LABORATORY REARING OF THE COCOA APHID TOXOPTERA AURANTII (BOY) AND SCREENING COCOA GENOTYPES FOR THEIR RESISTANCE TO THE APHID BY EMMANUEL AHIA CLOTTEY (B.Sc (Hons) Agriculture) A thesis presented in partial fulfilment o f the requirements for the de­ gree o f Master o f Philosophy in Entomology o f the University o f Ghana. Insect Science Programme* University o f Ghana Legon August 20Q1 * Joint interfaculty international programme for the training of Entomologists in West Africa. Collaborating Department: Zoology (Faculty o f Science) and Crop Science (Faculty o f Agriculture) University of Ghana http://ugspace.ug.edu.gh (C -364638 S B < % S ' A 5 \ U i i f t 6 o v „ University of Ghana http://ugspace.ug.edu.gh DECLARATION I hereby declare that, except for references to works of other scholars which have been duly cited, this thesis is the result of my own original research which has neither been presented in whole or in part for the award of a degree elsewhere. Dr. Beatrice Padi (SUPERVISOR) Dr. Paul K. Attah (SUPERVISOR) r. Ebenezer Owusu SUPERVISOR) (STUDENT) University of Ghana http://ugspace.ug.edu.gh ABSTRACT A study was conducted to develop a quick and efficient laboratory mass-rearing method for the cocoa aphid Toxoptera aurantii (Boy) and to screen cocoa genotypes for their resistance to the aphid. In a similar experiment, two cocoa genotypes were screened to determine their level of attractiveness to the cocoa capsid Sahlbergella singularis Hagl and the results compared with those from the aphid study. The study formed part of an on-going programme being conducted by the Cocoa Research Institute of Ghana (CRIG) to develop more vigorous, high-yielding cocoa geno­ types that are also resistant to diseases and insect pests. To raise adequate numbers of experimental aphids, a number o f leguminous and vegetable plants as well as tree crops, including cocoa (Y44), were evaluated for their suitability as rearing materials. Two aphid susceptibility evaluation methods were developed, one involving the use of very young seedlings of open pollinated Amazon cocoa (T85/799) and Trinitario (Y44) origins, and the other involving older seedlings of 25 clonal mate­ rials and 25 pair-wise crosses. Four cocoa progenies (Na32 x T60/887, P30 x P 30, ICS39 x Y44 and T79/501 x Pa 150) of different genetic sources were also screened to determine their levels of resistance to T. aurantii. In addition, tender shoots o f the T85/799 and Y44 were screened to determine their level of attractiveness to the cocoa capsid Sahlbergella singularis Hagl. using a laboratory “microtest” method and the results compared with findings from the aphid study. None of the leguminous and vegetable plants tested were suitable for rearing T. aurantii. O f the tree crops tested, cocoa (Y44) emerged as the best rearing material (producing 107 aphids/seedling in two weeks), followed by the Citrus varieties Medi­ terranean sweet (50 aphids/seedling) and Late Valencia (24 aphids/seedling), and University of Ghana http://ugspace.ug.edu.gh Coffea canephora (18 aphids/seedling). Cola nitida was found unsuitable. For both the Y44 and T85/799 young seedlings, aphid multiplication rate was highest on seedlings infested one week after cotyledon opening (179 & 91 aphids/ seedling, respectively) and lowest on those infested three weeks after cotyledon opening (38 & 9 aphids/seedling, respectively). There was positive correlation be­ tween the aphid infestation levels and the number of crinkled leaves produced (r =0.94 for Y44; r = 0.93 for T85/799). Thus, the number of crinkled leaves produced after eight weeks were highest on seedlings infested one week after cotyledon open­ ing (4 for Y 44 & 2 for T85/799) and lowest on seedlings infested three weeks after cotyledon opening (one for Y44 and zero for T85/799). The number o f aphids pro­ duced after one week and the number of crinkled leaves produced after eight weeks were scaled to provide criteria for determining aphid resistance. For the older cocoa seedlings, aphid multiplication rates differed significantly (P<0.01), with the clonal materials T85/799 x Pound 25 (58 aphids/seedlings) and T60/887 x Pound 7 (52 aphids/seedlings) emerging as the most susceptible whilst P30 (5 aphids/seedling) and T63/971 (7 aphids/seedling) were the least susceptible. With the pairwise crosses, (T85/799 x Pound 7) x (T60/887 x Pound 25) (62 aphids/ seedling) and IMC 85 x IMC 47 (57 aphids/seedling) were the most susceptible while Pound 10 x Pound 15 (4 aphids/seedling) and T60/887 x T63/971 (10 aphids/ seedling) were the least susceptible to aphid infestation. Results from the two evaluation methods (using very young and older seed­ lings) established differences in susceptibility levels among the four cocoa progenies screened. The Trinitario progeny Y44 x ICS39 was the most susceptible, followed by the Amazon progenies T79/502 x P a l50, Na32 x T60/887 and the Amelonado progeny, P30x P30, in decreasing order. It was concluded that young cocoa seedlings of Trinitario Y 4 4 , not more than University of Ghana http://ugspace.ug.edu.gh one week after cotyledon opening, were the best of all the materials tested for rear­ ing T. aurantii under insectary conditions at 24 ± 30C and relative humidity range of 72-85%. The comparable results obtained on the multiplication rate/crinkled leaf for­ mation from the aphid study and the number of capsid lesions recorded in the labo­ ratory microtest screening, strengthen the view by earlier workers that level of aphid resistance/ preference in cocoa types can be used as an index for determining their resistance to or preference by capsids and other sucking insects. It is anticipated that, at least, a few more of the cocoa types screened in the present study for aphid resistance/preference will be used in future studies to further confirm the view that cocoa types susceptible/resistant to 71 aurantii are also susceptible/resistant to capsids. University of Ghana http://ugspace.ug.edu.gh ACKNOW LEDGEMENT This publication could not have been possible without the help and support o f some individuals. I acknowledge with gratitude the relentless effort, advice and suggestions of Dr. Beatrice Padi, Head, Entomology Division, Cocoa Research Institute o f Ghana (CRIG) who supervised the work and arranged for my accommodation at Tafo. My heartfelt thanks go to my co-supervisors Dr. Paul Attah and Dr. Ebenezer Owusu, Senior Members, Zoology Department, University of Ghana for their guid­ ance, constructive criticisms and valuable suggestions during this work. I also wish to thank Dr. Adu-Ampomah, Head, Plant Breeding Division CRIG for his useful suggestions, Dr. Obeng Ofori, Crop Science Department, whose constant reminder gingered me up and Mr. & Mrs. J. E. Sarfo of CRIG for providing residential accom­ modations throughout my stay at Tafo. Special appreciation also goes to Mr. E. A. Dwomoh, Mr. Adu-Acheampong, Mr. A. Nkansah, all o f the Entomology Division, CRIG for their support; Mr. Victor Boadu my course mate with whom I stayed together at Tafo and Mrs. Gladys Essien o f the Ghana Universities Press who patiently and diligently typed the manuscript. I am also grateful to Cocoa Services Division (COCOBOD) for granting me study leave with pay and Prof. J. N. Ayertey, ARPPIS Coordinator who arranged for the Direct Support to training Organisation (DSO) scholarship for funding the pro­ gramme. Many thanks also goes to CRIG and CFC/ICCO/IPGRI project for provid­ ing all the cocoa types (clones and crosses) for the studies. Lastly, my most profound gratitude goes to the Lord God Almighty who has sustained me successfully through my studies. University of Ghana http://ugspace.ug.edu.gh DEDICATION This thesis is dedicated to my wife Olivia and four children, Daniel, David, Richard and Dorcas in appreciation of their love and understanding throughout the programme. vi University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS DECLARATION i ABSTRACT ii ACKNOWLEDGEMENT v DEDICATION vi TABLE OF CONTENTS vii LIST OF TABLES x LIST OF FIGURES xii LIST OF PLATES xiii LIST OF ABBREVIATION ix CHAPTERS 1 GENERAL INTRODUCTION 1 2.0 LITERATURE REVIEW 4 2.1 The cocoa plant 4 2.2 Economic importance of cocoa 4 2.3 Cocoa varieties grown in Ghana 5 2.4 Insect pest of cocoa 6 2.5 Cocoa mirids 6 2.5.1 Mirid species 7 2.5.2 Life cycle of mirids 7 2.5.3 Alternative hosts 8 2.5.4 Capsid damage 8 2.5.5 Chemical control 9 2.6 Non-chemical control measures 10 2.6.1 Biological control 10 2.6.2 Shade manipulation 11 2.6.3 Host plant resistance in insect management 11 2.7 Developing pest and disease tolerant/ resistant varieties of cocoa 12 2.8 The cocoa aphid 14 PAGES vii University of Ghana http://ugspace.ug.edu.gh 16 16 16 18 21 21 22 22 24 24 24 27 27 27 27 32 34 34 37 37 38 42 42 43 43 43 46 46 STUDY AREA The locality The climate The insectary DEVELOPMENT OF LABORATORY REARING METHOD FOR T H E COCOA APHID Introduction Materials and Methods Rearing of the aphid on leguminous and vegetable plants Rearing of the aphid on tree crops Feeding preference test Leaf disc test on feeding preference Investigation under semi-field conditions Statistical analysis Results Survival of aphids on the leguminous and vegetable plants Survival of aphids on the tree crops Feeding preference test The leaf disc method Survival of aphids on leguminous and vegetable plants under semi-field conditions Observations on the biology of T. aurantii Discussion DEVELOPMENT OF INFESTATION/EVALUATION METHOD AND ASSESSMENT OF APHID RESISTANCE IN COCOA Introduction Materials and methods Young cocoa seedlings Insectary screening of seedlings (aphids) Determination of the level of attractiveness of the two cocoa types to mirids using the "microtest" method Older cocoa seedlings ( pairwise crosses and clonal materials) viii University of Ghana http://ugspace.ug.edu.gh 5.2.2.1 Pairwise crosses 46 5.2.2.2 Clonal materials 49 5.3 Results 51 5.3.1 Very young cocoa seedlings of T85/799 and Y44 51 5.3.1.1 Number of aphids/aphid infestation level 51 5.3.1.2 Aphid infestation on growth of seedlings 54 5.3.1.3 Number o f crinkled leaves due to aphid infestation 56 5.3.2 Level of attractiveness of two cocoa types to mirids 58 5.3.3 Aphid multiplication rate on older seedlings 59 5.3.3.1 Pairwise crosses 59 5.3.3.2 Clonal materials 60 5.4 Discussion 61 5.4.1 Very young seedlings 61 5.4.2 Older seedlings: clones and pairwise crosses 64 6.0 SCREENING COCOA PROGENIES FOR THEIR RESISTANCE TO THE COCOA APHID TOXOPTERA AURANTII 67 6.1 Introduction 67 6.2 Materials and methods 68 6.2.1 Young germinating seedlings 68 6.2.2 Using flush leaves of older seedlings 69 6.3 Results 70 6.3.1 Young germinating seedlings 70 6.3.1.1 Number o f aphids per plant 70 6.3.1.2 Aphid infestation on growth of seedlings 73 6.3.1.3 Number of crinkled leaves due to aphid infestation 75 6.3.2 Aphid multiplication rate on older seedlings 77 6.4 Discussion 79 7.0 CONCLUSIONS 82 REFERENCES 84 APPENDICES 97 ix University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES 1. Daily survival of aphids recorded on seedlings of leguminous plants infested 3, 5,7, and 10 days after cotyledon opening 28 2. Daily survival of aphids recorded on seedlings of vegetable plants infested 3 ,5 ,7 , and 10 days after germination 29 3. Mean number of aphids produced in cocoa seedlings infested at different periods after cotyledon opening 3 0 4. Mean number of aphids produced on different perennials after two weeks of infestation 3 2 5. Distribution of aphids on leaf discs of different vegetable plants 34 6. Distribution of aphids on leaf discs of different leguminous plants 3 6 7. Nymphal developmental periods of T. aurantii on cocoa under insectary conditions 37 8. Mean number of aphids per seedling for Amazon T85/799 after eight weeks of infestation 51 9. Mean number of aphids per seedling for Trinitario Y44 after eight weeks of infestation 51 10. Mean number o f aphids produced after one week on two cocoa types infested at different periods (1,2 and 3 weeks) after cotyledon opening 52 11. Effect of aphid feeding on height (cm) of two cocoa types infested at different periods (1,2 and 3 weeks) after cotyledon opening 54 12. Effect of aphid feeding on girth (mm) of two cocoa types infested at different periods (1, 2 and 3 weeks) after cotyledon opening 55 13. Relationship between aphid density and number of crinkled leaves for PAGES University of Ghana http://ugspace.ug.edu.gh two cocoa types infested at different periods 14. Number of feeding punctures produced on two cocoa types by Sahlbergella singularis. 15. Rating of 25 pairwise crosses based on the mean number of aphids (multiplication rate) produced by one adult aphid for seven days 16. Rating of 25 clonal materials based on the mean number of aphids (multiplication rate) produced by one adult aphid for seven days 17a Mean number of aphids produced by T. aurantii reared on young seedling of four cocoa progenies after a week of infestation 17b Mean number of aphids per seedling produced from the first to eighth week (W,-Wg) after infestation on four cocoa progenies 18. Effect o f aphid feeding on height o f four cocoa progenies after eight weeks of aphid infestation 19. Effect of aphid feeding on girth of four cocoa progenies after eight weeks 20 Relationship between aphid density and the number of crinkled leaves for four cocoa progenies after 8 weeks of infestation 21. Mean number of aphids produced by one adult aphid on older seedlings of four cocoa progenies after 7 days 56 58 59 60 72 72 73 74 75 77 xi University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES I. Map of Ghana showing position of the study area (Tafo) 2a Mean monthly rainfall for ten years (1990-1999) and the year 2000 2b. Mean monthly relative humidity for ten years (1990-1999) and year 2000 3. Survival of aphids under starvation with time (hours) 26 4a. Mean number o f aphids produced after two weeks on cocoa seedlings infested at four different ages (3,5, 7 and 10) after cotyledon opening 4b. Relationship between age of seedling and the number of aphids produced after different days of infestation after cotyledon opening 5. Mean number of aphids produced on six different perennials infested with one adult aphid for seven days 6. Mean number of aphids produced on cocoa type T85/799 infested at different period (1, 2 and 3 weeks) after cotyledon opening 7. Mean number of aphids produced on cocoa type Y44 infested at different periods (1 ,2 and 3 weeks) after cotyledon opening 8. Mean number of crinkled leaves produced after eight weeks on two cocoa types infested at different periods (1 ,2 and 3 weeks) after cotyledon opening 9. Mean number o f aphids produced on young seedlings of four cocoa progenies for seven days 10. Mean number of crinkled produced after eight weeks of aphid infestation on four cocoa progenies II. Mean number of aphids produced on older seedlings of four cocoa progenies infested with one adult aphid for one week. University of Ghana http://ugspace.ug.edu.gh LIST OF PLATES 1. Cage used in aphid rearing 23 2. Arrangement of leaf disc in preference test 25 3. Arrangement of young seedlings o f two cocoa types in trays 44 4. Arrangement of twigs of two cocoa types in mirid attractive test (microtest) 47 PAGES xiii University of Ghana http://ugspace.ug.edu.gh LIST OF ABBREVIATIONS ANOVA — Analysis of Variance LSD . — Least Significant Difference CFC — Common Fund for Commodities ICCO — International Cocoa Organization IPGRI — International Plant Genetics Resource Institute The following are general codes for cocoa types (genotypes). T number — These are seedlings with pod progenies introduced from Trinidad T85/799 — Seedlings from pod 85 from Trinidad planted at stand 799 in the introduction plot. in 1944. Various seedlings obtained from a given pod are referred to by the pod number and stand number thus: NA P ICS PA IMC Imperial College Selections Iquitos Mixed Calabacillos Parinari (Pounds selections from the Parinari district) Nanay (Pounds selection from the Nanay pennisula) Amelonado cocoa SCI Y Anyinam (cocoa from Anyinam Farms) Posnettes “Sahlbergella Control” xiv University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE GENERAL INTRODUCTION Cocoa has been the backbone of the Ghanaian economy virtually throughout the last century. Despite problems currently besetting the cocoa industry, the crop still plays an important role in the economy of the country. The total area under cocoa cultiva­ tion is currently estimated at 1.2 million hectares and in 1999/2000 cocoa provided 22.5% of the total export earning (Asenso-Otchere, 2001). The cocoa industry em­ ploys sizeable percentage of the country’s labour force and production is steadily increasing. From a lowest level of 154,000 metric tons (m.t) in 1983/1984, produc­ tion has increased gradually reaching 409,000 m.t in 1997/1998 and 420,000 m.t in 1998/1999. Cocoa production is projected to reach 500,000 m.t by the year 2004/ 2005 and to 700,000 m.t by the year 2009/2010 (Anon, 1999). Cocoa will, therefore, continue to play a major role in the economy of Ghana. Cocoa capsids (Heteroptera: Miridae) were recognised as pest of cocoa in West Africa in 1910 (Dungeon, 1910) and are currently considered the most important pests of cocoa in Ghana. Yield losses due to capsid damage alone was estimated at 60,000 to 100,000 tonnes of dry cocoa beans (Stapley and Hammond, 1959) and may reach as high as 75% crop loss per annum if attacked farms are left unattended for more than three years (Anon, 1951). The recommended method for capsid control since 1957 has been by routine spraying with conventional insecticides applied four times a year at monthly intervals from August to December, omitting November (Gibbs et. al. 1968, Collingwood and Marchart, 1971) There is some indication that on mature cocoa, the frequency of 1 University of Ghana http://ugspace.ug.edu.gh insecticide application may be reduced to two per year if the tree canopy is good (ie closed) (Owusu-Manu, 1997). i Chemical control of cocoa capsids, however is plagued by several problems including the resurgence of resistance strains such as occurred with lindane in the early 1960’s (Dunn, 1963, Padi, 1997), the destruction of beneficial insects such as pollinators and natural pest control agents. Moreover, the high cost of inputs such as spraying machines, chemicals and labour and the cumbersome'spraying procedure (Adomako, 1990) has resulted in very low adoption rate of less than 3% (Padi e t al, 2000a). '{ It has, therefore, become necessary to look for alternative control methods that are environmentally friendly and devoid of the risks and problems associated with the current practice. The breeding of varieties of cocoa that are more resistant to capsids and other sucking insects offers an opportunity for reducing insect damage and the spread of insect transmitted diseases. The Cocoa Research Institute of Ghana has initiated a programme aimed at selecting high-yielding and vigorous cocoa types that are also resistant to the cocoa swollen shoot virus disease (CSSVD) and blackpod disease caused by Phytophthora species (Adu-Ampomah, 1994). Breeding for genotypes resistant to capsid attack will aid seedling establishment as well as mature tree performance in the field (Padi, 1997, and Adu-Ampomah et al., 1999). It has therefore, become necessary to incor­ porate into the breeding programme the development of capsid tolerant-or resistant cocoa types. Screening for capsid resistance is however, hampered by unavailability of capsids since they are extremely difficult to rear and maintain in sufficient numbers(under laboratory conditions (Raw. 1959; Youdeowei, 1964). Moreover, capsids occur in very low numbers in the field (Cotterell, 1926; Williams, 1954). It is however known University of Ghana http://ugspace.ug.edu.gh that resistance of cocoa to insect pests appears to be quite general and that cocoa types that are resistant or susceptible to one species of sucking insect tend to exhibit similar attributes to other sucking insects. For example, Bigger (1975) identified cocoa progenies including T85/799 x T17/359 that were susceptible to a range of sucking insects. He indicated that a cocoa variety which is resistant to Planococcoides njalensis may show resistance to a range of other sucking insects. The use of the cocoa aphid Toxoptera aurantii (Boy) as an indicator of general insect resistance in cocoa was suggested by Campbell (1990) based on the fact that aphids are more conspicuous than mealy bugs on cocoa and could be more convenient to monitor in screening programmes. The cocoa aphid T. aurantii occurs in large numbers in the field (Entwistle, 1972) and could be relatively easier to rear if a suitable rearing material is identified. Thus Eskes et. al. (1998) suggested that resistance to aphids may be easier to evalu­ ate than resistance to mirids and could be an indicator of capsid resistance in cocoa. The present study was, therefore, carried out with the following objectives: 1. To identify suitable breeding material for rearing sufficient numbers of the co­ coa aphid T. aurantii to be used in screening, for resistance as an index for capsid resistance. 2. To develop a reliable method of aphid infestation/evaluation for very young cocoa seedlings using beans from two types of open pollinated pods, Amazon (T85/799) and Trinitario (Y44) origins. 3. To develop a reliable method of aphid infestation/evaluation for older seedlings and clonal materials using young leaf flushes. 4 To screen four different cocoa progenies to determine their levels of resistance to the aphid. 3 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO LITERATURE REVIEW 2.1 The Cocoa Plant Cocoa, Theobroma cacao Linnaeus (Tiliales:Sterculiaceae) is the only species of the genus Theobroma cultivated commercially. The genus contains over twenty species all of which originated from the tropical rainforests o f equatorial America (Mossu, 1992). Theobroma cacao is classified into two main populations, Criollo and Forastero. The third population, the Trinitario, resulted from natural crosses between Criollo and Forastero types. The Criollo type is indigenous to the north and west of the Andes, whilst the Forastero is indigenous to the Amazon basin (Mossu, 1992). The bulk of cocoa grown in West Africa is Amelonado, a variety of Forastero (Wood, 1975). Commercial cultivation of cocoa in Ghana began with cocoa pods brought from Femado Po by a Ghanaian blacksmith, Tetteh Quarshie in 1879 (Hammond, 1962; Hill, 1963, Manu and Tetteh, 1987). However, cocoa is believed to have been intro­ duced earlier by the Basle and Dutch missionaries (Cardinal, 1931; Wanner 1963). 2.2 Economic importance of cocoa Cocoa has been the backbone of the Ghanaian economy virtually throughout the last century. Ghana was the world’s leading cocoa producer between 1910 and 1977 with market shares ranging from 30 to 40% of the world total production (Bateman, 1988). Until recently, cocoa contributed over 60% of the country’s foreign exchange (Onosaiye, 1987). In spite of the significant gains made by other sectors of the economy in recent times, cocoa continues to be the major source of revenue for the provision 4 University of Ghana http://ugspace.ug.edu.gh ! of socio-economic infrastructure in the country. The total area under cocoa cultiva­ tion is currently estimated at 1.2 million hectares and in 1999/2000 cocoa provided 22.5% of the total export earnings (Asenso-Otchere, 2001 ). The cocoa industry also employs a sizeable percentage of the country’s labour force. In 1965/66 production reached an all-time peak of 560,000 tonnes but this fell to a very low level of 159 tonnes in 1983/84 season resulting in Ghana losing her position to Cote d’Ivoire as the world’s largest producer of cocoa (Gill and Dufus, 1989). The ravages caused by insect pests and diseases and the low producer price, among other factors, contributed to the dwindling of the industry (Anon, 1995). 2.3 Cocoa varieties grown in Ghana The bulk of Ghana’s cocoa in the early years of production was the variety “Amelonado” which occupied about 80% of the total acreage. The rest were often referred to as local Trinitario. In many respects the Amelonados and Trinitarios are excellent varieties acceptable to cocoa growers and chocolate manufacturers for their flavour and other bean qualities. They, have however, major defects of slow growth in the early years, which leads to poor establishment under unfavourable conditions, and a long interval between planting and cropping and sensitivity to infection by cocoa swollen shoot virus ( CSSV) (Adu-Ampomah et al, 1996). In order to obtain the necessary variation needed for genetic improvement of the crop, Posnette intro­ duced semi-wild types from the headwaters of the Amazon basin called Upper Ama­ zon selections (Anon; 1946). As an interim measure to overcome the susceptibility to CSSV, “Mixed Amazons” were developed through multiplication of the Upper Ama- I zon selections to replace the Amelonados and Trinitarios. Although useful, the level of resistance and tolerance in the Mixed Amazons has been inadequate. To get superior varieties for farmers, the series II hybrids were later developed 5 University of Ghana http://ugspace.ug.edu.gh (Thresh et. al., 1988). These were the product of hybridization between selected Upper Amazon clones and selected Amelonado or local Trinitarios. The hybrids are comparable to the Mixed Amazons in many respects including response to CSSV, but they have the additional attributes of maturing early and having higher yield potential (Adu-Ampomah et. al., 1996). Recently, new inter-Amazon hybrids have been released to farmers which are superior to earlier genotypes (Adomako and Adu Ampomah, 2000) 2.4 Insect pest of cocoa Over 1,500 insect species have been recorded on cocoa but only a few are of eco­ nomic importance (Entwistle, 1972). In West Africa, the most important pests are the two species of cocoa capsids, Distantiella theobroma (Dist) and Sahlbergella singularis Hagl (Heteroptera: Miridae) and the mealybug vectors of CSSVD mainly Planococcoides njalensis (Laing) and Planococcus citri (Risso). Minor pests in­ clude the leaf eating caterpillars such as Earias biplaga, Anomis leona Schaus, the psyllid Tyora tessmani (Aulm) and the aphid Toxoptera aurantii (Boy), and the pod borer, Characoma stictigrapta Hmps. Others are the pod feeding bugs Bathycoelia thalassina (Heteroptera: Pentatomidae) and Pseudotheraptus devastans (Dist) (Heteroptera: Coreidae); stem borer Eulophonotus myrmeleon Fldr (Lodos, 1967; Owusu- Manu, 1974; Padi et. al. 1998); and termites (Ackonor et. al., 1993; Ackonor and Tei, 1995) which have in recent times become economically important in some parts of the country. 2.5 The cocoa mirids Cocoa mirids (Heteroptera: Miridae) belonging to the subfamily Bryocorinae, are the most harmful insects attacking cocoa trees throughout the world except the Car­ 6 University of Ghana http://ugspace.ug.edu.gh ibbean Islands (Lavabre, 1977). 2.5.1 Mirid species Mirids are sap-sucking insects which feed on green shoots, chupons and pods. In West Africa the species known to feed on cocoa include Sahlbergella singularis, Distantiella theobroma, Bryocoropsis laticollis and members of the genus Helopeltis (Collingwood, 1971). The most important species of mirids which cause most dam­ age to West African cocoa is Sahlbergella singularis Hagl which is the dominant species in Nigeria, Cameroun, Togo, and Sierra Leone. In Ghana; Distantiella theobroma. (Distant) has been the dominant species but, in recent times, S. singularis has emerged as dominant species (Padi and Adu-Acheampong, 2001). 2.5.2 Life cycle of mirids A detailed account of the biology of West African cocoa mirids is available in Cotterell (1926) and Anon (1945,1947). The tiny white eggs are laid embedded in the cortex of pods and the bark of shoots. The eggs carry two-threadlike filaments which prob­ ably aid in breathing (Cotterell, 1943). The maximum number of eggs laid by one female is 47 for S. singularis, and 192 for D. theobroma. The eggs take 12-18 days to hatch and the nymphs which emerge begin to feed soon after hatching (Hill, 1983). The average incubation period for S. singularis is 17 days, and 15 days for D. theobroma (Taylor, 1954). There are five nymphal instars, each with a duration o f 4— 5 days. The average period from egg to adult for S. singularis is 41 days, and for D. theobroma 39. There is a pre-maturity period for both species of about a week, and the longevity of the adults varies from 25-30 days. The proportion of females is greater than that of males (Raw, 1959). 7 University of Ghana http://ugspace.ug.edu.gh 2.5.3 Alternative Hosts Mirids have been found on other host plants (Box, 1944; Anon 1947). S. singularis has been recorded on 17 alternative hosts including Berria amonilla, Sterculiafoetida and several Cola species (C. nitida and C. acuminata) whilst D. theobroma is found on three alternative hosts namely Citrus spp, the silk cotton tree, Ceibapentandra, and the baobab tree, Adansonia digitata (Entwistle and Youdeowei, 1964; Entwistle, 1972; Padi et. al. 1996) 2.5.4 Capsid Damage Mirid damage usually starts where there is a break in the canopy which occurs from fallen trees, removal of swollen shoot diseased trees or from poor canopy develop­ ment (Williams, 1953). Canopy breaks attract mirids and may become foci for local population build-up. Thus young cocoa trees having open canopy are more vulner­ able to capsid attack. Capsids feed by inserting their mouthparts about 1.8-2.2 mm deep into the plant tissues. Toxic saliva is injected into the plant tissue after which the cell content is sucked leaving dark patches of drying tissue called feeding lesion. The feeding lesions are initially light brown but soon turn dark brown and eventually black (Goodchild, 1952). Extensive feeding on young pods (cherelles) causes them to wilt but damage to older pods is usually insignificant (Johnson, 1962). Feeding on stems causes wilting above the point of attack resulting in the death of terminal shoots. The damage ensuing from capsid attack is enhanced by the entry of parasitic fungi namely, Calonectria rigidiuscula Berk and Br. (Sacc) that spreads through the plant tissues causing die-back (Crowdy, 1947). This eventually results in the death of branches, breaks in canopy, poor yields, and eventually death of the cocoa tree (Are and Gwyne- Jones, 1974). Yield losses due to capsid damage was estimated at 60,000 to 100,000 University of Ghana http://ugspace.ug.edu.gh tonnes of dry cocoa beans annually (Hale, 1953; Stapley and Hammond, 1959) and may amount to as high as 75% crop loss if attacked farms are left unattended for more than three years (Anon, 1951). 2.5.5 Chemical Control Earlier attempts at capsid control involved the use of various emulsions and dusts. Dungeon (1910) used kerosene/soap emulsion as stem paint and Cotterel (1943) tested and found nicotine sulphate effective at 0.1 percent. However, the use o f nico­ tine sulphate was discontinued because it was found to be expensive and toxic to mammals. With the advent of motorized mistblowers in the mid 1950s, several con­ ventional insecticides, including dichlorodiphenyl trichloroethane (DDT) and lindane (Gammalin), were introduced for capsid control (Hammond, 1957; Stapley and Hammond, 1959; Owusu-Manu, 1985). Presently, the recommended method for capsid control involves the use of Gammalin 20 (Lindane) at 280 g a.i in 56 litres of water/ha and Unden 20 propoxur at 210g a.i in 56 litres of water/ha applied at 4-weekly intervals from August to December, omitting November (Padi, 1997). However, chemical control is plagued with several user problems and environmental hazards. These include the develop­ ment of pest resistance as occurred with lindane in the early 1960’s (Dunn, 1963; Marchart and Collingwood, 1969), and the destruction of beneficial natural enemies resulting in minor pests assuming major pest status as happened with Bathycoelia thalassina (Owusu Manu, 1974). Another problem related to the use of insecticides is the high cost of inputs such as spraying machines, chemicals and labour. Moreo­ ver, the spraying procedure is very tedious and, as pointed out by Adomako (1990), the cocoa farmer finds it difficult to support 25 kg of insecticide solution on his back during spraying. Due to these constraints, the adoption rate by farmers of the current University of Ghana http://ugspace.ug.edu.gh recommendations had been very low as evidenced by recent surveys (Donkor et. al., 1991; Henderson et. al, 1994; Padi et. al., 2000a). , 2.6 Non-chemical Control Measures The growing universal concerns against the use of conventional insecticides has cre­ ated the need to look for alternative measures that are environmentally friendly and less risky to users and consumers. According to Padi (1997), CRIG has embarked on the development o f the fol­ lowing alternative control methods: (i) The use of semio-chemicals including sex pheromones. (ii) Biological control with natural enemies (parasitoids, predators and pathogenic fungi) (iii) Cultural practices involving the use of suitable shade trees, shade manipulation and the avoidance/destruction of alternative capsid host plants such as Ceiba petandra, Citrus spp, Cola spp, and Adansonia digitata. (iv) The use of tolerant/resistant cocoa types. 2.6.1 Biological Control Studies in West Africa (Lodos, 1968; Entwistle 1972, Kumah 1975, 1976) have shown that developmental stages, including the egg stage of both D. theobroma and S. singularis, are affected by parasitic organisms which are either hymenopterous parasitoids or nematodes. Percentage parasitism has been generally lower in D. theobroma (about 4%) and higher in 4th instar nymphs of S. singularis (about 20- 59.5%) (King 1971). In West Africa, mirid predators recorded are several species of ants including Oecophylla longinoda and Macromischoides aculeatus Mayr, salticid 10 University of Ghana http://ugspace.ug.edu.gh spiders, reduviids, mantids, Grillidae and Pentatomidae, but most of these are facul­ tative parasites and their predatory potential remains to be exploited. Pathogenic fungi which attack capsids include fungi imperfecti, Baeuvaria bassiana, Bacillus and Aspergillus species (Anon, 1970, Lim Tong et. al. 1989, Collingwood 1971). Recent findings by Padi et. al. (2000 b) indicate that five Beauvaria isolates gave promising results against S. singularis. 2.6.2 Shade Manipulation Break in canopy increases light penetration and leads to emergence of young succu­ lent chupons and mirids are attracted to such spots which become foci for local population build-up. Shade manipulation, therefore, becomes vital and should be part of an integrated approach to mirid control. Collingwood (1971) indicated that well canopied cocoa is largely self protecting against serious capsid damage be­ cause of shade and high humidity within the canopy which restricts the develop­ ment of large population compared with more exposed cocoa with canopy breaks. Owusu-Manu (1997) has demonstrated that on mature cocoa with severe capsid damage and broken canopy at least two continuous years of 4 insecticide applica­ tions per year are required to restore the canopy. Where the cocoa canopy has al­ ready closed, 2 sprays per year are adequate. However, since excessive shade fa­ vours blackpod disease and suppresses yield, it is important to properly manage it. 2.6.3 Host plant resistance in insect management The development of resistant varieties usually takes a long period of experimenta­ tion. However, results already obtained indicate that the possibility of using plant resistance to insects deserve careful study for each major insect pest and many minor 11 University of Ghana http://ugspace.ug.edu.gh ones (Painter, 1951). The primary objective of programmes on insect resistance in crop plants is to develop cultivars that are resistant to an insect pest while maintain­ ing or improving their basic agronomic characteristics. Resistant varieties fall roughly into three groups namely (i) as the principal control method (ii) as an adjunct to other measures and (iii) as a safeguard against the release of more susceptible varieties than exist at the present time. Luginbill (1969) emphasised the importance of resist­ ant varieties in pest management and the millions of dollars saved by growers. Metcalf and Luckman (1982) pointed out that the most desirable features of plant resistance, among others are: 1. Specificity: Plant resistance is usually specific to a pest or complex of pest organisms and seldom has direct detrimental effect on beneficial insects. 2. Persistence: Most resistant varieties maintain high levels of resistance for a long time despite the occasional upsurge of biotypes. 3. Harmony with the environment: Since no unnatural elements are used, there is no likelihood of contaminating the environment or endangering man or wild life. 4. Ease o f adoption: Once developed, resistant varieties can easily be incorpo­ rated into normal farm operations at little or no extra cost and 5. Compatibility: Plant resistant is compatible with other tactics in pest manage­ ment, being an ideal adjuvant when resistance alone cannot maintain a pest below'the economic threshold. 2.7 Developing pest and disease tolerant /resistant varieties of cocoa Bigger (1974) indicated that it would be preferable to avoid the necessity of spraying cocoa with insecticides by the use of resistant varieties. However, according to Eskes and Lanaud (1996) cocoa cultivation is still largely based on traditional varieties 12 University of Ghana http://ugspace.ug.edu.gh domesticated more than 150 years ago and that less than one third consists of hybrid and clonal cultivars developed by breeding programmes initiated since the 1940’s, first in Trinidad and subsequently in Ghana, Brazil and other major cocoa producing countries. In Java, farmers planted more tolerant hybrids of Djale Roengo Criollo cross Trinitario against Helopeltis attack. Destroyed shoots of these hybrids are rapidly regenerated to prevent break in canopy (Hall, 1932). In West Africa SCI trees were identified among attacked trees at Asuansi in 1941 as being tolerant to capsid attack. This led to the selection of 54 cocoa clones on the basis of their resistance to capsids (Posnette 1946). Collingwood and Marchart (1971), in assessing capsid damage to the control plots of a systemic insecticide trial planted to five cocoa progenies, con­ cluded that the enormous differences, which could be shown between progenies in canopy damage, the tree deaths, were not due to differential attack by the capsid, but to the ability of the progeny to recover from the attack. The more vigorous Amazon crosses recovered more readily than Amelonado, and the Amelonado X Amazon cross was intermediate in response. Burle (1953) has reported resistance of particu­ lar trees, to attack by Sahlbergella singularis and Distantiella theobroma in the Ivory Coast, and Coolhause (1939) reported resistance to Helopeltis in Indonesia. Subsequent work in Ivory Coast showed a ten-fold difference in mean number of capsids recorded over two season between UPA 620 which had least and C409 which had most capsids. C5 and the hybrids tested were intermediate in response. A study by Bigger (1972) showed Distantiella to be more prevalent on series IIB hybrids and indicated that series II hybrids seemed more susceptible to attack by a range of suck­ ing insects and their attendant ants than T85/799 x T17/359. With regards to cocoa disease resistance, Adu Ampomah et. al. (1996) have indicated that due to improvement in the screening procedures coupled with large University of Ghana http://ugspace.ug.edu.gh scale introductions, some indications of sources of resistance to CSSVD have been identified especially among Pound’s Upper Amazon collection. Some progress has also been made in efforts to breed for cocoa types resistant to the black pod disease caused by Phytopthorapalmivora. T79/501 and Pa7/808 have been selected as po­ tential resistant parents and these have been planted in the seed gardens (Abdul- Karimu et. a l, 1996). In the case of P. megakarya some forty trees showing low levels o f infection have been identified. These materials are being evaluated to deter­ mine the nature of resistance. 2.8 The cocoa aphid Seven, or possibly eight species of aphids are known to attack cocoa but, all except Toxoptera aurantii (Boyer de Fonscolombe), are of rather casual occurrence and have other more important and usual host plants (Entwistle, 1972). T. aurantii oc­ curs in all the cocoa growing areas of the world and is quite common on cocoa in Ghana (Eastop, 1961). A brief account of the biology of T. aurantii on cocoa has been given in Cote d’ Ivoire by Alibert (1951) and in the British West Indies by Kirkpatrick (1955). Firempong (1975) worked on the biology of the aphid in Ghana. Apterae individuals are small, oval, brown-black or black with black-and-white banded antennae and black siphunculi and cauda. Alate individuals have a dark brown to black abdomen. On cocoa, the aphid is found on young flush leaves of apical shoots, on chupons, flower stalks and cherelles. Its seasonal occurrence is closely tied up with the flush­ ing cycle of cocoa. According to Firempong (1975), T. aurantii takes 6 days to become adult on cocoa seedlings and the temperature range most suited to its life function is 20°C - 25"C; with 22°C as the optimum temperature. Populations of the aphid comprise 14 University of Ghana http://ugspace.ug.edu.gh only females, reproduction being entirely parthenogenetic and the young produced viviparously. Broughton and Harris (1971) analysed the sound produced by T. aurantii which is the only aphid with audible stridulation. It attacks flower clusters causing abortion of the flowers and damages young leaves (Ingram, 1964; Gowdey, 1913). He considered the pest important enough to mention it annually in his report in Uganda. Saunders (1979) reported that heavily attacked leaves become cupped or rolled and flower development may be arrested. It’s pest status is most apparent in the nursery where it may severely retard the growth of growing plants if not controlled. Firempong (1975) indicated that severe attack on young cocoa causes crinkling in leaves. In addition to leaf crinkling; premature leaf fall, flower wilt, withering o f young stems and etiolation of affected plants may occur (Alibert-, 1951). In Peru it has been estab­ lished as a pest of Cocoa (Wille, 1944) but in Ghana and other W. African cocoa growing countries, T. aurantii has always remained a pest of minor importance on cocoa. • Due to their small size, parthenogenetic reproduction, high capacity for multi­ plication and world-wide distribution, aphids are ideal for studying many of the topi­ cal issues in ecology and plant breeding (Dixon, 1985; Maxwell and Jennings, 1980). 15 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE STUDY AREA 3.1 Locality The study was conducted at the Cocoa Research Institute of Ghana (C.R.I.G.) for­ mally known as the West Africa Cocoa Research Institute(WACRI) CRIG is located at New Tafo-Akim (Latitude 60 17'N and Longitude 0° 22E) about 40 km (24 miles) from Koforidua, the capital of Eastern Region, and 107 km (67 miles) away from Accra at an altitude of 220 metres above sea level (Fig.l). New Tafo is located southwest of the Mampong Scarp on the Togo hills within the equatorial climatic Zone (Udo, 1978). Keay (1959), giving it a broad classifica­ tion, puts it in the moist forest zone at low and medium altitude whilst Taylor (1952) locates it within the celtis - Triplochiton sub-division of the moist semi-decidous forest. 3.2 The Climate The rainfall pattern shows two wet seasons (March to July and September to No­ vember) with double maxima occurring in May/June and October (Church, 1957; Boateng, 1960). There are two peaks for the mean monthly temperature occurring from February to May and November to December and the minima occurring in July/ August. The average monthly relative humidity has a single peak occurring from July to October. The average monthly rainfall for the year 2000 was 127.0 mm, higher than that of the past ten years which averaged 119.3 mm (Fig 2a). Daily maximum and 16 University of Ghana http://ugspace.ug.edu.gh *119 6 ’ m * ■r*“ o« f r y .■ -*•«■ — • • ~ r w • ■/" */*— • — ------------------------- ") / UPPER / U. ' ^ « eolggtangg /U |PBR / EAST . ro