<%BANNER%>

Population Dynamics and within-Plant Distribution of Frankliniella Species and Orius insidiosus and Their Impact on Cott...

Permanent Link: http://ufdc.ufl.edu/UFE0021899/00001

Material Information

Title: Population Dynamics and within-Plant Distribution of Frankliniella Species and Orius insidiosus and Their Impact on Cotton Hardlock Disease
Physical Description: 1 online resource (98 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: cotton, frankliniella, hardlock, orius, thrips
Agronomy -- Dissertations, Academic -- UF
Genre: Agronomy thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Cotton hardlock, associated with Fusarium verticillioides, has become a very important disease of the crop, reducing yields significantly yearly in the Southeastern region of the US. The disease manifests in the failure of the fiber to fluff out as the boll opens at maturity and locks like wedges of an orange when broken apart. Thrips have been associated with the disease as vectors. Field studies were carried out to determine the relationship between thrips population fluctuations and the disease in two locations, Quincy and Marianna, Florida. Four species of Frankliniella thrips namely F. bispinosa, F. fusca, F. occidentalis and F. tritici were identified on the crop in both locations but F. tritici constituted > 98% of the adult population. No association was found between thrips and the disease. Field estimation of the relationship between thrips and a predatory minute bug, Orius insidiosus and the predatory role of this bug against the thrips on the crop were also conducted. No relationship was found between them and the predator was found not to effectively suppress the population of the thrips. Field evaluation of applications of an insecticide and a fungicide for control of the disease showed the former rather than the latter offered better management of the disease. The insecticide applications tend to significantly reduce thrips population which invariably led to a reduction in the disease and thereby led to increase in yield in most cases. These results demonstrate that thrips may be playing little or no role in the epidemiology of the disease and therefore lend support to the idea that, insecticides alone or combinations of insecticides and fungicides applications can be used to manage the disease.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Wright, David L.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021899:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021899/00001

Material Information

Title: Population Dynamics and within-Plant Distribution of Frankliniella Species and Orius insidiosus and Their Impact on Cotton Hardlock Disease
Physical Description: 1 online resource (98 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: cotton, frankliniella, hardlock, orius, thrips
Agronomy -- Dissertations, Academic -- UF
Genre: Agronomy thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Cotton hardlock, associated with Fusarium verticillioides, has become a very important disease of the crop, reducing yields significantly yearly in the Southeastern region of the US. The disease manifests in the failure of the fiber to fluff out as the boll opens at maturity and locks like wedges of an orange when broken apart. Thrips have been associated with the disease as vectors. Field studies were carried out to determine the relationship between thrips population fluctuations and the disease in two locations, Quincy and Marianna, Florida. Four species of Frankliniella thrips namely F. bispinosa, F. fusca, F. occidentalis and F. tritici were identified on the crop in both locations but F. tritici constituted > 98% of the adult population. No association was found between thrips and the disease. Field estimation of the relationship between thrips and a predatory minute bug, Orius insidiosus and the predatory role of this bug against the thrips on the crop were also conducted. No relationship was found between them and the predator was found not to effectively suppress the population of the thrips. Field evaluation of applications of an insecticide and a fungicide for control of the disease showed the former rather than the latter offered better management of the disease. The insecticide applications tend to significantly reduce thrips population which invariably led to a reduction in the disease and thereby led to increase in yield in most cases. These results demonstrate that thrips may be playing little or no role in the epidemiology of the disease and therefore lend support to the idea that, insecticides alone or combinations of insecticides and fungicides applications can be used to manage the disease.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Wright, David L.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021899:00001


This item has the following downloads:


Full Text

PAGE 1

1 POPULATION DYNAMICS AND WITHIN PLANT DISTRIBUTION OF Frankliniella SPECIES AND Orius insidiosus AND THEIR IMPACT ON COTTON HARDLOCK DISEASE By ENOCH ADJEI OSEKRE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Enoch Adjei Osekre

PAGE 3

3 To my father, Mr. E. A. Osekre, whose toil and sacrifices made this milestone possible

PAGE 4

4 ACKNOWLEDGMENTS I am greatly indebted to Dr. David L. Wright, for his supervision and support. I am very thankful to Dr. Joseph Funderburk for his inva luable assistance with technical guidance and direction. I am also very gratef ul to other members of my supervisory committee including Drs. Jim Marois, Richard Sprenkel, and Tom Sinclair. My thanks go to Drs. Tawainga Katsvairo and Duli Zhao for their support and encouragement, a nd Dr. D. J. Mailhot for sharing his expertise on the project with me. My gratitude also goes to my colleagues, Drs. Francis K. Tsigbey and Susan Bambo for their support, prayers and encour agement. My thanks go to Mr. Brian Kidd and Mr. Wayne Branch for their assistance in fiel d activities. I thank memb ers of the Extension Agronomy section of NFREC including Mr. Ke lly OBrien, Mr. Maynard Douglas, Mr. Ricky Beasley, Mr. Roosevelt Gordon, Ms. Debbie Dalton, and Ms. Youfu Huang. My sincere gratitude also goes to Mrs. Debbie Wright and the entire family for their wonderful support. I really appreciate the patience, support, sacrifice, understanding, commitment and prayers of my entire family back home. The glory goes to the Almighty God who gave me the spiritual direction, guidance and strength to su ccessfully complete this project.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES.........................................................................................................................9 ABSTRACT...................................................................................................................................11 CHAP TER 1 INTRODUCTION AND LITERATURE REVIEW..............................................................13 Cotton.....................................................................................................................................13 Some Innovations in Cotton Production ................................................................................. 15 Diseases of Cotton............................................................................................................. .....15 Cotton Boll Rots............................................................................................................... ......16 Boll Rots and Hardlock......................................................................................................... .18 Insects and Boll Rots.......................................................................................................... ....18 Thrips......................................................................................................................................19 Biology of Thrips....................................................................................................................20 Thrips and Cotton Hardlock................................................................................................... 20 Control of Boll Rots (Hardlock)............................................................................................. 22 Hardlock Management Using Natural Enemies of Thrips..................................................... 23 O. insidiosus ...........................................................................................................................24 Objective of the Study......................................................................................................... ...24 2 POPULATION DYNAMICS OF FRANKLINIELLA THRIPS .............................................25 Introduction................................................................................................................... ..........25 Temporal Abundance of Thrips.............................................................................................. 26 Seasonal Abundance of Thrips...............................................................................................26 Within Plant Distribution of Thrips........................................................................................27 Spatial Abundance of Thrips.................................................................................................. 28 Sex Distribution in Thrips......................................................................................................28 Materials and Methods...........................................................................................................30 Field Plots........................................................................................................................30 Thrips Sampling..............................................................................................................31 Thrips Population Dynamics on Hardlock Severity........................................................31 Data Analysis...................................................................................................................32 Results.....................................................................................................................................32 Population Dynamics of Franklinie lla Thrips.................................................................32 Within Plant Distribution of Frankliniella Thrips ...........................................................33 Population Dynamics of Thrips and Hardlock................................................................ 34 Discussion...............................................................................................................................35

PAGE 6

6 3 PREDATION OF FRANKLINIELLA TRITICI BY ORIUS INSIDIOSUS IN COTTON ................................................................................................................................54 Introduction................................................................................................................... ..........54 O insidiosus as an Effective Predato r of Frankliniella Thrips.............................................. 54 O. insidiosus as an Ineffectiv e Predator of Frankliniella Thrips............................................ 56 Materials and Methods...........................................................................................................58 Field Plots........................................................................................................................58 Thrips and Orius Sam pling.............................................................................................58 Data Analysis...................................................................................................................58 Results.....................................................................................................................................59 Discussion...............................................................................................................................61 4 FRANKLINIELLA THRIPS, COT TON HARDLOCK, COTTON SQUARES AND THEIR MANAGEMENT USING PESTICIDES.................................................................. 68 Introduction................................................................................................................... ..........68 Frankliniella Thrips, Hardlock and T heir Management................................................. 68 Pesticides and Abortion of Cotton Squares............................................................................ 70 Materials and Methods...........................................................................................................70 Field Plots........................................................................................................................70 Thrips and Orius Sam pling, Hardlock and Yield Assessment................................. 71 Effect of Pesticides on Square Abortion..................................................................71 Data Analysis...........................................................................................................72 Results.....................................................................................................................................72 Thrips and O. insidiosus Population Dyna mics and Pesticide Treatments..................... 72 Thrips and Pesticide Treatments and Hardlock........................................................ 73 Discussion...............................................................................................................................74 5 SUMMARY AND CONCLUSIONS.....................................................................................86 LIST OF REFERENCES...............................................................................................................90 BIOGRAPHICAL SKETCH.........................................................................................................98

PAGE 7

7 LIST OF TABLES Table page 2-1 Mean density of thrips on cotton leaf, square, flower, and boll in the experim ent conducted at two locations in Quincy, FL in 2005............................................................38 2-2 Mean density of thrips on cotton leaf, square, flower, and boll in the experim ent conducted at Quincy and Marianna, FL in 2006................................................................ 39 2-3 Mean density of thrips on cotton leaf, square, flower, and boll in the experim ent conducted at Quincy and Marianna, FL in 2007................................................................ 40 2-4 Percentage of total thrips found on cotton leaves, fruiting structures, and flowers in the experiment conducted at Quincy, FL in 2005..............................................................41 2-5 Percentage of total thrips found on cotton leaves, fruiting structures, and flowers in the experiment conducted at Quincy, FL in 2006..............................................................41 2-6 Percentage of total thrips found on cotton leaves, fruiting structures, and flowers in the experiment conducted at Marianna, FL in 2006..........................................................42 2-7 Percentage of total thrips found on cotton leaves, fruiting structures, and flowers in the experiment conducted at Quincy, FL in 2007..............................................................42 2-8 Percentage of total thrips found on cotton leaves, fruiting structures, and flowers in the experiment conducted at Marianna, FL in 2007..........................................................43 2-9 Mean number of adult and larval Fr ankliniella species thrips inhabiting the upper, mid-, and lower canopies in experiment conducted at Quincy, FL in 2006 and 2007...... 43 2-10 Mean number of adult and larval Fr ankliniella species thrips inhabiting the upper, mid-, and lower canopies in experiment conducted at Marianna, FL in 2006 and 2007....................................................................................................................................43 2-11 Mean number of adult and larval thrips Frankliniella spec ies thrips inhabiting the upper and lower flowers in experiments conducted at Quincy, FL in 2006 and 2007...... 44 2-12 Mean number of adult and larval thrips Frankliniella spec ies thrips inhabiting the upper and lower flowers in experiments conducted at Marianna, FL in 2006 and 2007....................................................................................................................................44 2-13 Mean densities of adult and larval F rankliniella thrips per plant part across 2005, 2006, and 2007 inhabiting the leaf, square, flower, and boll in experiments conducted at Quincy, FL....................................................................................................44

PAGE 8

8 2-14 Mean densities of adult and larval F rankliniella thrips per plant part across 2006 and 2007 inhabiting the leaf, square, flower, and boll in experiments conducted at Marianna, FL......................................................................................................................45 3-1 Mean densities of Frankliniella thrips and O. insidiosus per flower across 2005, 2006 and 2007 at Quincy and Marianna, FL. ............................................................................. 64 3-2 Weekly ratios of O. insidiosus to Fr ankliniella thrips inhabiting flowers in Quincy, FL.......................................................................................................................................64 3-3 Weekly ratios of O. insidiosus to Fr ankliniella thrips inhabiting flowers in Marianna, FL.......................................................................................................................................64 4-1 Mean densities of Frankliniella species th rips and O insidiosus per flower as affected by pesticide treatment at Quincy, FL in 2006...................................................... 77 4-2 Mean densities of Frankliniella thrip s species and O insidiosus per flower as affected by pesticide treatment at Quincy, FL in 2007...................................................... 77 4-3 Mean densities of Frankliniella thrip s species and O insidiosus per flower as affected by pesticide treatment at Marianna, FL in 2006.................................................. 77 4-4 Mean densities of Frankliniella thrip s species and O insidiosus per flower as affected by pesticide treatment at Marianna, FL in 2007.................................................. 77 4-5 Mean densities of Frankliniella species th rips and O insidiosus per flower across 2006 and 2007 in Quincy and Marianna, FL.....................................................................78 4-6 Effect of pesticides on hard lock and yield in Quincy, FL. ................................................78 4-7 Effect of pesticides on hardlo ck and yield in Marianna, FL. .............................................78 4-8 Effect of pesticides on abortion of cotton squares per day with aborted cotton squares collected from 3 m of two rows of cotton plants in Quincy and Marianna, FL................. 79

PAGE 9

9 LIST OF FIGURES Figure page 2-1 Mean densities (SEM) of Frankliniella thrips on cotton leaves from 18 May through 22 August of 2005, 2006 and 2007 in Quincy, FL............................................... 46 2-2 Mean densities (SEM) of Frankliniella thrips on cotton leaves from 23 June through September 1, 2006 in Marianna, FL..................................................................... 46 2-3 Mean densities (SEM) of Frankliniella thrips on cotton leaves from 21 May through August 13, 2007 in Marianna, FL........................................................................47 2-4 Mean densities (SEM) of Frankliniella thrips in cotton squares from 7 July through 25 August, 2006 in Marianna, FL......................................................................................47 2-5 Mean densities (SEM) of Frankliniella thrips in cotton squares from 9 July through 14 September, 2007 in Marianna, FL................................................................................ 47 2-6 Mean densities (SEM) of Frankliniella thrips in cotton squares from 28 June through 16 September of 2005, 2006 and 2007 in Quincy, FL......................................... 48 2-7 Mean densities (SEM) of Frankliniella thrips in cotton flow ers from 12 July through 29 August of 2005, 2006 and 2007 in Quincy, FL............................................... 49 2-8 Mean densities (SEM) of Frankliniella thrips in cotton flow ers from 7 July through 25 August in 2006 in Marianna, FL...................................................................................49 2-9 Mean densities (SEM) of Frankliniella thrips in cotton flow ers from 23 July through 11 September in 2007 in Marianna, FL................................................................ 50 2-10 Mean densities (SEM) of Frankliniella thrips on cotton bolls from 1 August through 29 August in 2005, 2006 and 2007 in Quincy, FL............................................... 50 2-11 Mean densities (SEM) of Frankliniella thrips on cotton bolls from 14 July through 11 August in 2006 in Marianna, FL...................................................................................51 2-12 Mean densities (SEM) of Frankliniella thrips on cotton bolls from 16 July through 27 August in 2007 in Marianna, FL...................................................................................51 2-13 Mean densities (SEM) of Frankliniella thrips relating fem ale and male occurrence in 2005, 2006 and 2007 in Quincy, FL..............................................................................51 2-14 Mean densities (SEM) of Frankliniella thrips relating fem ale and male occurrence in 2006 and 2007 in Marianna, FL..................................................................................... 52 2-15 Mean densities (SEM) of Frankliniella thrips relating fem ale and male occurrence in 2005, 2006 and 2007 in Quincy, FL..............................................................................52

PAGE 10

10 2-16 Mean densities (SEM) of Frankliniella thrips relating fem ale and male occurrence in 2006 and 2007 in Marianna, FL..................................................................................... 52 2-17 Regression of thrips with hardlock in Quincy, FL. ............................................................ 53 2-18 Regression of thrips with hardlock in Marianna, FL. ........................................................ 53 3-1 Mean densities ((SEM) of Frankliniella thrips and O. insidiosus in cotton flowers from 12 July Through 29 August, in 2005, 2006 and 2007 in Quincy, FL....................... 66 3-2 Mean densities ((SEM) of Frankliniella thrips and O. insidiosus in cotton flowers from 23 July through 11 September, 2006 in Marianna, FL............................................. 66 3-3 Mean densities ((SEM) of Frankliniella thrips and O. insidiosus in cotton flowers from 7 July through 25 August, 2007 in Marianna, FL..................................................... 67 4-1 Mean densities (SEM) of Frankliniella thrips and O. insidiosus in cotto n flowers as affected by pesticides from 12 July through 15 August 2006 in Quincy, FL.................... 80 4-2 Mean densities (SEM) of Frankliniella thrips and O. insidiosus in cotto n flowers as affected by pesticides from 26 July through 22 August 2007 in Quincy, FL.................... 81 4-3 Mean densities (SEM) of Frankliniella thrips and O. insidiosus in cotto n flowers as affected by pesticides from 23 July through 20 August 2006 in Marianna, FL................ 82 4-4 Mean densities (SEM) of Frankliniella thrips and O. insidiosus in cotto n flowers as affected by pesticides from 14 July through 11 August 2007 in Marianna, FL................ 83 4-5 Relationship between thrips and hardlock on per plot basis in Quincy, FL in 2006 (A) and 2007 (B). F = fungi cide (Thiophanate-methyl), I = insecticide (L ambdacyhalothrin), B = (fungicide + insect icide) and C = untreated control..............................84 4-6 Relationship between thrips and hardlock on per plot basis in Marianna, FL in 2006 (A2) and 2007 (B2). F = fungicide (Thiopha nate-m ethyl), I = insecticide (Lambdacyhalothrin), B = (fungicide + insect icide) and C = untreated control..............................84 4-7 Cotton boll showing symptoms of hardlock...................................................................... 85

PAGE 11

11 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy POPULATION DYNAMICS AND WITHIN PLANT DISTRIBUTION OF Frankliniella SPECIES AND Orius insidiosus AND THEIR IMPACT ON COTTON HARDLOCK DISEASE By Enoch Adjei Osekre May 2008 Chair: David L. Wright Major: Agronomy Cotton hardlock, associated with Fusarium verticillioides has become a very important disease of the crop, reducing yields significantly yearly in the Sout heastern region of the US. The disease manifests in the failure of the fiber to flu ff out as the boll opens at maturity and locks like wedges of an orange when broken apart. Thrips have been associated with the disease as vectors. Field studies were carried out to determine the relationship between thrips population fluctuations and the disease in two locations, Qu incy and Marianna, Florida. Four species of Frankliniella thrips namely F bispinosa, F. fusca F occidentalis and F tritici were identified on the crop in both locations but F tritici constituted >98% of the adult population. No association was found between thrips and the disease. Field estimation of the relationship between thrips and a predatory minute bug, Orius insidiosus and the predatory role of this bug against the thrips on the crop were also conducted. No relationship was found between them and the predator was found not to effectively suppress the population of the thrips. Field evaluation of applications of an insecticide and a fungici de for control of the disease showed the former rather than the latter offered better management of the disease. The

PAGE 12

12 insecticide applications tend to significantly re duce thrips population which invariably led to a reduction in the disease and thereby led to in crease in yield in most cases. These results demonstrate that thrips may be playing little or no role in th e epidemiology of the disease and therefore lend support to the idea that, insecticides alone or co mbinations of insecticides and fungicides applications can be us ed to manage the disease.

PAGE 13

13 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW Cotton Cotton belongs to the genus Gossypium which contains a great di versity of plant species ranging fro m herbaceous perennials to small trees This genus contains 49 species distributed throughout most tropical and subtro pical regions of the world. Cott on is not native to the USA. Its centers of diversity incl ude northwestern Australia, nort heastern Africa, the Arabian Peninsula, and western and north ern Mexico. The Spanish were th e first non-native people to experiment with cotton culture, having done so in Florida in 1556 (Smith and Cothren 1999). The most cultivated species of cotton in the world include Gossypium hirsutum L. and Gossypium barbadense L. (also referred to as New World species), accounti ng for >97% of the world fiber production. Co tton fibers of the G. hirsutum species range from 2 to 3 centimeters in length, whereas G. barbadense cotton produces long-stable fibe rs up to 5 centimeters in length (Smith and Cothren 1999). In Florida and the southeastern parts of the US, upland cotton is the main type grown. Cotton is a very important crop serving many purposes, including being a major natural fiber crop for cloth, providing ed ible oil and as a produc t for livestock food. Reproductive growth of cotton commences a bout 4 to 5 weeks after sowing, with the formation of floral buds in the apical part of the plant. This is followed in several weeks by flower opening (anthesis) which is usually follo wed shortly by the dehiscence of anthers soon after the petal open, and the start of fruit (bol l) development although high humidity or cool temperatures can delay this by 2 to 3 hours. Floral buds appear first as small, green, pyramidal structures known as squares. The first floral organs to be initiated are the sepal, followed by petals, stamens, and carpel (Smith and Coth ren 1999). The length of time between the first appearance of a given square (i.e., pinhead square size) and anthesis (white flower stage) is

PAGE 14

14 approximately 25 days (Tharp 1965). Flowering in cotton follows a predictable pattern, and is affected by environmental factors. Low night temperatures (below 32OC) cause flowering at a lower node as do longer photoperiods, e.g., 14-hour versus 8-hour days (Mauney 1968). The first flowers to open are usually at main-stem nodes 6 or 7 and at the first position along a fruiting branch. About 3 days elapse between the opening of a flower on a given fruiting branch and the opening of another. The same relative position on the next higher fruiting branch is usually separated by about 3 days (known as the vertical flowering interval ). On the other hand, the time interval for the development of two successive flowers on the same branch is about 6 days (horizontal flowering in terval). Upland cotton (G. hirsutum ) flowers are creamy white on the day of flowering, but petals turn pink-red the following day and abscise at the base 1 or 2 days later while Pima cotton ( G. barbadense ) flowers are yellow at anth esis but also turn pink. Cotton flowers open at or near dawn and remain open only a single day. Anthesis is usually followed shortly by the dehiscence of anthers soon after the petals open. Cotton can undergo self pollination in the absence of insect po llinators, and this process occurs shortly after anthesis; but when pollinators such as Bombus sp. are present, cross pollination can be significant (i.e., up to 50-80%). Fertilization, wh ich is primarily determined by temperature, could occur in as little time as 12 hours followi ng pollination, or it can be more than 24 hours. Temperature is the primary determining factor (Smith and Cothren 1999). Flowering continues until defoliation or when an unfavorable condition such as frost occurs. Cotton naturally sheds some squares and young bolls and commonl y sheds about 60% of its squares and young bolls under typical crop gr owing conditions (Smith and Cothren 1999). Cotton yield potential is significantly affected by various factors which initiate premature shedding of squares and young bolls (Smith and Cothren 1999). Excessive shedding of squares

PAGE 15

15 or bolls can result in signifi cantly altered plant morphology by effectively increasing vegetative growth. However, flowers are typical not sh ed (Smith and Cothren 1999). The typical development rate of cotton requires an average of 32 days in the S outheast to 50 days in the West from emergence to squares (floral bud) initiation, with the appearance of first white blooms three weeks later. Cotton yield is a combination of two major components: bo ll number and boll size or lint per boll. Some Innovations in Cotton Production Technologies in breeding and biotechnology have been applied in cotton to address pest problem s as well as environmental concer ns. The introduction of novel changes in Bt endotoxins and improved formulations resulted in the release in 1996 by Monsanto of Bt cotton that produces the Cry1Ac protein. Recent technologies ha ve led to the releas e of Bollgard II and Widestrike II Bt cotton that express Cry2Ab and a comb ination of Cry1Ac & Cry1F proteins, respectively. These are effective against important lepidopt erous insect pests of the crop. Umbeck et al. (1987) and Perlak et al. (1990) noted that the most important development in cotton production was the use of genetic engineering to manipulat e the cotton plant to express Bt endotoxin at levels that control le pidopterous pests. The use of the Bt cotton has contributed to the reduced number of sprays of chemical pe sticides, reduced labor cost and time, reduced production cost, reduced environmen tal pollution and reduced risk in cotton production. Another significant development was the incorporation of a gene conferring resistance to glyphosate in cotton (Roundup Ready). Diseases of Cotton Several d iseases affect cotton wherever the crop is grown. Important diseases of cotton include Verticillium wilt, Alternaria leaf spots, cotton root rots and cotton boll rots, and diseases caused by nematodes. Diseases reduce cotton yields, increase the cost of production and

PAGE 16

16 processing and reduce the value of fiber and seed. Boll rots have in recent times been recognized as a group of very important diseases of cotton contributing significantly to yield losses in the crop. Fusarium spp. and Alternaria spp. are among the most common pathogens associated with boll rots wherever the disease occurs in the USA. Cotton Boll Rots Many of the organism s in the boll rot comple x are probably responsible for discolored fibers in the harvest. This may occur if infest ation occurred near to boll maturity or if the infection is confined to a si ngle locule. The soilborne fungi which cause boll rot have been observed to survive on plant debris (Bell 1999). In the USA, Fusaria are reported to have replaced Collectotrichum spp. as the most common boll rot fungi (Bagga and Rannay 1969). In in fecting bolls, the first fungi to attack the aging boll directly are Collectotrichum spp. and Diplodia gossypium (Watkins 1981, Hillocks 1992). As the bolls near dehiscence, Fusarium especially F. semitectum and Alternaria spp. also may penetrate the bolls directly, although these pathogens more commonly enter through wounds or lesions first caused by Collectotrichum Fusarium spp., Diplodia spp. and other fungi have been associated with a progressive basal type of rot where bracts are infected first, followed by invasion through nectaries a nd base of boll (Bell 1999). Fusarium spp., among the most common isolated fungi boll rots, have been noted to have the capability to infect bolls abou t 35 days or older. Diseased bolls become dark brown with a white to salmon-pink overgrowth of the fungus As bolls age beyond 40 days they become progressively more susceptible to attack by se veral fungal pathogens (Bell 1999). Immature fiber exposed by injuries to the boll can be deterior ated by more than 100 different fungal species (Hillocks 1992). Infection with Fusarium spp. can occur through the fl ower and that infection can be found on flower parts as well as seed at 20, 40 and 53 days after flowering (Marois,

PAGE 17

17 Wright, Mailhot, unpublished data). This s uggests that two possible ways by which Fusarium infection can occur are infection through the flower as the pollen tube grows down the pistil, and infection through wounds on developi ng bolls created by insects. Most boll rot fungi produce air borne spores that come to rest on the bolls, bracts, or exposed fibers. The spores germinate and proceed to ramify through boll parts. Bracts tissue often dies before other boll parts and may serve as an important means of entry to the boll. The boll fungi apparently grow and sporulate prolifica lly on flowers, squares, and bolls that are shed from the plant and fall on the ground (Bell 1999). An average of more than 1 million conidia of Fusarium have been found per shed flower and square in Louisiana, and large numbers of both Diplodia and Fusarium were found in air samples collect ed over cotton fields (Snow and Sanders 1979). In an initial survey of boll rots in the cotton fields in the delta region of Louisiana and Mississippi in 1996, McLean and Lawrence (1998) isolated Fusarium spp. and Alternaria alternata from diseased bolls with a fr equency of 18 and 11%, respectively. Studies conducted in the USA by Sparni cht and Roncardi (1972) showed that Fusarium spp. do not penetrate directly through the peri carp. Under humid conditions, bolls decayed by F. moniliforme are known to be covered initially with a white or grey mycelium, which disappears when the fungus sporulates. Coni dia are produced in la rge numbers on the su rface of the bolls and have a pink color in mass. Fusarium species can be carried internally on cotton seed (Hillocks and Brettell 1993) and persist in the se edling or the soil surface and therefore provide a reservoir of inoculum to infect the lowe r bolls. Also, it appears that isolates of F. moniliforme and other Fusaria may be present in the vascular ti ssue of the healthy plants, becoming damaging only under certain conditions.

PAGE 18

18 Boll Rots and Hardlock There seem s to be different types of boll rots. Hillocks and Brettell (1993) indicated that discoloration of the seed cotton caused by the growth of microbial contamination occurred in the field under three distinct circumstances. One su ch circumstance involve s incomplete boll rot, where the boll is able to open normally but one or more locules fail to fluff out due to fungal or bacterial infection, which is known as tight lock (hardlock) in the USA. In this type, locks of infected bolls often fail to fluff (tight locks) or are only partially flu ffed and stained. The poorly fluffed locks are often knocked to the ground duri ng harvest. Hardlock is the condition in which individual locules within a bo ll remain compact and fail to op en normally (Marois et al. 2002). When picked and ginned, the stained fiber result s in lower fiber grades and, thus, decrease the value of the crop. Usually Fusarium spp. and D. gossvina infecting at the cracking stage may cause "tight-lock. Insects and Boll Rots Num erous arthropods are recogni zed as pests of cotton in the United States, but generally fewer than 25 are considered key pests of th e crop (Newsom and Brazzel 1968). Many of these pests have been persistent probl ems causing economic losses in co tton for over a century, in spite of the application of management strategies. The susceptibility of cotton plants to arthropod pests varies considerably across and within the various production regions of US. Insects occasionally transmit boll rot pathogens as well as provide wound for them to enter the boll (Watkins 1981, Hillocks 1992). Thrips have been associated with hardlock by creating wounds or entry points for the causal agent of the disease. The large amount of inoculum produced and carried in the environment and on plant debris suggests that even though the pathogen is seed-borne, that m eans of pathogen transmission could not be important in the epidemiology of the disease. Thus thrips role in the epidemiology of the disease is a matter of

PAGE 19

19 concern especially wh ere it is known that Fusarium sp., which is associated with the disease, usually infects through w ounds created by insects. Thrips Thrips, especially those belonging to the Frankliniella spp., are also considered im portant insect pests notably on seedling cotton. Several species of thrips, including tobacco thrips, F. fusca (Hinds), flower thrips, F. tritici (Fitch), and the west ern flower thrips, F. occidentalis (Pergande) have been noted to cause consider able damage to cotton. These thrips species commonly occur throughout all US cotton production regions. The adults are anthophilic (Cho et al. 2000, Hansen et al. 2003) and inhabit the fl owers of many of cultivated and uncultivated plants (Chellemi et al. 1994). Frankliniella spp. are the most abundant thrips on crops and the surrounding plant community in our agroeco system (Salguero-Navas et al. 1991). Thrips species generally have population char acteristics that include vagility, a short generation time, polyphagy, a tendency towards pa rthenogenesis, and possibly a competitive breeding structure that promotes aggregation and exploitation of localized optimal conditions (Mound and Teulon 1995). They have piercing-sucking, multi-purpose mouthparts. They use these to pierce leaves, flowers, seeds, pollen grai ns, and fruit, as well as to suck open liquids such as nectar, water, or insect secretions (Kir k 1997). This way of causing damage to plant parts give some unique characteristic symptoms includi ng streaks and discoloration of the petals, with dark flowers showing light streaks and light flow ers showing dark streaks on such plants after damage (Pfleger et al. 1995). Adult and larvae da mage cotton plants, but the larval stages are considered to cause more significant injury. Thri ps damage is most severe when feeding occurs in the apical meristem of the plant terminal, a nd the apical bud is destroyed, especially at the seedling stage. High densities of thrips occasio nally are found during the flowering stage, but

PAGE 20

20 plants of this age usually are c onsidered to tolerate thrips populations wi thout any substantial yield loss (Graves et al. 1987). Biology of Thrips Adult thrips are approxim ately 1 to 2 mm in length and generally appear in various colors ranging from yellowish-brown to dark brown. Thri ps are haplodiploid (males are haploids while females are diploids). Flower thrips are believed to overwinter as sexually mature females in soil and protected places of a plant such as curled l eaves. The life cycle consists of the egg, larva, prepupa, pupa and adult. Female adult thrips live up to 30 days and lay 2 to 10 eggs per day (Toapanta et al. 1996). Their developmen t is affected by temperature. At 20oC, development from egg to adult takes approximately 19 days, reducing to 13 days at 25oC. The female inserts eggs into soft plant tissues, in cluding flowers, leaves, stems a nd fruit using its ovipositor. The eggs hatch into larvae, which consist of two inst ars that feed and develop on the leaves, flowers and fruit. The prepupal and pupal stages often complete their development on the ground or growing medium, but pupation can also take plac e on the plant or in the soil. The pupa is a nonfeeding stage during which the wings and other adult structures develop. Thrips and Cotton Hardlock Som e insects are known to vector plant diseases and Frankliniella thrips are widely recognized as one of the most important vector s of plant diseases. In contrast, Lewis (1973) noted that there is limited information evidence to support the claim that thrips can vector bacteria and fungi. He listed f our economically important bacter ia that had been found on the bodies of thrips but considered the impact of thrips as disease vectors to be minor. Marullo (1997) also reported that some thrips species have been shown to feed on fungal spores, but fungi appear to be an uncommon food source for species of thrips in the family Thripidae. Bell (1999), however, noted that most fungi, apart from Diplodia sp., usually do infect only after the

PAGE 21

21 boll wall have been breached, either by insect damage or by preventing rupture of the suture. Thrips are believed to vector hardlock (Marois and Wright 2004) through similar means. Thrips have been associated (as vectors) with several important plant diseases such as tomato spotted wilt. Four species of Frankliniella thrips known to help vector hardlock are found in Florida. These species in northern Florida are three native species, including F. bispinosa, F. fusca, and F. tritici and a non-native species F. occidentalis The role of thrips in the epidemiology of hardlock has not been investigated. The determinati on of the role of thrips in the epidemiology of the disease would go a long way to identify control strategies for the disease. One way to determine the impact of thrips on the disease epidemiology is to evaluate the impact of the thrips population fluctuations on the di sease. There is the need to, first of all, even determine the abundance of the various Frankliniella thrips in this region and follow their population fluctuations during the cotton growing season since the occurrence of thrips species depends on various factors. Th rips species composition on co tton varies during the production season within a given region and is influenced by environmental conditions and alternate host in field borders (Leonard et al. 1999 ) Even though the role of thrips in the spread of the disease has caught the attention of some resear chers, there is little informa tion and research on the impact of population dynamics and with in-plant distribution of Frankliniella spp. on the disease spread and severity. Additionally, since the disease is associ ated with the bolls (which develop from the flowers), there is the need to determine the aggr egation pattern of thrips on the various plant parts to identify the temporal ag gregation of the insects on the plant parts. Results from these studies would show the relationshi p between thrips and hardlock. These issues are addressed in chapter 2, with the obj ectives to determine the most abundant Frankilniella species occurring

PAGE 22

22 during the cotton growing season, their aggregati on pattern on the various plant parts, and the impact of their population fluctuations on th e disease. This may provide a clue to the development of appropriate management strategi es for thrips and subsequent management of cotton hardlock. Control of Boll Rots (Hardlock) Cotton production in Florida had seen steady increases in volum e over the years until 2001 when consistent declines were recorded. One of the reasons accounting for the yield decline in recent times has been the emerging severity of hardlock disease. Hardlock was very severe in the Florida Panhandle in 2002, and the average yield of cotton was reduced from a five year average of 730 kg/ha to only 388 kg/ha, due almost entirely to hardlock (Marois and Wright 2004). Cotton diseases have been managed by the use of pesticides. Insecticides rather than fungicides are mostly employed to manage boll rots as fungicides are often not effective. Bell (1999) explained that fungicide a pplications generally do not control the disease because of the huge amounts of inoculum produced by the pathogen s in inaccessible places It seems the key for effective control involves the reduction of ear ly season inoculum, a voiding over-fertilization, improving air circulation in the plant canopy as well as applying insectic ides. Marois et al. (2002), however, reported about the effectivenes s of fungicides to re duce hardlock by noting that, in 2002 yields were almo st doubled by bloom time applicati on of fungicides in a Florida study. Insecticide and insect-resistant cultivars have been used to decrease boll rots by reducing insect wounds necessary for the entrance of certain fungal pathogens. The asso ciation of thrips to hardlock underscores the need to search for eff ective control strategies to control thrips to manage the disease. Thrips contro l using insecticides appears possi ble. The stage of the plant at which pesticides are applied seems important for e ffective thrips control. Leser (1985) and Carter

PAGE 23

23 et al. (1989) noted that successf ul management of thrips is usually accomplished primary at planting with insecticides treated seed or soil applied systemic insecticide. However, since hardlock affects the bolls, this control strate gy recommended by Leser ( 1985) and Carter et al. (1989) is not applicable. But, the report by Marois et al. (2002) about the effectiveness of fungicides to manage hardlock coupled with th e reported successes in the management of boll rots with insecticides underscores the need to expl ore the possibility of the use of insecticides or in combination with fungicides to manage hardlock. This would allow for consideration of the various pesticide combination options that coul d be used to manage the disease. This was addressed in chapter 4, w ith the objective of evaluating a f ungicide and an insecticide for the management of the disease. Hardlock Management Using Natural Enemies of Thrips Even though chem ical methods appear to be the immediate and widely used strategies to control pests, biological control has often been quite effective. Orius insidiosus (Say) (Heteroptera: Anthocor idae) is a natural en emy of thrips even though there have been controversies about their effectiven ess in the control of thrips. In managing hardlock, there is the need to consider various options and the use of natural enemies is one option worth considering. Local population of F. fusca were reported to have been near extinction in peanut by parasitism from Thripinema fuscum when levels of parasitism re ach 60 to 90% during late spring or early summer (Funderburk et al. 2002). In that report, Funderburk et al. (2002) noted that population of F fusca remained extinct until harvest of the peanut crop. O insidiosus has been reported to effectively suppress thrips population in greenhouses (Tavella et al. 1996). Some successes of this predator against Frankliniella thrips in field pepper have also been reported (Funderburk et al. 2000). But, it must also be st ated that there has be en varying degree of effectiveness of the predator against Frankliniella species. And, some researchers have also

PAGE 24

24 argued that the predator could not be effective against thrips (Mound and Teulon 1995). There was the need to evaluate the effectiveness of O insidiosus in suppressing thrips population so that this strategy could also be considered in the management of cotton hardlock disease, if found effective. This was investigated in Chapter 3, by determining the effectiveness of the predator against thrips. O. insidiosus O. insidiosus adults are sm all (about 3 mm long). Wings extend beyond the body tip. Nymphs are wingless, yellow-orange to brown in color. Both adu lts and nymphs feed by sucking juices from prey through the sucking mouthpart (rostru m). They are common on many agricultural crops and on pasture lands. They are commonly found in flowering plants and weeds during spring and summer when plant juice ab ounds. Their life cycle follows egg, nymphal and adult stages, with the nymph developing through five stages. Both mature adult and immature stages feed on many small prey including thrips, spider mites, insect eggs, aphids, and small caterpillars. Females lay 2 to 3 days after mating and can lay up to 300 eggs within plant tissue. Eggs take about 3 to 5 days to hatch, and deve lopment from egg to adult takes a minimum of 20 days under optimum conditions. Adults live about 35 days. They overwinter as an adult in leaf litter both inside and outside a farm. Objective of the Study The overall objectives of these studies were to determ ine the impact of the population dynamics of thrips and their within-plant di stribution on the diseas e epidemiology and to evaluate the potential of usi ng insecticides and/or fungicide s, or using natural enemies ( O insidiosus ) of thrips in managing the disease.

PAGE 25

25 CHAPTER 2 POPULATION DYNAMICS OF FRANKLINIELLA THRIPS Introduction The Thysanoptera are o pportunistic sp ecies exploiting intermittent occurring environments. Thrips in the genus Frankliniella (Thysanoptera: Thripidae) are ubiquitous, polyphagous pests of vegetables, fruits and ornamental cr ops (Hansen et al 2003). An even greater concern with Frankliniella thrips is the ability of some sp ecies to transmit many different pathogens. Not only do Frankliniella species feed on many species of host plants, they are able to feed in different microhabitats within a particular host plan t (Kirk 1997, Mound 1997). A complex of Frankliniella species occurs throughout northern Florida and the southeastern USA (Eckel et al. 1996, Chellemi et al. 1994, Puche et al. 1995). All of these species are highly anthophilic (Cho et al. 2000) and inhabit flower s of a variety of cultivated and uncultivated plants (Chellemi et al. 1994). Several researchers have undertaken studies on some aspects of th e ecology and chemical response behavior of thrips (Salguero-Nava s et al. 1991, Lewis 1997, Mi chelakis and Amri 1997, Pearsall and Myers 2000, Ramachandran et al 2001, Hansen et al. 2003, Whittaker and Kirk 2004). It is also worth noting that most of the innovative research on thri ps that had been done over the years had centered on population dynamics and biology of thrips on crops other than cotton. Salguero-Navas et al. (1991), Chellemi et al. (1994) Fritche and Tamo (2000), Funderburk et al. (2000), Pearsall and Myers (2000), Ramacha ndran et al. (2001), and Hansen et al. (2003) have all reported on such studies. A few other researchers bot h within and outside USA have, however, looked at th e population dynamics of thrips on cotton (Graves et al. 1987, Pickett et al. 1988, Atakan et al. 1998, Atakan an d Ozgur 2001). These studies have looked at the population abundance of Frankliniella species thrips at various times of the year.

PAGE 26

26 Temporal Abundance of Thrips The effect of tim e on thrips abundance in flow ers has been reported by a few workers, with slight variations. Gangloff (1999) reported that the effect of time of day on thrips abundance was never great. Continuing, he noted th at numbers of all stages of thrips generally climbed steadily throughout the day and were lowest between 0200 and 0600 hours. Atakan and Ozur (2001) found thrips numbers peaked ar ound 1200 hours; and Tappan (1986) a nd Kiers et al. (2000) also recorded peak numbers around 1200 hours and thereafter declined. Seasonal Abundance of Thrips Despite their sim ilarity in appearance and th eir overlapping host ranges, thrips species display different population dynamics (Cho et al. 2000, Ramachandran et al. 2001, Baez 2002, Reitz 2003). Reitz et al. (2003) repo rted that, in north Florida, F. occidentalis can be found year round and it is the most common sp ecies from winter to early spri ng when it is displaced by the increasingly abundant F. tritici and F. bispinosa They explained that although F. tritici is extremely abundant throughout the central and eastern parts of the US, it does not persist in central and southern Florida where F. bispinosa is the only abundant spec ies. Factors such as interspecific competition and differences in their ability to escape predation by Orius may be important. In their study on tomato in north Florida, Salguero-Navas et al. (1991) found large populations of F. occidentalis F tritici, and F. fusca between late April and early June, with greatest densities during May. They added that de nsities were low on other dates, especially during fall. In explaining this tr end, they indicated that population trends of thrips species were apparently unrelated to crop phenol ogy or the number of flowers per plant. Thus, in their study, they found that the larvae were more aggregat ed than adults. They added that colonization behavior of the tomato flower by F. occidentalis was different than that observed for F. tritici

PAGE 27

27 and F. bispinosa. Toapanta et al. (1996) reported that low numbers of thrips were recorded during the winter, but increased ra pidly during early spring, while Ch ellemi et al. (1994) reported the greatest number of F occidentalis F. tritici and F. bispinosa occurring in May when adults were found in the flowers of many wild plant species. Reitz (2003), who conducted studies in norther n Florida, reported that thrips rapidly colonize plants soon after the onset of floweri ng in early September, and that population peaked in mid-September and declined until the end of the month, and F. tritici was the predominant species. No F. fusca was collected in that month. Stavisky et al. (2002), who also conducted their studies on tomato in north Florid a, reported that the population of thrips increased in early May on tomato, peaked in mid-May, and declined in early June. Lewis (1997) reviewed the scien tific literature relating to peri ods of intense flight activity or mass flight of thrips. On this aspect, Sa lguero-Navas et al. (1991) and Funderburk et al. (2000) observed that such flights by F. occidentalis F. tritici and F. bispinosa are typical in the Florida geographical region during late April and early May when the adults disperse in large numbers from wild hosts and colonize crop fi elds. There are reported variations in the colonization behavior of the various Frankliniella spp. on various crops and there is the need to study these behaviors in different crops. This chapter addresses the thrips species abundance on cotton in the summer/fall season. Within Plant Distribution of Thrips Although it has been argued that seasonality is m ore important th an host plant phenology in determining abundance of thrips (Salguero-Nav as et al. 1991), host plant phenology also plays an important role in Frankliniella population dynamics, with the younger plants being able to support greater densities th an older plants. On the thrips pref erred position in terms of foliage, some workers found significantly more F. tritici and F. bispinosa adults in the upper canopy than

PAGE 28

28 in the lower canopy. Gillespie and Ve rnon (1990) noted that these diffe rences were as a result of differential aggregation of the sexe s of thrips spp. in the canopy. Terry (1997) also noted that the greater proportion of males observed in the upper canopy in his study may be related to aggregations. Males tend to aggregate in certain locations for mating, and females tend to remain in the flowers after mating. Females outnumbe red males by greater than a 4 to1 ratio in F. occidentalis (Hollingsworth et al. 2002). Spatial Abundance of Thrips Scientif ic literature shows that, ge nerally, the adults and larvae of F. occidentalis are most abundant in flowers of a variety of plants (G onzales and Wilson 1982, Pickett et al. 1988) but there have been reported varia tions in the pattern of distribu tion. Tavella et al. (1996) showed that 96% of adult and larval Frankliniella spp. occurred in the flowers of greenhouse grown pepper (Capsicum anuum L.). But, while Higgins (1992) found the majority of F. occidentalis adults in the flowers of gree nhouse-grown pepper and cucumber ( Cucumis sativa [L.]), he observed that the majority of larvae were f ound on the leaves. Studies by Funderburk et al. (2000) and Ramachandran et al. (2001) showed si milar patterns reported by Tavella et al. (1996). The leaves are preferred as a more stable sour ce of food for developing larvae in some species (Funderburk et al. 2002). Young leaves are expl oited by adults when the flowers are scarce (Teulon et al. 1991, Toapanta et al. 1996). The aggregation of thrips species in host plants seems to vary among the species and host plants, with lit tle information of thes e variations in cotton. The aggregation patterns of the various species on the plant parts were investigated in this chapter. Sex Distribution in Thrips On m ale to female distribution and ratio, Pear sall and Myers (2000) re ported that the sex ratio of 20 to 30% males was in accordance with the results of studies by Lewis (1973), who

PAGE 29

29 suggested that males would be expected to make up approximately 20% of the population in species such as western flower thrips in which reproduction is arrhenotokous. Gangloff (1999) reported that significantly fewer females and immature thrips occurred on onion foliage with flowers and buds compared with flowers with and without pollen. He also noted that fewer males were found on foliage compared with flowers. Ho wever, Salguero-Navas et al. (1991) indicated that sample location on plants did not influence density estimates of F. fusca and movement behavior of this species may differ from that of F tritici and F. occidentalis Variations in sex aggregation of thrips with in the vegetative and re productive structures need further investigations on various crops. Thus these varia tions emphasize the need to study the population dynamics of thrips on various crops, and how host plant architecture; fruiting and surface structures influence the within-plant distribution of these thrips. Additionally, quantifying the within-p lant distribution of Frankliniella thrips is important for the development of reliable and cost effective sampling protocols the basis for all decision-making in IPM programs (Atakan et al. 1996). Furthermore, thrips are believed (M arois and Wright 2004) to spread F. verticillioides (until 1998, was known as F moniliforme ) which is associated with hardlock in cotton by infecting flowers. It is estimated that the diseas e reduced cotton yield in the Panhandle of Florida by about 50-60% in 2002 (Wright et al. 2003). There is little information on the impact of thrips densities on the epidemiology of th e disease. This can be evaluate d by determining the impact of population dynamics of thrips in field cotton to es tablish if there is any relationship between the insects and the disease. This is based on the premise that high thrips numbers would lead to more damage than low thrips numbers and therefore create more entry points for infection by the causal agent of the disease. Information on the population dynamics of th rips and their within-

PAGE 30

30 plant distribution may contribute to the understanding of the role thrips play in the spread of this disease. This may provide a clue to the developm ent of appropriate control strategies for thrips and subsequent control of cotton hardlock. The hypotheses to be tested in this study were: a) more adult and larval thrips occur in the flowers than on the leaves b) more adult and larval thrips o ccur in the upper than the lower canopy c) periods of peaked thrips populations in the flowers would result in high Fusarium infection, leading to high incidence of hardlock. Materials and Methods Field Plots Two separate field experim ents were conduc ted to investigate the above hypotheses. Cotton (DPL-555 BG/RR) was sown on two fiel ds at the North Florida Research and Education Center (NFREC), Quincy, FL, during the summer/fall 2005; one each at the Front Office Block (FOB) and Walshfield sites. In 2006 and 2007 summer/fall, only one field was maintained in Quincy at the Walshfield site NFREC branch at Mari anna (approximately 64 kilometers from Quincy) was included as a seco nd location. The Quincy fields were part of a cotton-peanut-bahiagrass rotation study while the Marianna field ha d previously only bahiagrass maintained on it. Cotton plants were gr own according to normal production practices recommended by the University of Florida Exte nsion Services unless otherwise stated. Cotton rows were planted in a north-sout h direction. The field in each site consisted of four blocks in a randomized complete block design. Each field measured 100.6 m x 14.6 m with block size of 18 m x 14.6 m and between block distan ce of 4.6 m. Blocks consisted of 16 rows of plants, with a row spacing of 0.9 m. Planting was done after the applic ation of 5-10-15 (N-P-K) fertilizer at 225 kg/ha three days prior to planting. In 2005, ins ecticides (Karate Thiophanate methyl and Orthene) were applied in 190 liters/ha of water, with a gas pressurized mechanical sprayer to

PAGE 31

31 control southern green stink bugs (Nezara viridula (L)) and brown stink bugs (Euschistus servus (Say)) on August 11 and August 26 in Quincy. Dimethoate at 0.28 liters/A was also applied on September 9 to control aphids that year. Apart fr om these, no pesticides were applied again in that year and none were app lied throughout the study period in 2006 and 2007 in both locations. Thrips Sampling During the first four weeks after seedlings had emerged, above-ground parts of 40 plants (10 plan ts from each block) from each field were randomly collected weekly into separate 1.9-L plastic containers containing 70% ethyl alcohol and take n to the laboratory for processing. Thrips were extracted and adult species and their sexes were determined and counted under a stereomicroscope at 40x based on their taxonomic features. Larval thrips were counted as a group. From the fifth week onward, two leaves fro m the upper, middle, and lower canopies of 40 plants (10 plants from each block) from each fi eld were sampled and subjected to the same laboratory procedure described above. One each of cotton square, white flower and boll from 40 plants from each field were also collected and put into separate 60 ml wide-mouth HDPE sample bottles containing 70% ethyl alcohol. Anot her set of flowers (one flower each from the upper and lower canopies two flowers from each pl ant) from 40 plants from each field were separately put into sample bottles. The flowers we re inverted when being placed in the sample bottles such that insects inhabiting them got di slodged and dropped to the bottom of the bottle. These were taken to the laboratory for processi ng. Extracted insects in each bottle were also treated as described above, w ith respect to identification, se xing and counting. Flowers were randomly checked under the stereomicroscope to make sure all thrips were extracted. Thrips Population Dynamics on Hardlock Severity On the day of sa mpling each week, 100 white flowers from 100 plants (one from each plant) from each field (25 flowers from each block) were tagged with ribbons of different colors

PAGE 32

32 to identify the day of each week. At the time of assessing hardlock, only 40 open bolls for a given tagging date (between 8 to12 bolls from each block) were used for the assessment. Tags with dates having <40 open bolls as a result of severe boll or flower a bortion were not included in the assessment of hardlock. In addition, th e general hardlock severity for the season was determined by rating 40 randomly selected plants ( 10 plants from each block) in each field. The total number of locules with hard lock was divided by the total numbe r of locks in all the bolls on all 40 plants, giving a percent index for each fi eld. Hardlock was assessed two weeks before harvesting was done. Yield was also obtained by harvesting bolls after evaluating for hardlock. Data Analysis The data were subjected to analysis usi ng SAS (8.1) GLM procedure and analysis of variance perfor med at the 5% probability level. Tukeys procedure was used to separate means, and linear regression analysis was used for th e relationship between thrips and hardlock. Results Population Dynamics of Franklinie lla Thrips Frankliniella spp., F bispinosa F fusca, F occidentalis and F tritici were recorded in both Quincy and Marianna. However, F tritici accounted for >98% of the adult thrips population at both locations. At both locati ons, relatively high densities of F fusca were recorded on the leaves only during the first four weeks and dec lined greatly and the population nearly became extinct after that period (Figs. 2-1, 2-2 and 2-3). Very low densities of F bispinosa and F occidentalis were recorded in 2005 and 2006 and no F bispinosa were recorded in 2007 in both locations. Generally, all but F fusca were recorded on cotton square s in both locations and their respective densities per square per week were <1 throughout the sampling period (Figs. 2-4, 2-5 and 2-6). Likewise, weekly mean densities of F occidentalis F bispinosa, and larval thrips per leaf and flower were <1 during th e entire life of the plant. Adult thrips inhabited the leaves when

PAGE 33

33 there were no flowers but preferred the latter to the former when blooming began. Thrips population began to increase rapidly on the plants with the on set of bloom, peaking around midseason which also coincided with peak bloom (F igs. 2-7, 2-8 and 2-9). In both locations, peak densities of adult F tritici in flowers occurred around late July to mid-August (Figs. 2-7, 2-8 and 2-9). The highest weekly mean densities of adult F tritici recorded per leaf were 0.28, 0.28, and 0.33 in 2005, 2006 and 2007, respectively in Quincy while 0.93 and 0.33 per leaf were recorded in 2006 and 2007, respectively in Marianna. Mean densities of 26.8, 19.6 and 18.28 per flower of this species were recorded in the flowers in 2005, 2006, and 2007, respectively in Quincy while 33.7 and 24.18 were recorded in 2006 and 2007 in Marianna. Only adult F tritici and larval thrips were recorded on the bolls, and their densities per boll per week were <1 (Figs 2-10, 2-11 and 2-12). More larval thrips than adult F tritici occurred in the bolls in 2006 and 2007 in both locations. Within Plant Distribution of Franklinie lla Thrips Adult thrips preferred the fl owers to other parts of the pl ant. There were significant differences (P < 0.0001) in the mean densities of F tritici on various parts of the plant. Their mean densities in the flowers, over time, were si gnificantly more than that on the leaves, squares and bolls (Tables 2-1, 2-2 and 2-3). On the contrar y, significant differences in the mean densities of larval thrips in the various plant parts were mixed. In cert ain weeks the mean densities of F bispinosa and F occidentalis were significantly higher in the flowers than in any other plant part; otherwise no significant differences were ob served for the rest of the sampling period. Only adult F tritici and larval thrips were found in all pa rts of the plants. With a few exceptions, >90% adult thrips occur in th e flowers during blooming, with varying proportions on the leaves, squares and bolls (Tables 2-4, 2-5, 2-6, 2-7 and 2-8) More larval thrips inhabited the leaves at the initial stages of bloom but this proportion declined in fa vor of the flowers with time.

PAGE 34

34 Significantly (P < 0.0001, F2, 948 = 7.38) more adult F tritici were found in the upper canopy than the midand lower canopies in Marianna in 2006 and (P = 0.0346, F2,828 = 1.91) in 2007 (Tables 2-9 and 2-10). A similar trend was observed in Quincy. However, no significant differences were observed between the mida nd lower canopies. There were no significant differences in the mean densities of F occidentalis in the canopies in bo th locations. There was significantly (P = 0.0062, F2,828 = 2.4) more larval thrips in th e upper canopy than the other two canopy levels in both 2006 and 2007 in Marianna; but mixed results were obtained in Quincy. There was also significantly (P < 0.0001, F1,312 = 9.35) more adult F tritici in the upper than the lower flowers in Quincy in 2006 and (P < 0.0001, F1,232 = 10.19) in 2007 (Table 2-11). Similar observations were made in Marianna (P <0.0001, F1,392 = 9.86) and (P <0.0001, F1,232 = 9.89) respectively in these two year s (Table 2-12). Similar observations were obtained at both locations with regard to th e larvae. Mean densities of F occidentalis were significantly (P < 0.0001) more in the upper than the lower flower s in 2006 but not in 2007 in both locations. Sex aggregation pattern of the indivi dual species in the various plant parts are shown in Tables 2-13 and 2-14, and the sex distributions for all four species combined are presented in Figs. 2-13, 214, 2-15 and 2-16. The ratio of male to female per leaf ranged from about 1:1 to 1:10 in Quincy, and was 1:1 in Marianna, while ratios of about 1: 1 to 1:2 were obtained for the flowers in the former location and 1:1 in the latter (Figs. 2-15 and 2-16). Thus significantly more adult females aggregated in the flowers in Quincy but that was not the case in Marianna. Population Dynamics of Thrips and Hardlock There appears to be no association between thrips population dynam ics and hardlock disease. In Quincy, no association was observe d between thrips densit ies over time and the disease with very low R2 value of 0.09 obtained in both 2006 and 2007 (Fig. 2-17). Both were not significant with P-values of 0.62 and 0.63. There was also no association between them in

PAGE 35

35 both years in Marianna. The R2 values in the respective year s were 0.04 and 0.18 with P-values of 0.53 and 0.47, respectively (Fig. 2-18). The R2 values improved slightly when a multiple linear regression was performed on the dates of sampling, thrips densities and hardlock but these were still not significant. Across the years a nd locations, removing one outlier in each case increased the R2 values but still tested not significant. When a lin ear regression was performed on the yearly hardlock incidence and yearly m ean thrips densities, again, no association was obtained. Years with higher thrips densities did not translate into higher hardlock incidence. Discussion Sim ilar trends in the populati on dynamics of thrips in both Quincy and Marianna suggest the similarities of the conditions prevailing in both locations. F tritici appears to be the most abundant Frankliniella species on cotton during the summer/fa ll season. This is consistent with Reitz et al. (2002) that F occidentalis is displaced by F tritici and F bispinosa as the spring season approaches. Our results showing that >90% of the adult population are found in the cotton flowers are consistent with the findings of Tavella et al. (1996), who repo rted that about 96% of adult Frankliniella spp. was found in pepper flowers. Howeve r, our results showed a consistently lower percentage of larval thrips inhabiting the leaves of cotton in cont radiction to their report that about 96% of the larvae were found on the leaves. Funderburk et al. (2000) and Ramachandran et al. (2001) reported findings sim ilar to Tavella et al. (2000). The low densities of F bispinosa observed in our study could be attributed to migration to other crop or wild plants, or to other areas. Certai n changes in local conditions might also be contributing to the declining population of this species. Low densities of both adults and larvae in th e squares and bolls could be due to greater risks of exposure to predation that they face in these structures; additionally, the availability of the more nutritious food source (pollen) in the flowers serves to attract most of them. Thrips may

PAGE 36

36 also prefer the leaves to the squares and bolls becau se the leaves serve as a more stable plant part for survival in terms of food. Gene rally more thrips were recorded in Marianna than in Quincy and the presence of very large co tton and peanut fields adjacent to our field in Marianna may account for this. That may have offered opportunity for constant migration of thrips between the two fields. The fact that significantly more adult and larval thrips were mostly found in the upper than the lower canopies appear s to support the idea plant host phenology may be an important factor in thrips abundance. The availability of more succulent leaves in the upper canopy could be supporting more thrips than older ones in the lower canopies. Simila rly, significantly more densities of thrips inhabiting the upper flowers appear to show the nature of flight in the insects, that flight just above the top canopy could be more common than th at around the lower canopies. It is unclear why there were signi ficantly more females than males of F tritici in the flowers in Quincy but not in Marianna, contributing to the 1:2 and 1:1 male:female ratios observed respectively in the two locations. But the presumed constant migration of thrips between our field and the neighbor ing large cotton and peanut fi elds in Marianna may have contributed to this. That observa tion in Quincy is consistent w ith the explanation by Terry (1997) on sex aggregation in thrips. It appears that, in ad dition to the fact that Frankliniella thrips normally tend to produce more females than ma les, we agree with (Terry 1997) that males generally tend to spread out to ot her parts of the plants after mating while females tend to remain in the flowers to feed on the more nutritious polle n for the development of the eggs. The ratio of adult male to female thrips per leaf in our st udy in Quincy appears consistent with Pearsall (2002). Migration may have also contributed to the 1:1 male to female ratio observed for the leaves in Marianna, thus most of th e adults recorded were immigrants.

PAGE 37

37 Thrips may be playing some role in the epid emiology of hardlock. The debate has actually been the extent to which thrips influence the disease incidence and sp read. The very low R2 values and the fact that statistical tests prove not significant (ver y low P-values) show little or no contribution of the insects to the disease incidenc e in cotton in this study. It appears some factors make more important contributions to the disease incidence and severity. The dates of sampling for thrips appeared important in the epidemiology of the disease. Environmental factors seem to influence the disease more, as explained by Mailhot (2007) in his model that night temperatures could be a reliable factor in pred icting the disease severity. Thrips may create entry points for the causal agent of the disease but infection may not occur for subsequent development of the disease if environmental factors are not favorab le. Thus, no matter the th rips density in the flowers at any given time the most important dr iving factor for the development of the disease may be environmental. We therefore suggest th at the epidemiology of the disease hinges on the activities of all cotton flower inhabiting insects and environmental factors. It may explain why hardlock severity does not really depend on the seasonal mean densities of thrips per flower. Thus high thrips densities per season do not directly mean seve re hardlock and vice versa.

PAGE 38

38 Table 2-1. Mean density of thrips on cotton l eaf, square, flower, and boll in the experiment conducted at two locations in Quincy, FL in 2005. Mean densities of thrips per plant part FOB Walshfield Date/insect Leaves Squares FlowersB olls Leaves SquaresFlowers Bolls 12-Jul F. tritici 0.03b 0.08b 5.48a 0b 0.05b 7.5a F. bispinosa 0a 0a 0a 0b 0b 0.5a Larval thrips 0.01a 0.01a 0.03a 0b 0b 0.03a 19-Jul F. tritici 0.01b 0.05b 10.3a 0.05b 0.05b 19.03a F. bispinosa 0a 0a 0a 0b 0b 0.13a Larval thrips 0.08b 0.15b 0.53a 0b 0b 0.48a 26-Jul F. tritici 0.03b 0.05b 16.13a 0b 0.08b 26.8a F. bispinosa 0a 0.05a 0.05a 0b 0b 0.3a Larval thrips 0.13a 0.1a 0.25a 0b 0.3b 1.28a 1-Aug F. tritici 0.03b 0.03b 3.73a 0b 0b 7.75a F. bispinosa 0a 0a 0.15a 0a 0a 0a Larval thrips 0.05a 0.03a 0.15a 0.08b 0.18ab 0.4a 8-Aug F. tritici 0.03b 0.03b 5.17a 0.05b 0.03b 0b 3.83a 0.05b F. bispinosa 0a 0a 0a 0a 0b 0b 0.03a 0a Larval thrips 0.03a 0a 0.03a 0.05a 0.08a 0.75a 0.18a 0.05a 15-Aug F. tritici 0.03b 0.03b 1.15a 0.03b 0b 0.05b 5.33a 0.03b F. bispinosa 0a 0a 0a 0a 0a 0a 0.03a 0a Larval thrips 0.25a 0b 0.05b 0.03b 0.05b 0b 0.3a 0.03b 22-Aug F. tritici 0.05b 0.03b 4.03a 0.05b 0.05b 0.03b 2.4a 0.05b Larval thrips 0.03b 0.03b 0.2a 0.03b 0.03b 0.03b 0.13a 0.03bMeans followed by the same letter(s) in a row within a location are not significantly different at the 5% probability level

PAGE 39

39 Table 2-2. Mean density of thrips on cotton l eaf, square, flower, and boll in the experiment conducted at Quincy and Marianna, FL in 2006. Mean densities of thrips per plant part Quincy Marianna Date/insect Leaves Squares/bollsFl owers LeavesSquares/bolls Flowers 5-Jul 21-Jul F. tritici 0.4b 0.21b 4.1a 0.43b 0.13b 9.3a F. bispinosa 0a 0a 0.03a 0a 0a 0a Larval thrips 0b 0.2a 0.18a 0b 0.13a 0b 12-Jul 28-Jul F. tritici 0.18b 0.4b 6.7a 0.42b 0.13b 10.3a F. bispinosa 0a 0a 0a 0a 0a 0a Larval thrips 0.05a 0a 0.05a 0b 0b 1.15a 19-Jul 4-Aug F. tritici 0.15b 0b 16.6a 0.15b 0b 17.3a F. bispinosa 0a 0a 0a 0a 0a 0a Larval thrips 0b 0.13a 0.53a 0b 0b 1.75a 26-Jul 11-Aug F. tritici 0.12b 0.13b 14.7a 0.22b 0.05b 33.7a F. bispinosa 0a 0a 0a 0a 0a 0a Larval thrips 0.15b 0.12b 3.53a 0.78a 0.05b 0b 1-Aug 18-Aug F. tritici 0.15b 0.01b 15.1a 0.12b 0b 11.4a F. bispinosa 0a 0a 0a 0.03a 0a 0a Larval thrips 0.03b 0b 1.63a 0.75a 0b 0b 8-Aug 25-Aug F. tritici 0.28b 0.03b 4.1a 0.2b 0.05b 11.4a F. bispinosa 0a 0a 0a 0a 0a 0a Larval thrips 0.25a 0b 0.13b 0b 0.13a 0.08b 15-Aug 1-Sep F. tritici 0.24b 0.02c 4.03a 0.28b 0.13b 18.5a Larval thrips 0b 0b 0.13a 0.03b 0.18a 0.02b Means followed by the same letter(s) in a row within a location are not significantly different at the 5% probability level.

PAGE 40

40 Table 2-3. Mean density of thrips on cotton l eaf, square, flower, and boll in the experiment conducted at Quincy and Marianna, FL in 2007. Mean densities of thrips per plant part Quincy Marianna Date/insect Leaves Squares/bollsFl owers LeavesSquares/bolls Flowers 12-Jul 9-Jul F. tritici 0.33b 0.07b 3.7a 0.33b 0.4b 4.43a F. occidentalis 0b 0.13a 0.13a 0a 0a 0a Larval thrips 0.18a 0.08b 0.08a 0.08a 0.1a 0.1a 19-Jul 16-Jul F. tritici 0.03b 0.1b 3.75a 0.23b 0b 22.48a F. occidentalis 0a 0a 0a 0a 0a 0a Larval thrips 0.05a 0a 0a 2.23a 0.06b 2.4a 26-Jul 23-Jul F. tritici 0.05a 0.06b 9.45a 0.13b 0b 4.65a F. occidentalis 0a 0a 0.03a 0.03a 0a 0a Larval thrips 0.73a 0.02b 0.08b 0.9a 0.1b 0.5ab 1-Aug 30-Jul F. tritici 0b 0.03b 5.6a 0.2b 0.06b 24.18a F. occidentalis 0.03a 0a 0a 0a 0a 0a Larval thrips 0.28b 0.63a 0.1b 0.58a 0.3b 0.43a 8-Aug 6-Aug F. tritici 0b 0b 7.43a 0b 0.08b 8.35a F. occidentalis 0a 0a 0a 0a 0a 0a Larval thrips 0.2b 0.13b 0.4a 0.8b 0.26a 0.1c 15-Aug 13-Aug F. tritici 0b 0.08b 18.28a 0.03b 0b 6.51a F. occidentalis 0a 0a 0a 0a 0a 0a Larval thrips 0.08b 0.08b 0.23a 0.03c 0.21b 0.46a 22-Aug 20-Sep F. tritici 0.02b 0.05b 6.88a 0b 0.08b 1.58a Larval thrips 0b 0b 0.08a 0b 0.05a 0.2a Mean followed by the same letter(s) in a row within a location are not significantly different at the 5% probability level.

PAGE 41

41 Table 2-4. Percentage of total thrips found on co tton leaves, fruiting structures, and flowers in the experiment conducted at Quincy, FL in 2005. % of total thrips Adult Larval thrips Date Leaves Fruiting structures Flowers Leaves Fruiting structures Flowers (Squares/bolls) (Squares/bolls) Jun 28 59.1 40.9 --------Jul 5 66.7 33.3 --------Jul 12 0.6 0.4 98.4 75.1 0.3 24.6 Jul 19 0.5 0.4 99.1 11.6 12.4 76 Jul 26 0.2 0.3 99.5 12.8 22.3 64.9 Aug 1 0.7 0.2 99.1 14.3 22.9 62.8 Aug 8 0.6 0.3 99.1 14.3 25.1 60.6 Aug 15 0.7 1 98.3 39.2 7.7 53.1 Table 2-5. Percentage of total thrips found on co tton leaves, fruiting structures, and flowers in the experiment conducted at Quincy, FL in 2006. % of total thrips Adult Larval thrips Date Leaves Fruiting structures Flowers Leaves Fruiting structures Flowers (Squares/bolls) (Squares/bolls) Jun 21 15.8 84.2 --71.4 28.6 --Jun 28 34.8 65.2 --65.2 34.8 --Jul 5 9.5 5 85.5 91.5 0.1 8.4 Jul 12 2.1 1.8 96.1 85.1 0.2 14.7 Jul 19 6.3 0.1 93.6 8.3 0.1 91.7 Jul 26 1.1 0 98.9 3.9 3.2 92.9 Aug 1 1.1 0.7 98.2 7.2 14.5 78.3 Aug 8 3.8 2 94.2 29.4 41.2 29.4

PAGE 42

42 Table 2-6. Percentage of total thrips found on co tton leaves, fruiting structures, and flowers in the experiment conducted at Marianna, FL in 2006. % of total thrips Adult Larval thrips Date Leaves Fruiting structures Flowers Leaves Fruiting structures Flowers (Squares/bolls) (Squares/bolls) Jul 7 32 68 --37.562.5 --Jul 14 75.9 24.1 --68 32 --Jul 21 1.6 1.6 96.8 99.1 0.4 0.5 Jul 28 8.7 6.5 84.8 12.1 3.4 84.5 Aug 4 2.6 0.1 97.2 18.642.5 38.9 Aug 11 3.8 0.4 95.8 12 63.9 24.2 Aug 18 1.7 38.4 98.1 11.833.3 44.1 Aug 25 1 0.1 97.3 27 8.2 27.8 Table 2-7. Percentage of total thrips found on co tton leaves, fruiting structures, and flowers in the experiment conducted at Quincy, FL in 2007. % of total thrips Adult Larval thrips Date Leaves Fruiting structures Flowers Leaves Fruiting structures Flowers (Squares/bolls) (Squares/bolls) Jul 2 43.7 56.3 --32.7 67.3 --Jul 9 7.7 1.8 90.5 43.7 37.5 18.8 Jul 16 0.6 2.5 96.9 66.7 0 33.3 Jul 23 0.5 8.8 90.7 87.8 3 9.2 Jul 30 0.4 0.4 99.2 27.5 62.5 10 Aug 6 0.1 0.1 99.8 28.6 17.8 53.6 Aug 13 0 23.3 99.7 28.5 42.8 28.7 Aug 20 0.1 4.5 99.9 20.0 0 80

PAGE 43

43 Table 2-8. Percentage of total thrips found on co tton leaves, fruiting structures, and flowers in the experiment conducted at Marianna, FL in 2007. % of total thrips Adult Larval thrips Date Leaves Fruiting structures Flowers Leaves Fruiting structures Flowers (Squares/bolls) (Squares/bolls) Jul 9 38.7 61.3 --41.358.7 --Jul 16 6.3 7.7 84 27.336.4 36.3 Jul 23 1.6 0 99 42.611.5 45.9 Jul 30 3.1 0 96.9 0.6 6.7 92.7 Aug 6 0.8 0.2 90 44.223.1 62.7 Aug 13 0.3 0.9 98.8 6.766.7 26.6 Aug 22 0.2 0 99.8 2.621.1 76.3 Aug 27 1.4 4.5 94.1 16.716.7 66.6 Table 2-9. The mean number of adult and larval Frankliniella species thrips inhabiting the upper, mid-, and lower canopies in experi ment conducted at Quincy, FL in 2006 and 2007. Mean number of thrips per leaf 2006 2007 Leaf position tritici occidentalis larvae tri tici occidentalis larvae Upper canopy 0.23a 0.03a 0.22a 0.03a 0a 0.1a Mid-canopy 0.04b 0.01a 0.12ab 0.01a 0a 0.09a Lower canopy 0b 0.06a 0.06b 0.01a 0a 0.04a Means followed by the same letter(s) in a column within a year are not significantly different at the 5% probability level. Table 2-10. The mean number of adult and larval Frankliniella species thrips inhabiting the upper, mid-, and lower canopies in experi ment conducted at Marianna, FL in 2006 and 2007. Mean number of thrips per leaf 2006 2007 Leaf position tritici occidentalis la rvae tritic i occidentalis larvae Upper canopy 0.69a 0.04a 0.53a 0.06a 0a 0.33a Mid-canopy 0.09b 0a 0.12b 0.03ab 0a 0.16b Lower canopy 0.04b 0a 0.07b 0.01b 0a 0.18b Means followed by the same letter(s) in a column within a year are not significantly different at the 5% probability level.

PAGE 44

44 Table 2-11. The mean number of adult and larval thrips Frankliniella species thrips inhabiting the upper and lower flowers in experime nts conducted at Quincy, FL in 2006 and 2007. Mean number of thrips per flower 2006 2007 Flower position tritici occide ntalis larval thrips tritici occidentalis larval thrips Upper 13.35a 0.02a 1.45a 7.56a 0.01a 0.19a Lower 5.25b 0.01b 0.15b 2.71b 0a 0.11b Means followed by the same letter(s) in a co lumn within a year are not significantly different at the 5% probability level. Table 2-12. The mean number of adult and larval thrips Frankliniella species thrips inhabiting the upper and lower flowers in experiment s conducted at Marianna, FL in 2006 and 2007. Mean number of thrips per flower 2006 2007 Flower position tritici occide ntalis larval thrips tritici occidentalis larval thrips Upper 20.49a 0.73a 0.87a 12.37a 0.01a 0.34a Lower 6.62b 0.05b 0.23b 2.23b 0a 0.15b Means followed by the same letter(s) in a co lumn within a year are not significantly different at the 5% probability level. Table 2-13. Mean densities of adult and larval Frankliniella thrips per plant part across 2005, 2006, and 2007 inhabiting the leaf, square, fl ower, and boll in experiments conducted at Quincy, FL. Leaf Square Flower Boll tritici female 0.09.01a 0.06.01a 5.49.19a 0.02.01a tritici male 0.02a 0.03.01a 3.19.17b 0.01.01a occidentalis female 0.01a 0a 0a 0a occidentalis male 0a 0a 0a 0a fusca female 0.03a 0a 0a 0a fusca male 0.01a 0a 0a 0a bispinosa female 0a 0a 0.03.01a 0a bispinosa male 0a 0a 0a 0a Larval thrips 0.43.03 0.06.01 0.47.04 0.02.01 Mean density (SEM). Means followed by same letter between sexes of a species within a column are not significant different at the 5% probability level.

PAGE 45

45 Table 2-14. Mean densities of adult and larval Frankliniella thrips per plant part across 2006 and 2007 inhabiting the leaf, square, flower and boll in experiments conducted at Marianna, FL. Leaf Square Flower Boll tritici female 0.07.01a 0.06.01a 6.51.26a 0.01.01a tritici male 0.08.01a 0.02.01a 6.29.29a 0.01.01a occidentalis female 0.02a 0a 0.15.04a 0a occidentalis male 0.01a 0a 0.09.04a 0a fusca female 0.07a 0a 0.01a 0a fusca male 0.01a 0a 0a 0a bispinosa female 0a 0a 0a 0a bispinosa male 0a 0a 0a 0a Larval thrips 0.27.01 0.11.01 0.59.06 0.11.03 Mean density (SEM).

PAGE 46

46 Thrips per leaf 0 1 2 3 4 M a y 1 8 M a y 2 5 J u n 2 J u n 9 J u n 1 5 J u n 2 1 J u n 2 8 J u l 5 J u l 1 2 J u l 1 9 J u l 2 6 A u g 1 A u g 8 A u g 1 5 A u g 2 2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0 2 4 6 8 10 F. tritici F. fusca F. occidentalis Larval thrips 2005 2006 2007 Figure 2-1. Mean de nsities (SEM) of Frankliniella thrips on cotton leaves from 18 May through 22 August of 2005, 2006 and 2007 in Quincy, FL. J u n 2 3 J u n 3 0 J u l 7 J u l 1 4 J u l 2 1 J u l 2 8 A u g 4 A u g 1 1 A u g 1 8 A u g 2 5 S e p 1 Thrips per leaf 0.0 0.5 1.0 1.5 2.0 2.5 F. tritici F. fusca F. occidentalis Larval thrips Figure 2-2. Mean de nsities (SEM) of Frankliniella thrips on cotton leaves from 23 June through September 1, 2006 in Marianna, FL.

PAGE 47

47 M a y 2 1 M a y 2 8 J u n 4 J u n 1 1 J u n 1 8 J u n 2 5 J u l 2 J u l 9 J U l 1 6 J u l 2 3 J u l 3 0 A u g 6 A u g 1 3 Thrips per leaf 0 1 2 3 4 5 F. tritici F. fusca F. occidentalis Larval thrips Figure. 2-3. Mean densities (SEM) of Frankliniella thrips on cotton leaves from 21 May through August 13, 2007 in Marianna, FL. Jul 7Jul 14Jul 21Jul 28Aug 4Aug 11Aug 18Aug 25 Thrips per square 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 F. tritici Larval thrips Figure 2-4. Mean de nsities (SEM) of Frankliniella thrips in cotton squares from 7 July through 25 August, 2006 in Marianna, FL. J u l 9 J u l 1 6 J u l 2 3 J u l 3 0 A u g 6 A u g 1 3 A u g 2 0 A u g 2 7 S e p 4 S e p 1 1 S e p 1 4 Thrips per square 0.0 0.1 0.2 0.3 0.4 F. tritici F. occidentalis Larval thrips Figure 2-5. Mean densities (SEM) of Frankliniella thrips in co tton squares from 9 July through 14 September, 2007 in Marianna, FL.

PAGE 48

48 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 F. tritici Larval thrips Thrips per square 0.0 0.1 0.2 0.3 0.4 0.5 2005 2006 J u n 2 8 J u l 5 J u l 1 2 J u l 1 9 J u l 2 6 A i g 1 A u g 8 A u g 1 5 A u g 2 2 S e p 3 S e p 1 6 0.0 0.2 0.4 0.6 0.8 1.0 1.2 F. tritici F. occidentalis Larval thrips 2007 Figure 2-6. Mean de nsities (SEM) of Frankliniella thrips in cotton squares from 28 June through 16 September of 2005, 2006 and 2007 in Quincy, FL.

PAGE 49

49 0 10 20 30 40 50 60 F. tritici F. bispinosa Larval thrips Thrips per flower 0 10 20 30 40 F. tritici F. bispinosa Larval thrips Jul 12Jul 19Jul 26Aug 1Aug 8Aug 15Aug 22Aug 29 0 10 20 30 F. tritici F. occidentalis Larval thrips 2005 2006 2007 Figure 2-7. Mean de nsities (SEM) of Frankliniella thrips in cotton flowers from 12 July through 29 August of 2005, 2006 and 2007 in Quincy, FL. Jul 7Jul 14Jul 21Jul 28Aug 4Aug 11Aug 18Aug 25 Thrips per flower 0 10 20 30 40 50 60 F. tritici Larval thrips Figure 2-8. Mean de nsities (SEM) of Frankliniella thrips in cotton flowers from 7 July through 25 August in 2006 in Marianna, FL.

PAGE 50

50 Jul 23Jul 30Aug 6Aug 13Aug 20Aug 27Sep 4Sep 11 Thrips per flower 0 20 40 60 80 F. tritici F. bispinosa Larval thrips Figure 2-9. Mean de nsities (SEM) of Frankliniella thrips in cotton flowers from 23 July through 11 September in 2007 in Marianna, FL. 0.00 0.05 0.10 0.15 0.20 0.25 F. tritici Larval thrips Thrips per boll 0.0 0.1 0.2 0.3 Aug 1Aug 8Aug 15Aug 22Aug 29 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 2005 2006 2007 Figure 2-10. Mean densities (SEM) of Frankliniella thrips on cotton bolls from 1 August through 29 August in 2005, 2006 and 2007 in Quincy, FL.

PAGE 51

51 Jul 14Jul 21Jul 28Aug 4Aug 11 Thrips per boll 0.0 0.1 0.2 0.3 0.4 0.5 F. tritici Larval thrips Figure 2-11. Mean densities (SEM) of Frankliniella thrips on cotton bolls from 14 July through 11 August in 2006 in Marianna, FL. Jul 16Jul 23Jul 30Aug 6Aug 13Aug 20Aug 27 Thrips per boll 0 1 2 3 4 5 F. tritici Larval thrips Figure 2-12. Mean densities (SEM) of Frankliniella thrips on cotton bolls from 16 July through 27 August in 2007 in Marianna, FL. 0 0.2 0.4 0.6 0.8 1 1.2 200520062007Thrips per lea f Female Male Figure 2-13. Mean densities (SEM) of Frankliniella thrips relating female and male occurrence in 2005, 2006 and 2007 in Quincy, FL.

PAGE 52

52 0 0.5 1 1.5 2 2.5 3 2006 2007Thrips per lea f Female Male Figure 2-14. Mean densities (SEM) of Frankliniella thrips relating female and male occurrence in 2006 and 2007 in Marianna, FL. 0 1 2 3 4 5 6 7 2005 2006 2007Thrips per flowe r Female Male Figure 2-15. Mean densities (SEM) of Frankliniella thrips relating female and male occurrence in 2005, 2006 and 2007 in Quincy, FL. 0 1 2 3 4 5 6 7 8 9 10 2006 2007Thrips per flowe r Female Male Figure 2-16. Mean densities (SEM) of Frankliniella thrips relating female and male occurrence in 2006 and 2007 in Marianna, FL.

PAGE 53

53 2006 y = 0.0028x + 0.2768 R2 = 0.09 p = 0.62 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0 5 10 15 20 25 Thrips per flowerProportion hardlocked 2007 y = 0.0087x + 0.1687 R2 = 0.09 p = 0.63 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 024681012 Thrips per flowerProportion hardlocked Figure 2-17. Regression of thrips with hardlock in Quincy, FL. 2006 y = 0.0088x + 0.0943 R2 = 0.14 p = 0.53 0 0.05 0.1 0.15 0.2 0.25 0.3 02468101214161820 Thrips per flowerProportion hardlocked 2007 y = -0.0024x + 0.1665 R2 = 0.19 p = 0.47 0 0.05 0.1 0.15 0.2 051015202530 Thrips per flowerProportion hardlocked Figure 2-18. Regression of thrips with hardlock in Marianna, FL.

PAGE 54

54 CHAPTER 3 PREDATION OF FRANKLINIELLA TRITICI BY ORIUS IN SIDIOSUS IN COTTON Introduction Controlling thrips using insectic ides is quite difficult and expensive. Researchers have been considering other strategies that could comp lement chemical control. Several workers have undertaken studies on the viability of the use of na tural enemies to control thrips. A considerable number of studies have also looke d at the spatial distri bution of thrips and some of their natural enemies on various crops. In emphasizing the impor tance of spatial distribution, Atakan et al. (1996) stated that the degree of spatial distribution between pe sts and their natural enemies on plants is likely to influence the ability of these natural enemies to suppress pest populations. Predation is one of the most diffi cult interspecific interactions to estimate (Stuart and Greenstone 1990). Generalist predators like O. insidiosus may prefer a prey based on the preys occurrence in the preferred habitat of the predator (Cl outier and Johnson 1993), and the vulnerability of the prey (Lang and Gsodl 2001). It appears more studies are needed on the pred atory role of anthocorid s against thrips. This is against the backdrop of reported variations in the success of this biological control agent in many of these studies. Mound and Teulon (1995) argued that there is little quantifiable information to indicate density-dependent regu lation of thrips under natural conditions, and population attributes of rapid colonization and gr owth are still thought to possibly outstrip the capacity of natural enemies to regulate thrips population. O insidiosus as an Effective Predator of Frankliniella Thrips Extensive research has documented the capacity of natural enemies such as species of Orius to suppress population of F. occidentalis (Higgins 1992, Chambers et al. 1993, Nicoli 1997, Funderburk et al. 2000, Ramach andran et al. 2001). In supporti ng the idea of the ability of

PAGE 55

55 anthocorid predators to suppress thrips populatio n, Tavella et al. (1996) noted that anthocorid predators can be effective biolog ical control agents of thrips in greenhouses. Similar observation was made by Sabelis and van Rijn (1997) that the anthocorid O insidiosus has the intrinsic ability to suppress a coherent population of F occidentalis at a predator to prey ratio of 1:217 on plant species. Experiments in greenhouses suggested that successful biological control of F. occidentalis with O. insidiosus is possible on chrysanthemum (Beekman et al. 1991). In fact, Orius spp. are known as specialized thrips predators (Riudavets 1995). O. insidiosus is a generalist predator, and its population dynamics in soybean fields have been linked to both thrips population levels and soybean flowering (Isenhour and Marston 1981). Both nymphs and adults have been observed eating soybean aphids in the field (Rutledge et al. 2004) Reitz et al. (2001) found that larvae of F. occidentalis are significantly more vulnerable to predation by O. insidiosus than the adults. Funderburk et al. (2000) reported that O insidiosus was an effective predator of F. occidentalis F. tritici, and F. bispinosa in the flowers of field grow n pepper. They explained that predation by O. insidiosus not only resulted in suppression of adult and larvae of F. occidentalis in pepper flowers, but also in a decline of the population towards extinction. The suppression occurred during the period of great est annual population abundance of F. occidentalis F. tritici and F. bispinosa, when adults of these thrips were rapidly colonizing the flowers. Populations of F. occidentalis increased in the absence of predation, a nd densities were very great in the plots treated with synthetic insecticides. Van den Meiracker and Ramakers (1991) also re ported an almost complete elimination of F. occidentalis by O. insidiosus and that the predator persisted on sweet pepper for almost 6 months in the absence of the prey. F. occidentalis was reported to have been successfully

PAGE 56

56 controlled in Canadian pepper a nd cucumber in glasshouse using O. tristicolor (Gilkeson et al. 1990, Tellier and Steiner 1990). Jindr a et al. (1991) noted that Orius sp. can reliably control several pests, including spider m ites, white-flies, aphids and thri ps. Ramachandran et al. (2001) reported a slightly different observation wh en he noted that while the adults of F. occidentalis and larval thrips were significantly suppressed by O. insidiosus there was no significant suppression of the adults of F. tritici and F. bispinosa. Biological contro l agents, such as predatory anthocorids Orius spp. can provide effective control of F. occidentalis but are not successful on all crops or under all situations (Dissevelt et al. 1995, J acobson 1997, Jarosi et al. 1997). Biossot et al. (1998) found that declines in population abundance of F. occidentalis were not related to unfavorable temper ature or other envir onmental factors and that an increase in abundance of a native Orius sp. may have been responsible. Orius spp. have been identified as the dominant predator in cotton in the Southern Blackla nds (Sansone and Smith 2001). It seems that in spite of their seemingl y effective suppression of thrips species, Orius spp. are usually not able to completely eliminate them. In pepper, Funderbur k et al. (2000) observed that the thrips never went into extinction by the predation of O. insidiosus no matter the population size of the predator. Kawai (1995) also found th at populations of T. palmi never went extinct despite suppression by O insidiosus For sustainable biological c ontrol, this aspect of the relationship between thrips and the predator is desirable in helping to maintain the population of the predator. Coll and Ridgway (1995) studies indicated that O. insidiosus searches less effectively for F. occidentalis on tomato than on bean and pepper plants. O. insidiosus as an Ineffective Pred ator of Frankliniella Thrips It is worth noting that these reports about th e success of O. insidiosus to suppress the population of Frankliniella spp. were on crops other than cotto n. Contrary to the above reported successes of anthocorid predators as agen ts for the control of the population of Frankliniella

PAGE 57

57 spp., Parrella and Lewis (1997) concluded in their findings that natural enemies play a rather insignificant role in regulating th rips in field crops. Hulshof et al. (2003) argued that biological control of thrips is not easily achieved, and its success depends on the crop. They noted, with reference to other studies that, in cucumber, predator populations of Neoseiulus cucumeris (Oudemans) [Acari: Phytoseiidae] and O laevigatus (Fieber) [Hemiptera: Anthocoridae] invariably declined after predator release, nece ssitating repeated releases to achieve sufficient control, whereas on sweet pepper O insidiosus population remained consta nt even in the absence of thrips prey (van den Meiracker and Ramakers 1991, Chambers et al. 1993, van de Veire and Degheele 1992). Other authors including Loomans et al. (1997) and Parrela and Lewis (1997) concluded that natural enemies must not be important. Mound (1997) argued that the population attributes of thrips outstripped th e capacity of natural enemies to suppress them. That is, the attributes of rapid colonization and growth we re believed to outstrip the capacities of natural enemies to regulate opportunistic species of thrips. Relatively few natural enemies of thrips have been identified. Loomans et al. (1997) proposed that th e large size of potential natural enemies restrict their entry into the preferred microhabitats of thrips. It is therefore worthwhile to study how O. insidiosus relates to Frankliniella thrips and to determine whether they are effective predators of the latter in field cott on. This information, if obtained, may form a basis in deciding on approp riate strategies for thrips and hardlock management. The hypotheses tested under this study were that: a) there is a positive association between Frankliniella thrips and O. insidiosus b) O. insidiosus is an effective predator of Frankliniella thrips in field cotton

PAGE 58

58 Materials and Methods Field Plots The experim ental fields de scribed in chapter 2 were also used for this study. Thrips and Orius Sampling During the first four weeks after the seed lings had em erged, above-ground parts of 40 plants (10 plants from each bloc k) from each field were randomly collected weekly into separate 1-L plastic containers containi ng 70% ethyl alcohol and sent to the laboratory for processing. Thrips were extracted and adult species and their sexes were determined and counted under a stereomicroscope at 40x based on thei r taxonomic features. The number of O. insidiosus was also noted. Larval thrips were counted as a group. From the fift h week onwards, two leaves from the upper, middle, and lower canopies of 40 plants (10 plants from each block) from each field were sampled and subjected to the same labor atory procedure describe d above. One each of cotton square, white flower and boll from 40 rando mly sampled plants from each field were also collected and put into separate 60 ml wide-m outh high density polyethylene (HDPE) sample bottles containing 70% ethyl alcohol. The flow ers were inverted when being placed in the sample bottles such that insects inhabiting th em get dislodged and drop to the bottom of the bottle. These were taken to the laboratory for processing. Extracted insects in each bottle were also treated as described above, with respect to identification, sexing and counting. Flowers were checked under stereomicroscope to make sure all thrips and O. insidiosus were extracted. Data Analysis The data were subjected to analysis usi ng SAS (8.1) GLM procedure and analysis of variance perfor med at the 5% probability le vel. Correlation was used to determine the association between thrips and the predator. Ort hogonal contrast was used to compare the degree of population suppression between adult and larvae by the predator.

PAGE 59

59 Results Findings of our study showed that cotton fl owers serve as host to the bulk of Frankliniella thrips recorded on the plant. Ot her arthropods that we found inha biting the flowers were m ites, aphids, bugs and ants. Frankliniella thrips that were found in the flowers in Quincy and Marianna were same as previously reported in chapter 2. It would therefore be reasonable to suggest that the most abundant sp., F. tritici probably constituted the bulk of the larvae too. In most of the weeks during the study, th e mean densities of the predator, O. insidiosus inhabiting the flowers were significantly more than those found in other plant parts. Generally, initial (first week of sampling) mean densities of O. insidiosus in the flowers were very low (<0.1 per flower) and continued to remain low until about the third or fourth week before increasing to >0.1 per flower and then began to decline again. Th is was in direct contrast to that of F. tritici, for which densities increased rapidly in th e flowers after the first week. The population fluctuations of the thrips species and the predator in the flowers are shown in Figs. 3-1, 3-2 and 3-3. It a ppears the population fluc tuations of the predator and prey are linked. The highest weekly mean density of F. tritici recorded in 2005 in Quincy was 26.8 per flower, recorded a week after the peak density of 0.23 per flower for O. insidiosus The population of O. insidiosus declined thereafter to almost ex tinction by the sixth week before rebounding to 0.05 per flower by the se venth week. In both 2005 and 2006, F. tritici reached its peak density per flower a week or two after peak population of the predator, but in 2007 both F. tritici and O. insidiosus attained peak densities of 18.28 and 0.08 per flower, respectively in the sixth week, and declined to 3.23 and 0.01, resp ectively by the eighth week. In Marianna, F. tritici reached its highest mean density per flower of 33.7 a week before that of O insidiosus (0.13 per flower) in 2006. However, there were simultaneous peak densities of 24.18 and 0.06

PAGE 60

60 per flower of F. tritici and the predator respectively in the fourth week in 2007. The mean densities of the adult of the other Frankliniella species were <1 per flower throughout the study. The mean densities of the predator and F tritici across years are show n in Table 3-1. The ratio of the nymphal O. insidiosus to adult range from 1:2.5 to 1:4 in Quincy and 1: 3 to 1:5 in Marianna. The ratio of male to female F tritici ranged from 1:1 to 1:2 in Quincy but was about 1:1 in Marianna. Sex aggregation of the thrips spp. and the predator in the flowers is presented in Table 3-1. The predator/prey ratio over time is shown in Tables 3-2 and 3-3. The predator/prey ratio did not show any consistent improvement ove r time in both locations. In 2005, the lowest predator/prey ratio was 1:200 (fir st week) and the highest 1:17 (eighth week) in Quincy. In 2006, the lowest was 1:1700 (fifth week) and highest 1: 5 (eight week) in Quincy, and lowest of 1:1900 (seventh week) and highest of 1:92 (fifth week) in Marianna. In 2007, the lowest ratio was 1:700 (seventh week) and 1:49 (fifth week) in Quincy, and lowest of 1:700 (sixth week) and highest of 1:170 (third week) in Marianna (Tables 3-2 and 3-3). The high predator/prey ratios obtained towards the end of the sampling period (sev enth and eighth weeks) in 2005 and 2006 was possibly due to very low number of thrips in the flowers during th at period as there were very few flowers available on the plan ts. We believe this situation caused migration of thrips to neighboring peanuts and cotton fields which were in their peak bloom, so the low thrips numbers were not as a direct result of predation by O. insidiosus Generally, thrips numbers were comparativ ely lower in 2005 and that may have also explained the relatively higher predator/prey ratios in that year in Quincy. A positive correlation was obtained between O. insidiosus and thrips with R2 value of 0.38 in 2005; but no relationship was obtained between them in 2006 and 2007 with R2 values of 0.01 and 0.08, respectively in

PAGE 61

61 Quincy (Table 3-4). Even though the correlation was significant in 2005 (P = 0.019), it was not so in 2006 and 2007 (P-values of 0.43 and 0.17, respectively). R2 values of 0.01 and 0.5 were obtained in 2006 and 2007, respectivel y in Marianna. While this wa s significant in 2007 (P = 0.008), it was not so in 2006 (P = 0.561). Cont rast comparison of the predation of F. tritici and the larval thrips by the predator, O. insidiosus was not significant even though it appeared the larvae populations were comparably more suppressed than adult F. tritici, considering the relatively low numbers of the larvae r ecorded throughout the study period. Discussion Cotton flowers appear to be a good host for F. tritic i and the gradual decline in the population of the other Frankliniella species as summer approaches adds to its success during this period in this part of the Southeast. Pred ator-prey relationships ar e often very difficult to estimate in the field. The ability of a predator to effectively regulate the population of the prey depends on several factors includi ng the initial populat ions of the predator and prey, their fecundity and the host plant arch itecture. Several researchers (( Higgins 1992, Chambers et al. 1993, Nicoli 1997, Funderburk et al. 2000, Ramacha ndran et al. 2001) have reported of the ability of O. insidiosus to effectively suppress Frankliniella spp. in some crops. In most of the reports about these successes, the predator consistently reduced the thrips population over time. The effectiveness of O. insidiosus to suppress Frankliniella thrips has been variable and depended on the spec ies involved. Predatory insects like O insidiosus seem to do better as insect population regulat ors in greenhouses as compared to field as was suggested by Tavella et al. (1996) and demons trated by Beekman et al. (1991). Environmental factors and microhabitat conditions within plants in the gr eenhouse are different than field conditions and these also impact on the behavior of both pr edators and prey. Tommasi and Nacoli (1993) reported that the predator had the capacity to consume 12.5 of F occidentalis per day and in

PAGE 62

62 most cases the population of the pr ey was less than this predation rate, allowing e ffective control of this species of Frankliniella thrips. Most succe sses were reported on F. occidentalis than F. tritici and F. bispinosa, and among these three species, F occidentalis is the least active. In our study with cotton, the most abundant species was F. tritici, for which the predator has not been very successful due to the evasive nature of this adult thrips species. The predator seems to be more successful in consuming larvae than the adults of any of the Frankliniella species, as was demonstrated by Funderburk et al. (2000), and the less mobility of the larvae may account for this. We did not obser ve consistent suppression of F tritici populations over time, with a characteristic dip and rise pattern in th e predator/prey ratio in most cases. Predator/prey ratios in many of the sampling weeks were far lower than 1:217 for which Sabelis and van Rijn (1997) cited as intrinsical capacity ratio for O. insidiosus to effectively suppress populations of F. occidentalis Our study agrees with the report by Ramachandran et al. (2001) about the significant suppression of adult F. occidentalis and larval thrips but not F. tritici and F. bispinosa The rapid build-up of the population or colonization of F. tritici in the flowers and the low densities of O. insidiosus in the flowers in the first and the subsequent two or three weeks may account for their unsuccessful suppression of population of adult F. tritici in our study. We agree with Mound and Teulon (1995) that rapid population attribut es of rapid colonization and growth of thrips as was shown in our study possibly outstrip the capacity of O insidiosus to effectively control thrips populations. The seeming preference and consumption of the larvae to the adult by the predator as was shown by the low numbers of larvae recorded in our study is consis tent with Ramachandran et al. (2001) and Reitz et al. (2006). Th e adults are more active and th erefore could more likely evade predation than the larvae. If predator/prey ratio can serve as a reliable predictor of predators like

PAGE 63

63 O. insidiosus to suppress prey like Frankliniella thrips in the field, then the ratios obtained in our study suggest that O. insidiosus may not be effective in suppressing populations of F. tritici in cotton even though we observed O insidiosus consuming thrips in the flowers in the field. This is especially so when thrips numbers ar e quite high to overwhelm the ability of O insidiosus to suppress them. It appears at low thrips numbers, the predator is somehow effectiv e but at high thrips densities, it loses that ability. Coll and Ridgway (1995) noted that O insidiosus searches less effectively for F. occidentalis on tomato than on bean and pepper plants. This supports the idea that O. insidiosus may not be effective in suppressing populations of all thrips species on all crops and in all in stances. It appears the association between O. insidiosus and Frankliniella thrips is weak. In c onclusion, the ability of O. insidiosus to suppress populations of Frankliniella thrips depends on the host crop and the Frankliniella species involved and that it wa s not effective in suppressing the population of F. tritici in cotton in our study.

PAGE 64

64 Table 3-1. Mean densities of Frankliniella thrips and O. insidiosus per flower across 2005, 2006 and 2007 at Quincy and Marianna, FL. Quincy Marianna 2005 2006 2007 2006 2007 Tritici female 5.95.38 6.26.364.11.22 7.97.38 5.42.34 Tritici male 3.19.33 2.64.173.83.36 8.18.43 4.39.37 occidentalis female 0 0.01.010.03.01 0.29.07 0 occidentalis male 0 0.01.01 0 0.18.07 0 Fusca female 0 0.01.01 0 0.01.01 0 Fusca male 0 0 0 0 0 bispinosa female 0.08.02 0 0.010.01 0 0 bispinosa male 0 0 0 0 0 Larval thrips 0.36.05 0.76.100.28.05 0.06.08 0.58.09 orius nymph 0.13.02 0.05.010.05.01 0.05.01 0.03.01 orius adult 0.03.01 0.02.010.03.01 0.01.01 0.01.00 Table 3-2. Weekly ratios of O. insidiosus to Frankliniella thrips inhabiting flowers in Quincy, FL. Week 2005 2006 2007 1 1:200 1:23 1:49 2 1:85 1:140 3 1:200 1:130 1:190 4 1:100 1:140 1:110 5 1:130 1:1700 1:160 6 1:420 1:240 7 1:50 1:71 1:700 8 1:17 1:5 1:330 Table 3-3. Weekly ratios of O. insidiosus to Frankliniella thrips inhabiting flowers in Marianna, FL. Week 2006 2007 1 1:190 2 1:380 1:620 3 1:180 1:170 4 1:420 1:410 5 1:92 1:210 6 1:230 1:700 7 1:1900 8 1:830

PAGE 65

65 Table 3-4. Covariation of Frankliniella tritici and O. insidiosus in Quincy and Marianna, FL Association Correlation (r) p-value Quincy 2005 Positive 0.62 0.019 Quincy 2006 None 0.11 0.430 Quincy 2007 None 0.28 0.170 Marianna 2006 None 0.06 0.561 Marianna 2007 Positive 0.71 0.008

PAGE 66

66 0 10 20 30 40 50 60 0.0 0.1 0.2 0.3 0.4 0.5 F. tritici F. bispinosa Larval thrips O. insidiosus Thrips per flower 0 10 20 30 40 O. insidiosus per flower 0.0 0.1 0.2 0.3 0.4 F. tritici F. bispinosa Larval thrips O. insidiosus 2005 2006 2007 Jul 12Jul 19Jul 26Aug 1Aug 8Aug 15Aug 22Aug 29 0 10 20 30 40 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 F. tritici F. occidentalis Larval thrips O. insidiosus 2007 Figure 3-1. Mean densities ((SEM) of Frankliniella thrips and O. insidiosus in cotton flowers from 12 July Through 29 August, in 2005, 2006 and 2007 in Quincy, FL. Jul 23Jul 30Aug 6Aug 13Aug 20Aug 27Sep 4Sep 11 Thrips per flower 0 20 40 60 80 O. insidiosus per flower 0.00 0.05 0.10 0.15 0.20 0.25 0.30 F. tritici Larval thrips O. insidiosus Figure 3-2. Mean densities ((SEM) of Frankliniella thrips and O. insidiosus in cotton flowers from 23 July through 11 September, 2006 in Marianna, FL.

PAGE 67

67 Jul 7Jul 14Jul 21Jul 28Aug 4Aug 11Aug 18Aug 25 Thrips per flower 0 10 20 30 40 50 60 O. insidiosus per flower 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 F. tritici Larval thrips O. insidiosus Figure 3-3. Mean densities ((SEM) of Frankliniella thrips and O. insidiosus in cotton flowers from 7 July through 25 August, 2007 in Marianna, FL.

PAGE 68

68 CHAPTER 4 FRANKLINIELLA THRIPS, COT TON HARDLOCK, COTTON SQUARES AND THEIR MANAGEMENT USING PESTICIDES Introduction Frankliniella Thrips, Hardlock an d Their Management Cotton suffers from many seedling diseases. Kirkpatrick and Rothro ck (2001) noted that boll rots ranks second to the seed ling disease complex as the most important disease of cotton in the US. They noted that plant disease epidemics from boll (fruit) rotting pathogens contribute to reduced yield and quality in cotton, G. hirsutum, in most years. Cotton hardlock is classified as a subset of boll rots (Hillocks and Brettell 1993) ; or that the two are lin ked. Jones et al. (2000) observed a positive correlation between bolls with seed rot symptoms and the occurrence of hardlock bolls at boll opening, fo r which case the reduction in yield because of boll rots is loss of entire bolls. Roncandori et al. (1975) and Padgett et al. (2003) em phasized that epidemics occur when excess moisture and humidity are present just before and duri ng boll opening (August to September). Although the symptoms associated with hardlock and rotten bolls may differ, both conditions reduce seed-cotton yield. Fusarium is associated with hardlock. The disease is believed to result from infection by F. verticillioides at the time of bloom (Marois et al. 2002, Wright et al. 2004). Wright et al. (2003) reported that the disease signifi cantly reduced cotton yield in the Panhandle of Florida in 2002. Recent reports indicate that this has been the situ ation year after year. Some insect species have been associated with the disease; notable among these are thrips and stin k bugs. Pinckard et al. (1981) suggested that insects can play a role in predisposing bolls to invasion by pathogens, and that feeding may damage carpel walls and locules of bolls to the extent that boll opening is slowed and often imperfect. For instance, Barbour et al. (1990) determined that the proportion of harvestable locules per boll decreased as the dur ation of infestation a nd the number of punctures

PAGE 69

69 per boll by green stink bug, Acrosternum hilare (Say), increased. The identification of hardlock disease as a contributing factor of yield reduction in cotton, especi ally in Florida and other parts of the southeastern US, has contri buted to the intensification of research efforts for its control. Even though some reports (Roncandori et al. 1975, Bell 1999) have played down the effectiveness of fungicides to co ntrol boll rots, some of these ch emicals appear promising for the management of cotton hardlock, especially if they are combined with insecticides. Seebold et al. (2004) argued that the application of fungicides during bloo m may provide a reduction of hardlock severity. It may be possible to focus on the earlier flowers or first 4 weeks of bloom since these contribute the most to yield in some production system (Jenkins et al. 1990). Interestingly, insecticides rather than fungicides appear to have the potential to offer a better management of the disease. The target in this case is the insects, possibly thrips or other flower visitors that supposedly help vect or the disease. Thrips control us ing insecticides has been quite difficult due to factors such as resistance and rapid recolonization of the plant. Some insecticide trials to control the diseas e have been conducted with varying successes and implications. Farrar and Davis (1991) reported that controlling thrips with insecticides was an effective way to manage corn ear rot also caused by F. verticillioides In one toxicity effect study of selected insecticides on thrips and their predators, Studebaker and Kring (2000) noted that spinosad and methoxyfenozide had no lethal or sublethal effects on insidious flower bug. Cyhalothrin also did not produce sublethal eff ects but did cause significant mortality in the flower bug. It may be important to consider the ab ove factors when applying insecticides so that natural enemies can complement insecticidal cont rol. Applying pesticides which have minimal or no adverse effect on predatory in sects can allow plants host a number of naturally occurring predatory species, including anthoc orid bugs. It is needless to indi cate that all these factors may

PAGE 70

70 have to be factored in the overall strategy for th e management of hardlock There is therefore the need to investigate the efficacy and toxicity of di fferent pesticides to verify whether they can be incorporated into an integrat ed program for the management of cotton hardlock disease. Pesticides and Abortion of Cotton Squares Cotton initiates its reproduc tive stage with the form ati on of flowers buds commonly referred to as squares. The number of squares in itiated and carried to the end determines the yield of the crop. Smith and Cothren (1999) not ed that about 60% of squares are normally aborted as a result of the inability of the crop to maintain all into mature bolls. In addition to this natural cause of the abortion of the squares, va rious environmental, insect pests and diseases problems contribute to square abortion. Since yi eld of cotton is determined by the final number of bolls at the end of the s eason, various attempts are usually made to reduce the number of squares that are aborted. Primarily, these attemp ts are targeted at the biotic factors by the application of pesticides since genetic and environmental ones are not easy to control. There is very little information on the effect of pesticides on the abortion of squares and it is against this background that this aspect was also addressed in this study. The hypotheses tested in this st udy were that: (a) cyhalothrin (K arate) was effective against Frankliniella thrips and therefore can help manage hardlock, and (b) Karate and Topsin (Thiophanate methyl) have effect on square abortion. Materials and Methods Field Plots Two separate field experim ents were conducte d to investigate the above hypotheses. Cotton (DPL-445 BG/RR) was sown on two fields at the Quincy and Marianna branches of North Florida Research and Education Center (NFREC), during the summer/fall of 2006 and 2007. Cotton plants were grown according to no rmal production practices recommended by the

PAGE 71

71 University of Florida Extension Services unless ot herwise stated. The field in each site consisted of four blocks in a randomized complete block design. Each field measured 121 m x 30.5 m with block size of 55.4 m x 5.5 m, and 6.4 m maintained between blocks A treatment plot measured 7.6 m x 5.5 m, consisting of 6 rows of plants, with a row spacing of 0.9 m. Sowing was done after the application of 5-10-15 (N-P-K) fertilizer at 225 kg/ ha three days prior to planting. Weekly applications of either: 1) Lambda-cyhalothrin (Karate) (insecticide) in 17 ml of Karate/3 gal of water, or 2) Thiophanate-methyl (Topsin-M) (fungici de) in 52 ml/3 gal of water, or 3) A combination of the tw o (fungicide + insecticide). There was a control treatment where there was no pesticide applicati on. Application of the pesticides started at early bloom when 50% of plants had one ope n bloom. The pesticides were applied using a CO2, backpack sprayer delivering 24 gal/A with one 8002 nozzle/row at 25 psi. Thrips and Orius Sampling, Hard lock and Yield Assessment Weekly sampling of five white flowers from each treatment plot started for thrips and O. insidiosus a week following the first application of the pesticides The sampling and processing techniques were same as described in chapte r 2. Hardlock and yield were also assessed and determined, respectively, as previously described in chapter 2. Effect of Pesticides on Square Abortion Two field experim ents maintained at Quincy and Marianna and described previously in this Chapter were used. A distance of 3 m was demarcated with flag s in each treatment plot from which aborted squares were collected into labele d plastic Ziploc bags every three days at the Quincy plot and twice per week from the Marianna plot.

PAGE 72

72 Data Analysis The data were subjected to analysis usi ng SAS (8.1) GLM procedure and analysis of variance perfor med at the 5% probability level. Results Thrips and O. insidiosu s Population Dynamics and Pesticide Treatments Thrips species identified in the flowers in all the treated plots in Quincy and Marianna in both 2006 and 2007 were F occidentalis, and F tritici. F fusca was rarely identified and no F. bispinosa were identified. The predator, O insidiosus was also identified in all the treatments in both years. The population fluctuations of these insects are shown in Figs. 4-1, 4-2, 4-3 and 4-4. F tritici, as usual, constituted the bulk (>98%) of the adult populat ion across treatments across years. Averaging thrips population across the two years, F occidentalis constituted <1%, 19%, 45% and <1% of the adult population respectivel y in the fungicide alone, insecticide alone, insecticide plus fungicide and th e control treatments in Quincy. In Marianna, <1%, 7%, 7% and 1% respectively were obtained in those treatmen ts. In Quincy the highest weekly mean density per flower of adult F tritici in all the treatments were record ed on a single day (August 1) (midseason); population densities of 18.1, 5.3, 8.5 and 18.1 per flower were recorded in the fungicide alone, insecticide alone, fungicide plus insecticide a nd control treatments, respectively on that date. A similar trend was observed in 2007 with th e peak mean densities occurring on August 15; with densities of 16, 3.5, 1.35 and 12.3 per flower in the respective treatme nts. However, peak weekly mean densities in the respective treatments were recorded on different dates in Marianna. Peak populations per flower in the insecticide alone and th e fungicide plus insecticide treatments occurred on August 6 (16 and 21.5 respectively) in 2006. Peak populations per flower in the fungicide alone and th e control treatments were 38.2 (August 13) and 56.4 (August 20). In 2007, population peaks recorded in the fungicide al one, insecticide alone and control treatments

PAGE 73

73 respectively were 25.9, 4.7 and 2.5 per flower on July 21 and 18.9 in the fungicide plus insecticide treatment on July 28. With a few exceptions, the mean population per flower of F occidentalis were <1 in both locations in the fungicide alone and insecticide alone treatments. Similar trends were observed for the larval thrips and O insidiosus Generally, more thrips and O insidiosus were recorded in Marianna than in Quincy. Thrips and Pesticide Treatments and Hardlock Thrips populations were greatly suppressed by insecticide treatm ents in both 2006 and 2007 in both locations. T here were significant (P <0.0001) differences in the effect of the pesticides on thrips population (Tables 4-1, 42, 4-3 and 4-4). Insectic ide treatments reduced thrips population in the flowers by 77 to 82% in Quincy and 75 to 80% in Marianna. These reductions in thrips numbers possibly resulted fro m avoidance of those tr eated plants (flowers) by thrips. There were no significant differences between the insecticide alone and the fungicide plus insecticide treatments on the population of adult F tritici and the larvae in both locations and years. There was also no significant differe nce between the fungicide alone and the control treatments. In 2007 significantly (P = 0.01) more F occidentalis were recorded in the insecticide alone and the fungicide plus in secticide treatments in both locations. A similar observation was made in 2006 in Marianna but it was mixed in Quincy. There were no significant differences in the effect of the treatments on the population of O insidiosus in both locations in both years. Thrips densities per flower based on sex are shown in Table 4-5. The ratios of male to female F tritici across years and treatments were 1: 1.3 and 1:1.2 in Quincy and Marianna, respectively and that of O insidiosus nymph to adult were 1:4.5 and 1:5, respectively. Suppression of thrips populati on reflected in reduction of ha rdlock. Significant (P <0.001) differences in hardlock incidence among treatmen ts were obtained in both locations and years (Tables 4-6 and 4-7). However, while there were significant differences (P < 0.001) in the yield

PAGE 74

74 in both years in Quincy, there were no significan t differences in 2007 in Marianna. In assessing the relationship between thrips and hardlock across treatments, strong relationships were obtained in Quincy (R2 =0.42, P = 0.0065 and R2 = 0.8124, P = 0.001) in 2006 and 2007 (Fig. 45). No relationship was obtained in Marianna (R2 = 0.3123, P = 0.024 and R2 = 0.1915, P =0.09) (Fig. 4-6). Pesticides treatments did not affect the abortion of cotton square (Table 4-8). Discussion Hardlock disease still rem ains an important dise ase of cotton and the role of thrips in its spread has been controversial. However, resu lts from our studies a nd Mailhot (2007) involving pesticide treatments targeting insects and part icularly thrips have shown that reducing populations of these insects usua lly results in reduction in th e disease severity. Insecticide applications have proven more effective in reducing the disease severity than fungicides and this may be due to the sharp drop in thrips population as a result of the treatments. Weekly applications of insecticides keep the thrips population lo w throughout the season. Hardlock incidence in the fungicide alone and control tr eatments was similar because it appears fungicide applications are not able to significantly reduce the amount of inoculum present (Roncandori et al. 1975, Bell 1999). Combining a f ungicide and an insecticide has been suggested by some researchers (Seebold et al. 2004) as one way that could provide a better management of the disease even though our results showed that application of insecticide alone was equally effective. Controlling thrips is quite difficult due to resistance and recolonization but combining insecticides application with predation by natura l enemies may be helpful, especially if the insecticide used has no lethal or sublethal e ffects on the natural enemy (Studebaker and Kring 2000). In our study, we observed no significant differences in the population of the predator, O insidiosus in the treatments, which indicates that its population may not have been significantly

PAGE 75

75 affected by the insecticide treatments. The rela tively low numbers of the predator in our study could have also contributed to this. Combining th is observation of low densities of the predator in the flower and the fact that there were no sign ificant differences in the treatment effect on its population densities, we think the marked suppression of the thrips population was largely due to the application of insecticides a nd not directly from predation by O insidiosus Nevertheless, it may be useful to explore combining chemical control and use of natural enemies in managing such diseases. One interesting observation in our study was the significantly high numbers of F occidentalis recorded in the insecticide treated plots, which was also reported by Funderburk et al. (2000), as this has implications fo r chemical control of this species. F occidentalis appears to be resistant to Lambda cyhalothrin. This obser vation may be a strategy by the species to avoid predation in the insecticide plots. However, it remains unclear whether such a trait in this species is inheritable. Reduction in hard lock resulting from insecticid e application sometimes did not translate into yield and the reas on is not clear. The fact that insecticide application does not always reflect in significant redu ction in hardlock and the fact th at it does not always translate into yield increase, opens the de bate as to when and how much of pesticides should be applied since farmers would want to maximize pr ofits by reducing cost of pesticides. Based on our results, we suggest pesticide a pplication to control the disease should be considered in line with thrips abundance during the cotton growing season. Pesticides (especially insecticides) could be applied when conditions have been projected to favor the disease development and applications may also target the first four weeks of bloom (Jenkins et al. 1990) since bolls set during this period contribute most to the yield. Mailhot (2007) showed that cool,

PAGE 76

76 moist conditions may favor hardlock and his mode l may help schedule fungicide and insecticide applications. The relatively high R2 values and the significant regr ession values obtained in the relationship between thrips numbers and hardlock in our studies in Quincy but not Marianna tend to suggest that thrips may be playing some role in the disease spread. Even though more studies are needed to establish the strength of this rela tionship, thrips seem to play some role in the epidemiology of the hardlock. Sin ce, without doubt, environmental factors appear to play a key role in hardlock development (Mailhot 2007), we suggest that for seasons where favorable weather conditions for the disease are forecast, a nd there is the possibility of having high thrips population, hardlock incidence may be high and would therefore warrant immediate plans for its management. The fact that no significant differences were observed in the number of squares aborted among treatments show that pesticides do not significantly protect squares from abortion. Thus square abortion is influenced basically by normal physiological and environmental factors.

PAGE 77

77 Table 4-1. Mean densities of Frankliniella species thrips and O insidiosus per flower as affected by pesticide treatment at Quincy, FL in 2006. Treatment F. tritici F. occidentalis Larval thrips O. insidiosus Adult thrips Fungicide 9.41a 0.03b 1.11a 0.08a 9.43a Insecticide 2.04b 0.05b 0.05b 0.07a 2.1b Both 2.94b 0.13a 0.08b 0.05a 3.07b Control 8.87a 0.01b 1.03a 0.08a 8.89a Means followed by the same letter(s) in a column are not significantly different at the 5% level. Table 4-2. Mean densities of Frankliniella thrips species and O insidiosus per flower as affected by pesticide treatment at Quincy, FL in 2007. Treatment F. tritici F. occidentalis Larval thrips O. insidiosus Adult thrips Fungicide 9.75a 0.0c 0.44a 0.15a 10.28a Insecticide 1.85b 0.46b 0.1b 0.11a 2.41b Both 0.94b 0.95a 0.08b 0.18a 2.0b Control 10.16a 0.03c 0.45a 0.21a 10.61a Means followed by the same letter(s) in a column are not significantly different at the 5% level. Table 4-3. Mean densities of Frankliniella thrips species and O insidiosus per flower as affected by pesticide treatment at Marianna, FL in 2006. Treatment F. tritici F. occidentalis Larval thrips O. insidiosus Adult thrips Fungicide 27.87a 0.65b 0.73a 0.07a 28.25a Insecticide 6.19c 2.56a 0.32b 0.04a 8.81c Both 10.18b 2.85a 0.41b 0.06a 13.09b Control 24.77a 0.41b 0.53ab 0.07a 25.32a Means followed by the same letter(s) in a column are not significantly different at the 5% level. Table 4-4. Mean densities of Frankliniella thrips species and O insidiosus per flower as affected by pesticide treatment at Marianna, FL in 2007. Treatment F. tritici F. occidentalis Larval thrips O. insidiosus Adult thrips Fungicide 13.53a 0.03b 0.84a 0.08a 14.4a Insecticide 2.28b 0.1a 0.04b 0.06a 2.3b Both 1.83b 0.14a 0.11b 0.03a 2.09b Control 11.37a 0.02b 0.57a 0.08a 11.78a Means followed by the same letter(s) in a column ar e not significantly differe nt at the 5% level.

PAGE 78

78 Table 4-5. Mean densities of Frankliniella species thrips and O insidiosus per flower across 2006 and 2007 in Quincy and Marianna, FL. Quincy Marianna tritici female 3.24.16 7.93.32 tritici male 2.52.16 6.67.32 occidentalis female 0.12.02 0.53.06 occidentalis male 0.06.01 0.32.04 fusca female 0 0.01.01 fusca male 0 0 bispinosa female 0 0 bispinosa male 0 0 larval thrips 0.45.05 0.54.04 orius adult 0.09.01 0.05.01 orius nymph 0.02.01 0.01 Mean density (SEM) Table 4-6. Effect of pesticides on hardlock and yield in Quincy, FL. % Hardlock Yield (kg/ha) Treatment 2006 2007 2006 2007 Fungicide 29.1b 40.4a 1245a 1264b Insecticide 20.8c 17.1b 1362a 1721a Both 19.9c 12.3b 1158ab 1966a Control 47.8a 39.1a 953b 1240b Columns followed by same letter(s) are not signi ficantly different at 5% probability level. Weight (kg/ha) include lint + seed. Table 4-7. Effect of pesticides on hardlock and yield in Marianna, FL. % Hardlock Yield (kg/ha) Treatment 2006 2007 2006 2007 Fungicide 6.0a 14.4a 1189ab 2083a Insecticide 1.0b 11.6bc 1363a 2007a Both 1.8b 8.6c 1213ab 1961a Control 7.8a 17.4a 1169b 1937a Columns followed by same letter(s) are not signi ficantly different at 5% probability level. Weight (kg/ha) include lint + seed.

PAGE 79

79 Table 4-8. Effect of pesticides on abortion of cotton squares per day with aborted cotton squares collected from 3 m of two rows of cotton plants in Quincy and Marianna, FL. Numb er of squares aborted per day Quincy Marianna Treatment 2005 2006 2006 2007 Fungicide 43.5a 56.5a 19.3a 64.6a Insecticide 44.3a 58.8a 20.3a 61.8a Both 42.3a 61.3a 18.7a 58.2a Control 45.9a 69.2a 22.0a 57.8a Columns followed by same letter(s) are not signi ficantly different at 5% probability level.

PAGE 80

80 0 10 20 30 40 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 F. tritici F. occidentalis Larval thrips O. insidiosus Thrips per flower 0 2 4 6 8 10 O. insidiosus per flower 0.0 0.1 0.2 0.3 0.4 0 2 4 6 8 10 12 14 16 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Jul 12Jul 19Jul 26Aug 1Aug 8Aug 15 0 10 20 30 0.0 0.1 0.2 0.3 Thiophanate-methyl Lambda-cyhalothrin Thiophanate-methyl + Lambda-cyhalothrin Untreated Figure 4-1. Mean de nsities (SEM) of Frankliniella thrips and O. insidiosus in cotton flowers as affected by pesticides from 12 July through 15 August 2006 in Quincy, FL.

PAGE 81

81 0 5 10 15 20 25 30 35 0.0 0.1 0.2 0.3 0.4 0.5 0.6 F. tritici F. occidentalis Larval thrips O. insidiosus Thrips per flower 0 2 4 6 O. insidiosus per flower 0.0 0.1 0.2 0.3 0.4 0.5 0 1 2 3 4 5 0.0 0.2 0.4 0.6 Jul 26Aug 1Aug 8Aug 15Aug 22 0 5 10 15 20 25 0.0 0.1 0.2 0.3 Thiophanate-methyl Lambda-cyhalothrin Thiophante-methyl + Lambda-cyhalothrin Untreated Figure 4-2. Mean de nsities (SEM) of Frankliniella thrips and O. insidiosus in cotton flowers as affected by pesticides from 26 July through 22 August 2007 in Quincy, FL.

PAGE 82

82 0 20 40 60 80 0.00 0.05 0.10 0.15 0.20 0.25 F. tritici F. occidentalis Larval thrips O. insidiosus 0 10 20 30 0.00 0.02 0.04 0.06 0.08 0.10 Thrips per flower 0 10 20 30 40 50 O. insidiosus per flower 0.0 0.1 0.2 0.3 0.4 0.5 Jul 23Jul 30Aug 6Aug 13Aug 20 0 20 40 60 80 100 0.0 0.1 0.2 0.3 0.4 0.5 Thiophanate-methyl Lambda-cyhalothrin Thiophanate-methyl + Lambda-cyhalothrin Untreated Figure 4-3. Mean de nsities (SEM) of Frankliniella thrips and O. insidiosus in cotton flowers as affected by pesticides from 23 July through 20 August 2006 in Marianna, FL.

PAGE 83

83 0 10 20 30 40 50 60 0.00 0.05 0.10 0.15 0.20 0.25 F. tritici F. occidentalis Larval thrips O. insidiosus Thrips per flower 0 2 4 6 8 O. insidiosus per flower 0.0 0.1 0.2 0.3 0 1 2 3 4 5 0.00 0.05 0.10 0.15 0.20 Jul 14Jul 21Jul 28Aug 4Aug 11 0 10 20 30 0.0 0.1 0.2 0.3 0.4 0.5 Thiophanate-methyl Lambda-cyhalothrin Thiophanate-methyl + Lambda-cyhalothrin Untreated Figure 4-4. Mean de nsities (SEM) of Frankliniella thrips and O. insidiosus in cotton flowers as affected by pesticides from 14 July through 11 August 2007 in Marianna, FL.

PAGE 84

84 A y = 0.0192x + 0.1538 R2 = 0.421 P = 0.00650 0.1 0.2 0.3 0.4 0.5 0.6 0246810121416Thrips per flowerProportion hardlocke d F I B C B y = 0.0303x + 0.0775 R2 = 0.8124 p = 0.0010 0.1 0.2 0.3 0.4 0.5 0.602468101214 Thrips per flowerProportion hardlocke d F I B C Figure 4-5. Relationship between thrips and hardlock on per plot basis in Quincy, FL. A) In 2006. B) In 2007. F = fungicide (Thiophanate -methyl), I = ins ecticide (Lambdacyhalothrin), B = (fungicide + insectic ide) and C = untreated control. Ay = 0.0023x 0.0153 R2 = 0.3123 p = 0.024 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0510152025303540 Thrips per flowerProportion hardlocked F I B C B y = 0.0056x + 0.08 R2 = 0.1915 p = 0.09 0 0.05 0.1 0.15 0.2 0.25 0.3 05101520 Thrips per flowerProportion hardlocke d F I B C Figure 4-6. Relationship between th rips and hardlock on per plot basis in Marianna, FL. A) In 2006. B) In 2007. F = fungicide (Thiophanate -methyl), I = ins ecticide (Lambdacyhalothrin), B = (fungicide + insect icide) and C = untreated control.

PAGE 85

85 Figure 4-7. Cotton boll with symptoms of hardlock.

PAGE 86

86 CHAPTER 5 SUMMARY AND CONCLUSIONS Flower thrips, Frankliniella bispinosa (Morgan ); tobacco thrips, Frankliniella fusca (Hinds); western flower thrips, Frankliniella occidentalis (Pergande) and eastern flower thrips Frankliniella tritici (Fitch) were identified in both Mari anna and Quincy on cotton. However, F tritici accounted >98% of the adult population. F fusca was identified on the leaves usually in the seedling stage of the cotton plant, and rarely after the fourth week of sampling. Mean densities of F occidentalis F bispinosa and the larval thrips pe r leaf were <1 during the sampling period. Thrips population increased rapidly with the ons et of bloom, peaking around mid-season which also coincided with the peak of bloom. Adult thrips (about 90%) prefer flowers to other plant parts. Only adult F tritici and the larvae were found in all the parts of the plant. There was significantly more adult and larval thrips in the upper ca nopy than the midand lower canopies in Marianna but was mixed in Quincy. There was also significantly (P < 0.0001, F1,312 = 9.35) more adult F tritici in the upper than the lower fl owers in both locations. Mean densities of F occidentalis was significantly (P < 0.0001) more in the upper than the lower flowers in 2006 but not in 2007 in both locations. Th e ratios of male to female per leaf ranged from about 1:1 to 1:10 in Quincy, and about 1:1 in Marianna, while ratios of about 1:1 to 1:12 and about 1:1 in the flowers were obtained in the former and latter locations, respectively. In Quincy, no association was observed between thrips densities over time and cotton hardlock, with very low R2 value of 0.09 obtained in both 2006 and 2007, and these were not significant. No association was obtained in Marianna either in both years and the respective R2 values were 0.04 and 0.18 (not significant).

PAGE 87

87 The very low R2 values and the fact that statis tical tests proved not significant (demonstrating no association) seem to show little or no contribution of the insects to the disease spread. It appears other factors ot her than thrips make more contributions to the epidemiology of the disease. A series of field studies were also conducted to estimate the predation of F tritici by the predator, O insidiosus In most of the sampling w eeks, the mean densities of O insidiosus inhabiting the flowers were significantly more than those f ound in the other plant parts. However, generally, mean densities of O insidiosus per flower were very low (<0.3 predator per flower) in all the sampling weeks. In Quincy, th e highest weekly mean densities per flower of F tritici recorded were 26.8, 19.6 and 18.28 in 2005, 2006 and 2007 and that of O insidiosus were 0.23, 0.18 and 0.08. Similarly, densities of 33.7 and 24.18 per flower were obtained in 2006 and 2007 and that of O insidiosus were 0.13 and 0.06. The ratio of the nymphal O insidiosus to adult ranged from 1:2.5 to 1:4 in Quincy and 1:3 to 1:5 in Marianna. The ratios obtained for male to female F tritici ranged from1:1 to 1:2 in Quincy and about 1:1 in Marianna. The lowest predator/prey ratio obtained in Quincy was 1:1700 and the highest 1:5. In Marianna, the lowest was 1:1900 and highest 1:92. No correlation was obtained between the predat or and the thrips with R2 values of 0.38, 0.01, and 0.08 in 2005, 2006 and 2007, respectively in Quincy. R2 values of 0.01 and 0.5 were obtained in 2006 and 2007, respectively in Marianna. There was not a consistent suppression of the population of F tritici over time. It appears no a ssociation exists between O insidiosus and Frankliniella thrips. The predator, O insidiosus was not effective in s uppressing the population of F tritici in cotton in our study.

PAGE 88

88 In the studies on the effect of pesticides on thrips population and hardlock, Frankliniella thrips identified in all th e pesticide treatments were F tritici and F occidentalis ; and the larvae. F tritici constituted >98% of the adult population. O insidiosus was also identified in all the treatments. F occidentalis constituted about 1% of the adu lt population in the fungicide alone and control treatments and 19 and 45% in the insecticide alone and the fungicide plus insecticide treatments in Quincy. In Marianna, they constitu ted about 1% in the fungicide alone and control treatments and about 7% in th e insecticide alone and fungicide plus insecticide treatments. Generally, more thrips and O insidiosus were recorded in Mari anna than in Quincy. Insecticide treatments reduced th rips populations by about 77 to 82% in Quincy and 75 to 80% in Marianna. The insecticid e alone and the fungicide plus insecticide treatments were significantly more effective than the fungicide alone and the contro l treatments in reducing thrips population and hardlock. The ratios of male to female F tritici across years and treatments were 1:1.3 and 1:1.2 in Quincy and Marianna, respectively and that of O insidiosus nymphs to adult were 1:4.5 and 1:5, respectively. In assessing the relationship between thrips and hardlock across years, a strong one was obtained in Quincy (R2 = 0.42, P =0.0065 and R2 = 0.81, P = 0.001) in 2006 and 2007. A relatively weaker relati onship was obtained in Marianna (R2 = 0.31, P = 0.024 and R2 = 0.19, P = 0.09). Weekly applications of insecticides which usually resulted in suppression of thrips population proved more effective in reducing the di sease than fungicides a pplication. Insecticide applications did not significan tly suppress the population of O insidiosus Based on our results, we suggest pesticide application to control hardlock should be c onsidered in line with weather forecast during the planting season. Pesticide applications could be done when conditions have been forecast to favor the disease, and may also target the first four weeks of bloom since bolls

PAGE 89

89 set during this period contribute most to the yield. Since, w ithout doubt, environmental factors especially night temperatures play a key role in hardlock development, we suggest that for seasons where good weather conditions for the diseas e are forecast, and there is the possibility of having high thrips population, hardlock incidenc e may be high and would therefore warrant immediate plans for its management.

PAGE 90

90 LIST OF REFERENCES Atakan, E., A. F. Ozgur, and U. Kersting. 1998. Frankliniella occidentalis (Thysano ptera: Thripidae) on cotton in Cukurova Re goin. Sixth Internat ional Symposium on Thysanoptera, 27 April-1 May 1998, Antalya, Turkey. Atakan, E., M. Coll, and M. D. Rosen, 1996. With in-plant of thrips and their predators: Effect of cotton variety and developmental stage. Bull. Entomol. Res. 86: 641-646. Atakan, E., and A. F. Ozgur. 2001. Pre liminary investigation on damage by Frankliniella intonsa to cotton in the Cukurova region of Turke y. Thrips and Tospoviruses. pp. 133-140. In Proceedings, 7th International Sympos ium on Thysanoptera, 25-28 January, 2001. Ankara, Turkey. Baez, I. 2002. Population dynamics of flower thrips (Thysanoptera: Thripidae). Frankliniella and predator Orius insidiosus (Say) in tomato and pepper crops M. S. Thesis, Florida A & M University, Tallahassee, Florida. Bagga, H. S., and C. D. Ranney. 1969. Boll rot po tential and actual boll rot in seven cotton varieties. Phytopathol. 59: 255-6. Barbour, J. S., J. R. Bradley, and J. S. Bachel er. 1990. Reduction in yield and quality of damage by green stink bug (Hemiptera: Pentat omidae). J. Econ. Entomol. 83: 842-845. Beekman, M., J. J. Fransen, R. D. Oettingen, and M. W. Sabelis, 1991. Differential arrestment of the minute pirate bug, Orius insidiosus (Say) ( Hemiptera: Anthocoridae), on two plant species. Mededelingen van de Faculteit La ndbouwweschappen Rijksuniversiteit Gent 56: 273-276. Bell, A. A. 1999. Diseases of cotton. pp. 553-593. In C. W. Smith et al. (ed) Cotton Origin, History, Technology, and Production. John Wiley and Sons. USA. Boissot, N. B., B. Reynaud, and P. Letourmy. 1998. Temporal analysis of western flower thrips (Thysanoptera: Thripidae) population dynami cs on Reunion Island. Environ. Entomol. 27: 1437-1443. Carter, F. L., N. P. Tugwell, and R. R. Phillip s. 1989. Thrips control strategy: Effects on crop growth, yield, maturity and quality. p p. 295-297. In Proceedings, Beltwide Cotton Production Research Conference. 2-7 January 1989, Nashville, TN. Chambers, R. J., S. Long, and N. L Helyer. 1993. Effectiveness of Orius laevigatus for the control of Frankliniella occidentalis on cucumber and pepper in the UK. Biocontrol and Technol. 3: 295-307. Chellemi, D. O., J. E. Funderburk, and D. W. Hall. 1994. Seasonal abundance of flowerinhabiting Frankliniella species (Thysanoptera: Thripi dae) on wild plant species. Environ. Entomol. 23: 337-342.

PAGE 91

91 Cho, K., J. F. Walgenbach, and G. G. Ke nnedy. 2000. Daily and temporal occurrence of Frankliniella spp. on tomato. Appl. Ento mol. and Zool. 35: 207-214. Cloutier, C., and S. G. Johnson. 1993. Predation by Orius tristicolor (Hemiptera: Anthocoridae) on Phytoseiulus persimilis (Acarina: Phytoseiidae) on tes ting for compatibility between biocontrol agents. Environ. Entomol. 22: 477-482. Coll, M., and R. L. Ridgway. 1995. Func tional and numerica l responses of Orius insidiosus (Heteroptera: Anthocoridae) to its prey in different vegeta ble crops. Ann. Entomol. Soc. of Am. 88(6): 732-738. Dissevelt, M., K. Altena, and W. J. Ra vensberg. 1995. Comparison of different Orius species for control of Frankliniella occidentalis in greenhouse vegetable cr ops in The Netherlands. Med. Fac. Landbauww. Univ. Gent. 60: 839-845. Eckel, C. S., K. Cho, J. F. Walgenbach, G. G. Kennedy, and J. W. Moyer. 1996. Variation in thrips species composition in field crops and implications for tomato spotted wilt epidemiology in North Carolina. Entomologia Experimentalis et Applicata 78: 19-29. Farrar, J. J., and R. M. Davis. 1991. Relationshi ps among ear morphology, western flower thrips, and Fusarium ear rot of corn. P hytopathol. 81: 661-666. Fritsche, M. E., and M. Tamo. 2000. Influence of thrips prey species on the life-history and behavior of Orius albidipennis. Entolomogia Experimentalis et Applicata 96 : 111-118. Funderburk, J., J. Stavisky, a nd S. Olson. 2000. Predation of Frankliniella occidentalis (Thysanoptera: Thripidae) in field pepper by Orius insidiosus (Hemiptera: Anthocoridae). Environ. Entomol. 29(2): 376-382. Funderburk, J. E., J. Stavisky, C. Tipping, D. Gorb et, T. Momol, and R. Berger. 2002. Infection of Frankliniella fusca (Thysanoptera: Thripidae) in p eanut by the Parasitic nematode Thripinema fuscum (Tylenchidae: Allantonemetidae) Environ. Entomo l 31(3): 558-563. Gangloff, J. L. 1999. Population dynamics and insecticides resistance of onion thrips Thrips tabaci Lindeman (Thysanoptera: Thripidae) in onion. PhD dissertation, Cornell University, Ithaca, NY. Gilkeson, L. A., W. D. Morewood, a nd D. E. Elliot. 1990. Current stat us of biological control of thrips in Canadian greenhouses with Amblyseius cucumeris and Orius tristicolor IOBC/WPRS Bulletin 13: 71-75. Gillespie, D. R., and R. S. Vernon. 1990. Trap catch of western flower thrips (Thysanoptera: Thripidae) as affected by color and hei ght of sticky traps in mature greenhouses cucumber crops. J. Econ. Entomol. 83: 971-975.

PAGE 92

92 Gonzalez, D., and L. T. Wilson. 1982. A food web approach to economic threshold: A sequence of pest/ predaceous arthropods on California cotton. Entomophaga 27: 31-43. Graves, J. B., J. D. Powell, M. E. Farris, S. Micinski, and R. N. Story. 1987. Pest status of western flower thrips on cotton in Loui siana. pp. 229-231. In Proceedings, Beltwide Cotton Research Production Conference. National Cotton council and the Cotton Foundation. 4-7 January 1987, Dallas, TX. Hansen, E. A., J. E. Funderburk, R. R. Stuart, S. Ramachandran, J. E. Eger, and H. Mcauslane. 2003. Within-Plant distribution of Frankliniella species (Thysanopter a: Thripidae) and Orius insidiosus (Heteropera: Anthocoridae) in field pepper. Population Ecology 32 (5) : 1035-1044. Higgins, C. J. 1992. Western fl ower thrips (Thysanoptera: Thripidae) in greenhouses: Population dynamics, distribution on plan ts and association with predators. J. Econ. Entomol. 85: 1891-1903. Hillocks, R. J. (ed) 1992. Cotton diseases CAB International. Wallingford, UK. 415 pp. Hillocks, R. J., and J. H. Brettel. 1993. The association between honeydew and growth of Cladosporium herbarum and other fungi on cotton lint. Tropical Sci. 32(2): 121-129. Hollingsworth, R. G., K. T. Sewake, and J. W. Armstrong. 2002. Scouting methods for detection for of thrips (T hysanoptera: Thripidae) on Dendrobium orchids in Hawaii. Environ. Entomol. 31(3): 523-532. Hulshof, J., E. Ketoja, and I. Vannine n. 2003. Life history ch aracteristics of Frankliniella occidentalis on cucumber leaves with and wit hout supplemental food. Entomologia Experimentalis et Applicata 108: 19-32. Isenhour, D. J., and N. L. Martson. 1981. Seasonal cycles of O. insidiosus (Hemiptera: Anthocoridae) in Missouri soybeans. J. Kans. Entomol. Soc. 54: 129-142. Jacobson, R. J. 1997. Integrated pest management (IPM) in glasshouses. pp. 639-666. In T. Lewis (ed) Thrips as crop pests. CAB, New York. Jarosi, V., M. Kalias, L. Lapchin, J. Rochat, and A. F. G. Dixon. 1997. Seasonal trends in the rate of population increase of Frankliniella occidentalis (Thysanoptera: Thripidae) on cucumber. Bull. Entomol. Res. 87: 487-495. Jenkins, J. N., J. C. McCarty, and W. L. Parrott. 1990. Effectiveness of fruiting sites in cotton: Yield. Crop Sci. 30: 365-369. Jindra, Z., V. Taborssky, and P. Skoda. 1991. Spontaneous occurrence of a predatory bug Orius majusculus (Reut.) in glasshouses. Ochrany Rostlin (3-4): 207-209.

PAGE 93

93 Jones, M. A., J. D. Mueller, D. A. Kleupfel, M. J. Sullivan, J. T. Walker Jnr., M. E. Roof, J. M Stewart, and D. E. Linvill (eds). 2000. Pre liminary investigation on cotton seed rot in South Carolina. Clemson Univ. Stn. Bull. 675. 21 pp. Kawai, A. 1995. Control of Thrips palm i Karny (Thysanoptera, Thripidae) by Orius spp. (Heteroptera: Anthocoridae) on greenhouse aubergine. Appl. Entomol. Zool. 30: 1-7. Kiers, E., W. J. de Kogel, A. Balkema-Boomst ra, and C. Mollema. 2000. Flower visitation and oviposition behavior of Frankliniella occidentalis (Thysanoptera: Thripidae) on cucumber plants. J. Appl. Entomol. 124: 27-32. Kirk, W. D. J. 1997. Thrips feeding. pp. 21-29. In T. Lewis (ed) Thrips as crop pests. CAB. International. New York. Kirkpatrick, T. L., and C. S. Rothroc k. 2001. Compendium of cotton diseases, 2nd ed. Amer. Phytopathol..Soc. St. Paul, MN. Lang, A., and S. Gsodl. 2001. Prey vulnerability and active predator choice as determinants of prey selection: a carabid beetle and its prey. J. Appl. Entomol. 125 (1-2): 53-61. Leonard, B. R., J. B. Graves, and P. C. Ells worth. 1999. Insect and Mites. pp 553-593. In Smith, C. W. et al. (ed) Cotton Origin, Hist ory, Technology, and Production. John Wiley and Sons. USA. Leser, J. F. 1985. Thrips management: Probl ems and Progress. pp. 175-178. In Proceedings, Beltwide Cotton production Research Conf erence. 6-11 January, New Orleans, LA. Lewis, T. 1973. Thrips, their biology, ecology an d economic importance. Academic Press, London. Lewis, T. 1997. Chemical control. pp. 567-594. In Lewis T (ed) Thrips as Crop Pests. CAB International, Wallingford, UK. Loomans, A. J., M. T. Murai, and I. D. Gree n. 1997. Interactions with hy menopteran parasitoids and parasites nematodes. pp. 355-398. In T. Lewis (ed) Thrips as Crop pests. CAB International, Wallingford, UK. Mailhot, D. J. 2007. Relationship of flower thrips to hardlock of cotton. PhD. Dissertation, University of Florida. Florida, USA. Marois, J. J., and D. L. Wright. 2004. Etiology, epidemiology, and control of Fusarium hardlock of cotton in the southeast: the possibilitie s. pp.331-336. In Proceedings, Beltwide Cotton Production Research Conference, 59 January 2004, San Antonio, TX.

PAGE 94

94 Marois, J.J., D.L. Wright and P.J. Wiatrak. 2002. Association of Fusarium sp. with hardlock of cotton in the southeastern U.S. Proceedings, Beltwide Cotton Production Research Conference, 6-10 January 2002, Nashville, TN. Marullo, R. 1997. The relationship between sa prophagy and phytophagy in thrips. Bull. Lb. Agric. Filippo Silvestri.53: 19-24. Mauney, J. R. 1968. Morphology of the cotton plant. pp. 23-40. In F. C. Elliot et al. (ed) Cotton: Principles and Practice. Iowa St ate University Press, Ames, IA, McLean, K. S., and G. W. Lawrence. 1998. Premat ure boll rot associated with cotton in Louisiana. Plant Disease 82: 7-11. Michelakis, S. E., and A. Am ri, 1997. Integrated control of Frankliniella occidentalis in CreteGreece. Bull. OILB/SROP 20(4): 169-176. Mound, L. A. 1997. Biological Diversity. pp. 197-216. In T. Lewis (ed) Thrips as Crop Pests. CAB International, New York. Mound, L. A., and D. A. J. Teulon. 1995. Thysanopt era as phytophagus opport unists. pp 3-19. In B. L. Parker et al. (ed) Thrips Biolog y and Management. Plenum Press, New York. Newson, L. D., and J. R. Brazzel. 1968. Pests and their control. pp. 367-405. In E. C Elliot et al. (ed) Advances in Production and Utilization of Quality Cotton: Principles and Practices. Iowa State Univ. Press, Ames, IA. Nicoli, G. 1997. Biological control of exotic pests in Italy: Recent experiences and Perspectives. Bull. OEPP 27: 68-75. Padgett, G. B., D. J. Boquet, W. Rea, J. P. Ca ylor, and C. Shivers. 2003. Boll rot epidemics in Louisiana cotton. p. 211. In D. Richter (ed.) Proceedings, Beltwide Cotton Conference. 69 January 2003, Nashville, TN. Parella, M. P., and T. Lewis. 1997. Integrated pest management (IPM) in field crops. pp. 595614. In T. Lewis (ed) Thrips as Crop Pests. CAB Wallingford, Oxon, Illinois. Pearsall, I. A., and J. H. Myers. 2000. Popul ation dynamics of western flower thrips (Thysanoptera: Thripidae) in nectar ine orchards in British Columbia J. Econ. Entomol. 93(2):264-275. Perlak, F. J., R. W. Deaton, R. L. Armstrong, R. L. Fuchs, S. R. Sims, J. T. Greenplate, and D. A. Fischhoff. 1990. Insect resistan t cotton plants. Biotech. 8: 939-943. Pfleger, F. L., M. E. Ascerno, and R. Wawrzynski. 1995. Tomato spotted wilt virus. Minnesota Commercial Flower Gr owers Bull. 44(2): 1-7.

PAGE 95

95 Pickett, C. H., L. T. Wilson, and D. Gonzal ez. 1988. Population dynamics and within plant distribution of the western fl ower thrips (Thysanoptera: Thripidae), an early season predator of spider mites infesting cotton. Environ. Entomol. 17(3): 551-59. Pinckard, J. A., L. J. Ashworth, J. P. Snow, T. E. Russell, R. W. Roncadori, and G. L. Sciumbato. 1981. Boll rots, p. 20-24. In G. M. Watkins (ed.) Compendium of cotton diseases. Am. Phytopathol Soc. St. Paul, MN. Puche, H., R. D. Berger, and J. E. Funderburk. 1995. Population dynamics of Frankliniella species (Thysanoptera: Thripidae) thrips and progress of spotted wilt in tomato fields. Crop Protection 14: 577-583. Ramachandran, S., J. E. Funderburk, J. Stav isky, and S. Olson. 2001. Population abundance and movement of Frankliniella species and Orius insidiosus in field pepper. Agric. and Forest Entomol. 3: 129-137. Reitz, S. R., E. L. Yearby, J. E. Funderburk, J. Stavisky, M. T. Momol, and S. M. Olson. 2003. Integrated Management Tactics for Frankliniella Thrips (Thysanoptera: Thripidae) in field-grown Pepper. J. Econ. Entomol. 96: 1201-1214. Riudavets, J. 1995. Predators of Frankliniella occidentalis (Perg.) and Thrips tabaci Lind.: a review. Wageningen Agric. Univ. Paper 95: 43-87. Rutledge, C. E., R. J. ONeil, T. B. Fox, and D. A. Landis. 2004. Soybean aphid predators and their use in IPM. Ann, Entomol. Soc. Am. 97: 240-248. Roncadori, R. W., S. M. McCarter, and J. L. Crawford. 1975. Evaluati on of various control measures for cotton boll rot. Phytopathol. 65: 567-570. Sabelis M. W., and P. C. J. van Rijn. 1997. Pr edation by insects and mites. pp 259-354. In T. Lewis (ed) Thrips as crop pests. CAB Wallingford, Oxon, UK. Salguero-Navas, V. N., J. E. Funderburk, R. J. Beshear, S. M. Olson, and T. P. Mack. 1991. Seasonal patterns of Frankliniella spp. (Thysanoptera: Thripidae) in tomato flowers. J. Econ. Entomol. 84 (6): 1818-1822. Snow, J. P., and D. E. Sanders. 1979. Role of abscised cotton flowers, bolls and squares in production of inoculum by boll-rotting Fusarium spp Plant Disease Reporter 63: 288289. Sansone, C. G., and J. W. Smith, Jr. 2001. Indentifying predation of Helicoverpa zea (Lepidoptera: Noctuidae) eggs by Orius spp. (Hemiptera: Anthocoridae) in cotton by using ELISA. Environ. Entomol. 30(2): 431-438.

PAGE 96

96 Seebold, K., R. C. Kemerait, J. Marois, D. Wright B. Padgett, J. Mueller, K. McLean, and T. Estes. 2004. Results from the 2003 regi onal hardlock project. pp. 337-340. In Proceedings, Beltwide Cotton Production Rese arch Conference. Na tional Cotton Council Meetings, 6-10 January 2004, San Antonio, TX. Smith, C. W., and J. T. Cothren (eds). 1999. Cotton: Origin, History, Technology, and Production. John Wiley and Sons Inc. USA. 850. Spanicht, R. H., and M. W. Roncardi. 1972. Fusarium boll rot of cotton: pathogenecity and histopathology. Phytopathol. 62: 1381-1386. Stavisky, J., J. E. Funderburk, B. V. Brodbeck, S. M. Olson, and P. C. Andersen. 2002. Population dynamics of Frankliniella spp. and tomato spotted wilt incidence as influenced by cultural management tactic s in tomato. J. Econ. Entomol. 95(6): 1216-1221. Stuart, M. K., and M. H. Greenstone. 1990. Beyond ELISA: a rapid sensitive, specific Immunodot assay for identification of pr edator stomach contents. Ann. Entomol. Soc. Am. 83: 1101-1107. Studebaker, G. E., and T. J. Kri ng. 2000. Lethal and sublethal effect s of early season insecticides on insidious flower bug ( O. insodiosus ): An important predator in cotton. AAES Special Report 198. Proceedings, 2000 Cotton Research Meeting. Tappan, W. B. 1986. Relationship and sampling time to tobacco thrips (Thysanoptera: Thripidae) numbers in peanut foliage beds a nd flowers. J. Econ. Entomol. 79: 1359-1363. Tavella, L., A. Arzone, and A. Alma. 1996. Researches on Orius Laevigatus ( Fieb), a predator of Frankliniella occidentalis (Perg.) in greenhouse. A pr eliminary note. IOBC/WPRS Bull. 14: 65-72. Tellier, A. J., and M. Y. Steiner. 1990. Control of the western flower thrips, Frankliniella occidentalis (Pergande), with a native predator Orius t risticolor (White) in greenhouse cucumbers and peppers in Alberta, Canada. OIBC/WPRS Bull. 13(5):209-211. Terry, L. I. 1997. Host selection, communica tion and reproductive behavior. pp. 65-118. In T. Lewis (ed) Thrips as crop pest s. CAB International, New York. Teulon, D. A. J., and D. R. Penman. 1991. Effect s of temperature and diet on oviposition rate and development time of New Zealand flower thrips, Thrips obscuratus Entomol Expt. Appl. 60: 143-155 Tharp, W. H. 1965. The cotton plant. How it grow s and why it growth varies. Agric. Handbook No. 178, USDA, Washington, D. C.

PAGE 97

97 Toapanta, M., J. Funderburk, S. Webb, D. Chellemi, and J. Tsai. 1996. Abundance of Frankliniella spp. (Thysanoptera: Thripi dae) on winter and spri ng host plants. Environ. Entomol. 25: 793-800. Tommasini, M. G., and G. Nico li. 1993. Adult activity of four Orius species reared on two preys. Bull. Int Organ. Biological Contro l/Western Pacific Region 16: 181-184. Umbeck, P., G. Johnson, K. Barton, and W. Swain. 1987. Genetically transformed cotton ( Gossypium hirsutum L.) plants. Biotech. 5: 263-266. van den Meiracker, R. A. F., and P. M. Ramakers. 1991. Biological control of the western flower thrips, Frankliniella occidentalis on sweet pepper with th e anthocorid predator Orius insidiosus Meded. Fac. Landbouwet. Rijk s-univ. Gent 56/2a: 241-249. van de Veire, P., and D. Degheele. 1992. Biological control of the western Flower thrips, Frankliniella occidentalis (Pergande), (Thysanoptera: Th ripidae), glasshouse sweet pepper with Orius spp (Hemiptera: Anthocoridae). A comparative study between O. niger (Wolff) and O. insidiosus (Say). Biocontrol Sci. Tech. 2: 281-283. Watkins, G. M. (ed) 1981. Compendium of cott on diseases. Am. Phytopathol. Soc. St. Paul, MN. Wright, D. L., J. J. Marois, M. A. Vargas, a nd P. J. Wiatrak. 2003. Manage ment of hardlock in cotton in the Southeast. pp. 2076-2080. In Proc eedings, Beltwide Cotton Conference. 6-9 January. Nashville, TN. Wright, D. L., J. J. Marois, P. J. Wiatrak. 2004. History and overview of the hardlock problem in humid areas of the deep south. pp. 328-330. In Proceedings, Beltwide Cotton Production Research Conferen ce. National Cotton Council, 6-10 January. San Antonio, TX.

PAGE 98

BIOGRAPHICAL SKETCH Enoch Adjei Osekre was born in 1965 in Teshie-Accra, Ghana. He earned both his B.Sc. in Crop Science and his MPhil. in Insect Science from the Universi ty of Ghana, in 1990 and 1998, respectively. After earning his first degree, he worked with the Ghana Education Service as a Tutor at Teshie Presbyterian Senior High Sc hool until 1996, when he left to pursue the second degree. Enoch also worked as a Research Scientist with the Plant Genetic Resources Research Institute at Bunso, Ghana, from 1998 to 2004. In summer 2004, he entered the University of Florida in the Department of Agronomy on an a ssistantship to pursue a PhD degree. He worked on the impact of the population dynamics of thrips on cotton hard lock disease. After graduation, he plans to remain in Agricultural Research. E noch is married to Lawrenda Osekre, and they have two daughters: Portia, age 11; and Benedicta, age 8.