degrees of cellular degradation and gross enlargement caused by enzymatic action. Fragmented cell walls were thinner and stained lighter than cotton floral buds without damage. The authors observed that desiccation of abscised squares from tarnished plant bug feeding was irregular.
Plant bug feeding effects on crop yield and damage thresholds can be significantiy different depending upon location and sampling methods used from research studies
10 (Snodgrass, 1993; Guitierrez, 1995). Lygus thresholds for management decisions will vary depending upon cotton growing regions and production practices. Variation will occur due to difference in the length of growing seasons, rainfall patterns, planting times, and plant densities. Such variations should be taken into consideration (Wheeler, 2001).
The most generalized Lygus threshold of 10 bugs per 50 cotton sweeps (1 nymph = 2 adults) was established in early 1950 and has been previously reviewed and still accepted
(Stern ,1973) .Low densities or sub-economic densities of plant bugs infestations on cotton may result in increased yield due to the natural compensation that occurs when cotton tries to overcome damage. Teague et al. (2000) compared cotton response to square loss from tarnished plant bug feeding and manual removal. By artificially infesting cotton with tarnished plant bug nymphs and manually crushing squares, Teague observed direct evidence of compensatory response of cotton to tarnished plant bug.
Significant crop delay was observed when square retention was reduced by tarnished plant bug as measured by physiological cutout. The compensatory response of cotton must be understood in order to determine the true. According to Gutierrez et al. (1977),
Lygus hesperus is a serious pest only when present at extremely high densities or when cotton plants are under some kind of physiological stress. Mauney and Henneberry
(1984) emphasized the difficulty of assessing the various factors that are responsible for square damage and abscission, making published thresholds of plant bugs on cotton controversial. The relationship of Lygus densities with damage and yield is a complex issue that has been difficult for researchers to draw conclusions based on sound science.
11 Wheeler (2001) observed that thresholds in the mid-south and southeast are too
conservative and tend to overestimate Lygus bug importance as a cotton pest.
Rearing Lygus spp. in laboratory. According to Whitbey (1999), there are two
common methods for rearing and maintaining plant bugs in the laboratory. The first
method follows Beards and Leigh (1960) who reared Z. hesperus on green beans,
Phaseolus vulgaris L., which provided a food source and an oviposition site. Patana and
Debolt (1985) described a method that uses artificial diet packs where the food source is
enclosed in parafilm". These two methods of rearing Lygus with diet packs and on green
beans have a common problems in that insects oviposit on the same food source, that
becomes dry or infected with microbial growth which tends to reduce egg production and
recovery of adults. The green bean rearing method presents other problems, one of which
is determined by the source of green beans. Unless appropriate greenhouse space is
available, where pesticides use can be reduced, green beans must be purchased at the
market. Commercially produced green beans may come from locations where pesticides
use is not regulated or where Lygus parasites may be harbored. Although the beans are
washed, any pesticide residue can decimate a colony. Eggs of parasite species that remain
on the green beans and hatch can reduce Lygus densities as well. Fungal spores are not
uncommon on green beans once they have infiltrated a growth chamber they are difficult to eradicate.
Deboh (1982) created an artificial diet for rearing Lygus. Most of his effort aimed at developing artificial diets resulted in different degrees of success. The meridic diet by
Debolt maintained several generations of Z. hesperus, and was later improved by Cohen
12 (2000). Cohen's diet consisted of adding a semisolid slurry that accommodates solid and liquid feeding habits of Lygus. The basic ingredients, as described by Cohen are an entomophage component of cooked, whole chicken eggs, chicken egg yolks, sugar, and yeast combined with plant components of soy bean flour, wheat germ, lima bean meal and soy lecithin. Cohen's diet proved to maintain Lygus colonies longer than Debott's
(1982). There are problems with using artificial diet packs. The Lygus not only feed on the packs, they oviposit in them. If the diet packs dries out, high mortality to adults and eggs will occur. If the diet is not prepared in an aseptic environment, microbial growth will ruin it and begins to ferment, Lygus egg yield will be significantly reduced.
Extracting and sterilizing eggs from the diet pack is difficult and time consuming. The most efficient way to rear Lygus spp. is to use separate packs for the diet and oviposition.
However, this creates the problem of getting females to oviposit in the egg packs and not the diet packs. The diet pack method of rearing Lygus proposed by Cohen (2000), using separated gel packets for oviposition, is the most efficient way of maintaining Lygus spp in laboratory.
Objectives
Observations of plant bug densities, and distributions on the Texas High Plains
(Armstrong and Camelo, in press) demonstrated that Z. elisus is as relevant as Z. hesperus in terms of being a cotton pest. However, very little is known about Z. elisus biology on cotton or its damage potential. The research proposed here is to determine the
13 damage potential of Z. elisus and compare it with a known cotton pest, Z. hesperus. In both field and laboratory studies
Materials and Methodologv
Field experiments were conducted at the Texas Tech research farm located on the
northwest part of the Texas Tech University Campus, Lubbock, TX. Paymaster®
2326RR was used as a standard variety in 2000 and 2001. Laboratory experiments were
conducted on campus at the entomology laboratory. Agricultural Science building.
Rearing Lygus.
Lygus elisus and Lygus hesperus used in these experiments were reared in
laboratory from adults collected from the field. Most of these insects came from alfalfa
{Medicago sativa L.) and wild mustard {Sinapis arvensis Z.) from different South-Plains
sites. Insects were held in Im xlm Lumite® cages and fed artificial diet as described by
Cohen (2000). Insects were maintained in the laboratory so mortality from physical injury and disease could be reduced.
Artificial infestation procedures for field experiments.
Cotton plants were monitored regularly until squares of the 6' , 9' and 11' node, first position were one-third grown and ready for infesting. Artificial infesting procedures for one-third grown squares followed that used by Russel (1999) for caging Z. lineolaris on cotton bolls. Lygus were placed into 20 ml diet cups and transported to the field in a chilled ice cooler to eliminate heat stress. For each species (Z. elisus and Z. hesperus), one Lygus were enclosed in a 15 cm. x 11.5 cm. nylon mesh #280 bag, which also
14 enclosed a one-third grown square. Each bag had a drawstring that prevented Lygus from escaping closed the bags. The dates of infestation were recorded on a snap-on-tag® placed on the pedicel of each individual square. The experiments was designed as a randomized block and each treatment was replicated four (4) times. A control treatment, consisting of bags that enclosed a square but no Lygus. Ten squares from different plant were used in each block. Squares selected for artificial infesting were of a similar fruhing position and developmental stage of adjacent plants.
Lygus elisus and Lygus hesperus artificial infesting.
Infestation levels consisted of zero and one plant bug adults on each square. Lygus bugs were placed on squares and allowed to feed for 24 hours. The number of abscised squares was recorded at 3, 7, 14, 21, 28, and 35 days after infestation (DAI) and at the time of harvest. All harvestable bolls were individually collected and seed cotton weights were recorded. The data was analyzed with PROC ANOVA and means separated with
Fisher's Protected LSD (SAS Institute, 1989). Treatment affects were measured by correlating Lygus species with squares loss and cotton lint and seed yield. Damage potential can be estimated from this relationship.
Laboratory assessment.
In conjunction with field experiments, laboratory trials were developed to compare Z. hesperus and Z. elisus damage to one-third grown squares. Squares of identical age and position from the field study were artificially infested for 24hr then dissected to estimate the amount of injury. Damage assessment of Lygus elisus compared to Lygus hesperus feeding on cotton squares in laboratory were done following Maredia's
15 et al. (1994) technique of identifying resistant lines of cotton to plant bugs. According to the author, assessing anther damage is simple, rapid, and an effective means that entomologists and plant breeders can use to evaluate resistance and/or susceptibility of cotton to Lygus spp. It consists of slicing a square into two pieces at the point of
maximum diameter with a razor blade, and gently pressing the top end of the square
using the thumb and the forefinger, with a rolling action, the anthers will be removed
from the calyx and corolla and be exposed for observation of damage. In our studies,
Maredia's et al. (1994) technique was used to compare the feeding injury to one-third
grown cotton squares from the two different Lygus species (Z. hesperus and Z. elisus)
damage to anthers using one cotton genotype, from squares collected within the same
stage of development.
Feeding assessment was possible by setting cotton squares on Oasis® (Oasis
Craft Products, Kent, OH) to maintain the squares, and by placing one Lygus of either
species to feed on it. Individual squares were isolated and Lygus caged by placing 20 ml
diet cups (Bio-serve® supplies, Chicago, IL) over the top of a Lygus and a plastic film on
the surface to avoid contact of the insect with the Oasis. The experiments consisted of
three treatments (Z. elius, L. hesperus and a control) replicated ten times in a randomized
block design. Every time a field cage experiment was conducted, a laboratory experiment
was conducted on the same age and position (6'^ 9"" and 11* node, first fruit position)
square. Lygus were allowed to feed for 24 hours. Damage to each square was estimated
by removing the calyx and corolla to expose the anthers and revealing the surface area
that had anther tissue damage. The estimates were easily discernible because the tissue
16 affected by the Lygus feeding enzymes turns dark brown or is dissolved and no longer
present. The anthers were rated from 0 (no damage) trough 100 (maximum damage). The
damage estimations were analyzed with PROC ANOVA, and means separated with
Fisher's Protected LSD (SAS Institute, 1989).
Results
Field Experiments
Square abscission at the 6"^ node was significantiy greater on Lygus-infested
plants (F= 11.25, df=l 1, P=0.0053) in 2001 (Table 2.1 and Figure 2.1). In 2002,
abscission was significantly lower in the control when compared to Z. elisus-infested
plants, but abscission by Z. hesperus was intermediate (Table 2.2) (F=2.63, df= 11,
P=0.1358). Squares retained on the plant for 28 days after artificially infested (DAI) were
most likely to become harvestable bolls that resulted in seed-cotton yield (Figures 2.1 and
2.2).
Mean seed-cotton weights in 2001 and 2002 (Table 2.1) were not significantly
different (F=0.93, df=l 1, P=0.5237; F=0.29, df=l 1, P=0.9043, respectively). Bolls that remained on the 6'^ node accounted for a larger weight in Lygus-infcsted plants as a compensation for Lygus feeding. For early insect infestations that attack the first position,
6"^ node square, cotton has enough heat-units available to compensate to square losses.
Significant differences were observed in 2001 for the 9' node square abscission rates with greater abscission on both Lygus-infested treatments when compared to the control treatment (F=3.40, df=l 1, P=0.0842) (Table 2.2). Results for 2002 did not
17 showed the same trend, and no significant difference was recorded for any treatments
(F==2.57, dt=l 1, P=0.1408). Water stress from a lack of irrigation may have caused physiological stress that increased abscission, which masked the affects of insect damage from Zi;i,^/w-infested treatments (Figures 2.3. and 2.4.).
Yield at the 9 ^ node on Lygus elisus and Lygus hesperus-infested treatments in
2001 were statistically equal to each other but significantly lower than the control treatment (F=12.83, df=l 1, P=0.0037) (Table 2.2). However, due to high abscission from the control, results were not the same in 2002, and no differences were found in yield across treatments (F=0.31, df=l 1, P=0.8893) (Table 2.2).
Eleventh node natural abscission rate masked Zj'gw^-infested treatment affects. No significant difference was observed (Table 2.3) in 2001 and 2002 (F=3.00, df=l 1,
P=0.1256; F=0.32, df=ll, P=0.8844, respectively) with the control group averaging
83.75% abscission rate in both years. In 2002, abscission was gradually achieved, however final abscissions were similar when compared to 2001 (Figures 2.5 and 2.6).
Results of abscission rates at the 11"' node were reflected on yield observations and no significant difference was recorded among treatments for 2001 and 2002 (F=4.31, df=ll, P=0.8844; F=0.56, df=ll, P=0.7307; respectively) (Table 2.3).
Laboratory Trials
Injury estimates for 2001 were significantiy higher for both Zji/^u^-infested treatments when compared to the control group (Table 2.1) (F=14.35, df=29, P=<.0001).
Some injury (2.5%) was present in the control group in 2001 due to insect feeding in the
18 field previous to square removal from the field. For 2002, significant difference was observed in the control group compared to Z. e/ww^-infested squares; however, Z. hesperus-infested treatments were not significantly different from the conttol (Table 2.1)
(F=8.53, df= 29. P= 0.0013). The variance was high resulting in no significant damage from Z. hesperus.
The 2001 and 2002, 9' node, laboratory studies provided similar results in which both Z_ygw5-infested treatments were not significantly different from one another and
significantly greater than the control group (F=6.13, df=29, P=0.0064; F=8.27, df=29,
P=0.0016, respectively) (Table 2.2). Damage of the control group in 2002 (0.5%)
originated from insect feeding in the field previous to the square removal (Table 2.2).
Damage estimates done from laboratory studies on the 11"^ node squares, showed
Z. elisus damage significantly greater than both the control and the Z. hesperus treatments
(F=9.61, df=29, P=0.0007) (Table 2.3). For 2002, each treatment was significantiy different from the other with Z. e/Z^w^-infested treatment damage being greater than Z. hesperus, and Z. hesperus damage greater than the control group (F=18.18, df=29,
P=<.0001) (Table 2.3). High variance of the means within each treatment resulted in no significant differences for Z. hesperus and the control group in 2001 (Table 2.3).
Discussions
A certain percentage of cotton is expected to abscise from the plant before maturity due to physiological behavior and insect injury. According to Room and Heam
(1979) and Sadras (1995), the shed of small bolls and squares during cotton growth is a
19 natural feedback mechanism that alters the plant strain and is a complex nutritional and hormonal influence that is poorly understood (Teague et al., 2000). This mechanism can mask abscission caused by Lygus feeding. Field results for the 6"' node control group for
2001 averaged 12.5% abscised squares that did not result in seed-cotton. The two Lygus- infested treatments resuhed in significant higher abscission compared to the control. This finding was under more normal growing conditions where the cotton did not suffer severe drought stress as happened in 2002. Square abscission for 6"^ nodes in 2002 was affected by drought stress from a deficiency in irrigation. The irrigation well was not operating for
4-5 weeks, resulting in a higher than normal abscission from the control group, which resulted in no significant difference between the control group and the Z. hesperus-
infested squares. The Z. hesperus abscission treatment was not significantly different
from the Z. elisus treatment. The water deficiency is the most likely cause for a non- distinct separation of the 6^*^ node abscission in 2002.
Although there are significant differences for the 6^*^ node abscission for 2001, it
did not resuh in significant seed-cotton weights (Table 2.1). The control (non-infested) treatment weights averaged 3g more than the Z. hesperus treatment, which was 0.6g
lighter than the Z. elisus treatment. The average seed-cotton weight from the control for the 6'^ node was 1.04 g compared to 1.07 g for Z, hesperus and 1.12 g for Z. elisus. These results indicate a compensatory factor that occurs when a higher percent of 6' node squares are injured from either species of Lygus.
Ninth node, first position square abscission for the control group in 2001 was significantly higher than for Z. hesperus and Z. elisus-infested squares (Table 2.2). The
20 0%'erall abscission for the 9"^ node essentially doubled when compared to the 6'" node
abscission for 2001 (Tables 2.1 and 2.2). However, final abscission for 2002 was not
significantly different for any treatment group (Table 2.2). This lack of significance in
2002 is a probable consequence of the drought stress that resulted in excessive square
shed. Seed-cotton weights were significantly higher for the conttol compared to either
Lygus species, which were not significantly different for each other in 2001 (Table 2.2).
No differences were observed for seed-cotton weights in 2002. The final abscission
percentage for 2002 was not significant and no compensation appeared as reflected by
seed-cotton weights (Table 2.2).
Abscission was greater than 82% for all treatments and not statistically different
for the 11' node for both years (Table 2.3). Although there is no literature on the
percentage of yield that the 11* node conttibutes to total yield, h would appear that a
high percentage of 11'^ node first position squares do not contribute to the total yield, and
compensation did not occur based on the results of this study.
Laboratory feeding trials for the 6* node in 2001 estimated significantly higher
damage from both Zj^gw^-infested treatments when compared to the control group. Lygus
elisus damage averaged 34.8%, an average of 8% higher than the Lygus hesperus group
(Table 2.1). The control group averaged 1.25% of damage for both years, this damage was registered due to feeding on the squares in the field before the square was removed and brought to the laboratory. Results for 2002 showed Z. elisus causing significantly more damage than Z. hesperus and the control group (Table 2.1). No significant difference was observed when comparing the control group and the Lygus hesperus-
21 infested treatments, however variance from is likely to be the cause for the lack of
significant difference rather than a lower feeding damage from Z. hesperus.
Damage ratings for the ninth node, first position was estimated to range, from 21 to 29% damage for both years of the study and for both species of Lygus (Table 2.2). This was significantly higher than the control group, which averaged 0.25% damage. The data for 2001 and 2002 suggest that the feeding injury to the 9"" node squares is very similar for both Z. hesperus and Z. elisus.
Damage estimates for the 11''' node laboratory studies demonstrated that Z. elisus injury was significantly higher than Z. hesperus and the control for both years (Table
2.3). Z. elisus averaged 31.5% for both years while Z. hesperus averaged 14%. The control treatment for both years had no damage (0%).
Conclusions
Field Experiments
Yield results were not significantly difference across treatments in 5 out of 6 trials. However, comparison between damage potential of both Lygus species carmot be based specifically on yield outputs. According to Maredia (1994), ultimate production is influenced by many interacting factors that can mask plant bug damage and does not reflect susceptibility or resistance of the cotton plant to Lygus. Cotton has a natural ability to compensate for different stress environments and is able to maintain its potential yield.
Teague et al. (2000) explained that the compensatory response of cotton must be understood as an integrated approach and multiple factors should be considered. Teague
22 et al. (2000) compared cotton response to square loss from tarnished plant bug feeding
and manual removal and observed direct evidence of compensatory response of cotton to
or following insect injury. Significant crop delay was observed when square retention
was reduced by tarnished plant bug. Other studies have demonstrated that an increase in
yield is possible. In 1973, Stern observed that low densities or sub-economic densities of
plant bugs infestations on cotton could result in increased yield due to the natural
compensation that occurs when cotton tties to overcome damage. For the 2001 6"" node
studies, seed-cotton weights for both Lygus-infested treatments averaged 1.10 grams
while the control group averaged 1.04 grams. While fewer bolls were harvested for the
Zvg-w5--infested treatments, greater weight of individual bolls resulted in no significant
difference across treatments in yield as if cotton had compensated to Lygus-infested
treatments. For the 9* and 11'^ node studies in 2001 and 2002, yield results reflected
abscission significant differences across treatments.
In 5 out of 6 abscission trials (3 trials in 2001 and 3 in 2002), Lygus elisus and
Lygus hesperus had equal results when compared to one another and to the control group.
This shows a trend of both species causing similar damage to artificially infested cotton
squares. Abscission results for the 6'*^ node in 2001 showed a significant difference from both Lygus infested treatments for the control group that was repeated on the 9^*^ node study for 2001. However, 11"^ node squares natural abscission on the control group averaged above 83% masking the affects of the Lygus-infested txedXments both Z. elisus and L. hesperus had no significant difference among each other for both years.
23 Cotton squares were most likely to abscise 28 days after infestations in all treatments and node positions. After this period, one-third grown cotton squares that remained on the plant became young bolls. For the 6"' node abscission experiments in
2001, all treatments reached the final abscission rate on or before 28 days after infestation
(DAI). For the 2002, 6" node abscission experiments, the control and Lygus elisus treatments reached the final abscission rate 28 days after infestations while the Lygus hesperus increased only 2.5% from 28 DAI to 35 DAI where final abscission rate was reached. Lygus hesperus treatment for the 9"" node in 2002 reached final abscission after
28 days after infestation while Lygus elisus only increased 2.5% after 28 DAI reaching final abscission 35 days after infestation. The 2001 11'^ node squares reached final abscission rate in all treatments after 14 DAI.
Laboratory Trials
Lygus elisus damage estimates to cotton squares in the laboratory proved to be greater or equal to damage estimates caused by Lygus hesperus. In 3 out of 6 laboratory trials for 6* trough 11'^ node, Z. elisus feeding was significantly higher than Z. hesperus, and Z. elisus averaged across trials 29.1% of damage, while Z. hesperus averaged 19.8%.
These estimations are subjective, and based on observer bias. Some variation could occur if the experiment was repeated, however the results demonstrated that Z. elisus has the capacity of damage cotton anthers equally or greater than Z. hesperus.
24 Table 2.1. Mean abscission, yield, and square injury comparison of Z. hesperus and Z. elisus from field and laboratory trials from artificially infesting one-third grown, 6'*^ node, first position, cotton squares, 2001 and 2002. Field Experiments Laboratory Trials
Final square Yield (Seed-cotton) + Square injury abscission (%) + SEM SEM (%) + SEM
Treatments 2001 2002 2001 2002 2001 2002
Control 12.5+2.5a 22.5+4.8a 36.3+L9a 3Ll+2.1a 2.50+L7a 0.0+O.Oa
L. hesperus 22.5+4.8b 27.5+2.1ab 33.2+5.4a 29.9±3.7a 26.8+5.8 b 13.0+4.4a
L. elisus 27.5+6.3 b 32.5+4.8 b 32.6+4.6a 27.2+1.9a 34.8+4,7 b 27.0+6.7 b Column means followed by the same small letter are not significantly different, while those followed by a different letter are significantly different by Fisher's LSD, {P > 0.05).
25 35 r • ~ Control « hesperus —' elisus 27.5b 28 • T-
a 22.5b
a 21
B 2 14 12.5a •
S
CZ3
0 3 7 14 21 28 35 95 DAI
Figure 2.1. First position, 6^*^ node, cotton squares abscised after 24hr infestation of Lygus spp. at 3. 7, 14, 21, 28, 35 DAT, and at harvest for 2001.
26 40 r * Control — •— hesperus ~'^~ elisus
^ 32.5b 32 - T T- 27.5ab a -• u 24 22.5a a
16
!Z!
3 1/2
0 3 7 14 21 28 35 95 DAI
Figure 2.2. First position, 6"^ node, cotton squares abscised after 24hr infestation of Z>'gw5 spp. at 3, 7, 14, 21, 28, 35 DAT, and at harvest for 2002.
27 Table 2.2. Mean abscission, yield, and square injury comparison of Z, hesperus and Z. elisus from field and laboratory trials from artificially infesting one-third grown, 9' node, first position, cotton squares, 2001 and 2002. ^^^^^^^^ Field Experiments Laboratory Trials
Final square Yield (Seed-cotton) + Square injury abscission (%) + SEM SEM (%) + SEM
Treatments 2001 2002 2001 2002 2001 2002
Control 55.0+2.0a 62.5+7.5a 25,2+4,9a 16,1+2.2a 0.0+0.0a 0,50± 1.6a
L. hesperus 65.0±2,9b 65.0+10,4a 15.3+7,5b 14,2+3,6a 25.5± 6,5 b 25.5+7,7 b
L. elisus 67.5+4.8b 57.5+6,3a 15.3±5,4b 19,9+4.6a 21.0+6.9 b 29.0+5.4 b Column means followed by the same small letter are not significantiy different, while those followed by a different letter are significantly different by Fisher's LSD, {P > 0.05).
28 75 r
0 * 0 3 7 14 21 28 35 80 DAI
Figure 2.3. First position, 9^^ node, cotton squares abscised after 24hr infestation of Lygus spp. at 3, 7, 14, 21, 28, 35 DAT, and at harvest for 2001.
29 75 r
60 -
B
45 -
1/1 30 -
.Q
« 15 - S
0 0 3 7 14 21 28 35 80 DAI
Figure 2.4. First position, 9^t h node, cotton squares abscised after 24hr infestation of Z>'gw5 spp. at 3. 7, 14, 21, 28, 35 DAT, and at harvest for 2002.
30 Table 2.3. Mean abscission, yield, and square injury comparison of Z, hesperus and Z. elisus from field and laboratory trials from artificially infesting one-third grown, 11"^ node, first position, cotton squares, 2001 and 2002 Field Experiments Laboratory Trials
Final square Yield (Seed-cotton) + Square injury abscission (%) + SEM SEM (%) + SEM
Treatments 2001 2002 2001 2002 2001 2002
Control 82.5+4.8a 85,0+5.0a 7.1+ 1,5a 5.6+2.6a 0.0+0,0a 0.0+0.0a
L. hesperus 92.5+4,8a 87,5± 6.3a 3,2+1,9a 5,4+3.3a 13,0+3.9a 15.0+2.9b
L. elisus 82,5+6.3a 90.0+7.1a 6.5+1.9a 4.6+2,7a 39.5+10,6 b 23.5+3.9 c Column means followed by the same small letter are not significantly different, while those followed by a different letter are significantly different by Fisher's LSD, {P > 0.05).
31 100 Control — • —hesperus —"" — elisus g - • - 92.5a
(% ) ,— — g
V "—'• 82.5a o percentag e _E 7" B f •e 40 Vi
01 k.
g- 20
^1 L_ 1 I 1 1 1 3 7 14 21 28 35 65 n 1 DAI Figure 2.5. First position, 11"^ node, cotton squares abscised after 24hr infestation of Lygus spp. at 3, 7, 14, 21, 28, 35 DAT, and at harvest for 2001.
32 100 r * Control hesperus — •— elisus 90.0a -m 87.5a
0 3 7 14 21 28 35 65 DAI Figure 2.6. First position, 11' node, cotton squares abscised after 24hr infestation of Lygus spp. at 3, 7, 14, 21, 28, 35 DAT, and at harvest for 2002.
33 References
Armstrong, J. S. and Camelo, L. A. 2002. Damage comparison of Lygus Elisus and Lygus hesperus to one-third grown squares of Texas High Plains cotton. Proceedings beltwide cotton conference (In press).
Bailey, J.C. 1986. Infesting cotton with tarnished plant bug (Heteroptera: Miridae) nymphs reared by improved laboratory rearing methods. J. Econ. Entomol 79- 1410-1412.
Beards, G. W. and Leigh, T. F. 1960. A laboratory rearing method for Lygus hesperus Knight. J. Econ. Entomol. 53: 327-328.
Black, E. R. 1973. Economic thresholds studies of tarnished plant bug, Lygus lineolaris in cotton. PhD dissertation, Mississippi State University, pp 108.
Cohen, A.C. 2000. A review of feeding studies of Lygus spp. with emphasis on artificial diets. Southw. Entomol. 23: 111-119.
Cohen, A.C. 2000. New oligidic production diet for Lygus hesperus Knight and Z. lineolaris (Palisot de Beauvois) J. Entomol. Sci. 35 (3): 301-310.
Craig, C. Luttrell, R. G., Stewart, S. D. and Snodgrass, G. L., 1997. Host plant preferences of tamished plant bug: a foundation for trap crops in cotton. Proc. Beltwide Cotton Conf National Cotton Council, Memphis - TN. 1176-1181.
Debolt, J.W. 1982. Meridic diet for rearing successive generations of Lygus hesperus. Ann. Entomol. Soc. Am. 75: 119-122.
Diehl, J. W., Ellsworth, P. C. and Moore, L. 1998. Lygiis in cotton (1). Identification, biology and management. The University of Arizona Cooperative Extension. pp2.
Doederlim, T., Baugh, B., Leser, J. F., Boman, R. 2001. Plant response to different levels of pre-bloom square removal and its relevance to plant bug management at AG- CARES. http://lubbock.tamu.edu/2001book/cotton7.html.
Ellsworth, P. C. 2000. Lygus control decision aids for Arizona cotton. 2000 Arizona Cotton Report. The University of Arizona College of Agriculture. 269-280.
Gutierrez, A. P., Leigh, T. F., Wang, Y. and Cave, R. D. 1977. An analysis of cotton production in California: Lygus hesperus (Heteroptera: Miridae) injury - an evaluation. Can. Entomol. 109: 1375-1383.
34 Guttierrez, A. P. 1995. Integrated pest management in cotton. Integrated pest management. Chapman and Hall, London. 280-310.
Hearn, A. B. and Room, P. M. 1978/1979. Analysis of crop development for cotton pest management. Prot. Ecol. 1; 265-277.
Jubb, G. L. and Carruth, L. A., 1971. Growth and yield of caged cotton plants infested with nymphs and adults of Zv,i,'//.v hesperus. J. Econ. Entomol. 64(5): 1229-1236.
Layton, M. B. 1995. Tarnished plant bug: biology, thresholds, sampling, and status of resistance. Proc. Beltwide Cotton Conf National Cotton Council of America Memphis, TN. 131-134.
Layton, M. B. 2000. Biology and Damage of the Tarnished Plant Bug, Lygus Lineolaris, in Cotton. Southw. Entomol. 23: 7-20.
Leigh, T.F.; Kerby, T.A.; Wynholds, P.F. 1988. Cotton square damage by the plant bug, Lygus hesperus (Hemiptera: Heteroptera: Miridae), and abscission rates J. Econ. Entomol. 81(5): 1328-1337.
Mauney, J. R. and Henneberry, T. J. 1984. Cause of square abscission in Arizona. Crop, sci. 24: 1027-1030.
Maredia, K.M., Tugwell, N. P., Waddle, B. A. and Bourland, F. M. 1994. Technique for screening cotton germplasm for resistance to tamished plant bug, Lygus lineolaris (Palisot de Beauvois) Southw. Entomol. 19 (1): 63-70.
McGregor, E.A. 1927. Lygus elisus: a pest of the cotton regions in Arizona and Cahfornia. Technical Bulletin, USDA. 4: 1-15.
Muegge, M. A., Baught, B. A., Leser, J. F., Doederlein, T. A. and Boring, E. P. 2002. Managing cotton insects in the High Plains, Rolling Plains and Trans Pecos areas of Texas. Texas Cooperative Extension. Texas A&M University. E-6. 3-02. pp 17.
Pack, T.M., and Tugwell, N. P. 1976. Clouded and tamished plant bugs in cotton: A comparison of injury simtoms and damage fruit parts. Univ of Arkansa Agric. Exp. Stn. Bull. Report. Series 226: 1-17.
Patana, R. 1982. Disposable diet packet for feeding and oviposition of Lygus hesperus (Hemiptera: Miridae) J. Econ. Entomol. 75: 668-669.
Patana, R. and Debolt, J. W. 1985. Rearing Lygus hesperus in the laboratory. U.S. Dep. Agric. Agric. Res. Serv. ARS. 45: 1-9.
35 Russel, J. S., 1999. Effects of tarnished plant bug, Lygus lineolaris (Palisot de Beauvois), feeding on cotton boll abscission and yield. Thesis submitted to Louisiana State University Department of Entomology, pp 39.
Sadras, V. O. 1995. Compensatory growth in cotton after loss of reproductive organs. Field Crops Res. 40: 1-18.
Scales, A. L. and Furr, R. E. 1968. Relationship between the tarnished plant bug and deformed cotton plants. J. Econ. Entomol. 67: 602-604.
Schawrtz, M. D. and Fottit, R. G. 1998. Revision of the nearctic species of the genus Lygus hahn, with a review of the palaearctic species (Heteroptera: Miridae). pp 428.
Snodgrass, G.L. 1993. Estimating Absolute Density of Nymphs of Lygus lineolaris (Heteroptera: Miridae) in Cotton Using Drop Cloth and Sweep-Net Sampling Methods. J. Econ. Entomol. 86(4): 1116-1123.
Stern, V M. 1973. Economic thresholds. Annu. Rev. Entomol. 18: 259-280.
Stern, V. M. 1976. Ecological studies of Lygus bugs in developing a pest management program for cotton pests in the San Joaquin Valley, California. Lygus bug: Host plant interactions. University of Idaho Press, Moscow. 8-13.
Teague, T. G., Tugwell, N. P., Villavaso, E. J. 2000. Comparison of cotton plant response to square loss following manual removal for tarnished plant bug feeding from field trials in 2000. Unpublished Data.
Tugwell, P., Young, S. C, Dumas, B. A., Phillips, J. R. 1976. Plant bugs in cotton: importance of infestation time, cotton injury, and significance of wild hosts near cotton. University of Arkansas. Report Series 227: 1-24.
University of Califomia. 1996. Integrated pest management for cotton in the western region of the United States. 3305: pp 164.
Wheeler, A. G. 2001. Biology of the plant bugs (Hemiptera: Miridae) pp 507.
Whitbey, R.M. 1999. Green bean extract-induced oviposition site preference in laboratory reared western tamished plant bug (Heteroptera: Miridae). Envir. Entomol. 28 (2): 201-204.
36 Williams, L. III., Tugwell, N. P. 2000. Histological description of tarnished plant bug (Heteroptera: Miridae) feeding on small cotton floral buds. J. Entomol. Sci. 35(2): 187-195.
Williams, M. R. 1999. Cotton insect losses. Proc Beltwide Cotton Conf National Cotton Council. Memphis - TN. 783-809.
Williams, M. R. 2000. Cotton insect losses. Proc Beltwide Cotton Conf National Cotton Council. Atlanta-GA. 884-913.
Williams, M. R. 2001. Cotton insect losses. Proc Beltwide Cotton Conf National Cotton Council. Nashville, TN. In press.
37 CHAPTER III
DIETARY EVALUATION OF CYSTEINE PROTEINASE INHIBITORS scN
AND E-64 AGAINST SOUTHERN CORN ROOTWORM
(Diabrotica umiecimpunclala howardi Barber; INFESTING PEANUT
Abstract
The southern corn rootworm (Diabrotica undecimpunctata howardi Barber^) can
be a serious pest of Texas grown peanut. Recent research in plant biology has found a
broad class of plant-produced compounds collectively known as proteinase inhibitors.
Lepidopterans pests are susceptible to serine, while Coleopterans are more susceptible to
cysteine. This research is part of an effort for a more sustainable management system for
subterranean insect pests in peanuts. A laboratory protocol was developed in order to test
soyacystatin N (scN) and E-64 (synthetic inhibitor) against southern corn rootworm using
an artificial diet fed to young larvae. The main objective of this study is to develop the dietary assay for incorporating the soybean proteinase inhibitor (scN) and a synthetic
inhibitor (E64) into the diet of the southern corn rootworm and to determine the affects of the inhibitors on larvae development and mortality. If proved to be reliable, the technique could be used to assess other inhibitors and lectins against other insect pests of peanut.
Resufts from insect bioassays showed that soyacystatin N significantly increased mortality of SCRW larvae and affected larval development as measured by larval weight gain. The control group had a 1.15 mg weight gain per replicate that accounted to a 0.29 mg greater difference when compared to the higher scN dose (1.50mg/ml). The
38 soycystatin inhibitor proved to be a good candidate for a host plant resistance in peanuts against Diabrotica undecimpunclala howardi Barber larvae. E-64 also demonstrated significant mortality to Diabrotica undecimpunctata howardi Barber larvae when fed in artificial diet and to significant reduced surviving weight gains. The methodology can be used for screening cysteine inhibitors against different insect-pests for host-plant resistance.
Introduction
Diabrotica species from the order Coleoptera, are one of the most economically
important insect genera of insects in North America. It includes some of the most
destructive and aggressive pests that can cause damage to agricultural crops. Larvae of
Diabrotica undecimpunctata howardi Barber, commonly know as the southern corn
rootworm (SCRW), cause significant damage by feeding on roots and pods of major
crops. This insect can cause significant economic losses to corn {Zea mays), sweet potato
{Lpomoea batatas) and peanuts {Arachis hypogaea). The adult beetle also known as
spotted cucumber beetle, can present some problems to crops by feeding on the aerial
parts and by transmitting several economically important viruses (Edmonds et al. 1996).
Southem com rootworm is the primary insect pest attacking peanuts in the Eastern
growing regions of the United States such as Virginia and North Carolina (Herbert, 1996)
and is considered to be a secondary pest in Southern areas such as South Carolina,
Georgia, Alabama and Texas.
39 Peanuts have become a significant crop and a major economic commodity grown in the state of Texas. The state currently ranks second to Georgia, producing 320,000 to
370.000 acres annually. In 1999, 457,407 tons of peanuts were harvested (Lemon et al.,
2001) generating $ 560 million for farmers and $ 1.8 billion of general revenue.
"Runner," "Virginia," "Spanish," and "Valencia" are the four market-type varieties grown in the state. Over 70% of all the peanuts grown in the state of Texas come from the western Texas growing region. The crop has become a major alternative for low market prices of cotton and has increased significantly to the overall base economy for the state.
Soil and aerial insecticide applications, cultural control, and the introduction of more tolerant varieties are common ways of controlling SCRW in com (Mayo, 1980;
Brust, 1991), However, insect control tools for peanuts pests are more limited due to minor crop status. Insecticides labeled today for peanuts are of the organophosphate and carbamate groups, and have been targeted for review by the 1996 Food Quality
Protection Act. These insecticides have been found to exceed the risk cup, and will probably be unavailable for use in the future. One alternative for a more sustainable control option for the SCRW is the development of host plant resistance in peanut lines.
Insect resistance does exist in peanuts and has been identified for both lepidopteran and coleopteran pests. Southern com rootworm resistant peanuts has been discovered and incorporated into peanut breeding lines (Campbell et al., 1977; Petka et al., 1998) and one commercial line is available in the State of North Carolina. However, is used in the southeastem part of the U.S. and not suitable as a production variety for western Texas.
40 NC-6 demonstrated antibiosis against southern corn rootworm, however is not grown
acceptable to Texas. Smith et al. (1979) discovered peanut lines resistant to lesser
cornstalk borer in Texas, however very little research is available on host plant resistance
of peanuts against southern corn rootworm in Texas.
This effort of host-plant resistance in peanut would only be available through
conventional breeding techniques because genome transformation in peanuts is not
acceptable due to direct consumption by humans. Recent studies have indicated that the
expression of serine proteinase inhibitors can provide an improved level of resistance to
attack from lepidopteran pests (Boulter et al., 1990), whereas coleopteran pests including
Diabrotica species are susceptible to cysteine proteinase inhibitors as their major
digestive proteolytic enzymes (Murdock et al., 1987). The well-developed enhancement
of lepidopteran pest resistance conferred by the expression of serine protease inhibitors
would imply that the presence of cysteine protease inhibitor could be effective on
developing resistant lines against coleopteran pest insects.
Research developed by Koiwa et al. (2000) has proven that a soybean cysteine
proteinase inhibitor cultured in laboratory was highly active against western com
rootworm {Diabrotica virgifera virgifera), substantially inhibiting growth and
development of the larvae by attenuating digestive proteolysis. Such inhibitor was named
scN, and is derivative from soyacystatin N. Similar biology of WCR and SCRW and research by Edmonds et al. (1996) and Orr et al. (1994) implies that the southern corn rootworm would also be affected by scN ingestion. Screening for CPI (scN) in different peanut lines is in progress as a separated phase of this research and conducted at the
41 Norman Bourlag Center for Southern Crop improvement, Texas A&M University. In
case of significant inhibition of the larvae development and growth of SCRW by scN in
laboratory bioassays, peanut lines that have CPI (scN) will be used in a traditional
breeding program.
E-64 (trans-epoxysuccinyl-leucyl-agmatine), a tripeptide from TRP-64 strain of
Aspergillus japonicus is known to be a powerful and specific cysteine proteinase inhibitor
(Kitch and Murdock, 1986) and proved to be affective in different coleopteran pests
(Murdock et al 1988). E-64 will be used as a calibrator factor for the methodology on scN
studies.
Literature Review
Insect bioassays
Edmonds et al. (1996) tested cysteine proteinase inhibitor oryzacystatin against southem com rootworm by dissolving the proteins to be tested in 1 ml of sterile deionised water and incorporating into artificial diet that was prepared as recommended by Branson et al. (1975). The inoculated diet was divided into 2ml aliquots in small (7-10 ml) glass pots. Seven day-old larvae were cleaned and weighed before being employed in the bioassay system. Three larvae were placed in each pot, and the surface of the diet was scratched with a sterile needle to facilitate larval burrowing. Twenty-five or fifty larvae were used per treatment. Covered feeding pots were incubated in total darkness at 25°C.
After 9 days, the surviving larvae were counted and weighed. The data was analyzed using Mann-Whitney U-Test, for survival frequencies and unpaired Student's T-test.
42 Edmonds et al. (1996) observed that Oryzacystatin I, expressed as a fully functional fusion in E. Coli, was found to strongly inhibit larval gut protease activity. This recombinant oryzacystatin, incorporated into artificial diet at concentrations of lOmM and above, caused significant decreases in larval survival and weight gain. E-64 was also shown to cause a significant antimetabolic in vivo effect. Edmonds et al. (1996) results demonstrate the strong potential for cysteine proteinase inhibitors, such as oryzacystatin, as tools for exploitation in the control of the southern corn rootworm.
In 1994, Orr et al. tested mulcystain from Potato Tubers against southem com rootworm and western corn rootworm by dissolving and diluting the proteins in sterile water and applying (0.03 ml) to the surface of 0.25ml of artificial diet. Twenty four-well microtiter plates were used and the water was dried out before larvae infestation. The reduced use of diet resulted in a very thin (approx. 1.0 mm) layer of diet, which reduced the potential for larvae to burrow in the diet and avoid contact with the active proteins.
Eggs of both species were purchased from French Agricultural Research (Lamberton,
MN). Individual wells were infested with neonate larvae or a pre-weighed second instar southem or westem corn rootworm reared on com seedlings. Plates were incubated at
26°C in sterilized, sealed plastic containers for 6 days. Larvae were weighed after 1 week. Orr et al. (1994) observed that second instar western corn rootworm larvae had a higher proportional proteinase activity, which was attributable to cysteine proteinase
(92%) than southern corn rootworm larvae (75%). The presence of PMC in the diet caused a dose-dependent inhibition of growth in neonate southern com rootworm and second instar southern corn rootworm larvae. Neonate southern corn rootworm and
43 second instar western corn rootworm had similar sensitivity to the inhibitor (50%) inhibitor at 25 - 43.8 )Lig/cm"). In contrast to southern corn rootworm larvae, western corn rootworm growth was completely halted by PMC. Long-term exposure of SCRW larvae to PMC suggested that the larvae become less sensitive to the inhibitor during development. Hen egg cystatin (HEC) and T-PMC were unable to inhibit growth of either species but in SCRW. co-feeding of potato carboxypeptidase inhibitor (PCI) with T-PMC caused growth inhibition. These observations indicated that multicystatins such as PMC could be effective cystatins for use in controlling larvae of Diabrotica species in transgenic plants.
In 1990, Czapla and Lang conducted insect assays studying several lectins with different carbohydrate-specific binding sites in artificial diet to determine their anti- metabolic activity against neonate european corn borer {Osistinia nubialis Hubner), and southem com rootworm {Diabrotica undecimpunctata howardi Barber). Both species are major pests of corn. Genes of lectins with anti-insect activity are of great interest for enhancing insect resistance through plant transformation. Czapla and Lang's (1990) study allowed for dividing each culture tray into four treatment groups, each consisted of one row with six cells. Three lectins were screened per tray; each tray included a control treatment. Each cell was infested with two (2) neonate larvae and 12 insects per treatment maximum and in some only had three cells used, resuhing in a total of six insects per treatment. A Mylar film was affixed to the top of each tray to prevent escape and was punctured 12 times with insect pins to provide ventilation. Stoneville® medium was prepared as described except that only 90% of the original water was used. Diet
44 containing 0.5%) (5 mg/g) lectins was prepared with approximately 8.1g of this medium and adding 45mg of lectins (contained in 0.9 ml 0.1 MPBS buffer). The control treatment consisted of 0.9 ml PBS buffer added to 8. Ig medium. The medium was poured into six cells, infested, and covered. All lectins were tested using the topical bioassay. Larval weights and mortality were recorded at 7 days and compared with weights and mortality of the control group. Czapla and Lang (1990), concluded that several lectins where lethal to the european corn borer and to the southern corn rootworm.
Murdock et al. (1988) studied the effects of E-64 on cowpea weevil growth and development by preparing artificial seeds as presented by Shade et al. (1986) that consisted of cowpea flour thoroughly and evenly wetted with water or with an aqueous solution of E-64. The resulting paste is injected into a Teflon mold, frozen, and then lyophilized for 24 hours. Seeds for all treatments were divided into two equal lots and placed into two 0.5-liter glass oviposition jars for infestation. The jars contained 200 newly emerged adults (approx. 100 females) that were allowed to oviposit on the artificial seeds for 24 hours at 25°C and 60%) relative humidity. Five characteristics of cowpea weevil biology were evaluated for each treatment by measuring adult longevity, fecundity, duration of larval + pupal + unemerged aduh stages and percentage of hatched eggs not producing adufts. Murdock et al. (1988) observed that seeds at low levels (0.01-
0.25% by weight) prolonged developmental time and increased mortality of the larval cowpea weevil. The fecundity of females emerging from the artificial seeds was significantiy decreased by E-64 concentrations of 0.06% and higher.
45 Koiwa et al. (2000) conducted bioassays of western corn rootworm (WCR) using plastic trays (C-D International, NJ, USA) with 32 sets (columns) and 4 wells/set. Each 4 well set was a treatment block in a randomized complete block design. The diet used was a modification of Rose and McCabe (1973) and purchased from Bioserv, NJ, USA. The compounds were mixed with the diet prior to loading in each well. The surface of the diet was scratched with a sterile micro-spatula to facilitate feeding. Eggs of WCR were acquired from French Agricultural Research, (Lamberton, MN) and surface-sterilized prior to incubation at 26°C for 2 days. Three neonates were transferred to each well. The mean weight and mortality of each treatment after 12 days post-infestation determined westem corn rootworm response to scN. Regression was used to examine the dose response curve. The work here followed that of Koiwa et al. (2000) who established that soybean cysteine proteinase inhibitor N (soyacystatin N, scN) substantially inhibits development and growth of western corn rootworm.
Rearing southern corn rootworm
Rearing Diabrotica species in laboratories in the Unites States has been done for decades because of the economic importance of these genera to agricultural crops.
Rearing with artificial diets that substitute natural food sources has many advantages, including less labor and time for conducting screening trials through the diet. Bacterial contamination was the major problem with rearing rootworms on artificial diets until
Marrone et al. (1985) improved published procedures. Branson et al. (1975), Rose and
Mccabe (1973), Guss and Krysan (1973), Sutter et al. (1971), and Skelton and Hunter
46 (1970) are some of the attempts to improve laboratory rearing of SCRW before Marrone et al. (1985). Marrone discovered that the main source of contamination of the artificial
diet was associated with eggs. He then developed procedures and tested different
sterilants to sterilize the eggs and improve viability. After modifying published
procedures, 30%) of larvae reared on the improved diet reached 18mg (pupa weight) in 10
days, with < 8%o bacterial contamination of the diet. Although these improvements helped
improved rearing procedures, bacterial contamination can be a major source of reducing
the rearing capability of rootworms, including SCRW.
The artificial diet developed by Rose and McCabe (1973) is widely used and
commercialized for SCRW rearing. Rose and McCabe (1973) developed diets for both
the adult and larval stage of southern corn rootworm. The diet developed for the adult
stage is composed of vanderzant-adkisson special wheat germ for insects (40.0 g), agar
(8.0 g), 10%) KOH (5.0 ml), sliced sweet-potato (40.0 g), alfalfa meal (10.0 g), regular
com oil (3.0 ml), vanderzant modification vitamin mixture for insect diets (2.0 g), active
dry yeast (2.0 g), ascorbic acid (0.5 g), and distilled water (390.0 ml). This diet proved to
be very efficient and used to rear an adult colony for over a year. Rose and McCabe
(1973) also developed an artificial diet for the larval stage of the southem com rootworm
that was composed of vanderzant-adkisson special wheat germ (25.0 g), agar (3.1 g),
distilled water (215.0 ml), methyl paraben (0.20 g), sorbic acid (.10 g), 10% KOH (2.5
ml), formalin (0.15 ml), regular corn oil (1.0 ml), streptomycin (0.030 g), tetracycline
(0.030 g), penicillin "G" (0.075 g), bacitracin (0.030 g) and vanderzant modification
vitamin mixture for insect diets (2.0 g). Over 40% of the newly hatched larvae placed on
47 the artificial diet successfully completed larval development to pupae in 20 to 21 days.
This diet described by Rose and McCabe (1973) is widely used for southern corn rootworm rearing in laboratory and listed in catalogs of insect diet supply companies as their base reference for the commercialized formula.
Objectives
The major objective of this study is to develop a dietary assay for incorporating the soybean proteinase inhibitor (Soyacystatin N, scN) and E-64 into the artificial diet of southern corn rootworm and to determine the affects of the inhibitors on larval growth and development as a potential for host plant-resistance. The methodology, if proved efficient, can be used to test other cysteine proteinase inhibitors and lectins against a broad spectrum of insect pests. If soyacystatin N (scN) proves to be significantly active against southern corn rootworm, cysteine proteinase inhibitor (scN) detection in wild peanut lines at Texas A&M University, aiming a traditional breeding program to develop peanut varieties resistant or more tolerant to SCRW feeding will be done as a continuation of this research.
Materials and Methodologv
Purification of recombinant soyacystatin N, and E-64 acquisition
Purification of scN was done in the laboratory of Dr. Keyan Zhu-Salzman at the
Normal Bourlag Technology Center, Texas A&M University, College Station, Texas.
Bacteria expressed recombinant scN was obtained following Koiwa et al. (2000).
48 Bacterial strain BL21 harboring a construct containing the cDNA encoding scN was
grown in laboratory at 37°C. Recombinant protein production was then induced with
isopropyl-D-thiogalactoside overnight at 18"C, Cells were disrupted by sonication, and
recombinant proteins were purified via a Ni2 + chelate affinity column (Amersham
Pharmacia Biotech). Purified proteins were then dialyzed against distilled water and
lyophilized. The E-64, [trans-Epoxysuccibyl-l-Leucylamido-(4-Guanidino)Butane]
synthetic inliibitor was acquired from Sigma® Chemical Company, product number E-
3132.
Southern corn rootworm feeding bioassay
Eggs of southern corn rootworm were acquired from French Agricultural
Research (Lamberton, MN) and incorporated to humidified sterile filter papers for
hatching at 27°C. Hatching took from 9 to 12 days where f' instar larvae were
transferred to artificial diet when enough larvae were harvested to conduct the
experiment. Artificial diet for rearing SCRW larvae (Rose and McCabe, 1973) was
purchased from Bio-serv® (Frenchtown, NJ) and prepared following company
instructions. Well plates (4 x6 wells on 13 x 16 cm, cells culture clusters. Coming
Incorporated, Coming, NY) were used instead off diet cups due to the need for using
smaller quantities of diet per well (1ml/ well), without allowing the diet to dry. After
preparation of the diet, antibiotics were added (ampicillin at 50mg/ml of diet and
tetracycline at 15mg/ml of diet) to slow down bacteria development in the diet. Four
experiments were completed at Texas Tech University, entomology laboratories. Two
49 experiments for each inhibitor, held at the same conditions, 27° C, 40% R.H., and complete dark were completed. Southern corn rootworm larvae were reared for seven days on regular artificial diet before they were transferred into diet inoculated with
inhibitors. Assays were arranged as completely randomized designs, with each treatment
replicated four times. The inhibitors we topically applied using a solution prepared with
sterile, distilled water. Diet was allowed to cool and become solid before the application
of inhibitors. The surface of the diet was punctured with a sterile needle too allow
penetration of the compounds into the diet and facilitate feeding by larvae. Soyacystatin
N treatment arrangements included a control, 0.25, 0.75 and 1.5mg of scN/ml of diet.
Four larvae were placed in each well and replicated four times, each well consisted of one
replication. Parafilm® was used to cover well plates with small holes inserted for air
circulation. Feeding bioassays were held at 27°C for 7 days. E-64 experiments followed
the same methodology as the scN trials and doses of 0.5, 1.0 and 2.0 mg of E-64/ml were
evaluated.
Larval mortality was assessed daily and larval weight gain was measured by
weighing the larvae at the beginning and at the end of the experiment. The data was
analyzed with PROC ANOVA, with means separated by Fisher's Protected LSD (SAS
Institute 1989). Treatment affects were measured by correlating treatment doses with
larval mortality and by regressing weight gain of surviving larvae with dosage of
inhibitor
50 Results
The insect-bioassay system used one-week-old southern corn rootworm larvae
and measured mortality and weight gain per treatment over a seven-day period. No
cannibalism was observed in these assays. Results include the combined values from two
separate experiments for each inhibitor (scN and E-64) (Table 3.1 and Figures 3.1 - 3.6).
Soyacystatin N (scN) was incorporated into the diet at concentrations of 0.25,
0.75, 1.50 mg/ml of diet and a control set was present. Results showed that larval
mortality significantiy increased {P < 0.05) by scN ingestion in all treatments (F= 11.10,
df= 31, and P=<.0001) (Table 3.1). Soyacystatin N at 1.50 mg/ml (Higher concentration)
had 71.88%) final mortality, more than double when compared to the control group and
significantiy different when compared to both the control group and scN at 0.25mg/ml.
Soyacystatin N at 0.75 mg/ml resulted in 50.0% final mortality and was significantiy
greater than the control group and statistically similar to 0.25 and 1.50 mg/ml
concentration treatments. Finally, scN at 0.25 mg/ml was significantly greater than the
control group with 37.5%o of final mortality. All treatment doses increased southem com
rootworm larvae mortality and were significantly greater than the control group. Larval
development measured by weight gain showed no differences after 4 days of feeding on
scN inhibitor (Figure 3.1). However, after 7 days, all treatments showed reduced weight
gain when compared to the control group (Figure 3.2). Soyacystatin N incorporated into the diet of SCRW resulted in a dose response and a significant reduction in weight gain of surviving larvae after seven days. There was a significant linear relationship between
51 doses of scN and weight gain of surviving larvae as showed by linear regression analysis
(F= 7.09, df= 31, P=0.0124, R^ = 0.223, Y= 0.97 - 0.57x) (Figure 3.5).
E-64 was incorporated into the diet at concentrations of 0.50, 1.00 and 2.00
mg/ml of diet. Results showed that larval mortality significantly increased {P < 0.05) by
E-64 ingestion in the higher dose treatment (2.00 mg/ml), whereas both 0.50 mg/ml and
1.00 mg/ml were statistically similar to the control and the higher dose treatment (F=
3.77, df=31, P= 0.0218) (Table 3.1). E-64 at 2.00 mg/ml was significantiy greater than
the control group and resulted in 53.13% final mortality, over twice that of the control
group. E-64 at 0.50 mg/ml and 1.00 mg/ml were significantly similar to the control and
the higher treatment and resulted in 31.25% and 40.63% final mortality, respectively.
Larval weight gain was not different after 4 days of feeding on E-64 (Figure 3.3).
However, after seven days, all treatments showed reduced weight gain when compared to
the control treatment (Figure 3.4). E-64 incorporated into the diet of SCRW resulted in a
dose response and a significant reduction in weight gain of surviving larvae after seven
days feeding on diet infested with inhibitor. There was a significant linear relationship
between doses of E-64 and weight gain of surviving larvae as showed by regression
analysis (F= 4.39, df= 31, P= 0.0447, R^ = 0.127, Y= 0.59 -0.37x) (Figure 3.6).
Discussions
Soyacystatin N (scN) significantly increased mortality of southern corn rootworm larvae feeding on infested diet. The higher dose (1.50 mg/ml) accounted for 71.9% of final mortality, more than double than the control group. All treatments were significantly
52 greater than the control group, which averaged 31.3% of final mortality. Cannibalism did
not occur on insect assays and results represent the combined values of two separated
experiments for each compound (scN and E-64). Larval development assessed by
measuring weight gain of surviving larvae showed no significant differences after 4 days
of feeding however, differences were significant after 7 days. The control group had a
1.15 mg weight gain per replicate that accounted to a 0.29 mg greater difference when
compared to the higher scN dose (1.50mg/ml). Treatment dose at 0.75 mg/ml had 0.61mg
of weight gain that accounted for 0.54 mg greater than the control group while the 0.25
mg/ml treatment had a 0.89 mg of weight gain, 0.26 mg greater than the control
treatment. The higher treatment if scN (1.50 mg/ml) had a higher weight gain than the
0.75 mg/ml dose. Variation of individual larval weights and topical application of the
treatments might have contributed to this.
E-64 incorporated on artificial diet significantly increased southern com
rootworm larval mortality. The higher treatment dose was significantly different from the
treatment group and statistically similar to the others treatment doses (0.50 mg/ml and
1.00 mg/ml). The 2.00 mg/ml treatment dose had 53.13% of final mortality that was
doubled the value from the control group that had 25% of final mortality. Intermediate
treatment groups (0.50 mg/ml and 1.00 mg/ml) were statistically similar to the control
and the higher dose treatment however, numerical difference was observed and both
treatments were greater than the control group and smaller than the higher dose treatment
(31.25%) and 40.63% mortality, respectively). Larval development assessed by weight gain of surviving larvae demonstiated no statistical difference after 4 days of feeding,
53 however significant differences were observed after 7 days. Each treatment doses had
weight gains less than half when compared to the control group. 2.00 mg/ml E-64
treatment dose had 0.26 mg, 1.00 mg/ml dose had 0.25 mg, and 0.50 mg/ml had 0.40 mg
of weight gain. A significant linear relationship between existed between E-64 dosage
incorporated into SCRW diet and weight gain of surviving larvae (F= 4.39, df= 31, P=
0.0447, R^ = 0.127, Y= 0.59 -0.37x) (Figure 3.6).
Conclusions
Soyacystatin N (scN), cysteine proteinase inhibitor, proved to significantly affect
southem corn rootworm larvae development and increase mortality in laboratory
bioassays. The inhibitor can be considered a good candidate for a host-plant resistance program against Diabrotica undecimpunctata howardi Barber on peanuts. E-64, synthetic
inhibitor caused significant increase in southern corn rootworm mortality in laboratory bioassays and affected larval development assessed by weight gain. There was a
significant linear relationship between E-64 dose and weight gain of surviving larvae and a curve-response to dose. E-64 demonstrated that the methodology is reliable and can be used to assess other protein inhibitors against a broad spectrum of insect pests.
54 Table 3.1. Effect of scN and E-64 on mortality of one-week-old southern corn rootworm larvae reared on artificial diet.
Soyacystatin N E-64
Treatments Final Mortality (%) Treatments Final Mortality (%)
Control 31.3 + 4.09a Control 25.0±8.18a
0.25 mg/ml 37.5 +6.68 b 0.50 mg/ml 31.25 ±4.09ab
0.75 mg/ml 50.0 ±4.72 be 1.00 mg/ml 40.63 ±4.57ab
1.50 mg/ml 71.9 + 5.66C 2.00 mg/ml 53.13±7.38b Column means followed by the same small letter are not significantly different, while those followed by a different letter are significantly different by Fisher's LSD, {P < 0.05).
55 1.25
1.00
0.75
t 0.50 c CS 4> 0.25
0.00 Control scN 0.25mg/ml scN 0.75mg/ml scN 1.50mg/ml Treatments
Figure 3.1. Weight gain (mg + SEM) of one-week-old southern coon rootworm larvae after 4 days of feeding on artificial diet containing scN.
56 Control scN 0.25mg/ml scN 0.75mg/ml scN 1.50mg/ml Treatments
Figure 3.2. Weight gain (mg + SEM) of one-week-old southern coon rootworm larvae after 7 days of feeding on artificial diet containing scN (F= 7.09, df= 31, P=0.0124, Y= 0.97-0.57X).
57 0.35
0.00 Control E64 0.5mg/ml E64 l.Omg/ml E64 2.0mg/ml Treatments Figure 3.3. Weight gain (mg + SEM) of one-week-old southern coon rootworm larvae after 4 days of feeding on artificial diet containing E-64.
58 1.15
0.00 Control E64 0.5mg/ml E64 l.Omg/ml E64 2.0mg/ml Treatments
Figure 3.4. Weight gain (mg + SEM) of one-week-old southern coon rootworm larvae after 7 days of feeding on artificial diet containing E-64 (F= 4.39, df= 31, P= 0.0447, Y= 0.59-0.37X).
59 2.00
1.60-
DO B 1.20
C3
•SO.80
0.40
0.00 0.0 0.25 0.75 1.50 Treatments (mg/ml of scN)
Figure 3.5. Linear relationship of SCRW larval weight gain from two individual trials after seven days of assays on scN. (F= 7.09, df= 31, P=0.0124, R' = 0.223, Y= 0.97 - 0.57x).
60 1.20
0.00 0.00 0.50 1.00 2.00 Treatments (mg/ml of E-64)
Figure 3.6. Linear relationship of SCRW larval weight gain from two individual trials after seven days of assays on E-64. (F= 4.39, df= 31, P=0.0447, R^ = 0.127, Y= 0.59 - 0.37x).
61 References
Bouher, D., Edwards, G. A., Gatehouse, A. M. R., Gatehouse, J. A., and Hilder, V. A. 1990. Additive protective effects of different plant-derived insect resistance genes in transgenic tobacco plants. Crop Protection. 9: 351-354.
Branson, T. F., Guss, P. L., Krysan, J. L., and Sutter, G. R. 1975. Corn rootworms: laboratory rearing and manipulation. Technical Bull. U.S Dept. Agric, ARS-NC- 28: pp 18.
Brust, G. E. 1991. Soil moisture, no-tillage and predator effects on southern corn rootworm survival in peanut agroecosystems. Entomologia Experimentalis et Applicata. 58: 109-121.
Campbell, W. V., Wynne, J. C, Emery, D. A., and Mozingo, R. W. 1977. Registration of NC-6 peanuts. Crop Sci. 17:346.
Czapla, T. H. and Lang, B. A. 1990. Effect of plant lectins on the larval development of European corn borer (Lepidoptera: Pyralidae) and southern com rootworm (Coleoptera: chrysomelidae). J. Econ. Entomol. 83: 2480-2485.
Edmonds, H. S., Gatehouse, L. N., Hilder, V. A. and Gatehouse, J. A. 1996. The inhibitory effects of the cysteine protease inhibitor, oryzacystatin, on digestive proteases and on larval survival and development of the southern corn rootworm {Diabrotica undecimpunctata howardi). Entomologia Experimenalis et Applicata 78: 83-94.
Guss, P. L. and Krysan, J. L. 1973. Maintenance of the southem corn rootworm on a dry diet. J. Econ. Entomol. 66: 352-353.
Herbert, D. A. Jr., Ang, B. N., and Hodges, R. L., 1996. Attractants for adult southern corn rootworm (Coleoptera: Chrysomelidae) monitoring in peanut fields and relationship of trap catch to pod damage. J. Econ. Entomo. 89(2): 515-524.
Kitch, L., and Murdock, L. L. 1986. Partial characterization of a major midgut thiol proteinase from larvae of Callosobruchus maculates (F.). Archiv. Insect Bio- chem. Physiol. 3: 561-575.
Koiwa, H., Shade, R. E. Zhu-Salzman, K., D'unzo, M. P. Murdock, L. L. Bressan, R. A., and Hasegawa, P. M. 2000. A plant defense cystatin (soyacystatin) targets cathesphin L-like digestive cysteine proteinase (PrCALS) in the larval midgut of western corn rootworm. FEBS letter. 471: 67-70.
62 Lemon, R. G., Lee. T. A., Black, M., Grichar, W. J., Baughman, T., Dotray. P., Trostie, C, McFarland, M., Bauman, P., Crumley, C, Russell, J. S., Norman, G. 2001. Texas peanut production guide, Texas Agric. Extension Service. Texas A&M University. B-1514. 04-01. pp 84.
Marrone, P. G., Ferri, F. D. and Mosley, T. R. 1985. Improvements in laboratory rearing of the southern corn rootworm, Diabrotica undecimpunctata howardi Barber (Coleoptera: Chrysomelidae), on an artificial diet and corn. J. Econ. Entomol. 78: 290-193.
Mayo, Z. B., 1980. Influence of planting dates on the efficacy of soil insecticides applied to control lai'vae of the western and northern corn rootworm. J. Econ. Entomol. 73:211-212.
Murdock, L. L., Brookhart, G., Dunn, P. E., Foard, D. E., Kelley, S., Kitch, L., Shade, R., Schukle, R. H., and Wolfson, j. L. 1987. Cysteine digestive proteinase in Coleoptera. Comp. Biochem. Physiol. 87B: 783-787.
Murdock, L. L., Shade, R. E. and Pomeroy, M. A. 1988. Effects of E-64, a cysteine proteinase inhibitor, on cowpea weevil growth, development and fecundity. Environ. Entomol. 17: 467-469.
Orr, L. G., Strickland, J. A. and Walsh, T. A. 1994. Inhibition of Diabrotica larval growth by a multicystatin from potato tubers. J. Insect Physiol. 40: 893-900.
Petka, W. J. Jr., Herbert, D. A. Jr., and CoffeU, T. A. 1998. Peanut resistance to southern corn rootworm. U.S. Dept. Agric. ARS. Technical Bulletin, pp 3.
ose, R. I., and McCabe, J. M. 1973. Laboratory rearing techniques for southem com rootworm. J. Encon. Entomol. 66: 398-400.
Shade, R. E., Murdock, L. L., Foard, D. E., and Pomeroy, M. A. 1986. An artificial seed system for bioassay of cowpea weevil growth and development. Environ. Entomol. 15: 1286-1291.
Skelton, T. E. and Hunter, P. E., 1970. Laboratory rearing and reproduction of the spotted cucumber beetie. J. Econ. Entomo. 63: 756-757.
Smith, J. W. Jr., and Holloway, R. L., 1979, Lesser cornstalk borer larval density and damage to peanuts. J. Econ. Entomol. 72: 535-537.
Smith, J. W. Jr., Lazaro, P., and Smith, O. D., 1980. Greenhouse screening peanut germ plasm for resistance to the lesser cornstalk borer. Peanut Science 7:68-71.
63 Sutter, G. R., Krysan, J. L., and Guss, P. L. 1971. Rearing the southern corn rootworm on artificial diet. {Diabrotica undecimpunclala howardi Barber). J. Econ. Entomol. 64: 65-67.
64 APPENDIX A
CHAPTER II SAS PROGRAMS AND OUTPUTS FOR MEAN SEPARATION
TESTS AND RAW DATA TABLES, 2001 AND 2002 SEASONS
65 6" Node abscission 2001
1 OPTIONS LS=80; 2 DATA ONE; 3 INPUT TRT REP Y; 4 CARDS;
NOTE: The data set WORK.ONE has 12 observations and 3 variables. NOTE: DATA statement used: real time 0.55 seconds cpu time 0.16 seconds
17 PROC ANOVA; 18 CLASS TRT REP; 19 MODEL Y=TRT REP; 20 MEANS TRT/LSD; 21 RUN;
OPTIONS LS==80 ; DATA ONE; INPUT TRT REP Y; CARDS; 1 1 10 1 2 20 1 3 10 1 4 10 2 1 20 2 2 30 2 3 20 2 4 20 3 1 30 3 2 30 3 3 20 3 4 30 PROC ANOVA; CLASS TRT REP; MODEL Y==TR T REP MEANS TRT/LSD; RUN;
The SAS System 6
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
TRT 3 12 3
REP 4 12 3 4
66 Number of observations 12
The SAS System 7
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y Sum of Source DF Squares Mean Square F Value Pr>F
Model 5 625.0000000 125.0000000 11.25 0,0053
Error 6 66.6666667 11.1111111 Corrected Total II 691.6666667
R-Square CoeffVar Root MSE Y Mean
0.903614 16.00000 3.333333 20.83333
Source DF Anova SS Mean Square F Value Pr > F
TRT 2 466.6666667 233.3333333 21.00 0.0020 REP 3 158.3333333 52.7777778 4.75 0.0501
The SAS System 8
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0.05 Error Degrees of Freedom 6 Error Mean Square 11.11111 Critical Value of t 2.44691 Least Significant Difference 5.7674
67 Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 27.500 4 3 A A 22.500 4 2
B 12.500 4 I
6"'Node abscission 2002
22 OPTIONS LS=80;
NOTE: There were 12 observations read from the data set WORK.ONE. NOTE: PROCEDURE ANOVA used: realtime 2:41.20 cpu time 0.43 seconds
23 DATA ONE; 24 INPUT TRT REP Y; 25 CARDS;
NOTE: The data set WORK.ONE has 12 observations and 3 variables. NOTE: DATA statement used: real time 0.04 seconds cpu time 0.03 seconds
38 PROC ANOVA; 39 CLASS TRT REP; 40 MODEL Y=TRT REP 41 MEANS TRT/LSD; 42 RUN;
OPTIONS LS==80 ; DATA ONE; INPUT TRT REP Y; CARDS; 1 1 20 1 2 20 1 3 20 1 4 30 2 1 30 2 2 30 2 3 20 2 4 30 3 I 30 3 2 40 3 3 30
68 3 4 30 PROC ANOVA; CLASS TRT REP; MODEL Y=TRT REP; MEANS TRT/LSD; RUN;
The SAS System 9
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
TRT 3 12 3
REP 4 12 3 4
Number of observations 12
The SAS System 10
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y Sum of Source , DF Squares Mean Square F Value Pr > F
Model 5 291.6666667 58.3333333 2,63 0.1358
Error 6 133.3333333 22.2222222 Corrected Total !1 425.0000000
R-Square CoeffVar Root MSE Y Mean
0.686275 17.14198 4.714045 27.50000
Source DF Anova SS Mean Square F Value Pr > F
TRT 2 200.0000000 100.0000000 4.50 0.0640
69 REP 3 91.6666667 30,5555556 1,38 0.3376
The SAS System 11
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0.05 Error Degrees of Freedom 6 Error Mean Square 22.22222 Critical Value of t 2,44691 Least Significant Difference 8,1564
Means with the same letter are not significantly different,
t Grouping Mean N TRT
A 32.500 4 3 A B A 27.500 4 2 B B 22.500 4 1
6* Node yield 2001
43 OPTIONS LS=80;
NOTE: There were 12 observations read from the data set WORK.ONE. NOTE: PROCEDURE ANOVA used: realtime 2:33,14 cpu time 0.05 seconds
44 DATA ONE; 45 INPUT BLK TRT Y; 46 CARDS;
NOTE: The data set WORK.ONE has 12 observations and 3 variables. NOTE: DATA statement used: real time 0.04 seconds cpu time 0.04 seconds
59 PROC ANOVA; 60 CLASS BLK TRT;
70 61 MODEL Y=BLK TRT; 62 MEANS TRT/LSD; 63 RUN;
OPTIONS LS=80; DATA ONE; INPUT BLK TRT Y; CARDS; 1 1 36,4 1 -I 36,0 3 20.2 I 36,7 1 1 17.8 2 3 31.0 3 1 31.4 3 2 36.0 3 3 41.0 4 1 40.7 4 2 43.1 4 3 38.1 PROC ANOVA; CLASS BLK TRT; MODEL Y=BLK TRT; MEANS TRT/LSD; RUN;
The SAS System 12
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
BLK 4 12 3 4
TRT 3 12 3
Number of observations 12
The SAS System 13 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y
71 Sum of Source DF Squares Mean Square F Value Pr > F
Model 5 297.5183333 59,5036667 0,93 0,5237
Error 6 385,6683333 64,2780556 Corrected Total II 683,1866667
R-Square CoeffVar Root MSE Y Mean
0.435486 23.55737 8,017360 34,03333
Source DF Anova SS Mean Square F Value Pr > F
BLK 3 265,8466667 88,6155556 1.38 0.3367 TRT 2 31.6716667 15.8358333 0.25 0,7892
The SAS System 14
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0,05 Error Degrees of Freedom 6 Error Mean Square 64.27806 Critical Value of t 2.44691 Least Significant Difference 13.872
Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 36.300 4 1 A A 33.225 4 2 A A 32.575 4 3
6"' Node yield 2002
43 OPTIONS LS=80;
NOTE: There were 12 observations read fi-omth e data set WORK.ONE. NOTE: PROCEDURE ANOVA used:
72 realtime 2:33,14 cpu time 0,05 seconds
44 DATA ONE; 45 INPUT BLK TRT Y; 46 CARDS;
NOTE: The data set WORK,ONE has 12 observations and 3 variables. NOTE: DATA statement used: real time 0.04 seconds cpu time 0.04 seconds
59 PROC ANOVA; 60 CLASS BLK TRT; 61 MODEL Y=BLK TRT; 62 MEANS TRT/LSD; 63 RUN;
OPTIONS LS^=80 ; DATA ONE; INPUT BLK TRT Y; CARDS; 1 1 34,03 I 2 35.16 1 3 23.52 2 1 33.77 2 2 19.07 2 3 30.97 3 1 31.62 3 2 31.71 3 3 30.88 4 I 24.99 4 2 33.57 4 3 25.47 PROC ANOVA; CLASS BLK TRT; MODEL Y==BL K TRT; MEANS TRT/LSD; RUN;
The SAS System 22 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
73 BLK 4 12 3 4
TRT 3 12 3
Number of observations 12
The SAS System 23
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y Sum of Source DF Squares Mean Square F Value Pr>F
Model 5 54.3305833 10,8661167 0,29 0.9043
Error 6 227.5456833 37.9242806
Corrected Total II 281.8762667
R-Square CoeffVar Root MSE Y Mean
0,192746 20,83077 6,158269 29,56333
Source DF Anova SS Mean Square F Value Pr > F
BLK 3 30,72026667 10.24008889 0.27 0.8451 TRT 2 23.61031667 11.80515833 0.31 0.7437 The SAS System 24
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0.05 Error Degrees of Freedom 6 Error Mean Square 37.92428 Critical Value oft 2.44691 Least Significant Difference 10.655
74 Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 31.103 4 1 A A 29.878 4 2 A A 27.710 4 3
6* Node Laboratory 2001
86 OPTIONS LS=80;
NOTE: There were 12 observations read from the data set WORK.ONE. NOTE: PROCEDURE GLM used: realtime 1:33.51 cpu time 0.07 seconds
87 DATA ONE; 88 INPUT TRT Y; 89 CARDS;
NOTE: The data set WORK.ONE has 30 observations and 2 variables. NOTE: DATA statement used: real time 0.04 seconds cpu time 0.02 seconds
120 PROC ANOVA; 121 CLASS TRT; 122 MODEL Y=TRT; 123 MEANS TRT/LSD; 124 RUN;
OPTIONS LS=80; DATA ONE; INPUT TRT Y; CARDS; 0 15 0 0 0 0 10 0 0 0 2 7.5 2 25 2 15
75 2 15 2 10 2 20 2 35 2 33 2 37.5 2 70 3 40 3 40 3 30 3 40 3 30 3 35 3 17.5 3 70 3 25 3 20 PROC ANOVA; CLASS TRT; MODEL Y=TRT; MEANS TRT/LSD; RUN;
The SAS System 25
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
TRT 3 12 3
Number of observations 30
The SAS System 26
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y Sum of Source DF Squares Mean Square F Value Pr > F
Model 2 5645.85000 2822,92500 14.35 <.0001
Error 27 5312.22500 196.74907
Corrected Total 29 10958.07500
R-Square CoeffVar Root MSE Y Mean
76 0.515223 65.69896 14.02673 21,35000
Source DF Anova SS Mean Square F Value Pr > F
TRT 2 5645,850000 2822,925000 14,35 <.000I
The SAS System 27
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0.05 Error Degrees of Freedom 27 Error Mean Square 196.7491 Critical Value of t 2.05183 Least Significant Difference 12.871
Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 34.750 10 3 A
A 26.800 10 2
B 2.500 10 1
6* Node Laboratory 2002
125 OPTIONS LS=80; NOTE: There were 30 observations read from the data set WORK.ONE. NOTE: PROCEDURE ANOVA used: real time 54.37 seconds cpu time 0.07 seconds
126 DATA ONE; 127 INPUT TRT Y; 128 CARDS;
NOTE: The data set WORK.ONE has 30 observations and 2 variables. NOTE: DATA statement used: real time 0.05 seconds cpu time 0.03 seconds
159 PROC ANOVA; 160 CLASS TRT;
77 161 MODEL Y=TRT; 162 MEANS TRT/LSD; 163 RUN;
OPTIONS LS=8(); DATA ONE; INPUT TRT Y; CARDS; 0 0 0 0 0 0 0 0 0 0 5 25 15
2 45 2 0 2 5 2 5 2 20 2 5 3 60 3 5 3 25 3 5 3 20 3 15 3 30 3 35 3 10 3 65 PROC ANOVA; CLASS TRT; MODEL Y=TRT; MEANS TRT/LSD; RUN;
The SAS System 28 14:53 Tuesday, December 17, 2002
78 The ANOVA Procedure
Class Level Information Class Levels Values
TRT 3 12 3
Number of observations 30
The SAS System 29
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y
Sum of
Source DF Squares Mean Square F Value Pr > F
Model 2 3646.666667 1823.333333 8,53 0,0013
Error 27 5770,000000 213,703704
Corrected Total 29 9416,666667
R-Square CoeffVar Root MSE Y Mean
0,387257 109.6396 14.61861 13.33333
Source DF Anova SS Mean Square F Value Pr > F
TRT 2 3646,666667 1823,333333 8,53 0.0013
The SAS System 30
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate. Alpha 0.05 Error Degrees of Freedom 27 Error Mean Square 213.7037 Critical Value of t 2.05183 Least Significant Difference 13.414
79 Means with the same letter are not significantly different,
t Grouping Mean N TRT
A 27,000 10 3
B 13,000 10 2 B B 0.000 10 I
80 Table A.l. Lint and seed weight combined (gms) for cotton bolls, first position, 6'^ node infested for 24h with Lygus in the field during the 2001 season. Treatments Replications Boll Control L hesperus L elisus 1 5.30 4.12 2.65 - 3.57 2.49 0.00 3 3.48 2.58 0.00 „ , 4 3.62 5.08 3.66 '^^P ' 5 0.00 4.36 4.03 6 4.47 0.00 4.90 7 4.08 4.86 0.00 8 4.63 3.29 4.93 9 4.07 3.60 0.00 \0 3J9 5,66 0.00 1 5.13 0.00 4.08 2 4.65 0.00 3.98 3 3.19 0.00 4.65 4 4.04 2.45 0.00 R^P- 5 3.98 4.04 4.73 6 0.00 0.00 3.63 7 5.63 4.69 0.00 8 3.64 2.51 3.92 9 3.40 0.00 2.52 10 3,01 4,12 3.49 1 5.02 4.69 0.00 2 0.00 5.37 5.31 3 3.41 3.78 2.69 4 2.96 5.13 3.36 Rep 3 5 0.00 1.60 5.03 6 2.91 0.00 6.40 7 4.48 3.21 4.11 8 4.10 3.72 3.16 9 3.45 4.33 4.70 10 5,03 AAJ 6.23 1 0.00 0.00 6.01 2 3.85 7.55 5.87 3 5.54 0.00 0.00 4 5.65 5.14 4.86 Rep 4 5 3.59 4.36 5.18 6 4.70 3.84 0.00 7 3.42 6.14 0.00 8 4.95 5.22 5.25 9 5.61 5.89 5.29 10 3.42 4.95 5.64
81 Table A.2. Percent damage estimates from artificially infesting cotton squares from first position, 6al l node with Lygus for 24h (2001) Treatments
Replications Control L. hesperus L. elisus
R 1 0.0 10.0 40.0
R2 15.0 25.0 40.0
R3 0.0 15.0 30.0
R4 0.0 15.0 40.0
R5 0.0 10.0 30.0
R6 0.0 20.0 35.0
R7 10.0 35.0 20.0
R8 0.0 33.0 70.0
R9 0.0 40.0 25.0
RIO 0.0 75.0 20.0
82 Table A.3. Lint and seed weight combined (gms) for cotton bolls, first position, 6"" node infested for 24h with Lygus in the field during the 2002 season. Treatments Replications Boll Control L. hesperus L. elisus 1 4.41 4,09 0.00 2 4.62 0.00 5.99 3 0.00 0.00 0.00 4 4.95 2.24 3.46 Rep 1 5 4.35 4.07 3.26 6 4.57 5.47 6.82 7 2.27 5.16 3.99 8 4,35 5.83 0.00 9 4,51 4.41 0.00 10 0.00 3.89 0.00 1 0.00 0.00 4.47 2 4.51 0.00 4.25 3 1.29 4.82 4.15 4 4.82 1.80 4.42 Rep 2 5 5.22 0.00 5.51 6 4.97 0.00 1.74 7 3.64 4.55 3.70 8 4.29 5.56 2.73 9 5.03 2.34 0.00 10 0.00 0.00 0.00 1 1.60 4.64 0.00 2 2.24 3.56 4.90 3 5.62 2.08 6.11 4 4.97 2.75 4.81 Rep 3 5 0.00 0.00 0.00 6 0.00 4.98 1.52 7 5.24 4.85 0.00 8 2.60 3.84 4.13 9 4.40 5.01 4.30 10 4.95 0.00 5.11 1 0.00 4.84 5.42 2 5.16 5.64 1.88 3 3.35 4.68 0.00 4 4.12 1.17 4.66 Rep 4 5 2.41 4.25 4.26 6 3.02 2.17 3.82 7 0.00 5.93 2.18 8 3.08 0.00 0.00 9 0.00 4.89 3.25 10 3.85 0.00 0.00
83 Table A.4. Percent damage estimates from artificially infesting cotton squares from first osition, 6a h node with Lygus for 24h (2002). Treatments
Replications Control L. hesperus L. elisus
RI 0.0 5.0 60.0
R2 0.0 25.0 5.0
R3 0.0 15.0 25.0
R4 0.0 5.0 5.0
R5 5.0 45.0 20.0
R6 0.0 0.0 15.0
R7 0.0 5.0 30.0
R8 0.0 5.0 35.0
R9 0.0 20.0 10.0
RIO 0.0 5.0 65.0
84 9"'Node abscission 2001
164 OPTIONS LS=80;
NOTE: There were 30 observations read from the data set WORK.ONE, NOTE: PROCEDURE ANOVA used: real time 4:32.98 cpu time 0.06 seconds
165 DATA ONE; 166 INPUT TRT REP Y; 167 CARDS;
NOTE: SAS went to a new line when INPUT statement reached past the end of a line, NOTE: The data set WORK,ONE has 12 observations and 3 variables, NOTE: DATA statement used: real time 0.05 seconds cpu time 0.02 seconds
181 PROC ANOVA; 182 CLASS TRT REP; 183 MODEL Y=TRT REP; 184 MEANS TRT/LSD; 185 RUN;
OPTIONS LS^=80 ; DATA ONE; INPUT TRT REP Y; CARDS; 1 1 60 1 2 55 1 3 55 I 4 50 2 1 60 2 2 60 2 3 70 2 4 70 3 1 70 3 2 60 3 3 70 3 4 70 PROC ANOVA; CLASS TRT REP; MODEL Y==TR T REP; MEANS TRT/LSD; RUN;
85 The SAS System 34
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
TRT 3 12 3
REP 4 12 3 4
Number of observations 12
The SAS System 35 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y
Sum of Source DF Squares Mean Square F Value Pr > F
Model 5 425.0000000 85.0000000 3.40 0.0842
Error 6 150.0000000 25.0000000 Corrected Total 11 575.0000000
R-Square CoeffVar Root MSE Y Mean
0.739130 8.000000 5.000000 62,50000
Source DF Anova SS Mean Square F Value Pr > F
TRT 2 350,0000000 175,0000000 7.00 0,0270 REP 3 75.0000000 25.0000000 1.00 0.4547
The SAS System 36
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0,05 Error Degrees of Freedom 6
86 Error Mean Square 25 Critical Value of t 2,44691 Least Significant Difference 8,6511
Means with the same letter are not significantly different,
t Grouping Mean N TRT
A 67,500 4 3 A
A 65,000 4 2
B 55,000 4 1
9* Node abscission 2002
164 OPTIONS LS=80; NOTE: There were 30 observations read from the data set WORK.ONE, NOTE: PROCEDURE ANOVA used: real time 4:32,98 cpu time 0,06 seconds
165 DATA ONE; 166 INPUT TRT REP Y; 167 CARDS;
NOTE: SAS went to a new line when INPUT statement reached past the end of a line. NOTE: The data set WORK.ONE has 12 observations and 3 variables. NOTE: DATA statement used: real time 0.05 seconds cpu time 0.02 seconds
181 PROC ANOVA; 182 CLASS TRT REP; 183 MODEL Y=TRT REP; 184 MEANS TRT/LSD; 185 RUN;
OPTIONS LS=80; DATA ONE; INPUT TRT REP Y; CARDS; 1 1 70 1 2 50 I 3 80 I 4 50
87 2 1 60 2 2 40 2 3 90 2 4 70 3 1 60 3 2 60 3 3 70 3 4 40 PROC ANOVA; CLASS TRT REP; MODEL Y=TRT REP; MEANS TRT/LSD; RUN;
The SAS System 45
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
TRT 3 12 3
REP 4 12 3 4
Number of observations 12
The SAS System 46
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y Sum of Source DF Squares Mean Square F Value Pr > F
Model 5 1750.000000 350,000000 2,57 0.1408
Error 6 816.666667 136.111111 Corrected Total 11 2566.666667
R-Square CoeffVar Root MSE Y Mean
0.681818 18.91892 11.66667 61.66667
Source DF Anova SS Mean Square F Value Pr > F
TRT 2 116.666667 58.333333 0.43 0.6699
88 •^EP 3 1633.333333 544.444444 4.00 0.0701
The SAS System 47
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type 1 comparisonwise error rate, not the experiiTientwise error rate.
Alpha 0.05 Error Degrees of Freedom 6 Error Mean Square 136.1111 Critical Value of t 2,44691 Least Significant Difference 20,186
Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 65.000 4 2 A A 62.500 4 1 A A 57.500 4 3
9* Node yield 2001
207 OPTIONS LS=80;
NOTE: There were 12 observations read from the data set WORK.ONE. NOTE: PROCEDURE ANOVA used: realtime 1:34.04 cpu time 0.14 seconds
208 DATA ONE; 209 INPUT BLK TRT Y; 210 CARDS;
NOTE: The data set WORK.ONE has 12 observations and 3 variables, NOTE: DATA statement used: real time 0,05 seconds cpu time 0.03 seconds
223 PROC ANOVA; 224 CLASS BLK TRT;
89 225 MODEL Y=BLK TRT; 226 MEANS TRT/LSD; 227 RUN;
OPTIONS LS=80; DATA ONE; INPUT TRT REP Y; CARDS; I 1 13.2 1 2 2.3 1 3 9,7 2 I 36,7 2 2 36,7 2 3 29,6 3 1 28,1 3 2 12,9 3 3 17.3 4 1 22.9 4 2 9.1 4 3 4.7 PROC ANOVA; CLASS TRT REP; MODEL Y=TRT REP; MEANS TRT/LSD; RUN;
The SAS System 51
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
BLK 4 12 3 4
TRT 3 12 3
Number of observations 12
The SAS System 52 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y
90 Sum of Source DF Squares Mean Square F Value Pr > F
Model 5 1441.775000 288.355000 12.83 0.0037
Error 6 134,885000 22,480833 Corrected Total 11 1576,660000
R-Square CoeffVar Root MSE Y Mean
0,914449 25,49137 4,741396 18,60000
Source DF Anova SS Mean Square F Value Pr>F
BLK 3 1178.420000 392.806667 17.47 0.0023 TRT 2 263.355000 131.677500 5.86 0.0389
The SAS System 53
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type 1 comparisonwise error rate, not the experimentwise error rate.
Alpha 0,05 Error Degrees of Freedom 6 Error Mean Square 22.48083 Critical Value oft 2.44691 Least Significant Difference 8.2037
Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 25.225 4 I
B 15.325 4 3 B B 15.250 4 2
91 9"' Node yield 2002
228 OPTIONS LS=80;
NOTE: There were 12 observations read from the data set WORK.ONE. NOTE: PROCEDURE ANOVA used: realtime 2:04.78 cpu time 0.08 seconds
220 DATA ONE; 230 INPUT BLK TRT Y; 231 CARDS;
NOTE: The data set WORK.ONE has 12 observations and 3 variables, NOTE: DATA statement used: real time 0.04 seconds cpu time 0.03 seconds
244 PROC ANOVA; 245 CLASS BLK TRT; 246 MODEL Y=BLK TRT; 247 MEANS TRT/LSD; 248 RUN;
OPTIONS LS==80 ; DATA ONE; FNPUT BLK TRT Y; CARDS; 1 1 11.3 I 2 9.5 1 3 22,4 2 1 20,2 2 2 24,8 2 3 9,6 3 1 19,4 3 2 10,5 3 3 16.5 4 1 13.5 4 2 12 4 3 31.1 PROC ANOVA; CLASS BLK TRT; MODEL Y==BL K TRT; MEANS TRT/LSD; RUN;
The SAS System 54 14:53 Tuesday, December 17, 2002
92 The ANOVA Procedure
Class Level Information
Class Levels Values
BLK 4 12 3 4
TRT 3 12 3
Number of observations 12
The SAS System 55 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y Sum of Source DF Squares Mean Square F Value Pr > F
Model 5 108.6400000 21.7280000 0,31 0.8893
Error 6 418.5666667 69,7611111 Corrected Total 11 527,2066667
R-Square CoeffVar Root MSE Y Mean
0.206067 49.91421 8.352312 16.73333
Source DF Anova SS Mean Square F Value Pr > F
BLK 3 41.25333333 13.75111111 0,20 0.8947 TRT 2 67.38666667 33.69333333 0.48 0.6390
The SAS System 56
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0,05 Error Degrees of Freedom 6 Error Mean Square 69,76111
93 Critical Value oft 2,44691 Least Significant Difference 14.451
Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 19.900 4 3 A A 16.100 4 I A A 14.200 4 2
9"'Node Laboratory 2001
249 OPTIONS LS=80;
NOTE: There were 12 observations read from the data set WORK.ONE, NOTE: PROCEDURE ANOVA used: realtime 1:16,90 cpu time 0.06 seconds
250 DATA ONE; 251 LNPUTTRTY; 252 CARDS;
NOTE: The data set WORK.ONE has 30 observations and 2 variables. NOTE: DATA statement used: real time 0.03 seconds cpu time 0.02 seconds
283 PROC ANOVA; 284 CLASS TRT; 285 MODEL Y=TRT; 286 MEANS TRT/LSD; 287 RUN;
OPTIONS LS=80; DATA ONE; INPUT TRT Y; CARDS; 0 0 0 0 0 0 0
94 I 0 1 0 1 0 2 50 2 5 2 5 2 45 2 5 2 50 2 10 2 50 2 15 2 20 3 5 3 50 3 30 3 40 3 60 3 5 3 5 3 5 3 5 3 5 PROC ANOVA; CLASS TRT; MODEL Y=TRT; MEANS TRT/LSD; RUN;
The SAS System 57 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
TRT 3 12 3
Number of observations 30
The SAS System 58
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y Sum of Source DF Squares Mean Square F Value Pr > F
Model 2 3705.00000 1852.50000 6.13 0.0064
95 Error 27 8162.50000 302.31481
Corrected Total 29 11867,50000
R-Square CoeffVar Root MSE Y Mean
0,312197 112,1755 17.38720 15.50000
Source DF Anova SS Mean Square F Value Pr > F
TRT 2 3705.000000 1852,500000 6.13 0.0064
The SAS System 59
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0.05 Error Degrees of Freedom 27 Error Mean Square 302.3148 Critical Value of t 2.05183 Least Significant Difference 15.955
Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 25,500 10 2 A
A 21,000 10 3
B 0.000 10 1
9* Node Laboratory 2002
288 OPTIONS LS=80; NOTE: There were 30 observations read from the data set WORK.ONE. NOTE: PROCEDURE ANOVA used: real time 41.43 seconds cpu time 0.08 seconds
289 DATA ONE; 290 INPUT TRT Y; 291 CAFIDS;
NOTE: The data set WORK.ONE has 30 observations and 2 variables,
96 NOTE: DATA statement used: real time 0,04 seconds cpu time 0,04 seconds
322 PROC ANOVA; 323 CLASS TRT; 324 MODEL Y-TRT; 325 MEANS TR F LSD; 326 RUN;
OPTIONS LS=80; DATA ONE; INPUT TRT Y; CARDS; 0 0 0 0 0 0 0 0 5 0 2 80 2 15 2 10 2 0 2 10 2 15 2 35 2 50 2 10 2 30 3 20 3 5 3 10 3 60 3 20 3 35 3 25 3 50 3 35 3 30 PROC ANOVA; CLASS TRT; MODEL Y=TRT; MEANS TRT/LSD; RUN;
97 The SAS Systein 60 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
TRT 3 12 3
Number of observations 30
The SAS System 61
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y Sum of Source DF Squares Mean Square F Value Pr > F
Model 2 4831,66667 2415,83333 8,27 0.0016
Error 27 7885.00000 292,03704
Corrected Total 29 12716.66667
R-Square CoeffVar Root MSE Y Mean
0.379948 93.21322 17.08909 18,33333
Source DF Anova SS Mean Square F Value Pr>F
TRT 2 4831.666667 2415.833333 8.27 0.0016
The SAS System 62
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate. Alpha 0.05 Error Degrees of Freedom 27 Error Mean Square 292.037 Critical Value of t 2.05183 Least Significant Difference 15.681
98 Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 29.000 10 3 A A 25.500 10 2
B 0.500 10 I
99 Table A.5. Lint and seed weight combined (gms) for cotton bolls, first position, 9'^ node
TrpatmpntQ Replications Boll Control L. hesperus L. elisus 1 0.00 2.30 0.00 2 5.54 0.00 0.00 3 0.00 0.00 0.00 4 0.00 0.00 Rep 1 0.00 5 4.43 0.00 0.00 6 0.00 0.00 0.00 7 0.00 0.00 0.00 8 0.00 0.00 4.58 9 3.19 0.00 0.00 10 0.00 0.00 5.16 1 0.00 3.64 3.59 2 0.00 5.19 0.00 3 0.00 6.27 5.47 4 5.09 4.79 5.24 Rep 2 5 0.00 2.64 0.00 6 4.68 5.09 0.00 7 0.00 4.67 5.38 8 0.00 0.00 0.00 9 3.19 4.41 5.30 10 4,78 0.00 4.66 1 6.24 0.00 0.00 2 0.00 3.53 0.00 3 3.69 0.00 1.79 4 5.17 0.00 0.00 Rep 3 5 5.24 0.00 0.00 6 3.80 0.00 0.00 7 3.92 4.53 5.73 8 0.00 0.00 5.19 9 0.00 4.84 4.62 10 0.00 0.00 0.00 1 4.04 0.00 0.00 2 5.50 0.00 4.73 3 0.00 0.00 0.00 4 0.00 4.78 0.00 Rep 4 5 0.00 0.00 0.00 6 0.00 0.00 0.00 7 5.76 0.00 0.00 8 4.21 0.00 0.00 9 0.00 0.00 0.00 10 3.38 4.36 0.00
100 Table A.6. Percent damage estimates from artificially infesting cotton squares from first position, 9"-ith' node with Lygus for 24h (2001). Treatments
Replications Control L. hesperus L. elisus
RI 0.0 50.0 5.0
R2 0.0 5.0 50.0
R3 0.0 5.0 30.0
R4 0.0 45.0 40.0
R5 0.0 5.0 60.0
R6 0.0 50.0 5.0
R7 0.0 10.0 5.0
R8 0.0 50.0 5.0
R9 0. 15.0 5.0
RIO 0.0 20.0 5.0
101 Table A.7. Lint and seed weight combined (gms) for cotton bolls, first position, 9'^t"h node infested for 24h with Lygus in the field during the 2002 season. Treatments Replications Boll Control L. hesperus L. elisus 1 0.00 0.00 0.00 2 0.00 0.00 6.35 3 2.73 0.00 3.08 4 4.94 0.00 0.00 Rep 1 5 3.67 4.70 0.00 6 0.00 0.00 4.57 7 0.00 2.89 4.53 8 0.00 1.90 3.84 9 0.00 0.00 0.00 10 0.00 0.00 0.00 1 0.00 4.66 4.48 2 0.00 0.00 0.00 3 0.00 0.00 0.00 4 6.13 4.08 0.00 Rep 2 5 0.00 0.00 0.00 6 6.04 5.24 0.00 7 3.87 3.77 5.13 8 0.00 2.90 0.00 9 4.18 4.14 0.00 10 0.00 0.00 0.00 1 0.00 0.00 0.00 2 0.00 0.00 0.00 3 0.00 0.00 0.00 4 4.88 0.00 5.00 Rep 3 5 0.00 0.00 5.20 6 5.10 4.00 5.07 7 4.63 0.00 0.00 8 0.00 2.11 0.00 9 0.00 4.36 1.26 10 4.79 0.00 0.00 1 4.10 0.00 3.03 2 0.00 0.00 4.87 3 4.07 0.00 0.00 4 3.33 0.00 5.55 Rep 4 5 0.00 0.00 4.10 6 2.02 0.00 2.46 7 5.52 5.80 4.73 8 6.06 6.22 0.00 9 0.00 0.00 6.40 10 0.00 0.00 0.00
102 Table A.8. Percent damage estimates from artificially infesting cotton squares from first position, 9"^ node with Ly^us for 24h (2002). Treatments
Replications Control L. hesperus L. elisus
R 1 0.0 80.0 20.0
R^ 0.0 15.0 5.0
R3 0.0 10.0 10.0
R4 0.0 0.0 60.0
R5 0.0 10.0 20.0
R6 0.0 15.0 35.0
R7 0.0 35.0 25.0
R8 0.0 50.0 50.0
R9 5.0 10.0 35.0
RIO 0.0 30.0 30.0
103 11"'Node abscission 2001
327 OPTIONS LS=80;
NOTE: There were 30 observations read from the data set WORK.ONE. NOTE: PROCEDURE ANOVA used: realtime 5:44.34 cpu time 0.05 seconds
328 DATA ONE; 329 INPUT TRT REP Y; 330 CARDS;
NOTE: The data set WORK.ONE has 12 observations and 3 variables. NOTE: DATA statement used: real time 0.03 seconds cpu time 0.03 seconds
343 PROC ANOVA; 344 CLASS TRT REP; 345 MODEL Y=TRT REP; NOTE: SCL source line. 346 MEANS TRT/LSD TUCKEY DUNCAN;
1 WARNING 1-322: Assuming the symbol TUKEY was misspelled as TUCKEY. 347 RUN;
OPTIONS LS^=80 ; DATA ONE; INPUT TRT REP Y; CARDS; 1 1 70 I 2 90 I 3 90 1 4 80 2 1 90 2 2 100 2 3 100 2 4 80 3 1 80 3 2 100 3 3 80 3 4 70 PROC ANOVA; CLASS TRT REP; MODEL Y= TRT REP; MEANS TRT/LSD; RUN;
104 The SAS System 63
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
TRT 3 12 3
REP 4 12 3 4
Number of observations 12
The SAS System 64 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y
Sum of Source DF Squares Mean Square F Value Pr > F
Model 5 1025.000000 205,000000 4,61 0.0449
Error 6 266,666667 44,444444 Corrected Total II 1291,666667
R-Square CoeffVar Root MSE Y Mean
0.793548 7.766990 6.666667 85.83333
Source DF Anova SS Mean Square F Value Pr > F
TRT 2 266,6666667 133,3333333 3.00 0.1250 REP 3 758.3333333 252,7777778 5,69 0,0345
The SAS System 65
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
105 Alpha 0,05 Error Degrees of 1-rcedom 6 Error Mean Square 44,44444 Critical Value oft 2.44691 Least Significant Difference 11.535
Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 92.500 4 2 A A 82.500 4 1 A A 82.500 4 3
11* Node abscission 2002
348 OPTIONS LS=80;
NOTE: There were 12 observations read from the data set WORK.ONE. NOTE: PROCEDURE ANOVA used: realtime 58.14 seconds cpu time 0.10 seconds
349 DATA ONE; 350 INPUT TRT REP Y; 351 CARDS;
NOTE: The data set WORK.ONE has 12 observations and 3 variables, NOTE: DATA statement used: real time 0.03 seconds cpu time 0.03 seconds
364 PROC ANOVA; 365 CLASS TRT REP; 366 MODEL Y=TRT REP; NOTE: SCL source line. 367 MEANS TRT/LSD TUCKEY DUNCAN;
1 WARNING 1-322: Assuming the symbol TUKEY was misspelled as TUCKEY. 368 RUN;
OPTIONS LS=80; DATA ONE; INPUT TRT REP Y;
106 CARDS; I 1 70 1 2 90 1 3 100 1 4 90 2 1 100 2 2 80 2 3 70 2 4 90 3 1 70 3 2 100 3 3 90 3 4 100 PROC ANOVA; CLASS TRT REP; MODEL Y=TRT REP; MEANS TRT/LSD; RUN;
The SAS System 66
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
TRT 3 12 3
REP 4 12 3 4
Number of observations 12
The SAS System 67
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y Sum of Source DF Squares Mean Square F Value Pr > F
Model 5 341.666667 68.333333 0.32 0.8844
Error 6 1283.333333 213.888889 Corrected Total 11 1625.000000
R-Square CoeffVar Root MSE Y Mean
107 0.210256 16,71422 14,62494 87.50000
Source DF Anova SS Mean Square F Value Pr > F
TRT 2 50.0000000 25.0000000 0.12 0.8917 REP 3 291,6666667 97,2222222 0,45 0,7237
The SAS System 68
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0,05 Error Degrees of Freedom 6 Error Mean Square 213,8889 Critical Value oft 2,44691 Least Significant Difference 25,304
Means with the same letter are not significantly different,
t Grouping Mean N TRT
A 90,00 4 3 A A 87.50 4 1 A A 85.00 4 2
11* Node yield 2001
369 OPTIONS LS=80;
NOTE: There were 12 observations read from the data set WORK.ONE, NOTE: PROCEDURE ANOVA used: realtime 53.28 seconds cpu time 0.10 seconds
370 DATA ONE; 371 INPUT BLK TRT Y; 372 CARDS;
NOTE: The data set WORK.ONE has 12 observations and 3 variables. NOTE: DATA statement used:
108 real time 0,03 seconds cpu time 0,03 seconds
385 PROC ANOVA; 386 CLASS BLK TRT; 387 MODEL Y=BLK TRT; 388 MEANS TRT/LSD; 389 RUN;
OPTIONS LS =80; DATA ONE; INPUT BLK TRT Y; CARDS; 1 1 9,4 1 2 5.1 1 3 3.6 2 I 4,9 2 2 0 2 3 3,5 3 1 4.2 3 2 0 3 3 7.7 4 1 9,8 4 2 7.6 4 3 11.2 PROC ANOVA; CLASS BLK TRT; MODEL Y= BLK TRT; MEANS TRT/LSD; RUN;
The SAS System 72
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
BLK 4 12 3 4
TRT 3 12 3
Number of observafions 12
The SAS System 73 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
109 Dependent Variable: Y
Sum of Source DF Squares Mean Square F Value Pr > F
Model 5 113.9583333 22.7916667 4.31 0.0519
Error 6 31.7183333 5.2863889 Corrected Total II 145.6766667
R-Square CoeffVar Root MSE Y Mean
0.782269 41.17997 2,299215 5,583333
Source DF Anova SS Mean Square F Value Pr > F
BLK 3 78.49666667 26.16555556 4,95 0,0461 TRT 2 35.46166667 17,73083333 3,35 0,1052
The SAS System 74
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0.05 Error Degrees of Freedom 6 Error Mean Square 5,286389 Critical Value oft 2,44691 Least Significant Difference 3,9782
Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 7,075 4 1 A A 6.500 4 3 A A 3.175 4 2
110 11* Node \icld 2002
390 OPTIONS LS=80;
NOTE: There were 12 observations read froin the data set WORK.ONE. NOTE: PROCEDURE ANOVA used: real time 57.87 seconds cpu time 0.07 seconds
391 DATA ONE; 3^)2 INPUT BLK TRT Y; 393 CARDS;
NOTE: The data set WORK.ONE has 12 observations and 3 variables, NOTE: DATA statement used: real time 0.03 seconds cpu time 0.03 seconds
406 PROC ANOVA; 407 CLASS BLK TRT; 408 MODEL Y=BLK TRT; 409 MEANS TRT/LSD; 410 RUN;
OPTIONS LS=8i0 ; DATA ONE; INPUT BLK TRT Y; CARDS; I 1 8.2 1 2 0.0 1 3 0.0 2 1 2.7 2 2 0.0 2 3 8.4 3 1 0.0 3 2 13.5 3 3 0,0 4 1 11.6 4 2 8.2 4 3 10,0
PROC ANOVA 5 CLASS BLK TRT; MODEL Y=BLK TRT; MEANS TRT/LSD; RUN;
The SAS System 75 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
111 Class Level Information
Class Levels Values
BLK 4 1234
TRT 3 12 3
Nuinber of observations 12
The SAS System 76 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y
Sum of Source DF Squares Mean Square F Value Pr > F
Model 5 96.0450000 19.2090000 0.56 0,7307
Error 6 206,5316667 34.4219444 Corrected Total 11 302.5766667
R-Square CoeffVar Root MSE Y Mean
0.317424 112,4669 5,867022 5,216667
Source DF Anova SS Mean Square F Value Pr > F
BLK 3 93,68333333 31,22777778 0,91 0,4911 TRT 2 2.36166667 1.18083333 0.03 0.9665
The SAS System 77
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0,05 Error Degrees of Freedom 6 Error Mean Square 34,42194 Critical Value of t 2,44691 Least Significant Difference 10.151
112 Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 5.625 4 1 A A 5.425 4 2 A A 4.600 4 3
if'Nodel.aboratorv 2001
411 OPTIONS LS=80;
NOTE: There were 12 observations read from the data set WORK.ONE. NOTE: PROCEDURE ANOVA used: real time 59.47 seconds cpu time 0.06 seconds
412 DATA ONE; 413 INPUT TRT Y; 414 CARDS;
NOTE: The data set WORK.ONE has 30 observations and 2 variables. NOTE: DATA statement used: real time 0.03 seconds cpu time 0.03 seconds
445 PROC ANOVA; 446 CLASS TRT; 447 MODEL Y=TRT; 448 MEANS TRT/LSD; 449 RLTN;
OPTIONS LS=SO; DATA ONE; INPUT TRT Y; CARDS; 0 0 0 0 0 0 0 0 0 0 2 5 2 5
113 -> 45 1 5 -) 5 2 10 1 15 1 20 ~> 10 ~i 10 25 3 100 -> 20 -> 50 3 45 -> J 80
3 60 3 5 3 5 PROC ANOVA; CLASS TRT; MODEL Y=TRT; MEANS TRT/LSD; RUN;
The SAS System 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values
TRT 3 12 3
Number of observations 30
The SAS System 82
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y Sum of Source DF Squares Mean Square F Value Pr > F
Model 2 8105.00000 4052.50000 9.61 0.0007
Error 27 11382.50000 421.57407
Corrected Total 29 19487.50000
114 R-Square CoeffVar Root MSE Y Mean
0.415908 117.3273 20.53227 17.50000
Source DF Anova SS Mean Square F Value Pr > F
TRT 2 8105.000000 4052.500000 9.61 0.0007
The SAS System 83
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
Alpha 0.05 Error Degrees of Freedom 27 Error Mean Square 421.5741 Critical Value oft 2.05183 Least Significant Difference 18.841
Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 39.500 10 3
B 13.000 10 2 B
B 0,000 10 I
11* Node Laboratory 2002
450 OPTIONS LS=80; NOTE: There were 30 observations read fi-omth e data set WORK,ONE, NOTE: PROCEDURE ANOVA used: realtime 44.91 seconds cpu time 0.07 seconds
451 DATA ONE; 452 INPUT TRT Y; 453 CARDS;
NOTE: The data set WORK.ONE has 30 observations and 2 variables. NOTE: DATA statement used: real time 0.03 seconds cpu time 0.03 seconds
484 PROC ANOVA;
115 485 CLASS TRT; 486 MODEL Y=TRT; 487 MEANS TRT/LSD; 488 RUN;
OPTIONS LS=80; DATA ONE; INPUT TRT Y; CARDS; 0 0 0 0 0 0 0 0 0 0 2 20 2 10 2 10 2 10 2 35 2 5 2 15 2 10 2 25 2 10 3 50 3 10 3 30 3 10 3 15 3 15 3 25 3 30 3 20 3 30 PROC ANOVA; CLASS TRT; MODEL Y=TRT; MEANS TRT/LSD; RUN; The SAS System 84
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values TRT 3 12 3
116 Number of observations 30
The SAS System 85 14:53 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: Y
Sum of Source DF Squares Mean Square F Value Pr > F
Model 2 2831.666667 1415,833333 18,18 <,0001
Error 27 2102.500000 77.870370
CoiTected Total 29 4934,166667
R-Square CoeffVar Root MSE Y Mean
0.573890 68.76171 8,824419 12,83333
Source DF Anova SS Mean Square F Value Pr > F
TRT 2 2831,666667 1415.833333 18,18 <.0001
The SAS System 86
14:53 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for Y NOTE: This test controls the Type 1 comparisonwise error rate, not the experimentwise error rate.
Alpha 0.05 Error Degrees of Freedom 27 Error Mean Square 77.87037 Critical Value oft 2.05183 Least Significant Difference 8.0973
Means with the same letter are not significantly different.
t Grouping Mean N TRT
A 23.500 10 3
B 15.000 10 2
C 0.000 10 1
117 Table A.9. Lint and seed weight combined (gms) for cotton bolls, first position, 11'*^ node infested for 24h with Lygus in the field during the 2001 season. Treatments '^^T'"^-^^'°"s Boll Control L. hesperus L elisus 1 0.00 0.00 0.00 2 0.00 0.00 0.00 3 3.58 0.00 0.00 P 4 0.00 0.00 3.64 '^^P ' 5 0.00 0.00 0.00 6 1.78 0.00 0.00 7 0.00 0.00 0.00 8 0.00 5.05 0.00 9 0.00 0.00 0.00 H) 4X11 (XOO 0.00 1 0.00 0.00 3.49 2 0.00 0.00 0.00 3 0.00 0.00 0.00 4 0.00 0.00 0.00 R^P- 5 0.00 0.00 0.00 6 0.00 0.00 0.00 7 0.00 0.00 0.00 8 0.00 0.00 0.00 9 0.00 0.00 0.00 10 4,90 0,00 0.00 1 0.00 0.00 0.00 2 0.00 0.00 0.00 3 0.00 0.00 0.00 4 0.00 0.00 0.00 Rep 3 5 0.00 0.00 0.00 6 0.00 0.00 0.00 7 4.15 0.00 3.84 8 0.00 0.00 0.00 9 0.00 0.00 3.86 10 0,00 0,00 0.00 1 0.00 5.10 4.32 2 0.00 0.00 0.00 3 0.00 0.00 0.00 4 4.91 0.00 0.00 Rep 4 5 1.91 0.00 3.58 6 0.00 0.00 0.00 7 0.00 0.00 0.00 8 0.00 0.00 0.00 9 0.00 0.00 0.00 10 0.00 2.51 3.32
118 Table A. 10. Percent damage estimates from artificially intfesting cotton squares from first position, 11"' node with Lygus for 24h (2001). Treatments
Replications Control L. hesperus L. elisus
R 1 0.0 5.0 25.0
R2 0.0 5.0 100.0
R3 0.0 45.0 20.0
R4 0.0 5.0 50.0
R5 0.0 5.0 45.0
R6 0.0 10.0 80.0
R7 0.0 15.0 5.0
R8 0.0 20.0 60.0
R9 0.0 10.0 5.0
RIO 0.0 10.0 5.0
119 Table A. 11. Lint and seed weight combined (gms) for cotton bolls, first position, 11t h
; ; Treatments Keplications Boll Control L. hesperus L. elisus 1 0.00 0.00 0.00 0 0.00 0.00 0.00 3 3.92 0.00 0.00 4 0.00 Rep 1 0.00 0.00 5 0.00 0.00 0.00 6 4.26 0.00 0.00 7 0.00 0.00 0.00 8 0.00 0.00 0.00 9 0.00 0.00 0.00 10 0.00 0.00 0.00 1 0.00 0.00 0.00 2 0.00 0.00 0.00 3 0.00 0.00 0.00 4 2.67 0.00 0.00 Rep 2 5 0.00 0.00 0.00 6 0.00 0.00 0.00 7 0.00 0.00 4.74 8 0.00 0.00 0.00 9 0.00 0.00 0.00 10 0.00 0.00 3.64 1 0.00 0.00 0.00 2 0.00 0.00 0.00 3 0.00 0.00 0.00 4 0.00 0.00 0.00 Rep 3 5 0.00 4.73 0.00 6 0.00 0.00 0.00 7 0.00 4.45 0.00 8 0.00 4.29 0.00 9 0.00 0.00 0.00 10 0.00 0.00 0.00 1 0.00 0.00 0.00 2 0.00 4.84 0.00 3 0.00 0.00 4.59 4 0.00 0.00 0.00 Rep 4 5 0.00 0.00 0.00 6 0.00 0.00 0.00 7 5.52 0.00 0.00 8 6.06 0.00 5.39 9 0.00 3.37 0.00 10 0.00 0.00 0.00
120 Table A. 12. Percent damage estimates from artificially intfesting cotton squares from
Treatments
Replications Control L. hesperus L. elisus
RI 0.0 20.0 50.0
R2 0.0 10.0 10.0
R3 0.0 10.0 30.0
R4 0.0 10.0 10.0
R5 0.0 35.0 15.0
R6 0.0 5.0 15.0
R7 0.0 15.0 25.0
R8 0.0 10.0 30.0
R9 0.0 25.0 20.0
RIO 0.0 10.0 30.0
121 APPENDIX B
CHAPTER III SAS PROGRAMS AND OUTPUTS FOR MEAN SEPARATION
TESTS, LINEAR REGRESSION ANALYSES AND RAW
DATA TABLES FOR scN AND E-64 TRIALS
122 PROC ANOVA (Mean separation of final mortality on scN trials)
NOTE: There were 32 observations read from the data set WORK.ONE. NOTE: PROCEDURE PLOT used: real time 1: 11.06 cpu time 0.05 seconds
572 Options ls=72; 573 data one; 574 input trt rep data; 575 cards;
NOTE: The data set WORK.ONE has 32 observations and 3 variables. NOTE: DATA statement used: real time 0.03 seconds cpu time 0.03 seconds
608 proc sort; by trt rep;
NOTE: There were 32 observafions read from the data set WORK.ONE. NOTE: The data set WORK.ONE has 32 observations and 3 variables. NOTE: PROCEDURE SORT used: real time 0.03 seconds cpu time 0.03 seconds
609 proc glIm ; classes trt; 610 model data= trt; 611 means trt/Isd; 612 run;
Options ls= 72- data one; input trt rep1 data; cards; 1 1 25 1 2 25 1 3 25 1 4 25 2 1 50 2 2 50 2 3 25 2 4 0 3 1 50 3 2 50 3 3 50 3 4 50 4 1 100 4 2 75 4 3 75 4 4 75
123 1 I 50 1 2 50 1 3 25 1 4 25 2 1 50 2 2 25 1 •; 50 2 4 50 3 1 50 3 2 75 3 3 25 3 4 50 4 I 50 4 2 50 4 3 75 4 4 75 proc sort; b> trt rep; proc anova; class trt; model data=trt; means trt/lsd; run; The SAS System 90 14:53 Tuesday, December 17, 2002
The ANOVA Procedure Class Level Information
Class Levels Values trt 4 1234 Number of observations 32
The SAS System 91 14:53 Tuesday, December 17, 2002 The ANOVA Procedure Dependent Variable: data Sum of Source DF Squares Mean Square F Value
Model 3 7714.84375 2571.61458 11.10
Error 28 6484.37500 231.58482
Corrected Total 31 14199.21875
Source Pr > F
Model <.0001
Error
Corrected Total
124 R-Square CoeffVar Root MSE data Mean
0.543329 31.93267 15.21791 47.65625
Source DF Anova SS Mean Square F Value t"^ 3 7714.843750 2571.614583 11,10 Source Pr > F trt <,0001 The SAS System 92 14:53 Tuesday, December 17, 2002
The ANOVA Procedure t Tests (LSD) for data
NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate. Alpha 0.05 Error Degrees of Freedom 28 Error Mean Square 231.5848 Crifical Value of t 2.04841 Least Significant Difference 15.586 Means with the same letter are not significantly different.
t Grouping Mean N trt A 71.875 8 4
B 50.000 8 3 B C B 37.500 8 2 C C 31.250 8 1
LINEAR REGRESSION (scN trials)
NOTE: There were 32 observafions read from the data set WORK.ONE. NOTE: PROCEDURE GLM used: realtime 1:13.44 cpu time 0.09 seconds
613 option ls=80; 614 data one; 615 input dose weight; 616 cards;
NOTE: The data set WORK.ONE has 32 observations and 2 variables. NOTE: DATA statement used: real time 0.02 seconds cpu time 0.02 seconds
649 proc reg;
125 650 title "scN TEST REGRESSION"; 651 model weight=dose;
NOTE: 32 observations read. NOTE: 32 observations used in computations. NOTE: There were 32 observations read from the data set WORK.ONE. NOTE: PROCEDURE REG used: real time 0.04 seconds cpu time 0.04 seconds
652 proc plot; 653 plot weight*dose 654 run;
option ls=Sil; dat£1 one; input dose weight; cards; 1.58 1.38 1.25 1.46 1,20 0,64 0,74 0,95 T 0,83 2 1,52 2 0.26 2 0.91 2 0.28 2 1.48 2 -0.55 2 1.14 3 0.64 3 -0.55 3 1.20 3 1.17 3 -0.55 3 0.65 3 -0.55 3 0.51 4 -0.55 4 -0.55 4 1.02 4 -0.55 4 1.68 4 0.59 4 -0.55 4 0.74 proc reg;
126 title "scN riiS'l RlXiRliSSlON": model weight=dose; proc plot; plot weight*dose; run;
scN TEST REGRESSION 93 14:53 Tuesday, December 17, 2002
The REG Procedure Model: MODELl Dependent Variable: weight
Analysis of Variance
Sum of Mean Source DF Squares Square F Value Pr > F
Model 1 4.05132 4,05132 8.62 0.0063 Error 30 14.10077 0.47003 Corrected Total 31 18.15209
Root MSE 0.68558 R-Square 0.2232 Dependent Mean 0.60687 Adj R-Sq 0,1973 CoeffVar 112.96957
Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr>|t|
Intercept 1 1.40250 0.29687 4.72 <.0001 Dose 1 -0.31825 0.10840 -2.94 0.0063
127 scN TEST REGRESSION 94 14:53 Tuesday, December 17, 2002
Plot of weight*dose. Legend: A = 1 obs, B = 2 obs, etc. 2.0
1.5
A A 1.0 A A weight
0.5
0.0
-O.S
-1.0
dose
PROC ANOVA (Mean separation of final mortality on E-64 trials)
NOTE: There were 30 observations read from the data set WORK.ONE. NOTE: PROCEDURE ANOVA used: realtime 10:54.13 cpu time 0.06 seconds
489 Options ls=80; 490 data one; 491 input trt rep data; 492 cards;
128 NOTE: The data set WORK.ONE has 32 observafions and 3 variables, NOTE: DATA statement used: real time 0,02 seconds cpu time 0,02 seconds
525 proc sort; by trt rep;
NOTE: There were 32 observations read from the data set WORK,ONE, NOTE: The data set WORK.ONE has 32 observations and 3 variables. NOTE: PROCEDURE SORT used: real time 0.03 seconds cpu time 0.03 seconds
526 proc glm; classes trt; 527 model data= trt; 528 means trt Isd; 529 run;
Options ls=8n; data1 one; input trt rep data: cards; 1 1 0 1 -> 0 1 -1 0 I 4 50 2 1 25 2 2 25 2 3 25 2 4 50 3 1 50 3 2 50 3 3 25 3 4 25 4 1 50 4 2 50 4 3 50 4 4 100 1 1 25 1 2 50 1 3 50 1 4 25 2 1 25 2 2 25 2 3 50 2 4 25 3 1 50 3 2 50 3 3 50 3 4 25
129 4 1 50 4 2 50 4 3 25 4 4 50 proc sort; by trt rep; proc anova; class trt; model data=trt; means trt/lsd: run; The SAS System 17:06 Tuesday, December 17, 2002
The ANOVA Procedure
Class Level Information
Class Levels Values trt 4 12 3 4
Number of observafions 32 The SAS System 2 17:06 Tuesday, December 17, 2002
The ANOVA Procedure
Dependent Variable: data
Sum of Source DF Squares Mean Square F Value Pr > F
Model 3 3593.75000 1197.91667 3.77 0.0218
Error 28 8906.25000 318.08036
Corrected Total 31 12500.00000
R-Square CoeffVar Root MSE data Mean
0.287500 47.55949 17.83481 37.50000
Source DF Anova SS Mean Square F Value Pr > F
trt 3 3593.750000 1197.916667 3.77 0.0218 The SAS System 3
17:06 Tuesday, December 17, 2002
The ANOVA Procedure
t Tests (LSD) for data NOTE: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.
130 Alpha 0.05 Error Degrees of Freedom 28 Error Mean Square 318.0804 Critical Value oft 2,04841
Least Significant Difference 18,266
Means with the same letter are not significantly different,
t Grouping Mean N trt
A 53,125 8 4 A B A 40,625 8 3 B B 31.250 8 2 B B 25.000 8 1
LINEAR REGRESSION (E-64 trials)
530 option ls=80; 531 data one; 532 input dose weight; 533 cards;
NOTE: The data set WORK.ONE has 32 observations and 2 variables. NOTE: DATA statement used: real time 0.03 seconds cpu time 0.03 seconds
566 proc reg; 567 fifie "E-64 TEST REGRESSION"; 568 model weight=dose;
NOTE: 32 observations read, NOTE: 32 observations used in computations, NOTE: There were 32 observations read fi-omth e data set WORK,ONE. NOTE: PROCEDURE REG used: real time 0,22 seconds cpu time 0,10 seconds
569 proc plot; 570 plot weight*dose; 571 run; opfion ls=80; data one; input dose weight; cards;
131 0 0,58 0 0.10 0 1.79 0 0.25 0 0.61 0 0.59 0 1,24 0 1.13 0,5 0.89 0.5 -0.49 0.5 -0.07 0.5 -0,49 0.5 -0.49 0,5 -0.49 0,5 3,02 0.5 0,52 -0.49 0.16 -0,49 0,41 0.47 -0.04 -0,22 0,20 2 0,21 -1 -0.06 2 0.12 -) -0,49 2 -0,49 2 0.67 2 0.52 2 -0.49 proc reg; title 'E-64 TEST REGRESSION"; model weight=dose; proc plot; plot weight*dose; run;
E-64 TEST REGRESSION 17:06 Tuesday, December 17, 2002
The REG Procedure Model: MODELl Dependent Variable: weight
Analysis of Variance
Sum of Mean Source DF Squares Square F Value Pr > F
132 Model I 2.35156 2.35156 4.39 0.0447 Error 30 16.07639 0.53588 Corrected Total 31 18.42795
Root MSE 0.73204 R-Square 0.1276 Dependent Mean 0.27125 Adj R-Sq 0.0985 CoeffVar 269.87580
Parameter Estimates
Parameter Standard Variable DF Estimate Error t Value Pr > I
Intercept I 0.59200 0.20048 2.95 0.0061 dose 1 -0.36657 0.17499 -2.09 0.0447
E-64 TEST REGRESSION 5 17:06 Tuesday, December 17, 2002
Plot of weight*dose. Legend: A = 1 obs, B = 2 obs, etc.
3.0 A
2.5
133 2.0
weight
1.5
A A 1.0
A
0.5 A A A
A A
A A 0.0 A
-0.5 0.0 0.5 1.0 1.5 2.0 dose
Table B.l. Number of larvae dead in each replication for CPI (scN) test number 01.
Rep Trt 1 DAI 2 DAI 3 DAI 4 DAI 5 DAI 6 DAI 7 DAI (n=4)
RI 0 0 0 1 1 1 1 Tl R2 0 0 0 1 1 1 2 Control R3 0 0 0 1 1 1 1
R4 0 0 0 1 1 1 1
RI 0 0 0 2 2 2 2 T2 R2 0 0 0 2 2 2 3 0.25 mg/ml R3 0 0 0 1 1 1 2 diet R4 0 0 0 0 0 0 1
134 RI 0 0 0 1 2 2 2 T3 R2 0 0 0.75 1 2 2 2 4 mg/ml R3 0 0 0 2 2 2 3 diet R4 0 0 0 2 2 2 2
RI 0 0 0 4 4 4 4 T4 R2 0 0 1 3 3 3 4 1.50 mg/ml R3 0 0 0 3 3 3 3
diet R4 0 0 2 3 3 3 4
Table B.2. Larval weight (mg) of living larvae from each replication for CPI (scN) test number 01. Weight Weight Weight Treatments Rep ODAI 4DAI 7DAI
RI 2.44 6.90 6.38 Tl R2 3.03 5.26 3.85 Control R3 2.33 2.97 5.37
R4 1.67 3.03 6.02
RI 1.32 4.75 2.74 T2 R2 1.74 1.75 2.06 0.25 mg/ml diet R3 1.04 1.01 1.61
135 diet „. R4 2.39 4.70 4.35 RI 2.35 4.04 2.36 T3 R2 0.75 mg/ml 1.88 2.35 0.00 diet R3 3.50 3.60 1.74
R4 2.00 1.44 3.42
RI 2.53 2.03 0.00 T4 R2 2.44 2.14 0.00 1.50 mg/ml diet R3 1.85 1.53 1.56
R4 4.00 4.60 0.00
Table B.3. Number of larvae dead in each replication for CPI (scN) test number 02.
Rep Trt 1 DAI 2 DAI 3 DAI 4 DAI 5 DAI 6 DAI 7 DAI (n=4) RI 0 0 0 2 2 2 3 Tl R2 0 0 1 2 2 2 2 Control R3 0 0 0 1 1 2 2
R4 0 0 0 1 1 2 2
RI 0 0 0 2 2 2 2 T2 R2 0 0 1 1 1 1 3 0.25 mg/ml R3 0 0 0 2 2 4 4
136 mg/ml R4 0 0 1 2 2 2 3 diet
RI 0 0 0 0 2 2 4 T3 R2 0 0 0.75 0 3 3 3 3 mg/ml R3 0 0 0 1 1 4 4 diet R4 0 0 1 2 2 2 3
RI 0 0 0 2 2 2 3 T4 R2 0 0 1 2 2 2 3 1.50 mg/ml R3 0 0 0 3 3 4 4
diet R4 0 0 3 3 3 3 3
Table B.4. Larval weight (mg) of living larvae in each replication for CPI (scN) test number 02. Weight Weight Weight Treatments Rep ODAI 4DAI 7DAI
RI 2.39 1.86 1.74 Tl R2 1.15 0.92 2.36 Control R3 2.49 2.05 2.56
R4 1.05 2.24 2.98
RI 1.83 1.85 1.65 T2 R2 2.30 2.03 2.02
137 0.25 mg/ml R3 1.79 1.45 0.00 diet R4 1.95 1.78 1.68
RI 3.41 1.01 0.00 T3 R2 0.75 mg/ml 3.10 1.19 1.19 diet R3 3.25 3.18 0.00
R4 2.86 2.06 1.05
RI 2.13 1.03 2.22 T4 R2 1.25 1.56 1.13 1.50 mg/ml diet R3 1.17 1.17 0.00
R4 0.97 1.13 1.28
Table B.S. Number of larvae dead in each replication for E-64 test number 01.
Rep Trt 1 DAI 2 DAI 3 DAI 4 DAI 5 DAI 6 DAI 7 DAI (n=4) RI 0 0 0 0 0 1 2 Tl R2 0 0 0 0 0 1 2 Control R3 0 0 0 0 0 1 3
R4 0 0 0 0 2 2 2
RI 0 0 0 0 1 2 3 T2 R2 0 0 0 0 1 2 4
138 0.50 R3 0 0 0 0 1 0 2 mg/ml R4 0 1 1 1 2 diet 3 4
RI 0 0 0 0 2 2 3 T3 R2 0 0 0 1 1.00 2 2 2 mg/ml R3 0 0 0 0 1 2 2
diet R4 0 0 1 1 1 3 3
RI 0 0 0 0 2 3 3 T4 R2 0 1 1 1 2 2 3 2.00 mg/ml R3 0 1 1 1 2 3 3
diet R4 0 0 0 0 4 4 4
Table B.6. Larval weight (mg) of living larvae in each replication for E-64 test number 01. Weight Weight Weight Treatments Rep ODAI 4DAI 7DAI
RI 1.98 2.16 2.13 Tl R2 1.11 1.23 1.18 Control R3 0.85 1.51 2.28
R4 1.17 1.31 1.47
RI 1.25 1.17 1.38
139 T2 R2 0.67 0.94 0.00 0.50 mg/ml R3 1.76 1.44 0.84 diet R4 2.15 2.91 0.00
RI 1.87 1.38 0.96 T3 R2 1.63 0.80 1.00 mg/ml 0.90 diet R3 1.21 1.40 0.55
R4 2.27 1.28 0.69
RI 2.11 1.27 0.70 T4 R2 1.83 1.59 0.43 2.00 mg/ml diet R3 2.61 1.52 0.61
R4 1.84 1.81 0.00
Table B.7. Number of larvae dead in each replication for E-64 test number 02.
Rep Trt 1 DAI 2 DAI 3 DAI 4 DAI 5 DAI 6 DAI 7 DAI (n=4) RI 0 0 1 1 1 1 3 Tl R2 0 0 0 0 2 2 2 Control R3 0 0 0 1 2 2 2
R4 0 0 1 1 1 2 3
RI 0 0 1 1 1
140 "pT R2 0 0 0 0 1 2 3 0.50 R3 0 0 1 1 2 3 3 mg/ml diet R4 0 0 0 0 1 2 2
RI 0 0 1 2 2 2 4 T3 R2 0 0 0 0 2 2 1.00 3 mg/ml R3 0 0 0 1 2 3 4
diet R4 0 0 0 0 1 3 3
RI 0 0 1 1 2 3 4 T4 R2 0 0 2 2 2 2 3 2.00 mg/ml R3 0 0 0 0 1 2 2
diet R4 0 0 0 0 2 3 3
Table B.8. Larval weight (mg) of living larvae in each replication for E-64 test number 02. Weight Weight Weight Treatments Rep ODAI 4DAI 7DAI
RI 0.94 1.15 1.10 Tl R2 1.14 2.21 2.15 Control R3 2.63 3.15 3.46
R4 1.48 1.52 1.62
141 RI 4.07 3.78 0.00 T-> R2 5.35 5.65 0.25 0.50 mg/ml diet R3 4.11 4.01 3.51
R4 1.75 2.11 2.01
RI 1.05 0.92 0.00 T3 R2 1.11 0.74 1.00 mg/ml 0.65 diet R3 1.79 1.75 0.00
R4 1.62 0.98 0.90
RI 1.81 2.01 0.00 T4 R2 2.18 2.32 1.16 2.00 mg/ml diet R3 3.84 3.95 2.03
R4 1.47 1.42 0.17
142