University of Ghana http://ugspace.ug.edu.gh BOOK NUMBER Qj-5£ ■iTjhen Balmje Liibrnary iii University of Ghana http://ugspace.ug.edu.gh EFFECTS OF MAIZE VARIETY AND SEASON ON POPULATION DYNAMICS OF THE LARGER GRAIN BORER, PROSTEPHANUS TRUNCATUS (HORN) (COL.: BOSTRICHIDAE) AND THE MAIZE WEEVIL, SITOPHILUS ZEAMAIS MOTS.(COL.: CURCULIONIDAE) AND GRAIN LOSSES IN THE TRADITIONAL ‘EWE’ STORAGE BARN IN GHANA BY BERNARD AGYEMAN BOATENG B.Sc. (HONS.) AGRICULTURE A THESIS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON, IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF PHILOSOPHY (M. PHIL.) IN CROP SCIENCE (ENTOMOLOGY) AUGUST, 1996 University of Ghana http://ugspace.ug.edu.gh ---351S75 (5x5^6. B5. Kooin University of Ghana http://ugspace.ug.edu.gh ABSTRACT The effects of three maize varieties and two storage seasons on the population dynamics of Prostephanus truncatus and Sitophilus zeamais were observed in the traditional4Ewe’ barn on the field. Two local varieties, Dzolokpuita and Abutia and an improved variety, Abeleehi, were stored with the husk on, during the long season, and Abutia only during the short season, of the 1994/1995 storage period. Destructive sampling techniques were applied to obtain data at monthly intervals for eight months. Maize variety showed significance (P<0.05) on P. truncatus density in the first four months of the long season but was barely so during the late storage phase. However, it did not influence the population dynamics of S. zeamais. Trend analysis showed significant increase in densities of P. truncatus over time especially during the late phase of the long season with a maximum densities of 131.1, 43.7, and 16.9 adults per kg. t grain for Abutia , Abeleehi and Dzolokpuita, respectively. S. zeamais however peaked at between 280 and 350 adults per kg grain on all varieties after only three months of storage, then stabilized at about 250 insects per kg. grain for the rest of the season. Maize variety also influenced weight loss with Dzolokpuita faring better. Good husk cover and relatively harder grains of Dzolokpuita may explain the lower P. truncatus density and weight loss recorded for this variety. Season did not to affect P. truncatus density but it influenced S. zeamais dynamics, and weight loss levels. The relative economic value loss also increased with time with value loss of 21.1%, 21.5%. and 19.4% for Abutia, Abeleehi and Dzolokpuita, respectively after eight months of storage. University of Ghana http://ugspace.ug.edu.gh DECLARATION I do hereby declare that, except for references to work of other researchers which have been duly cited, this thesis consists entirely of my original research work conducted at the Ministry of Food and Agriculture Research Station, Kpeve, Volta Region and that no part of it has been presented for another degree elsewhere. BERNARD AGYEMAN BOATENG (Student) Prof. J. N. AYERTEY (Principal Supervisor) ft/iLi). {AaL . v-hln Dr. W. G. MEIKLE (Co-Supervisor, IITA-Benin) University of Ghana http://ugspace.ug.edu.gh iii DEDICATION This thesis is dedicated to my parents, Mr. and Mrs. Onwona-Agyeman, my brother, Tony, and my sisters Naana, Mother, Maty and Nana Yaa for their love and prayers. University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENTS Thanks and praise to Almighty God for giving me life to complete this work. I am veiy much indebted to the following personalities with whose invaluable help and diverse contributions this work has been successfully completed. -Professor J. N. Ayertey, my Principal Supervisor, for his patience, the independence he offered me, concern, constructive criticism and fatherly advice. -Dr. W. G. Meikle, my Co-Supervisor, for his untiring help in setting up the experiment, productive comments, concern, encouragement throughout this study, and above all for “keeping me from drowning” time and again. Yes sir! I appreciate that very much. -Messrs. John Tsra-kasu, Head of the Ministry of Food and Agriculture (MOFA) Research Station, Kpeve, his deputy R. Adasi and also Bonsi; Nana Boatemaa, Peace, Irene, and all the other workers of the Research Station for their support, hospitality and encouragement. -Messrs. Pascal Degbey and Ben Azoma for their priceless assistance rendered during sampling and processing of materials. God bless you, guys. -Miss Julia Compton, Mr. Anthony Ofosu, Sammy, Ken, Hilarious, Simon and all the workers of the ODA/MOFA LGB Project at Ho for technical advice and support -Mr. Kwame Vowotor, Dr. Christian Borgemeister and Monsieur Richard Oussou. Thank you for every help provided me in Cotonou. You really made my stay worthwhile. -Dr. I. Ofori and Isaac Osei Akoto (ISSER) for advice during the statistical analyses. -Mr. F. K. Saalia of the Nutrition and Food Science Department for sacrificing to save me from my desperate search for a computer for my work. University of Ghana http://ugspace.ug.edu.gh -My colleagues: Rosina, Adam, Addison, Mike, Pody, Freakie, Frank Boakye Antwi, Kweku Boison, Christianne Reuse and Leona Ursula Hug for their moral support. -My room-mate, Godfned Kwadwo Addo for his brotherly love and support. -Miss Joycelyn Owusu-Ensaw for her prayers, special support and concern. -Profound gratitude also goes to the Volta River Authority (V.R.A) for the fellowship offered me as well as the DTA-LGB Project and DANIDA for the sponsorship of the University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS ABSTRACT.......................................................................... .........................................i DECLARATION........................................................................................................... DEDICATION.............................................................................................................. “ ACKNOWLEDGEMENTS........................................................................................... & TABLE OF CONTENT................................................................................................ ™ LIST OF TABLES............................................................................. -......................... « LIST OF FIGURES...................................................................................................... xi LIST OF PLATES........................................................................................................xii 1.0 GENERAL INTRODUCTION................................................................. ...............* 1.1 Maize: Production And Importance In Ghana....................................................... 1 1.2 Farm Level Maize Storage And Problems..............................................................2 1.3 Need For Improved Rural Storage......................................... .............................. * 3 1.4 Objectives............................................................................................................ 6 2.0 LITERATURE REVIEW................................ ...................................................... 7 2.1 The Larger Grain Borer, Prostephanus truncatus (Horn)....................................... 7 2.1.1 Taxonomy and Description................................ ................................................7 2 2.1.2 Geographical Distribution.................. ...................................... ...................... 7 2.1.3 Prostephanus truncatus in Ghana............................................................... . 8 2.1.4 Hosts................................................................................................................ 9 2. 1.5 Host Selection....... ........................................................................................ 10 2.1.6 Life History of P. truncatus.............................................................................11 2.1.7. Pest Status and Losses............................... .....................................................13 2.1.8. Population dynamics in rural stores.................................................................14 2.2 The maize weevil, Sitophilus zeamais.....................................................................16 2.2.1 Taxonomy and Description..............................................................................16 2.2.2 Distribution and pest status...... ................................................................... 17 2.2.3 Bionomics.......................................................................................................18 2.3. The Predator, Teretriosoma nigrescens Lewis (Coleoptera: Histeridae)................ 19 2.3.1 Taxonomy and Morphology.............................................................................19 2.3.2. Biology And Ecology......................................................................................20 2.3.3 Classical biological control of P. truncatus using T. nigrescens........................ 21 2.4.1 Rapid Loss Assessment Method....................................................................... 23 2.4.2 Visual Scales...................................................................................................24 2.4.3 Analysis and interpretation of results from visual scale data...............................25 2.4.4 Visual scales and economic value loss of stored commodity.............................. 27 University of Ghana http://ugspace.ug.edu.gh 3.0 MATERIALS AND METHODS........................................................................... 3.1 Experimental stores............................................................................................. 28 3.2 Experimental design: Maize varieties...................................................................30 3.3 Sampling for baseline data.................................................................................. 30 3.4 Sampling Procedure........................................................................................... 31 3.5 Sampling for weight loss..................................................................................... 32 3.6 Rapid loss assessment......................................................................................... 33 3.7 Rearing out......................................................................... ............................ . 34 3.8 Study site.......................................................................................................... 34 3.9 Analyses of results.............................. ......................................... ..................... 34 4.0. DESCRIPTION OF EXPERIMENTS..................................................................36 4.1 Effect of maize variety on population dynamics of P. truncatus and S. zeamais.... 36 4.1.1 Introduction.................................................................................................... 36 4.1.2 Materials and Methods.....................................................................................36 4.1.3 Results and Discussion....... ............................................................................. 37 4.1.3.1. Insect species recorded on the stored maize............................................... 37 4.1.3.2 Population dynamics of P. truncatus and S. zeamais in the long season...... 40 4.1.3.3 Spatial Distribution Within Stores...............................................................49 4.1.3.4. Distribution of ovipositon sites of P. truncatus and S. zeamais on maize cobs...........................................................................................................51 4.2 Weight losses among different varieties in the long season...................................... 54 4.2.1 Introduction:...................................................................................................54 4.2.2 Materials And Methods............... ................................................................... 55 4.2.3 Results And Discussion................................. ..................................................55 4.2.3.1 Weight Loss..............................................................................................55 4.2.3.2 Relationship Between Pests And Losses..................................................... 58 4.2.3.3 Association Of P. Truncatus And S. Zeamais With Some Secondary Pests. 60 4.2.3.4 Dynamics of T. nigrescens population....................................................... 63 4.3 Comparison of pest dynamics and weight loss between long and short seasons....... 68 4.3.1 Introduction....................................................................................................68 4.3.2. Materials And Methods...................................................................................68 4.3.3 Results And Discussion....................................................................................69 4.3.3.1 Insect Pests Of The Short Season...............................................................69 4.3.3.2 Population densities of P. truncatus and S. zeamais in the short season.......70 4.3.3.3 Weight losses in the short season................................................................73 4.4 Economic and Weight loss measurement using the Visual scale of damage............. 74 4.4.1 Introduction.................................................................................................... .. 4.4.2 Materials And Methods...................................................................................75 University of Ghana http://ugspace.ug.edu.gh viii 4.4.2.1 Weight and Economic Loss Assessment......................................................75 4.4.3 Results And Discussion.................................................................................... 77 4.4.3.1 Structure of cob damage............................................................................77 4.4.3.2 Diy weight loss by visual scales..................................................................80 4.4.3.3 Economic Value Loss................................................................................ 84 5.0 SUMMARY AND CONCLUSIONS......................................................................88 6.0 LITERATURE CITED......................................................................................... 91 APPENDIX............................................................................................................... 106 University of Ghana http://ugspace.ug.edu.gh List of Tables Table 1: Insect species found in the maize stores............................................................... ...... Table 2: Densities of Prostevhanus truncatus adults recorded on cobs of three maize varieties stored during the long season (October 1994 to May 1995)..........................42 Table 3 : Densities of Sitophilus zeamais adults recorded on cobs of three maize varieties stored during the long season (October 1994 to May 1995)....................................... 45 Table 4: Cob characteristics of three maize varieties used in experiment.... .............................47 Table 5: Grain hardness of the three maize varieties used in the experiment.......................... 47 Table 6: Densities of Prostephanus truncatus adults recorded in the sections of a bam compartment.......................................................................... .................................50 Table 7 Densities of Sitophilus zeamais adults recorded in the sections of a bam compartment.............................................................................................................51 Table 8: Percentage weight loss suffered by the three maize varieties stored during the long season (October 1994 to May 1995)........................................................................ 56 Table 9: Correlation coefficients of percentage weight loss with Prostephanus truncatus and Sitophilus zeamais density................................................................................ 59 Table 10: Correlation coefficients of Prostephanus truncatus with some secondary insect pests.......................................................................................................... 61 Table 11: Correlation coefficients of Sitophilus zeamais with some secondary insect pests.......................................................................................................... 63 Table 12: Correlation coefficients of T. nigrescens with P. truncatus and loss........................65 University of Ghana http://ugspace.ug.edu.gh Table 13: Changes in the ratio of T. nigrescens (Tn) to P. truncatus adults in the stores in the long and short seasons..................................................................................... Table 14: Density of P. truncatus. recorded on Abutia variety stored during the long and short seasons...................................................................................................... 71 Table 15: Density of S. zeamais. recorded on Abutia variety stored during the long and short seasons...................................................................................................... 71 Table 16: Percentage weight loss suffered by three varieties stored in bams during die long and short seasons..............................................................................................74 Table 17: Damage development of maize cobs of Abutia variety stored during the long season............................................................................................................... 78 Table 18: Damage development of maize cobs of Abeleehi variety stored during the long season.................................................................................................................78 Table 19 Damage development of maize cobs of Dzolokpuita variety stored during the long season................................................. ......................................................... 79 Table 20: Percentage weight loss using the visual scale for the three maize varieties stored during the long season..................................................................................... 83 Table 21 Percentage relative economic value loss using the visual scale for the three maize varieties stored during the long season.........................................................................86 University of Ghana http://ugspace.ug.edu.gh List of Figures Figure 1: Diagram of the experimental store showing compartments......................................... 31 Figure 2: Transverse section of a bam compartment showing the vertical sections.....................33 Figure 3: Densities of Prostephanus truncatus adults recorded on three maize varieties stored during the long and short seasons............................................................................... 41 Figure 4 Densities of Sitophilus zeamais adults recorded on three maize varieties stored during the long and short seasons............................................................................... 44 Figure 5: Percentage moisture content of the three maize varieties stored during the long short seasons................................................... .............................. ................... .......46 Figure 6 : Ovipositional distribution of Sitophilus zeamais and Prostephanus truncatus on maize cobs...........................................................................................................53 Figure 7: Percentage weight loss suffered by three maize varieties stored during the long and short seasons.............................................................................................................. 57 Figure 8 : Densities of Teretriosoma nigrescens adults collected from the three maize varieties stored during the long and short seasons.........................................................67 Figure 9: Percentage weight loss ( by visual scale of damages) for the three maize varieties stored during the long season....................................................................................... 82 Figure 10: Percentage relative economic value loss using the visual scale for the three maize varieties stored during the long season..........................................................................85 University of Ghana http://ugspace.ug.edu.gh List of Plates Plate 1: Visual Scale showing damage classes of maize cobs University of Ghana http://ugspace.ug.edu.gh 1 1.0 GENERAL INTRODUCTION 1.1 Maize: Production and importance in Ghana Maize (Zea mays ) is one of the most important food crops in Ghana. It is grown in all the agroecological zones of the country, although the intensity of cultivation and production systems differ in these ecological zones. Between 1986 and 1989, about 49% or 620,000 ha. of land area allocated to cereals was planted with maize (CIMMYT, 1990). Maize grows best on well-drained loamy soils, and is also grown in low lying areas and valley bottoms as a way of getting early crop outside the main growing seasons (NARP, 1993). However, the major maize belt of Ghana is the forest-savanna transition zone, which largely comprises the Ashanti, Brong-Ahafo, and parts of Eastern and Volta regions. Here, maize features as a major food crop with respect to production and monocropping of the crop accounts for 60% of the land area cropped to maize, with mixed cropping providing the remaining 40% of maize area (GGDP, 1991). Depending on agroecological location, there may be one or two cropping seasons of maize in Ghana. Although, rainfall can be erratic, especially with regard to onset, duration and distribution, two rainfed crops of maize a year are successfully produced in most agroecological regions of the southern half of the country, where rainfall is bimodal. There are two rainy seasons in the important production areas of the forest belt and the forest-savanna transition zone. These are the long (major) and the short (minor) seasons. Rainfall in the long season begins in March/April and ends in July/August while that of the short season normally begins in September and ends in October/November. With supplemental irrigation, three crops per year are easily obtainable (NARP, 1993) although this is scarcely done. The dry spell in August allows for harvesting of the long season crop and land preparation for the minor season. In the Northern parts of Ghana covering University of Ghana http://ugspace.ug.edu.gh 2 the Guinea savanna areas, the unimodal rainfall pattern spanning, over five months fropn May to September, provides for only one crop. In some years, rainfall, or more importantly the lack of it, in concert with incidence of Lepidopteran stem and cob borers, reduce the success rate of the short season crop. In spite of these drawbacks, maize production continues to grow nation-wide. The average annual yield increased from 141,000 metric tones in 1983 to 533,000 metric tonnes in 1987. It further increased to 961,000 metric tonnes in 1993 (PPMED, 1993) and is currently estimated to be 1,0 0 0 ,0 0 0 tonnes. Maize constitutes an important part of the diet of rural and urban Ghanaians and many traditional dishes use of maize have been compiled by MOA/WFED (1990). To the urban dweller it is a politically sensitive crop and a popular food prepared from it, 'Kenkey' was once brought to Parliament where the size of a ball was used as a measure of the state of the economy (NARP, 1993). It is an important source of protein in Ghana, ranking only after fish and legumes in terms of annual protein production (Twumasi- Afriyie et al., 1992). The importance of maize during the lean season is considerable since it stores longer than root crops. 1.2 Farm level maize storage and problems Maize production is largely done by small-scale farmers whose farm sizes are usually less than five hectares (CIMMYT, 1990). Large scale production of the crop is done mainly in the maize belt of Ejura-Techiman-Wenchi-Odumasi area in the forest-savanna transition zone. The annual yield of maize is usually beyond the immediate needs of the small scale farmer. This, in addition to other factors discussed by Nyanteng and van University of Ghana http://ugspace.ug.edu.gh 3 Apeldoorm (1971) calls for the storage of the surplus. At farm level maize is stored in cribs, bams, mud granaries, baskets, platforms, mudpots or gourds mostly as unshelled maize; husked (undehusked) or dehusked. Rawnsley (1969) and Nyanteng (1972) have reviewed the structures used for storage at the farm level, hi the Volta Region, where the work was conducted, the 'Ewe' bam is the most predominantly used store type, with about 90% of all stores being variations of this type (Addo, pers. com.). Proper storage protects the produce from damage by storage pests such as insects, mites, rodents and fungi. Insects are undoubtedly the most important pests of stored maize in Africa. Hall (1970) reported weight losses of 20% in stored maize due to the maize weevil Sitophilus zeamais Motschulsky in Ghana, and 50% in Uganda due to Tribolium castaneum (Herbst). De lima (1979) also reported 4-5% weight loss in stored maize in Kenya due to Sitotroga cerealella. However, with the outbreak of the larger grain borer(LGB), Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) on the African continent, weight loss figures have risen considerably. When P. truncatus becomes established in stores, it has enormous destructive potential which often surpasses that of S. zeamais and S. cerealella (Hoppe, 1986; Novillo, 1991). In Ghana, the pest status of P. truncatus has greatly increased, becoming most severe in the Volta Region, which shares a common border with Togo where the pest was first reported in West Africa (Krall, 1984). 1.3 Need for improved rural storage Previously, traditional maize varieties, usually stored on-farm with husks intact, often suffered only moderate losses from the maize weevil, Sitophilus zeamais (Dobie, 1977; McFarlane, 1988). Slaveys of traditional stores carried out in the late 1970s and early 1980s in eastern and southern Africa revealed average weight losses ranging from 2 to 6 % University of Ghana http://ugspace.ug.edu.gh 4 (Adams, 1977; De Lima, 1979; Golob, 1981a & b). Most improved varieties have however been reported to be more susceptible to pests across Africa; Benin (Anon, 1989), Cameroun (Almy and Asanga, 1988), Ghana (Badu-Apraku et al., 1992), Kenya (Giles and Ashman, 1971). Indeed, this apparently high susceptibility of traditionally-stored improved varieties to S. zeamais attack has been a major constraint to the adoption of these varieties in traditional maize-growing regions of West Africa (Kossou et a l, 1993). With the accidental introduction of the larger grain borer, which attacks stores of small scale farmers in the rural areas, the food security of these farmers, which is already precarious, is further threatened. Efforts to improve on-farm storage techniques have had varying degrees of success. Insecticide-based techniques for protecting maize in small traditional granaries are arguably the most commonly used control method. This has been further intensified following P. truncatus outbreak on the African continent. Chemical control methods based on the direct application on cobs of the binary insecticide dust, consisting of an organophosphorus and a pyrethroid (usually pirimiphos-methyl and permethrin) (Golob, 1988), or (pirimiphos-methyl and deltamethrin) (Biliwa, 1988) have been found to control both P. truncatus and other storage insects in Tanzania and Togo, respectively. Although these techniques have been used with some success in some parts of East Africa (Markham et a l, 1994a) they have not always been acceptable to farmers. The reasons are many and varied. The high humid conditions in fanners stores in the tropics cause the active ingredients of most insecticides to break down (Markham, 1981). Insect pests often become resistant to synthetic pyrethroids, and Golob et al, (1990) have already observed resistance in laboratory populations of P. truncatus. Also, problems with cost and availability of appropriate chemicals cause farmers to use unsuitable pesticides, and with University of Ghana http://ugspace.ug.edu.gh 5 their poor insecticide-use culture (where pesticides are applied unevenly or at inappropriate doses) accelerate the development of resistance by the insect pests. There is, therefore, the need for an improved small-scale, on-farm storage system based on methods that can be well integrated with traditional methods with a reduced dependence on chemical control methods (Markham et al, 1994c). The use of resistant crop varieties and biological control measures present cheaper and ecologically-sustainable alternatives for managing losses caused by storage insects. But the effectiveness of these control alternatives depends greatly on the knowledge of the pests’ ecology in the storage environment as well as the inter-relationships with other pests and natural enemies. The most important pests of stored maize in Ghana are S. zeamais and P. truncatus. Recently, a histerid predator, Teretriosoma nigrescens Lewis (Coleoptera: Histeridae) has been released in an effort to control P. truncatus. There are, however, indications that the prospect for complete classical biological control of P. truncatus may not appear very bright (Markham et a l, 1994a). There is, therefore, the need to examine how other processes, such as susceptibility of the maize substrate, which is one of the key factors that could influence the relationship between the predator and prey, can augment biological control. However, there is virtually no information on how the differential susceptibility of maize varieties in the countiy influence the population changes and distribution of these major pests in rural stores. Also very little is known about how results of laboratory screening tests for resistance translate in pest population dynamics and losses in the field. University of Ghana http://ugspace.ug.edu.gh 6 1.4 Objectives The general objectives of these field studies were: 1. To understand the effects of relative varietal susceptibility of maize on population dynamics of Prostephanus truncatus and Sitophilus zeamais in maize stores and on losses. 2. To compare population densities of the pests and losses incurred over the long and short seasons. 3. To understand how the economic value of stored maize changes during storage Specific questions addressed include the following; 1. What effect does maize variety have on insect densities and on grain losses, and whether this effect changes over time? 2. What important variables explain the loss rate? 3. What is the distribution of P. truncatus and S. zeamais in the entire store, and the ovipositional distribution on cobs? 4. How is T. nigrescens distributed on maize and is it associated with lower P. truncatus densities? 5. How are important secondary insects associated with primary pests? 6 . What are the differences in insect densities and losses between long and short seasons? 7. How does the economic value of the maize change over time, and is the change different for different varieties? University of Ghana http://ugspace.ug.edu.gh 7 2.0 LITERATURE REVIEW 2.1 The Larger Grain Borer, Prostephanus truncatus (Horn) 2.1.1 Taxonomy and description. Prostephanus truncatus was first described by Horn (1878) as Dinoderus truncatus Horn. The genus Prostephanus was erected by Lesne (1879). There are other species of Prostephanus but P. truncatus is the only one known to be associated with stored products (Hodges, 1986). Horn (1878), Lesne (1898) and Fisher (1950) provided an elaborate key for identification of P truncatus adults. Shorter identification keys are provided by Kingsolver (1971) and Hodges (1982). Spilman (1984) gives detailed descriptions of the larvae and pupae but unfortunately there are no identification keys for these stages. The greatest distance between the ventrally sclerotised lateral structures of the frontoclypeus has been used by Subramanyam et al (1985) for instar recognition However, Haines (1981) gives a description of all the developmental stages. The clypeal tubercles (Shires and McCarthy, 1976) may be used to sex the adults, while for pupae it can be determined according to the size and shape of the genital papillae (Bell and Watters, 1982). 2.1.2 Geographical distribution The distribution of P. truncatus is considered veiy phenomenal, compared to some major storage pests, in that it is very discrete and delimited (Markham et a/., 1991). Although, the insect was originally described from specimens collected in California, it appears to be indigenous to Mexico and parts of Central America where it sporadically achieves a primary pest status of local importance on maize (Markham et al, 1991). The Larger grain borer outbreak in Africa apparently resulted from separate introductions; first in Tanzania (Dustan and Magazini, 1981) then Togo (Krall, 1984) University of Ghana http://ugspace.ug.edu.gh 8 and again into Guinea (Kalivogui and Miick, 1991). From Tanzania the insect has spread to the nearby countries of Kenya (Kega and Warui, 1983), Burundi (Gilman, 1984), Malawi and Rwanda (GTZ, unpublished data cited by Hodges, 1994). The Togo introduction has apparently spread to Benin (Krall and Favi, 1984), Ghana (Dick et a l, 1991) and Nigeria (Pike et a l, 1992). 2.1.3 Prostephanus truncatus in Ghana As a consequence of an earlier introduction of the Larger Grain Borer into Togo in 1984, a survey was carried out in 1986 in Ghana on the Ghana-Togo border to determine the presence of the beetle in the country (Ofosu, pers. com.), but the results proved negative. It was not until 1989 when a second survey revealed that LGB was present in villages in the Volta region over a 150km stretch of the border with Togo; from Kpetoe in the south to Ameyoe in the north (Dick et a l, 1989). Another survey in April 1991 revealed considerably higher population build-up at places where the insect had been found in the earlier survey (Ayertey and Brempong-Yeboah, 1991). Two years later, a Ghana LGB Project survey showed that LGB was present in all districts of the Volta Region (GLGBP, 1993) A national monitoring programme, with 101 permanent trapping sites, has revealed the incidence of the insect in every region of the country (Robin Boxall pers. com.). The most seriously affected parts of the country, apart from the Volta Region, are the Saboba district of the Northern Region, the Dangbe West district of Greater Accra and the areas of Eastern regions bordering the Volta Region (Addo, 1994). University of Ghana http://ugspace.ug.edu.gh 9 2.1.4 Hosts Although P. truncatus has been reported to attack a wide range of substances (Mushi, 1984) not all can be considered as satisfactory hosts for the beetle. Intensive research efforts in recent years however continue to reveal the expanding host range of the beetle. P. truncatus has been known to infest both stored maize (Giles and Leon, 1975; Hodges Qt a l, 1983a) and maize in the field before harvest (Giles, 1975), although there is much variation in the extent to which this occurs (Henckes, 1992; Wright et a l, 1993; Tigar et a l, 1994). It also infests dried cassava (Manihot esculenta Crantz) as well (Nyakunga, 1982; Hodges et al, 1985). P truncatus develops and reproduces on both of these crops. Reproduction was also recorded on soft wheat and chickpeas (Howard, 1983). More recently, previously par-boiled dried tubers of white yam (Discorea rotundata Poir) and sweet potato (Ipomoea batatas (L.) Poir) have been shown in the laboratory to support reproduction of the insect (Li, 1988; Jalloh, 1989). With the advent of pheromone-baited flight-traps to monitor the pest, it has been observed that there are large populations of P. truncatus in the natural environment away from maize fields and stores (Rees et a l, 1990; Nang’ayo et a l, 1993). Laboratory investigations in Togo (Helbig et a l, 1990) and Kenya (Nang’ayo et al, 1993) have shown that the pest is capable of developing and reproduction on dried twigs of a range of tree species. In Kenya, this includes about 15 tree species from the families Leguminosae, Burseraceae and Anacardiaceae. The woody stems of cassava and Flamboyant (Delonix regia ) have also been shown to support the reproduction of Larger grain borer (Detmers et al, 1990), a capacity that the authors attribute to the high starch content of these materials. University of Ghana http://ugspace.ug.edu.gh 10 Very little infonnation is available so far on the natural origins of the Larger grain borer. Maize cannot, as pointed out by Markham et al. (1991), be the natural host of the insect since the plant itself does not exist in nature and seems to have been selected by man only within the last 5000 years. Chittenden (1911) concluded that roots and tubers were the natural hosts of P. truncatus, after finding that various root crops imported into the United States from Mexico and Guatemala were infested with the pest (Riley, 1894). The problem is to clarify the field conditions under which naturally occurring root and tubers would become sufficiently diy to provide a satisfactory substrate for the development of P. truncatus (Markham et al., 1991). The laboratory demonstration by Li (1988) of development of the pest in acorns (from English oak, but not on those from scarlet oak) prompted Markham et al. (1991) to speculate that the great diversity of oak species in the area of origin of the insect might represent a natural host for the Larger grain borer. 2.1.5 Host selection Very little is known about the host selection behaviour of P. truncatus. It has been postulated by Markham et al. (1994c) that initial infestation levels are determined by seasonal migration patterns and host finding behaviour of P. truncatus, in relation to local farming practices (such as tuning of harvest, selection of cobs for storage, choice and management of the storage structure, among others). However, Hodges (1994) suggested that host selection, particularly in the case of maize and cassava, may occur largely by chance. This resulted from the observation that P. truncatus adults are attracted over short distances to maize grains (Detmers, 1990) and dried cassava (Wright et a l ., 1993) but there is no long-range attraction to these stored products in the field (Tigar et al., 1994; Wright et al., 1993). Pike et al., (1994) identified hexanoic University of Ghana http://ugspace.ug.edu.gh 11 acid and nonanal as the major active component of maize and cassava volatiles, respectively. It is thought that adult males, arriving at a suitable site make test burrows, which are normally 1-2 cm long (Hodges, 1986) in search of food. The beetles leave these burrows and try again elsewhere if no suitable food is found. However, once suitable food is located, the male secretes a pheromone to attract females and other males (Dendy et a l, 1989; Cork et a l, 1991) which also arrive to exploit both food and mates. Aggregation of both male and female P. truncatus may occur in this way (Hodges, 1994) leading subsequently to reproduction, pest damage and population increase. These activities depend, of course, on the biotic potential of the particular P. truncatus population in relation to local climate, maize variety, activities of natural enemies and competition with other storage pests. 2.1.6 Life History of P. truncatus The developmental biology of P. truncatus has been studied, on maize (Shires, 1977, 1979, 1980; Bell and Watters, 1982; Howard, 1983) and on cassava (Nyakunga, 1982). There are a number of apparent discrepancies (reviewed in Hodges, 1986) among the various laboratory studies. This variation probably reflects differences in aspects of the experimental technique, such as form and consistency of food substrate presented (Dick, 1988). But Subramanyam and Hagstrum (1991) suggest a contribution could have come from the different strains used by these researchers. The optimum conditions for development on maize are approximately 32°C and 70- 80% relative humidity (r.h.). Under these conditions, Bell and Watters (1982) observed University of Ghana http://ugspace.ug.edu.gh 12 that the life cycle was completed in 24-25 days on tightly packed ground maize or on whole grain. This contrasts sharply with Shires (1979, 1980) who observed a developmental period of 35.4 days on loosely packed maize flour. P. truncatus is relatively long-lived as an adult and does not lay eggs as rapidly or as haphazardly as shorter-lived stored product beetles. The eggs are laid within maize grains, in blind end tunnels created by the adults. Grain stability influences oviposition rate (Howard, 1983). Under optimum conditions, larvae hatch after 4.1 days and the mean larval period is 16.1 days (Bell and Watters, 1982). There are three larval instar stages (Bell and Watters, 1982; Subramanyam et a l, 1985). The mean pupal period for P. truncatus lasts 4.7 days. The mean development period on yellow ’American No. 3' maize and blocks of dried cassava at 27°C and 70% r.h were 39.2 days and 43.1 days respectively, suggesting that maize is more suitable for development of P. truncatus than cassava (Nyakunga, 1982). Adult longevity is also variable, although differences observed were not statistically significant. Shires (1980) found that on maize flour at 32°C and 80% r.h, females generally outlived males; the mean life expectancies being 61.1 days and 44.7 days, respectively. However, the opposite was found to be the case when Howard (1983) maintained adults at 25°C and 70% r.h on maize flour and grain. The differences in female longevity observed may be correlated with differences in oviposition rates; females in Howard’s studies having a much higher rate and hence perhaps shorter life span. P. truncatus is tolerant to diy conditions. Maize with moisture contents in the range of 9-10.6% have been found to be heavily infested. This was demonstrated by Young et University of Ghana http://ugspace.ug.edu.gh 13 at, (1962) and confirmed in laboratory studies by Shires (1979), Hodges and Meik (1984), and in the field by Giles and Leon (1975) and Hodges et al (1983a). 2.1.7. Pest status and losses. Larger Grain Borer is a pest with high damage potential. The insect infests cobs shortly before harvest and during storage, but appears to spread during the storage season. Hodges (1984) found in Tanzania that the pest could be found in 20% of stores, immediately after harvest and at the end of a 5-month storage period, 80% of the granaries were infested by P. truncatus. Also the timing of onset of attack may vary from year to year. Hodges (1986) noted that at one location in a particular year, the infestation may be extremely severe, but at the same place and time the following year, few if any, beetles may be found. The reasons for this seemingly sporadic pest incidence are not clearly understood, but might include weather influences, exhaustion of food resources the previous season or density-dependent mortality factors. Weight losses after six months of storage of maize cobs have increased from 7.1% before the advent of P. truncatus to 30.2% afterwards, and up to 44.8% over 8 months of storage has been recorded in Togo (Pantenius, 1988). As the cobs affected were unfit for human consumption, Pantenius (1988) therefore concluded that real loss could be some 10% higher. In Tanzania, after being in store for 3-6 months, maize cobs showed weight losses of up to 34%, averaging 9% (Hodges et ah, 1983a). Comparison with earlier published data suggests a major increase in losses following the introduction of the Larger grain borer (Laborius et a l, 1989). Keil (1988) reported particularly severe losses also in Tanzania, averaging 17.9% after six months and 41.2% after 8 months of storage in maize cobs. University of Ghana http://ugspace.ug.edu.gh 14 In the Neotropics, the pest status has been the subject of considerable debate. Giles (1977) reported the occurrence of weight losses of 40% after 24 weeks with heavy infestation of P. truncatus in Nicaragua. Also Hoppe (1986) has reported serious losses of over 30%, particularly associated with heavy P. truncatus infestation from Honduras. However, lower loss figures reported by Boye (1988) from Costa Rica; averaged 5% after six months of storage during the rainy season and 12% after six months of storage during the diy season. As Markham et al.( 1991) rightly pointed out, the impression that losses are generally higher in Africa than in the Neotropics persists (Boye et a l, 1988; Laborius et al, 1989; Laborius, 1990a), despite serious local losses that had been reported (see also Barnes et a l, 1959; Ramirez, 1959). The most obvious cause of this loss is the conversion of the maize grain into flour by the boring adult beetles (Li, 1988). Until a large population of larvae is established, Hodges (1986) pointed out, feeding activity of the larvae would represent a secondary source of loss. P. truncatus also causes extensive damage to dried cassava chips. In a trial in Tanzania, Hodges et a l, (1985) recorded losses of up to 70% in fermented and 50% in unfermented cassava after only 4 months of storage. 2.1.8. Population dynamics in rural stores. The optimum developmental conditions of P. truncatus are similar to those of other storage pests. Thus the physical and biological constraints that affect the dynamics of these pest populations would similarly affect P. truncatus. However, with its comparatively limited geographical distribution and erratic incidence in both space and University of Ghana http://ugspace.ug.edu.gh 15 time, it is not clearly certain which factor might determine its incidence. Until recently, data on the dynamics of P. truncatus with respect to different environments, storage practices and pest associations have been scanty. Most of these studies, as reviewed by Markham et al (1991) were conducted in the Neotropics, but they nevertheless provide useful information which can help in providing a better understanding of the behaviour of this pest in West African stores. Markham et al (1991) noted that in almost all these studies, the maximum numbers of the three primaiy pests; Sitophilus zeamais Mots., Sitotroga cerealella (Olivier) and P. truncatus occur at different points in the storage cycle. Data on the interactions between these species in their natural storage environment are scarce. Results of a farm storage survey in Honduras indicated that 90% of the samples infested by P. truncatus were also infested by S. zeamais, although there was no relationship between the numbers of the two species present in individual samples (Hoppe, 1986). Studies in the laboratory have also provided some insight into competitive interactions between P. truncatus and S. zeamais. (Howard, 1983, Haubruge and Verstraeten, 1987). In traditional stores, there are differing relationships between secondary pests and primaiy pests. The incidence of Cathartus quadricollis (Guerin) and Carpophilus spp. where present, appears to be independent of that of P. truncatus. The dynamics of Gnatocerus maxillosus (Fabricius) on the other hand correlates closely with P. truncatus abundance ( Novillo, 1991 cited in Markham et al 1991). A close association was also observed between the incidence of Tribolium castaneum (Herbst) and P. truncatus in Africa (Hodges et al, 1983) but Novillo (1991, cited in Markham University of Ghana http://ugspace.ug.edu.gh 16 et a l, 1991) reported the apparent displacement of T. castaneum by G. maxillosus later in the season. A small number of natural enemies have been found associated with P. truncatus in the Neotropics, the most important among them is the predator Teretriosoma nigrescens Lewis, a species not known to be present in Africa. T. nigrescens has since been introduced into East and West Africa. In African maize stores, a number of species of natural enemies are associated with the indigenous pests, but it is not clear at present whether any of the indigenous species of insect predators or parasitoids may have any effect on P. truncatus as a host. Previous infestation of the wooden storage structure has been noted to influence the population dynamics of P. truncatus by hastening recolonization of the maize and early establishment of the beetle on the new grain placed in storage. 2.2 The maize weevil, Sitophilus zeamais 2.2.1 Taxonomy and description The maize weevil, Sitophilus zeamais Motschulsky is a coleopteran stored grain pest (Family Curculionidae, subfamily Calandrinae). There are three species in this genus. The others are the rice weevil, S. oryzae (L.) and the granary weevil, S. granarius (L.); the taxonomy of this genus has been revised by Floyd and Newsom (1959). In the tropics and subtropics, the important species are the S. zeamais, which is more common on maize, and S. oryzae , which is frequently associated with small grains. These two species are difficult to separate morphologically. Their body size ranges University of Ghana http://ugspace.ug.edu.gh 17 from 2.4 to 4.5 mm but S. zeamais is invariably the bigger, whilst body colour varies from light brown to black. Both have four reddish patches on the elytra but those of S. zeamais are more clearly defined (Apert, 1987). They have an elongated characteristic rostrum with capitate, elbowed antennae. However, the species can be separated by examination of the internal genitalia (Halstead, 1964). The dorsal surface of the aedeagus of S. zeamais has two parallel grooves while that of S. oryzae is smooth and convex (Proctor, 1971). In the female S. zeamais , the prongs of the Y-shaped sclerite are pointed at the ends and the gap between is wider than their combined width. However, in S. oryzae these prongs are rounded and the gap between them is narrower than their combined width (Anon, 1989). Again, the arrangements and patterns made by the pronotal pits has been used in separating these two species (Fisher, 1987). 2.2.2 Distribution and pest status Sitophilus zeamais is cosmopolitan in its distribution, and is found in the tropics and all warm climates. S. zeamais attacks all kinds of cereals but is predominantly found on maize (Longstaff, 1981; Ayertey and Akibu, 1982). S. zeamais is a strong flier and can readily fly from stocks of stored grain over distances of 400-800 meters (Chestnut, 1972) into growing maize. At harvest time all stages of development of this insect are present. Adults first emerge just before harvest and emergence continues for some time thereafter (Giles and Ashman, 1971). Most studies have shown that among several factors affecting distribution and numbers of S. zeamais in the maize growing crop, the proximity and intensity of a source of infestation (usually infested cereals in store) and the degree of kernel coverage by the husk appear to be most important (Giles and Ashman, 1971; Dobie, 1977; Golob, University of Ghana http://ugspace.ug.edu.gh 18 1984). Colonising weevils, which are predominantly males, aggregate on cobs (Markham, 1981; Dix and All, 1985). This behaviour is thought to be due to male- produced aggregation pheromones. Before the accidental introduction of the Larger grain borer, the maize weevil used to be one of the most important primary pests of stored grain in Ghana, causing estimated weight losses of 7-20% (Anon, 1978; Hall, 1970) in stored maize, and in some cases up to 35% after 6 months of storage (Wheatly, 1973 cited in Yoshida, 1983). Generally, most damage and weight loss figures due to S. zeamais reported in local maize stores have been low. Schulten (1975) for instance, reported weight losses of 1-2% and 10% in local and improved varieties respectively. In some cases the introduction of higher yielding maize varieties has exacerbated storage pest problems, especially in the rural areas, as these improved varieties store poorly in rural granaries compared to the traditional varieties. Similar trends were reported from studies in other African countries (see Kossou et a l, 1993). It is important to note that the extent of damage and weight loss recorded is often influenced by time of storage as well as form of storage. 2.2.3 Bionomics Aspects of the basic developmental biology of the maize weevil have been carried out on shelled kernels by Schoonhoven et a l, (1974), Tipping (1986) and Urrelo and Wright (1989), and more recently on unshelled maize by Kossou et a l (1992, 1993) and Vowotor et al (1994, 1995a). In a study of Walgenback and Burkholder (1987) it was observed that in S. zeamais populations, several matings may take place, with copulation lasting for about 4 .8 hours University of Ghana http://ugspace.ug.edu.gh 19 on the average. Eggs are laid throughout the adult life of the insect. The eggs are laid individually into small cavities chewed into the surface of the grain by the female S. zeamais and covered with a waxy secretion which when hardened is referred to as 'egg plug’. At all moisture contents above 10% and between 20 - 32°C at 70% r.h (Anon 1984; Okelana and Osuji, 1985) oviposition rates are at their maximum. Optimum conditions for development are 25 - 27°C and 70% r.h. (Dick, 1988). Under these optimal conditions, the apodous larva takes 6-7 days to hatch and there are four larval instars. The larva chews through the grain kernel, leaving a characteristic tunnel and it is this chewing activity of the larva which causes weight loss of the stored maize (Urrelo and Wright, 1989; Shade et al., 1990). The total length of the larval and pupal stages are 18-22 days and 6-7 days, respectively. The newly developed adults chew their way out, leaving characteristic emergence holes. The total development period of S. zeamais under optimal conditions has been shown to vary from 31-37 days, depending on the variety of maize used as food as well as storage form (Urrelo and Wright, 1989; Dick, 1988; Vowotor et al., 1995a). Shorter developmental periods have been reported in susceptible varieties (Dobie, 1974; Kossou et al., 1993) and lower numbers of weevils have emerged from cobs than from shelled grains (Kossou et a l, 1992; Vowotor, 1992). 2.3. The predator, Teretriosoma nigrescens Lewis (Coleoptera: Histeridae) 2.3.1 Taxonomy and morphology Teretriosoma nigrescens, first described by Lewis (1891), belongs to the family Histeridae, which has over 3700 species, but Hinton (1945) identified only 14 species as associated with stored products worldwide. However, until recently, evidence of the University of Ghana http://ugspace.ug.edu.gh 20 presence of T. nigrescens was found only in Mexico and Central America (Haines, 1981; Boye, 1988; Rees et al., 1990) in association with a pest of erratic importance, P. truncatus , on farm stored maize (Hoppe, 1986). It has now been introduced into Tanzania, Togo (Rees, 1991), Benin and Ghana as part of biocontrol efforts in Africa against P. truncatus. T. nigrescens, like all other histerid beetles is about 2.3mm in size (Leliveldt, 1990) with unusually hard chitinous cuticle which is shiny black. The hard elytra do not completely cover the abdomen, leaving two abdominal segments uncovered. The antennae are jointed and thicken up toward the tip like a club. The tibia of the first pair of legs is toothed and is a feature used to define the species (Poschko, 1994). External sexual dimorphism is not apparent in T. nigrescens so sex is determined after dissection. 2.3.2. Biology and ecology Early investigations into the development cycle of T. nigrescens by Rees (1985) revealed that the developmental time from egg to adult to be 56 days at 26°C 70% r.h. while Leliveldt and Laborius (1990) observed complete development in an average of less than 50 days at 30°C and 70% r.h. However, Oussou et a/.(1995) observed a significantly shorter overall development period of 23.5 days at 30°C and 70% r.h. According to Rees (1985) the relatively large eggs (1.1 x 0.5 mm) laid singly amongst the commodity, hatch in about 7 days at 27°C. The compodeiform larvae, 2-3 mm long on eclosion, have flattened head bearing large sickle-shaped mandibles. The head and prothorax are heavily sclerotized. There are two larval instars lasting between University of Ghana http://ugspace.ug.edu.gh 21 about 12 days (Oussou et a l 1995) and 26 days (Leliveldt, 1990) all together at 30°C, 70% r.h. and 30°C, 75% r.h., respectively. Rees (1985) observed that the fully grown larva, 1.0 -1 .2 cm long, pupates within a chamber in presumably already damaged grain, fashioned with its mandibles. The pupation stage takes 11-18 days at 30°C (Leliveldt and Laborius, 1990; Oussou et a l, 1995). T. nigrescens is a long-lived insect During long-term experiments, T. nigrescens imagines reached an age of over 20 months. They were still able to reproduce after 16.5 months (Poschko, 1994). T. nigrescens is a polyphagous predator with a distinct preference for P. truncatus as a host. Although it has shown the ability to reproduce on two species of Dinoderus , on Sitophilus oryzae, S. zeamais, Rhizopertha dominica, and Tribolium castaneum , comparative experiments however confirmed that P. truncatus is the host preferred by this predator (Rees, 1987; 1991; Poschko, 1994). Like other histerids, T. nigrescens hunts aggressively for its prey. Indeed, large numbers of adult T. nigrescens have been caught in maize fields in traps baited with Prostephanus pheromone, indicating that this predator is capable of locating P. truncatus over much greater distances (Boye, 1989). Poschko (1994) found that T. nigrescens is not purely a predator but can also make use of plant food. However, Poschko (1994) felt that the fear that it could be a pest on stored food itself was unfounded, since the losses to plant materials were low enough to be neglected. 2.3.3 Classical biological control of P. truncatus using T. nigrescens. The conventional response to P. truncatus outbreak in Africa, according to Markham et al. (1994a) has gone through a familiar cycle; attempted eradication, containment by University of Ghana http://ugspace.ug.edu.gh 22 statutory and other regulatory measures and testing of insecticide-based strategies for long-term management of the pest A considerable success was reported by Golob (1988) integrating the application of a binary insecticide with modification of traditional storage practices in East Afiica. But in West Africa, the humid conditions have apparently limited the adaptability of this strategy (Markham et a l, 1994b). Classical biological control, especially with T. nigrescens, amongst the limited arthropod natural enemies of P. truncatus, has clearly been an appealing alternative. The body similarity in dimensions and in environmental requirements of T. nigrescens and P. truncatus suggest that the predator is well adapted to the habits and habitat of P. truncatus. Significantly, T. nigrescens can perceive the aggregation pheromone produced by P truncatus and is therefore able to search specifically for its prey. Also since P. truncatus could be found in a wide range of habitats, the use of T. nigrescens , an insect that appears to be closely associated with its prey both within and outside the storage environment, may offer the only practical means to control it T. nigrescens adults and larvae consume both eggs and larvae of P. truncatus . Laboratory studies on the predatory behaviour of T. nigrescens by Leliveldt and Laborius (1990) showed that one T. nigrescens adult consumes an average of 5.7 eggs or 4.9 larvae of the pest during a 24 hour period in the dark, although Rees (1985) found an average of 1.7 P. truncatus larvae per day. Results from further experiments on suppression of the growth of P. truncatus population are very promising. On maize in jars in the laboratory, Rees (1985) found that at a predator-prey ratio of 1:10, numbers of prey declined in all treatments containing predators after 8 weeks, whereas numbers of prey in controls increased by at least 10-fold. University of Ghana http://ugspace.ug.edu.gh 23 Again in Costa Rica, Boye (1988) reported a decline of 87% of P. truncatus population on shelled maize by the predator after 110 days, compared to the control. In a corresponding test using maize cobs, the population of P. truncatus was reduced by 72%. By using T. nigrescens, the loss of shelled maize and cob maize could be reduced by 76% and 62% respectively whereas damage to the grain and the cobs caused by P. truncatus was reduced by 47% and 21% respectively (Boye, 1988). Similar results were achieved in Togo according to Poschko (1994, citing Helbig et al, 1992) in small stores of maize under semi-practical conditions within a storage period of 9 months. T. nigrescens reduced the number of P. truncatus adults by 46.5% and damage to the grain by about 42% compared to control The mean population of T. nigrescens increased within the experimental period from 67 to 876 adults per 100 cobs. The predator had been readily maintained in the laboratoiy, where it proved hardy and long-lived. It is felt that the voracious predation by both adults and larval T. nigrescens would outweigh any disadvantages from the predator's somewhat slow development time and low reproduction rate. Overall, the potential of T. nigrescens as a control agent looks sufficiently promising to justify field releases, which in the end provides one of the best complementary control strategies in a holistic approach to the control of P. truncatus. Field releases have, however, been made in Togo, Benin, and Ghana. 2.4.1 Rapid loss assessment method. The primary aim of undertaking measurement of post harvest food losses is to establish the seriousness of the situation in order to determine whether intervention measures are University of Ghana http://ugspace.ug.edu.gh appropriate, and where feasible, to appreciate the impact of measures designed to reduce them, for example, new storage technologies, new varieties, etc. Several methods exist for the assessment of losses in stored grains (see Harris and Linblald, 1978; Boxall, 1986; Ratnadass and Fleurat-Lessard, 1991). However, most of these are unreliable, labour intensive and relatively slow, such that a number of researchers (Pantenius, 1988; FAO-PFL, 1990) have called for the development of simpler methods which are suitable for use in a rapid rural appraisal situation. 2.4.2 Visual scales In response to the needs of a specific farm storage survey, a rapid loss assessment method using visual scales of loss was developed by the Ghana Ministry of Food and Agriculture / United Kingdom Overseas Development Agency Larger Grain Borer Project (MOFA/ODA LGBP). This method has been described in Compton et al. (1992). However, the method was originally developed for use in surveying farm stores in central Togo (Compton 1991; Stabrawa, 1992). The objectives of the survey were: a) to assess levels of post-harvest loss in maize cobs and dried cassava chips due to insect pests; b) to evaluate the relative importance of the larger grain borer or LGB (Prostephanus truncatus), a post - harvest pest new to central Togo; and c) to identify those aspects of production and storage practice which were associated with lower storage losses. The use of visual scale of damage for rapid survey was appealing because of practical constraints encountered, namely; limited time and facilities, unavailability of data on the University of Ghana http://ugspace.ug.edu.gh condition of the commodities prior to storage, and the unreliability of many standard loss assessment measures such as the standard count - and-weigh and the converted percentage damage methods with particular reference to the nature of damage to maize cobs by P. truncatus. Four damage classes numbered from 1 (no damage) to 4 (severe damage) were defined from several fumigated cobs which had been previously attacked by a variety of insects (mainly stalkborer larvae, grain weevils and larger grain borer). Similarly five damage classes were distinguished for dried cassava chips mainly holed by Dinoderus sp. and P. truncatus. Representative cobs and cassava chips from each damage class, including the highest and lowest damage in each class, were photographed for use as a reference standard. The scales do not measure mould damage. The scales were used in the Togo survey to score the damage in maize cobs and cassava chips sampled from farmers' stores. Full details of sampling methods and other data collection techniques can be found in Compton (1991) and Stabrawa (1992). 2.4.3 Analysis and interpretation of results from visual scale data. Different ways of analyzing visual scale data have also been described by Compton et al.(1992). Basically, analysis can be performed on the original (ranked) data to investigate the relationship between damage levels and other survey variables be they qualitative factors (such as crop variety or store type) or quantitative factors (such as insect numbers or length of storage period). Moreover, by calibrating the scales the original data can be used to estimate weight loss (and other parameters) in stored produce. University of Ghana http://ugspace.ug.edu.gh Log linear modeling or cross-tabulation, in conjunction with chi-square tests of association can be used to investigate the effects on damage of qualitative factors (Compton et al., 1992). However, the effects of quantitative variables can be studied by means of rank correlation or ordinal logistic regression (McCullagh, 1980). Multiple ordinal logistic regression was used in the Togo survey, for instance, to investigate the relationship between damage levels in maize and density of the two main pests; P. truncatus and Sitophilus sp. (Compton, 1991; Stabrawa 1992). In using visual scales for weight loss estimation the scale should be calibrated against a widely used, reliable standard weight loss assessment measure to determine the mean weight loss for each damage class in the scale. Where practical constraints require modification of the loss assessment procedure it should be carried out. For instance, standard count - and - weigh technique for assessing grain losses caused by several insects species has been realized to underestimate weight loss of softer maize varieties stored on cobs when attacked by P. truncatus. Therefore, Compton et al. (1992) calibrated their visual scale against a modified version of the standard count-and-weigh method described in Compton et al. (1995) which incorporates destroyed and missing grains for accuracy of the mean weight loss values of each damage class. However, it is interesting to point out that Meikle (pers. com.) working with four maize varieties (two local Mid two improved) found no significant differences among varieties in the ‘whole- kemel loss per cob’ for the first six months. After eight months of storage, TZSR-W (an improved and the hardest variety, and the one with the highest P. truncatus density) had about 80% of the number of kernels on cobs from the first month, but Poza-Rica, also improved and softer, had 97%. Therefore, the standard count and weigh technique may not be that bad. University of Ghana http://ugspace.ug.edu.gh 27 2.4.4 Visual scales and economic value loss of stored commodity. Insofar as market value is related to the appearance of the commodity, it should be possible to devise and calibrate visual scales to give an estimate of economic value. In most areas fanners and traders use an informal visual grading system in the sale of shelled maize. Thus, in valuing stored maize, understanding how the fanning family selects cobs and grains for sale and consumption is very important (Compton et al, 1992). The Ghana MOFA/UK ODA LGB Project is attempting to develop the visual scale to give a rapid estimate of economic value loss (Compton, pers. com.). The principle is the same as obtaining the weight loss. However, two measures of loss are calculated here, for each damage class. These are; (a) the average loss in volume of grain produced since grain is sold by volume in "standard" market bowls, b) die average loss in price for the grade of grains produced by shelling cobs in each class. The value loss is then calculated as: l-((l-% price loss) x (1 - % volume loss)). Several cob samples are needed for this calibration and it should be done over the entire season since price changes a lot in the course of a season, especially during the lean season. University of Ghana http://ugspace.ug.edu.gh 28 3.0 MATERIALS AND METHODS 3.1 Experimental stores The maize store structure used in this experiment was that of raised platform version of the traditional 'Ewe' bam (Fig. 1). In this structure a 2 m x 2 m split bamboo (Bambusa vulgaris Schrad ex Wendland) platform is supported by eight wooden legs of fan palm (Borassus aethiopum Mart.) on the exterior and one post in the center. This platform is lm from the ground. Chicken wire is attached to each leg above the platform thus dividing a store into eight equal compartments. Below the platform rodent guard is fixed onto each of the legs, about 0.6 m from the ground. On the platform, maize cobs were stacked with the husk-on in a circular fashion to mimic a typical traditional ‘Ewe’ bam. Large cobs were carefully stacked on the exterior with the remaining cobs filling the inside column into a compact cylinder. A store was about 1.8 m in diameter and approximately 0.8 m in height. As the chicken wire provided support, a compartment could be completely emptied and sampled without affecting the rest of the store. The advantage of this design is that it provides thorough sampling (in the exterior as well as the inner perimeter) of a traditional store. A thatch roof of available grass species, Imperata cylindrica (L.) Beauv., covered the maize against sun and rain. There were sixteen of these stores and each store was three metres apart within a block. Four extra stores were built to serve as replacement stores. University of Ghana http://ugspace.ug.edu.gh Fig 1 D w g iw of the experimental store showing comparing hbff* TIk post at the comei closest to the observer has been cul away When the store is completely stacked with maize cobs, it is cylindrical (Adapted from Meikle ei al.t 19%) University of Ghana http://ugspace.ug.edu.gh 30 3.2 Maize varieties The trial was set up for maize grown in both the long (major) and short (minor) seasons. For the long season maize, three maize varieties were used. These were Abeleehi (an improved variety), Abutia and Dzolokpuita (local varieties). These varieties which have gone through fanners' selection for many years were obtained from fanners fields, hi September 1994, the maize varieties were stacked in the stores in a randomised complete block design with four replications, thus occupying twelve of the sixteen stores. The remaining four stores were stacked with short season maize (Abutia) in Januaiy 1995. the long season trial lasted for eight months while short season maize stayed in the bams for five months. 3.3 Sampling for baseline data At the very beginning of the storage season, fifty cobs per variety were sampled. These were dehusked, shelled and sieved and all adult insects were collected, identified and counted. The grain was evaluated for weight loss (using the standard count and weight method) and for moisture content (using the T)ole' grain moisture tester). Again, 10 cobs per variety were weighed as units, their cob characteristics were measured, then dehusked and shelled. All components of the cobs were weighed to measure differences in varieties, including relative grain weight per cob. University of Ghana http://ugspace.ug.edu.gh 31 3.4 Sampling procedure A compartment was sampled from each store every four weeks and this was done only once for a compartment. The compartment to be sampled was selected at random (by assigning each compartment in each store a unique number and using a random number generator). A compartment was considered to consist of three main vertical sections; the surface (with the larger, carefully stacked cobs), the middle and the inner centre of the store) sections as shown in Fig. 2 below. Fig. 2: Transverse section of a barn compartment showing the vertical sections Thirty cobs were selected and removed from each section. The cobs on the surface were selected by marking with a permanent marker. Care was taken that selected cobs did not touch each other. While the compartment was being dismantled layer by layer from the top, each marked cob was removed along with a cob from the middle. When a third of the surface and middle cobs had been removed, ten cobs were picked randomly from the inner section. This procedure was followed until the whole compartment was completely University of Ghana http://ugspace.ug.edu.gh 32 dismantled. Cobs sampled were kept separately in labelled plastic bags while those were kept apart according to their vertical zone. Thirty replacement cobs of the same variety from the replacement store were added to each vertical zone. These cobs were then used to re-construct the compartment after sampling to ensure that the store integrity and bulk were maintained. Sampled cobs were then taken from the field to the laboratory. 3.5 Sampling for weight loss To determine grain weight loss twenty-six out of the thirty cobs sampled four-weekly from each section per store were used. They were dehusked, and shelled inside the plastic bags into which they were sampled to ensure that all insects as well as maize grains were retained. The insects were sieved out using sieve wire of mesh size 4 mm2 and then put into small plastic bags before being frozen for later identification. The weight and moisture content of the maize grains were then taken. Two subsamples of 1000 grains were used for diy-weight loss assessment, using the standard count-and-weigh technique (Boxall, 1986). In this technique, the grain subsample was separated into damaged and undamaged grains and grains with any kind of boreholes were regarded as damaged. Both the damaged and undamaged grains were counted by hand with a ’safespoif hand Tally counter and weighed with the electronic scale (OHAUS Advanced type). Rodent damage was not important because the stores were protected with guards; except for a few rare cases of lizard damage. Few mouldy cobs were observed. However, few grains with very low fungal infection that also showed insect boreholes were counted as damaged. The percent diy- weight loss was calculated using the following equation (Boxall, 1986); University of Ghana http://ugspace.ug.edu.gh 33 % Weight loss = ((UNd - DNu/TJ(Nd + Nu)) x 100 Where: Nd = number of damaged grains Nu = number of undamaged grains D = weight of damaged grains U = weight of undamaged grains. 3.6 Rapid loss assessment Visual scale of damage developed by the Ghana Larger Grain Borer Project (GLGBP) for rapid loss assessment was also used to calculate weight loss mid economic value loss of the maize varieties in store for only the long season. This was done on all the twenty-six cobs used for the standard count and weigh method described earlier (see section 3.5). After dehusking, each cob was classified into one of the six damage classes on the visual scale (see plate 2). Then total number of cobs in each damage class was recorded against the class (see section 4.4.2.1 for details). Weight loss was calculated from the formula: E N iM i/N Where: Ni = Number of cobs in ith class Mi = Average weight loss for ith class N = Total number of cobs University of Ghana http://ugspace.ug.edu.gh 34 3.7 Rearing out Two cobs picked randomly from the thirty cobs sampled per section from each store were placed separately into labelled paper bags. Each was dehusked and insects associated with it identified and counted. Each cob was then shelled and the grains put separately in rearing jars. These bottles were examined once a week for four weeks and all emerging adult insects were collected. This provided information on the distribution of the opposition sites of P. truncatus and S. zeamais populations as well as the variety and density of parasitoids. 3.8 Study site This study was carried out from September 1994 until June 199S at the Ministry of Food and Agriculture (MOFA) research station, Kpeve, Volta Region, Ghana, within the research activities of the ETA Larger Grain Borer Project 3.9 Analyses of results Several analysis packages were used for various sections of the work. They were StatView 4.0 (Abacus Concept 1992) and SAS Statistical tools (SAS, 1987). The graphs were also drawn with Deltagraph Professional 2.0 and Microsoft Excel 4.0. The data were analysed after the following transformations were made: loglO (x + 1) for adult insect numbers and arcsin ( (x +l)1/z) for weight and economic losses. For analysis purposes 1he long season was divided into two storage phases: early (sampling occasions 1-4) and late (sampling occasions 5-8) (see Borgemeister et al., 1994). Repeated measure anova models (a = University of Ghana http://ugspace.ug.edu.gh 35 0.05) were used to evaluate treatment effects on insect densities and losses across time, separately within each phase. University of Ghana http://ugspace.ug.edu.gh 36 4.0. DESCRIPTION OF EXPERIMENTS 4.1 Effect of maize variety on population dynamics of P. truncatus and S. zeamais 4.1.1 Introduction Infestations of farm-stored maize by insect pests notably P. truncatus and S. zeamais pose economic and food security threats to Ihe small-scale rural farmer since they decrease quantity, quality and marketability of maize. S. zeamais has for many years consistently been a major primary pest in farm-stored maize in most African countries before the accidental introduction of P. truncatus into this continent. However, P. truncatus, apparently a potentially dangerous pest seems to be displacing S. zeamais in terms of economic importance. While efforts are being made to develop control strategies against these destructive pests, there are still gaps in knowledge on the ecology of these pests, especially in the West African farm stores. There is therefore the need to obtain information on the abundance and distribution patterns of these pests and how they are affected by maize variety and season. 4.1.2 Materials and Methods. At every sampling occasion all the insects collected from grains shelled from the 26 cobs sampled from each section per store were put into small plastic bags. These plastic bags were frozen to kill all insects before contents were sieved and all insects identified and counted. ( Refer to section 3.4). University of Ghana http://ugspace.ug.edu.gh 37 4.1.3 Results and Discussion 4.1.3.1. Insect species recorded on the stored maize A total number of twenty insect species including sixteen pests, two predators and two parasitoids were collected in both the long and short storage seasons. They belonged to the orders Coleoptera, Lepidoptera, Hymenoptera and Heteroptera (see Table 1). In October, one month into the long storage season, only S. zeamais, Cathartus quadricollis (Guerin), Palorus subdepressus (Wollaston), Carpophilus dimidiatus (Fabricius) and the predatory bug, Xylocoris flavipes (Reuter) were found in all the stores. But seven months on, all the insect pests and natural enemies listed in Table 1 were present in all the stores with the singular exception of the predator T. nigrescens, which was present in only seven of the stores. Even though P. truncatus was present in four of the maize stores in October, the mean densities were very low (see Table 2). By the middle of the storage season, i.e. in February, the number of Prostephanus - infested stores increased to eight with half of these stores containing T. nigrescens. All the stores were infested with P. truncatus at the end of the long storage season although the densities varied tremendously, giving credence to the observation that P. truncatus spreads mainly during the storage season (Hodges, 1984). For instance, 547 adult insects were collected on 78 cobs from a store while only 12 adults were collected on the same number of cobs from a different store stocked with same variety. It should also be pointed out that two stores stocked with Abeleehi and Dzolokpuita varieties stood uninfested by P. truncatus until the eighth month, although adjacent stores were very heavily infested. University of Ghana http://ugspace.ug.edu.gh 38 Eldana saccharina (Walker) was very rare. Occasionally only a few larvae were found in samples early in the storage season. But Mussidia nigrevenella (Ragnot) was common early in the storage season. University of Ghana http://ugspace.ug.edu.gh 39 Table 1. Insect species found in the maize stores. ORDER/FAMILY SPECIES COLEOPTERA Curculionidae Sitophilus zeamais Motschulsky Tenebrionidae Tribolium castaneum (Herbst) Palorus subdepressus (Wollaston) Gnathocerus maxillosus (Fabricius) Alphitobius diaperinus (Panzer). Nitidulidae Carpophilus dimidiatus (Fabricius) Bostrichidae Prostephanus truncatus (Horn) Dinoderus minutus Fabricius Xyloperthella picea(Om.) Rhizopertha dominica (Fabricius) Silvanidae Cathartus quadricollis(G\ierin) Oryzaephilus surinamensis (Linnae) Cucujidae Cryptolestesferrungineus (Stephens) Scolytidae Stephanoderes coffeae Hagedom Histeridae Teretriosoma nigrescens Lewis LEPIDOPTERA Pyralidae Mussidia nigrevenella (Ragnot) Eldana saccharina Walker HYMENOPTERA Pteromalidae Anisopteromalus calandrae (Howard) Choetospila elegans Westwood Evanidae Unidentified species HETEROPTERA Anthocoridae Xylocoris flavipes (Reuter) University of Ghana http://ugspace.ug.edu.gh 40 4.1.3.2 Population dynamics of P. truncatus and S. zeamais in the long season. In all the treatments P. truncatus numbers were very low during the first four months of storage (Fig. 3), although temporal analysis showed significant increase (Fi>9=5.38, P=0.0046) (see Appendix 2) in P. truncatus numbers between the first and the fourth sampling occasions. Maize variety showed significance (F2̂ =7.75, P=0.011) in the early phase of storage with the Abutia having the highest numbers of P. truncatus. There was also a significant interaction between maize variety and sampling occasion. During the late phase, the density of P. truncatus increased for all varieties. However, differences among maize varieties were significant (F2>9=4.3, P=0.049). This may be due to the sporadic distribution of P. truncatus (Markham et al, 1991) which results in high within-treatment variance, thus, making significant differences difficult to detect. Throughout the storage season, granaries stocked with Abutia variety recorded the highest densities of P. truncatus. After the second sampling occasion, P. truncatus densities were significantly higher on Abutia than on any other variety and remained so throughout the storage season (Table 2) even though much of these densities could be explained by the high density found in a single store of Abutia variety. No significant differences could be detected between Abeleehi and Dzolokpuita, which recorded the lowest densities of P. truncatus throughout the storage period. University of Ghana http://ugspace.ug.edu.gh 41 Fig. 3 Density of P. truncatus adults recorded on three maize varieties stored during the long and short seasons 140 T 120 - - 100 - •abeleehi abutia dzolokpuita abutia/s Oct Nov Dec Jan Feb Mar Apr May Jun Month of Sampling Number of P. truncatus /kg. grain University of Ghana http://ugspace.ug.edu.gh 42 Table 2: Densities of Prostephanus truncatus adults recorded on cobs of three maize varieties stored during the long season (October 1994 to May 1995). Sampling Occasion Mean density of P. truncatus * Abutia Abeleehi Dzolokpuita 1 (Oct. 1994) 0.6a 0.2a 0.0a 2 (Nov. 1994) 1.1a 0.5a 0.0a 3 (Dec. 1994) 0.9b 0.0a 0.0a 4 (Jan. 1995) 3.9b 0.1a 0.1a 5 (Feb. 1995) 29.0b 4.6a 0.2a 6 (Mar. 1995) 24.6b 5.1a 1.7a 7 (Apr. 1995) 109.3b 15.4a 6.5a 8 (May 1995) 131.1b 43.7a 16.9a * Means followed by same letter within a row are not significantly different (P>0.05) from each other by the Scheffe’s test. Insect numbers determined from 3 pooled 26-cob samples per bam. Each value is a mean of 4 bams. P. truncatus density also showed significant differences over time in the late phase (F3>27=25.11, P=0.0001) (see Appendix 2). The population significantly increased (Fi>9=37.04, P=0.0002) from the fifth to the eighth sampling occasion. The maximum densities of 131.1, 43.7 and 16.9 insects per kg. grain were recorded at the eighth sampling period for Abutia, Ableehi and Dzolokpuita respectively. No interactions between maize variety and time of sampling was detected during the late phase. University of Ghana http://ugspace.ug.edu.gh 43 Across all varieties, densities of S. zeamais were higher than those of P. truncatus (cf. Fig. 3 and 4) although, densities were not significantly different among varieties (P> 0.05) throughout the long season (see Appendix 3). However, S. zeamais population displayed significant differences over time during the early phase. Densities of S. zeamais on all treatments recorded in the first three sampling occasions increased sharply (Fi,9=87.55, P=0.0001), peaking in December at 281.7, 349.0 and 348.5 for Abutia, Abeleehi and Dzolokpuita varieties, respectively (Table 3). Thereafter the densities remained relatively stable up to April, fluctuating between 200 and 300 adult insects per kg of maize grains. Apart from Abeleehi, which recorded decreasing density between the months of April and May, Abutia and Dzolokpuita recorded increasing densities; an increase of 49 and 82 insects per kg of maize grains for the two varieties, respectively. It is interesting to note that unlike P. truncatus densities, which were significantly highest on Abutia, after the November sampling, no significant differences were found among varietal treatments for S. zeamais except on the first sampling occasion (Table 3) The reason for this significant difference at the beginning of the storage season may be parity due to field infestation of S. zeamais, particularly among cobs with damaged or poor husk cover (Markham, 1981) University of Ghana http://ugspace.ug.edu.gh 44 Fig. 4 Density of S. zeamais adults recorded on three maize varieties stored during the long and short seasons 350 300 250 •abeleehi 200 abutia dzolokpuita abutia/s 150 0 -I------------ 1-------------1-------------1-------------1-------------1-------------1-------------1-------------1 Oct Nov Dec Jan Feb Mar Apr May Jun Montth of Sampling Number of S. zeamais /kg. grain University of Ghana http://ugspace.ug.edu.gh 45 Table 3: Densities of S. zeamais recorded on cobs of three maize varieties stored in barns during the long season (October 1994 to May 1995). Sampling Occasion Mean density of S. zeamais*. Abutia Abeleehi Dzolokpuita 1 (Oct 1994) 169.30b 81.70a 69.01a 2 (Nov. 1994) 232.50a 259.74a 212.28a 3 (Dec. 1994) 281.69a 349.03a 348.47a 4 (Jan. 1995) 217.24a 297.38a 251.68a 5 (Feb. 1995) 242.32a 263.28a 213.32a 6 (Mar. 1995) 226.19a 243.45a 242.52a 7 (Apr. 1995) 255.21a 277.80a 231.43a 8 (May 1995) 303.99a 250.84a 313.49a * Means followed by the same letter within a row are not significantly different (P>0.05) from each other by the Scheffe’s test. Insect numbers determined from 3 pooled 26-cob samples per bam. Each value is a mean of 4 bams. University of Ghana http://ugspace.ug.edu.gh 46 Fig. 5 Percentage moisture content of three varieties of maize stored during ttie long and short seasons abutia abeleehi dzolokpuita abutia(ss) Month of Sampling Moisture content of grain (%) University of Ghana http://ugspace.ug.edu.gh 47 Table 4: Cob characteristics of three maize varieties used in experiment Character Mean Weight/length (g./ cm)* Abutia Abeleehi Dzolokpuita Core Weight 17.4a 16.0a 23.2b Husk leaves number 8.0a 8.1a 10.2b Grain number 343.4a 331.3a 357.2a Tip length 10.9a 10.0a 11.7a ’'‘Means followed by the same letter within a row are not significantly different (P>0.05) from each other by the Scheffe’s test. Table 5: Grain hardness of the three maize varieties used in the experiment1 Maize variety Coefficient of Hardness2 Abutia 0.8 Dzolokpuita 1.5 Abeleehi 1.8 1 Grain hardness increases with magnitude of the coefficient of hardness 2 (Adapted from Addo, 1994. See appendix 1) University of Ghana http://ugspace.ug.edu.gh 48 The levelling - off period beginning in December for S. zeamais, which contrasts sharply with the increasing P. truncatus after this period, coincided with the dry and cool weather conditions of the harmattan, normally experienced during this time of the year. Low ambient humidity reduces grain moisture content and this may lower survivorship and reproductive rate of S. zeamais (Longstaff, 1981). Moisture content during this period remained at about 12% (Fig. 5). However, P. truncatus tolerates lower grain moisture content (Hodges et al, 1983a). It might also be due to P. truncatus immigrating into stores in response to environmental cues; volatiles from maize (Pike et al, 1994), pheromones (Hodges, 1994), temperature, etc. (Fadamiro and Wyatt, 1995) and other factors. Hence, the moisture content may merely be correlated with other factors that more directly influence infestation. The number of husk leaves per cob (Table 4) showed Abutia to be inferior to the other varieties. From the values obtained by Addo (1994) who performed grain hardness tests using, among others, the varieties used in this experiment, Abutia ranked lowest with coefficient of hardness of 0.8 (Table 5). These two factors combined may explain why Abutia experienced the highest P. truncatus levels, although other husk cover characteristics such as strength of husk and length of husk beyond the tip of the cob have also been identified as providing important barrier between P. truncatus and maize grains on the cob (Keil, 1988; Addo, 1994). Howard (1983) and Li (1988) recorded reduced oviposition rate when P. truncatus was maintained on flinty and popcorn maize varieties. Li (1988) further explained that this was due to the higher energy cost of tunnelling the harder maize since the female had to lay eggs in blind-ending tunnels thereby leading to reduced fecundity and consequently population size is reduced on the harder maize variety. University of Ghana http://ugspace.ug.edu.gh With respect to S. zeamais numbers, however, the same assertion cannot be made since these varietal characteristics apparently did not influence it. Interestingly, other husk characteristics measured did not show any significant difference between varieties. However, the apparent lower degree of varietal effect on the S. zeamais was observed in laboratory studies by Howard (1983), although other works by Eden (1952 ), Dobie (1974), Kossou et al (1993) observed the contrary. 4.1.3.3 Spatial Distribution Within Stores. There were no clear-cut trends in die vertical distribution of P. truncatus within stores (Table 6), although differences in P. truncatus densities have been noticed regarding horizontal layers (C. Borgemeister, pers. com.). The inner layer generally appeared to record lower densities than any of the outer layers with the exception of the April sampling, although these differences were not significant. This suggests that the P. truncatus adults may, probably, have colonised the stores from outside rather than the cobs themselves being previously infested from the field before storage. The baseline field insect data as well as the rearing out may confirm this view because no P. truncatus adults were found in them. The reason for this distribution trend remaining so for most sampling occasions was probably due to the low and late infestations and establishment of P. truncatus in the stores. The few adults that infested the surface and middle layer cobs had to live and multiply in these cobs before spreading to the inner layer. It might also have to deal with S. zeamais infestation: a careful look at S. zeamais distribution in the different sections of the stores showed that most S. zeamais adults were in the inner layer (Table 7). This might suggest P. truncatus adults were not attracted to such cobs or could not reproduce well in cobs previously infested with S. zeamais because of the aggressive behaviour of S. zeamais larvae (Sharifi and Mills, 1971; Vowotor et ah, 1995b). It may also suggest that University of Ghana http://ugspace.ug.edu.gh 50 the P. truncatus individuals preferred the outer cobs for reasons that are not clear, although the small and insignificant differences (P>0.05), the low numbers of P. truncatus recorded and the significant interaction between variety and sampling occasion in the early storage phase (see 4.1.3.1) make it difficult to draw definite conclusions. Table 6: Density of P. truncatus adults recorded in the sections of a bam compartment* Sampling occasion Surface Middle Inner 1 (Oct.) 0.2a 0.4a 0.2a 2 (Nov.) 0.4a 1.0a 0.1a 3 (Dec.) 0.4a 0.3a 0.2a 4 (Jan) 0.6a 1.9a 1.0a 5 (Feb.)) 9.9a 13.6a 10.2a 6 (Mar) 10.8a 8.5a 12.2a 7 (Apr.) 48.4a 39.7a 43.0a 8 (May) 61.6a 78.4a 51.8a * Means followed by the same letters within a row are not significantly different (P>0.05) by Scheffe’s test. Similarly, there were small differences between S. zeamais densities in the various vertical layers (Table 7), and no general pattern was discernible. Although a cursoiy look at the data reveals slightly higher densities in the inner layer where there were perhaps, suitable stable environmental conditions than the outer two layers, these small differences were not significant. The middle layer, however, recorded significantly lowest densities (P<0.05) during the fifth and the seventh sampling occasions. University of Ghana http://ugspace.ug.edu.gh 51 Table 7: Density of & zeamais adults recorded in the sections of _______ a barn compartment _____________ Sampling occasion Surface Middle Inner 1 (Oct.) 98.2a 120.3a 101.6a 2 (Nov.) 192.7a 241,4a 270.4a 3 (Dec.) 312.6a 332.3a 334.3a 4 (Jan) 297.8a 219.2a 249.3a 5 (Feb.) 260.8a 179.0b 279.1a 6 (Mar) 272.7a 210.7a 228.9a 7 (Apr.) ... 265.1a 195.9b 303.5a 8 (May) 324.5a 257.8a 286.1a * Means followed by the same letters within a row are not significantly different (P>0.05) by Scheffe’s test. 4.1.3.4. Distribution of opposition sites of P. truncatus and S. zeamais on maize cobs. In order to understand the competitive preference for oviposition sites on maize cobs by one species in the presence or absence of the other species, cob by cob examinations were carried out. Cob samples were kept singly in jars and reared out to obtain about one generation (four weeks) of insects. Most cobs on which P. truncatus was found to be present also contained S. zeamais, although the relative numbers varied with the cobs on which they were found (Fig. 6). The populations of P. truncatus and S. zeamais were aggregated on cobs and did not occur at uniform average densities. Adult insect emergence for all treatments followed a similar trend; late incidence of low initial densities of P. truncatus which were aggregated onto few cobs, then increase in number of infested cobs and P. truncatus density with S. zeamais numbers generally decreasing. Although environmental conditions may have affected S. zeamais numbers, inter- and intra-specific competition for oviposition sites on the cobs may also have played a part. At the grain level, S. zeamais has been found to University of Ghana http://ugspace.ug.edu.gh 52 out-compete P. truncatus (Vowotor et a l , 1995b) but the destructive nature of P. truncatus on cobs can prevent S. zeamais from finding any suitable oviposition site. This was shown clearly in Fig. 6 where high numbers of emerging P. truncatus adults were associated with low S. zeamais numbers. University of Ghana http://ugspace.ug.edu.gh 53 Samp. occ. 1 Samp. occ. 5 .CO 120 -r I 100 <► % 80 - 60 CO o 40 o 20 0 -CH- H 0.5 10 20 30 Samp. occ. 2 Samp. occ. 6 .CO 150 -r 40 CD eCTJ 4 30

9=17.5, P=0.0008) (see Appendix 4). Weight loss also increased over time (F3>27=10.06, P=0.0001). This increase was gradual as significant differences were not observed between successive sampling periods but between the weight losses of the first and fourth sampling occasions (Fi,p=8L65, P=0.0001). Significant interaction of weight losses sustained by the maize varieties and time was observed (F6̂ 7=3.17, P=0.041). Similarly, weight losses significantly increased over time (F3>27:=9.72, P=0.0002) in the late phase (see Appendix 4). Varieties also showed significant differences (P< 0.05). However, there was no interaction between variety and time of sampling. University of Ghana http://ugspace.ug.edu.gh 56 Table 8: Percentage weight loss suffered by three maize varieties stored in barns during the long season (October 1994 to May 1995). Sampling Occasion Mean percentage grain weight loss* Abutia Abeleehi Dzolokpuita 1 (Oct 1994) 2.6b 1.9ab 1.56a 2 (Nov 1994) 4.8a 2.3a 2.2a 3 (Dec 1994) 4.2b 5.3b 1.8a 4 (Jan 1995) 3.9a 4.6a 3.3a 5 (Feb 1995) 8.2b 6.3ab 4.1a 6 (Mar 1995) 8.2b 4.5a 5.2ab 7 (Apr 1995) 6.3a 8.0a 6.1a 8 (May 1995) 11.3b 11.2b 7.4a * Means followed by the same letter within a row are not significantly different (P>0.05) by the Scheffe’s test. Each value is a mean of 4 bams. University of Ghana http://ugspace.ug.edu.gh 57 Fig.7 Percentage weight loss suffered by three maize varieties stored during the long and short seasons ■abeleehi - abutia dzolokpuita abutia/s Month of Sampling Weight loss (%) University of Ghana http://ugspace.ug.edu.gh 58 After eight months of storage the maximum weight loss recorded were 11.3%, 11.2% and 7.4% for Abutia, Abeleehi and Dzolokpuita respectively (Table 8). Dzolokpuita recorded significantly lowest grain loss at almost every sampling occasion compared with the other two varieties. Indeed, it had the lowest score in seven out of the eight sampling occasions. It is interesting to note that of the two local varieties, Dzolokpuita variety performed better than Abutia. The fact that the grains of Dzolokpuita are harder than Abutia (Table 5 and Appendix 1) might partly explain this observation (Howard, 1983; Li, 1988). Differences observed both graphically (Fig 7) and statistically (Table 8) were not consistent. This might probably be due to the sampling technique. One compartment of a store was sampled at each occasion (see section 3.2) so any spatial heterogeneity in grain loss, for instance, one side experiencing higher losses than another because of peculiar aggregation of insects within a given store, may be reflected in the treatment means. 4.2.3.2 Relationship Between Pests Density And Losses. Defining the relationship between pest densities and damage is a veiy important basis for sound pest management (Stem, 1973). Consequently, multiple regression analyses were conducted on percentage grain weight loss for each variety for all sampling occasions using P. truncatus and S. zeamais as independent variables to establish this relationship for all the maize varieties. The analyses showed that P. truncatus and S. zeamais mean densities could not explain the weight losses recorded during storage, except in October (i.e. after a month’s storage) for Abutia variety, where S. zeamais regression showed significance (F2>9=3.457, P=0.0294). A significant standardised regression coefficient of 0.704 was recorded for S. zeamais indicating that S. zeamais density played a major role in explaining the variance in weight loss registered than did P. truncatus. University of Ghana http://ugspace.ug.edu.gh 59 Table 9. Correlation coefficients of percentage grain weight loss with P. truncatus and S. zeamais density8. Weight Loss P. truncatus S. zeamais L, 0.133 0.437 U 0.009 0.512 l3 0.238 0.359 u 0.076 0.317 u 0.462 0.191 u 0.229 0.279 u 0.159 0.189 u 0.327 0.166 a: Li = Percentage weight loss at ith sampling occasion, r values > 0.325 have P < 0.05. 36 observations were used in this computation. Correlation coefficients of the pests and losses were therefore used to estimate the relationship. Coefficients of the primaiy pests were considered since they generally initiate damage and cause loss to previously undamaged maize grains. It should however be pointed out that later in the storage season, when several grains have already been damaged, and are vulnerable to attack, the contribution of these opportunistic secondary pests to grain weight loss cannot be ignored. University of Ghana http://ugspace.ug.edu.gh 60 Generally, there was a positive correlation between P. truncatus and S. zeamais densities and weight loss (see Table 9) throughout the storage season. P. truncatus density had significant correlation with weight loss at the fifth (0.462) and eighth (0.327) sampling occasions respectively, although they were not particularly strong. Similarly, S. zeamais density produced positive significant correlation with weight loss, although they were also not too strong. However, they occurred early in the storage season. The pattern of significance of the correlation coefficients for the two insect pests conform to findings in which S. zeamais density is often higher and cause most damage than P. truncatus early in the season, a situation which reverses later in the storage season (Novillo, 1991; Markham et al; 1994c; Borgemeister et al., 1994). 4.2.3.3 Association Of P. truncatus And S. zeamais With Some Secondary Pests. In the tropics secondary pests tend to be abundant in rural maize stores and evidence of differing relationships with primary pests have been reported (Novillo, 1991; Markham et a l , 1991). University of Ghana http://ugspace.ug.edu.gh 61 Table 10 Correlation coefficients of P. truncatus with some secondary insect pests*. P.trunc3. Cath.b Pal.c Crypt.d Others m 0 .0 29 0 .029 0.152 -0.128 Pt2 ** ** ** ** Pt3 0.467 0.461 0.954 0.132 Pt4 0.023 0.235 0.148 0.088 Pt5 0.336 0.505 0.555 0.560 Pt6 0.371 0.589 0.494 0.540 Pt7 0.265 0.167 0.495 0.330 Pt8 0.148 0.178 0.156 0.357 *: Pti = P. truncatus density at zth sampling occasion. **= No P. truncatus adults recorded r values > 0.232 have P< 0.05. 72 observations were used in this computation. a: P. truncatus, b: Cathartus quadricollis; c: Palorus subdepressus d: Cryptolestesferrugineus. In this study the association of secondary insect pests with P. truncatus and S. zeamais were examined at the cob level. P. truncatus association with the most abundant secondary pests, namely C. quadricollis, P. subdepressus and Cryptolestes ferrugineus were positive and generally significant after three months of storage up to and including the seventh month (see Table 10). The eighth month, which recorded the highest density of P. truncatus, irrespective of the maize variety (see Fig. 3 and Table 2) however showed a reduced association with each of these secondary pests. It appears the other secondary pests, namely, Tribolium castaneum, Gnathocerus maxillosus and Carpophilus dimidiatus University of Ghana http://ugspace.ug.edu.gh 62 which were not abundant and were therefore lumped together as ’'others” (see Table 10), displaced C. quadricollis, P. subdepressus and C. ferrugineus as the season progressed and P. truncatus density increased. In fact, this group ("others") recorded significant correlation coefficients of 0.56, 0.54, 0.33 and 0.357, respectively from the fifth to the eighth sampling occasion of the long storage season. The reason for this apparent displacement is not clear. Probably, inter- and intra-specific competition coupled with an increase in the moisture content of the grains favoured some of these secondary pests. Schulten (1976) reported that at high grain moisture content, Carpophilus spp can act as primary pests. Admittedly, the moisture content indicated (about 30%) was quite high it nevertheless points to the fact that some of these secondary insects which tolerate high grain moisture content and take advantage of it in interspecific competition. Also the tolerance of T. castaneum to high temperatures and relative humidities (Beckett et al 1994) and its widely known predation behaviour (Markham et al., 1991) may also explain this observation. In relation to S. zeamais, there was a significantly close association with C. quadricollis, P. subdepressus, and C. ferrugineus throughout the major storage season (see Table 11). The group labelled "others” also showed similar association with S. zeamais. University of Ghana http://ugspace.ug.edu.gh 63 Table 11: Correlation coefficients of S. zeamais density with densities of some secondary insect pests*. S. zeamais Cath* Palc Cryptrf Others Szl 0.665 0.631 0.381 -0.030 Sz2 0.658 0.753 0.460 0.730 Sz3 0.596 0.468 0.253 0.458 Sz4 0.649 0.843 0.595 0.806 Sz5 0.639 0.872 0.822 0.843 Sz6 0.617 0.829 0.753 0.780 Sz7 0.415 0.521 0.493 0.418 Sz8 0.642 0.840 0.734 0.840 *: Szi = S. zeamais density at 2th sampling occasion, r values > 0.232 have P< 0.05. 72 observations were used in this computation. b: Cathartus quadricollis, c: Palorus subdepressus d: Cryptolestes ferrugineus. 4.2.3.4 Dynamics of T. nigrescens population Teretriosoma nigrescens, which is an effective predator of Prostephanus truncatus under both laboratory (Rees, 1985; Leliveldt and Laborius, 1990) and semi-field conditions (Helbig, 1995) has been now introduced into Ghana. This study also examined its association with, and influence, if any, on P. truncatus population. Densities of T. nigrescens recorded on the three maize varieties were quite low (Fig. 8). However, the numbers followed the trend of the P. truncatus population dynamics (see Fig. 3). T. nigrescens adults are able to specifically pursue P. truncatus because they can University of Ghana http://ugspace.ug.edu.gh 64 also perceive the male-produced aggregation pheromone secreted by P. truncatus to attract mates (Rees et al., 1990). Thus, most of the T. nigrescens were recorded on the Abutia variety and to a lesser extent on Abeleehi. Dzolokpuita recorded the least densities at all sampling occasions. No differences were observed within sampling occasions during the early phase. However, there were significant increases (F3,9=5.3, P=0.005) during the late phase of the long season. Maize variety never had any effect on the T. nigrescens dynamics. The highly positive significant correlation observed between 71 nigrescens and P. truncatus densities (Table 12) during the season is indicative of the close association of the predator and its prey, even in the midst of other storage insects (see Table l)in the open stores under field conditions (Boye, 1990; Rees et al., 1990). This association became closer with time as both populations increased. The positive correlation observed over time suggests a numerical response of the predator to P. truncatus density, even though T. nigrescens could not establish early to have a restrictive overall effect on the P. truncatus population. University of Ghana http://ugspace.ug.edu.gh 65 Table 12: Correlation coefficients of T. nigrescens with P. _________ truncatus and Loss______________________ T. nigrescens________P. tmncatus____________Loss Tnl * * Tn2 * * Tn3 0.609 0.125 Tn4 0.438 -0.08 Tn5 0.709 0.359 Tn6 0.767 0.259 Tn7 0.835 0.222 Tn8 0.801 0.253 * No T. nigrescens recorded. Tni = T. nigrescens density at ith sampling occasion. Loss = diy weight loss of maize grain by standard count and weigh method, r values > 0.325 have P < 0.005 36 observations were used in this computation Table 13 Changes in the ratio of I nigrescens (Tn) to R truncatus ________adults in the stores in the long and short storage seasons. Tn: P. truncatus at sampling occasion Season 1 2 3 4 5 6 7 8 Long 1 * * 1.2 3.1 5.3 6.9 14.2 30.1 Short 1 ** 1.85 * * 0.4 *No T. nigrescens recorded **No T. nigrescens and P. truncatus recorded This is well illustrated by the changes in the predator-prey ratio (Table 13). The ratio decreased from 1:1.2 after three months of storage to 1:30.1 at the eighth sampling occasion. An increasing predator-prey ratio could either be due to increasing P. truncatus University of Ghana http://ugspace.ug.edu.gh 66 density relative to T. nigrescens, or a decreasing T. nigrescens density relative to P. truncatus population. However, a look at T. nigrescens and P. truncatus densities over time (Fig. 3 and Fig. 8) coupled with the positive correlation (Table 12) show that the densities of T. nigrescens and P. truncatus increased with time. Therefore, the probable indications of increasing predator-prey ratio may be that the per capita growth rate of P. truncatus population outstripped that of T. nigrescens. The very high magnitude of the ratios, particularly at the seventh and eighth sampling occasions suggests very low densities of T. nigrescens. Boye (1988) recorded mean ratios of T. nigrescens: P. truncatus of 1:8.4 and 1:7.2 after eight months of storage of maize in a farm in Costa Rica, where T. nigrescens effectively controlled P. truncatus in stores, during the 1985/86 and 1986/87 seasons respectively. These values were higher than the 1.30.1 recorded in the storage bams at Kpeve. University of Ghana http://ugspace.ug.edu.gh 67 Fig. 8 Density of T. nigrescens adults collected on the three maize varieties stored cjurlng the long and short seasons — abcleehl ■ — abulia —▲— ckoluK^uild c iL / u l ia / s Month of Sampling Number of 7. nigrescens f kg. grain University of Ghana http://ugspace.ug.edu.gh 68 4.3 Comparison of pest dynamics and weight loss between long and short seasons 4.3.1 Introduction Rainfall is bimodal in the southern half of the country, and in fact in most parts of the West African sub-region.. For Ghana, it covers the forest savanna transition areas and the forest belt which are important maize growing areas. As has been noted already (refer to section 1.0) the first rains normally begin from March/April to July/August and the second from September to November. The diy period, usually occurring in August for 2 - 4 weeks, separates these two rainfall seasons. There is another longer diy period which occurs from the end of November to the beginning of March. The seasonality of rainfall is exploited for the successful cropping of two rain-fed crops of maize in the these areas. The long (or major) season maize is harvested and stored during the dry spell in August, through the second season rains under conditions of high ambient relative humidity whereas the short (or minor) season crop is harvested and stored under dry - season or harmattan conditions, which are more favourable for grain drying. A trial was therefore conducted to investigate the population dynamics of P. truncatus and S. zeamais during the short season storage as well as grain losses they cause. These results were compared with those of the long season trial. 4.3.2. Materials and Methods. The stores used were similar to those used in long season experiment described in section 3.1. Only one variety, Abutia, was used in this study. The experimental design and sampling procedures were as described in sections 3.2 and 3.3. All measurements taken on University of Ghana http://ugspace.ug.edu.gh 69 the long season samples were also conducted on the short season samples (see sections 3.3 - 3.7). The trial was terminated in June after five months of storage. This was because by this time the storage conditions in the stores had become similar to those in the stores stocked with long season maize. 4.3.3 Results And Discussion 4.3.3.1 Insect Pests Of The Short Season One month into the storage season most of the major insect pests recorded in the long season (see table 1) were also registered in the short season with the exception of P. truncatus and its predator, T. nigrescens. In addition Rhizopertha dominica which was not recorded in appreciable numbers in the long season was found in a store. The first P. truncatus and T. nigrescens adults were encountered after two months in store. Although, the number of Prostephanus-mfQstQd stores as well as its density increased with the progress of the short season (see Fig. 3), no T. nigrescens adults were recorded in the subsequent until the last sampling period, when two out of the four stores recorded T. nigrescens incidence. Even here, the density was very low as a store recorded only one T. nigrescens adult on 78 cobs sampled. S. zeamais and R. dominica commonly occurred in all stores, after five months of storage,. Tolerance of R. dominica to high temperatures and varying relative humidities is well documented (Becket et al., 1994). By the end of the short season storage, almost all other insects encountered in the long season had been recorded. University of Ghana http://ugspace.ug.edu.gh 70 4.33.2 Population densities of P. truncatus and S. zeamais in the short season. The population density of P. truncatus at all sampling occasions was very low (see Table 14). After its first occurrence at the second sampling occasion, P. truncatus density remained very low; 0.15 to 0.13 adults per kg. maize grains during subsequent sampling occasions until the fifth, when it increased to 4.17 adult pests per kg. of maize grains. The patterns of build-up of P. truncatus infestation in both the long and short seasons stores were similar. There was a relatively large increase in the mean density of the fourth to fifth sampling occasions for the two seasons, although at different rates, The difference between the resulting densities was not statistically significant (P<0.05). However, it is interesting to observe this kind of sudden population density ’’jump” in both seasons even though sampling dates were different. Studying the grain moisture content patterns for the varieties (Fig. 3), the increase might partly be due to the drop in the moisture content of the grains between the fourth and fifth sampling occasions especially for the short season storage. The moisture content of the long season maize was already decreasing because of the harmattan conditions. P. truncatus is known to tolerate diy grains (Hodges, 1986). University of Ghana http://ugspace.ug.edu.gh 71 Table 14: Density of P. truncatus recorded on Abutia variety stored during the long and short seasons Sampling Occasion_________ Mean density of P. truncatus on Abutia* Lone season Short season 1 0.6+ 0.3a 0.00b 2 l.l+_0.6a 1.4+0.9a 3 1.0+ 0.5a 0.2+0.1a 4 3.4+ 1.2a 0.1+1.0b 5 29.0+17.7a 4.2+ 1.3a * Means (+ S.E.) followed by same letters within a row are not significantly different (P> 0.05). Insect numbers determined from 3 pooled 26-cob samples per bam. Each value is a mean of 4 bams + S.E. Table 15: Density of S. zeamais recorded on Abutia variety stored ________ during the long and short seasons.__________________ Sampling occasion Mean density of S. zeamais on Abutia* Lone Season Short Season 1 169.3 + 13.7a 56.5+ 12.0b 2 232.5 + 17.8a 88.8+ 17.4b 3 281.7 + 29.6a 105.2+27.1b 4 217.3 ± 35.1a 168.3 + 21.0a 5 242.3 + 30.1a 185.9 + 20.8a * Means (± S.E) followed by same letters within a row are not significantly different (P> 0.05). Insect numbers determined from 3 pooled 26-cob samples per bam. Each value is a mean of 4 bams ± SE. University of Ghana http://ugspace.ug.edu.gh 72 Nevertheless, differences were observed in the densities of P. truncatus in stores. P truncatus appeared in stores earlier in the long season than in the short season. Furthermore, at the fourth sampling occasion there was a significant difference between the mean P. truncatus densities (t=3.23, P< 0.05). The population of S. zeamais increased progressively from the first to the fifth sampling occasions (Fig. 4). Comparing the densities of S. zeamais adults per kg. grain for the long and short seasons within sampling occasions (see Table 15) revealed significant differences (P < 0.05) during the first three samplings, with t-values of 7.58, 7.12 and 5.42 for the first, second and third sampling occasions, respectively. S. zeamais density during the short season were significantly lower. It is interesting to note that in stores stocked during the short season, S. zeamais increased throughout the harmattan season while the density in stores stocked during the long season had levelled - off (see Fig. 4). Part of the difference in the density patterns may have been due to density dependent effects, such as greater deterioration of the maize in the long-season trial, competition among beetles for oviposition sites, etc. resulting in S. zeamais leaving these stores to colonise the short season stores. As with the density of P. truncatus, the density of S. zeamais at every sampling occasion in the short season was lower than that recorded for the long season. University of Ghana http://ugspace.ug.edu.gh 73 4.3.3.3 Weight losses in the short season. Percentage grain weight losses over time are shown in Fig 7. From the losses at harvest at 0.5%, losses increased to 3.4% after five months of storage. This was significantly lower than was recorded for the long season (see Table 8). Subtracting the initial loss at harvest leaves about 2.9% as real storage loss which is good enough loss level. This value was lower than what occurred in store after five months of storage in the long season: 7.7%. Losses recorded in the short season were consistently lower than those recorded in the long season at all sampling periods. The differences were also statistically different (P< 0.05) in four out of five occasions (see Table 16). It is interesting to note that the short season stores, although stocked near already infested stores of the long season still suffered less losses, apparently from low infestation of the primary insect pests especially P. truncatus. Nevertheless, multiple regression analyses conducted on the percentage grain loss with P. truncatus and S. zeamais as independent variables revealed that most of the losses to the grains were caused by S. zeamais. All regression ANOVA F-tests and standardized regression coefficients were also significant for S. zeamais with the exception of the fifth sampling occasion. The low density of P. truncatus may explain its minor importance in this analysis. University of Ghana http://ugspace.ug.edu.gh 74 Table 16: Percentage weight loss suffered by the Abutia variety in the long and short seasons. Sampling Occasion Mean Losses suffered by Abutia* Long Season Short Season 1 2.6+ 0.1 a 0.6 +_0.1b 2 4.8 + 1.3a 1.2 +_0.2b 3 4.2 + 0.5a 1.2 + 0.2b 4 3.9 +_0.5a 2.0 + 0.6a 5 8.1 + 0.9a 3.4 ±_0.5b *Means (+ S. E.) followed by same letters within a row are not significantly different (P>0.05) by t-test. Each value is a mean of 4 bams. 4.4 Economic and weight loss measurement using the visual scale of damage 4.4.1 Introduction Conventionally, storage losses are determined by quantitative methods. However, in rural maize trade, where maize is sold by volume, maize price is often determined by the quality of the grains. A visual scale of damage developed by the UK Overseas Development Agency (ODA)/ Ghana Ministry of Food and Agriculture (MOFA) Larger Grain Borer Project was therefore used to assess the relative economic value loss of maize stored in the long season. Here, dry-weight loss and the pattern of cob damage development in storage were observed using the visual scale. University of Ghana http://ugspace.ug.edu.gh 75 4.4.2 Materials And Methods 4.4.2.1 Weight and economic loss assessment The visual scale for rapid loss assessment developed by the Ghana larger Grain Borer Project (GLGBP, 1994) consists of six categories of damage numbered from 1 (no damage) to 6 (severe damage). Representative cob samples photographed as standard reference (see Plate 1) were used in assigning the maize cobs into their respective classes. This was applied to all the batches of twenty-six cobs used for the standard count and weigh loss assessment described earlier (see section 3.5). Here, after dehusking, each cob was classified into one of the six damage classes on the visual scale. The total number of cobs in each damage class was then recorded against the class. Weight loss and economic loss were calculated using the formula: ENiMt/N Where: Ni = number of cobs in tth class Mi = Average weight (or economic) loss for tth class N = Total number of cobs The average weight loss values or coefficients used in the calculation were: 0%, 2%, 10%, 30%s 50% and 90% for the damage classes 1, 2, 3, 4, 5 and 6 respectively. These values were obtained from calibration curves, by the GLGB Project at Ho. Similarly, the coefficients used for the economic loss were 0%, 0%, 12%, 47%, 75%, and 100% for the damage classes 1 2, 3, 4, 5 and 6 respectively. University of Ghana http://ugspace.ug.edu.gh I’lale 1: damage classes o f maize cobs :8 ! = UMDAMAGED COB I’M: I:’If;!l. • ■ ■ m CLASS 3 IL-i-f'fl !>(• IhIi if' P L » L1 II ?. ii i : -i ! 1 CLASS 5 >ASS 6 = TO*TALLY DAMAGED COB ■i!)’ ... ’ i-1 *: jyfoiMa** - -- University of Ghana http://ugspace.ug.edu.gh 77 4.4.3 Results And Discussion 4.4.3.1 Structure of cob damage A critical observation of the damage to cobs across all varieties (Tables 17, 18 and 19) revealed that throughout the storage season about 46% of the cobs did not suffer any damage at all, i.e., they were found in the damage class 1, and almost none in class 6. However, as the storage season progressed the proportion of cobs in damage classes 2 and 3 decreased while those in classes 4 and 5 generally increased. Variety-wise, it could be seen that the damage development of Abeleehi and Dzolokpuita appeared similar while both differed from that of Abutia with slow increases in the damage levels of the two. A mean frequency of about 20 cobs were always recorded in classes 4 and 5 during the first seven months of storage compared with about 25 for Abutia. Moreover, a relatively higher proportion of cobs in damage class 3 was registered for Abeleehi and Dzolokpuita than for Abutia. However, during the last month of storage, over 30% of the cobs were found in this damage class for all varieties. The allocation of consistently fewer Abutia cobs into damage class 3, an occurrence that became more severe later in the season although proportion of cobs in class 2 decreased, may be evidence of the destructive nature of P. truncatus. It suggests that damage by this insect is very high. This is because the progression of cobs from class 2 through 3 to classes 4 and 5 could not be detected with a sampling frequency of one month. Abeleehi also exhibited this lower and decreasing proportion of class 3 cobs later in the season when P. truncatus density on the cobs increased (see Table 2). University of Ghana http://ugspace.ug.edu.gh 78 Table 17: Damage development of maize cobs of Abutia variety stored during the long season Damage Class Sampling occasion 1 2 3 4 5 6 1 No data was taken for this month 2 52.00 17.25 10.50 16.65 4.25 0.00 3 48.50 18.00 9.50 20.00 4.25 0.00 4 48.00 20.00 9.00 13.75 9.50 0.00 5 50.25 12.00 7.75 21.25 8.75 0.00 6 47.25 13.35 8.25 17.75 13.50 0.00 7 53.75 14.25 3.75 20.25 8.25 0.00 8 51.50 10.75 3.75 12.75 L9.25 0.00 Table 18: Damage development of maize cobs of Abeleehi variety stored during the long season Damage Class Sampling occasion 1 2 3 4 5 6 1 No data was taken for this month • 2 39.75 27.00 14.25 8.00 0.50 0.00 3 43.00 23.25 17.00 13.50 3.25 0.00 4 43.75 22.00 11.50 15.25 7.50 0.00 5 47.25 17.75 13.50 18.00 3.50 0.00 6 43.25 18.25 9.00 19.50 10.25 0.00 7 50.75 10.75 10.50 19.50 8.75 0.00 8 47.50 10.00 7.50 20.75 ri4.50 0.00 University of Ghana http://ugspace.ug.edu.gh 79 Table 19: Damage development of maize cobs of Dzolokpuita variety stored during the long season. Damage Class Sampling occasion 1 2 3 4 5 6 1 No data was taken for this month • 2 45.25 30.75 12.75 9.00 2.25 0.00 3 41.75 28.25 16.00 10.25 3.75 0.00 4 42.25 26.50 12.50 16.25 2.25 0.00 5 41.25 24.25 10.75 17.25 2.50 0.00 6 44.75 18.75 10.00 19.75 6.50 0.75 7 43.25 19.25 11.25 16.50 9.50 0.00 8 42.25 17.50 14.25 18.00 13.25 0.00 But it is striking to note that about 50% of the cobs for the Abutia variety that experienced a particularly high density of primaiy insects, especially R truncc&us (see TaHes 2 and 3), remained undamaged (i.e. allocated into class 1) and again no cob entered the class 6 category. Reasons for these observations are not very clear. It is possible that those insects that left these infested cobs, as a result of food and oviposition site limitation, were rather attracted to already infested cobs, probably due to aggregation pheromone secretion (Trematera and Girgenti, 1989; Cork et a l , 1991). It is also possible that with the generally low P. truncatus infestation during the season, comparison is not clear since the insect densities were calculated on per kilogramme basis rather than per cob. However, errors committed during the placement of damaged cobs into categories because of the fine distinctions between damage classes, cannot be discounted. This is a disadvantage of using visual scales. University of Ghana http://ugspace.ug.edu.gh 80 4.4.3.2 Dry weight loss by visual scales The trend in weight loss obtained using the visual scales (Fig. 9) was similar to that obtained from the standard count and weight data, although the magnitude of the values were larger (cf. Table 8 vrs. Table 20). The maximum weight loss values recorded after eight months of storage by the visual scale method were 13.4%, 13.0%, and 12.0% respectively for Abutia, Abeleehi and Dzolokpuita (Table 20). These values were between 1.8% and 4.6% higher than what was obtained using the standard count and weigh method. The higher magnitude of weight loss figure by using the visual scales method may be due to the fact that the coefficients of the various damage classes used in calculating this weight loss were calibrated against the modified count and weigh method (see Compton et al., 1992). This modified version of the count and weigh takes into account grains that get completely destroyed as a result of insect pest activities as well as those grains that crumble during shelling. These ‘missing grains’ do not get considered by the standard count and weigh method. Thus, the standard count and weigh method usually underestimates diy weight loss of maize varieties with softer grains as well as those that suffer severe infestations of primaiy insect pests. Varietal effects showed significance (F2,9 =5.06, P=0.034) (see Appendix 5). However, mean separation by Sheffeis test revealed significance (P<0.5) only at the second sampling occasion for the early storage phase. There was also a significant difference (P<0.5) at the fifth sampling occasion for the late storage phase (see Table 20). Weight loss by this method also changed with time (F3,27 =7.25, P=0.001). There was a significant increase in dry weight loss after two months in storage (FI,9 =18.14, P=0.002) i.e. between the second and third successive samplings, and again between the fourth and fifth successive University of Ghana http://ugspace.ug.edu.gh 81 samplings (FI,9 =8.31, P=0.018). After five months in storage, grain loss generally increased for all varieties but differences were not significant. However, a significant increase in weight loss (FI,9 =6.58, P=0.03) was observed between the seventh and eighth sampling occasions. This might be due to the significant increase in P. truncatus during that penod (see section 4.3.2). No interaction between variety and time of sampling was observed. University of Ghana http://ugspace.ug.edu.gh 82 Fig.9 Percentage weight loss (by visual scale of damage) of three varieties of maize stored during the long and short seasons - abutia abeleehi dzolokpuita abutia(ss) Month of Sampling University of Ghana http://ugspace.ug.edu.gh 83 Table 20: Percentage weight loss using the visual scale for the three maize varieties stored during the long season. Sampling occasion Mean weieht loss (visual scale) (%)* Abutia Abeleehi Dzolokpuita 1 No data was taken 2 7.7a 4.5b 5.5b 3 8.6a 7.3a 6.5a 4 9.2a 9.1a 7.2a 5 10.6a 8. lab 7.4b 6 11.9a 11.1a 10.0a 7 9.8a 10.4a 10.1a 8 13.4a 13.0a 12.0a ** AM /Te„-a--n- -s- follow--e--d3 Uby..+ t1h__e_ _s__a_m___e letters within a row are not significantly different (P>0.05) by Scheffe’s test. University of Ghana http://ugspace.ug.edu.gh 84 4.4.3.3 Economic Value Loss Maize is sold on the open market with high price variability among markets, depending on the location of the market with respect to the main production centers (see Henckes, 1991; Ctnnptttti and Magrath, 1994). Maize prices also fluctuate during the course of the Storage season, with extremely high prices usually in the pre-harvest lean season (between April and June). Price increases during this period could be as high as 200% (Magrath and Compton, 1994). Kartey (1992) analysing retail prices of a unit bowl (about 2.5 kg.) of maize between 1986-1991 in the Volta region, for instance, found that the storage of maize was profitable each year, since the nominal price increased significantly during the period. Taking inflation into consideration, Magrath (1994) agreed there were significant real price increases over the same period. With the out-break of P. truncatus in the region, making maize storage more risky, it is important that economic losses in maize storage are determined to provide information for sound decision-making by the farmers. Such information comes in handy when one is deciding on, for instance, whether to bring down a store because of post-harvest pest damage or hold on to the maize stock, in spite of the infestation, and wait for higher prices during the lean season, particularly in the deficit areas where prices are driven more by availability than by quality (Henckes, 1991; Magrath, 1994). The economic value loss increased during both phases of storage in die long season (Fig. 10). Although Abutia recorded the highest value loss at almost every sampling occasion (Table 21), it was Abeleehi which experienced the greatest economic loss of 15.5% in storage. University of Ghana http://ugspace.ug.edu.gh 85 Fig. 10 Percent economic value loss of three varieties of maize stored during the long and short seasons abutia abeleehi dzolokpuita abutia/s Month of Sampling University of Ghana http://ugspace.ug.edu.gh 86 Table 21: Percentage relative economic value loss using the visual scale for the three maize varieties stored during the long season. Sampling occasion Mean economic value loss (visual scale) (%)* Abutia Abeleehi Dzolokpuita 1 No data was taken 2 12.2a 6.0b 7.5b 3 13.6a 10.8a 9.4a 4 14.5a 14.2a 10.9a 5 17.4a 12.7ab 11.4b 6 19.4a 17.8a 15.8a 7 16.2a 16.8a 16.0a 8 22.1a 21.5a 19.4a * Means followed by the same letters within a row are not significantly different (P>0.05) by Scheffe’s test. It recorded the least value loss of 6.0% after two months in store, with Abutia and Dzolokpuita registering 12.2%, and 7.5% respectively. However, at the end of the long season storage period, Abeleehi suffered a cumulative value loss of 21.5%, compared with 22.1% and 19.4% for Abutia and Dzolokpuita respectively. Significant differences (P<0.05) among maize varieties, as well as sampling occasions, were observed in both University of Ghana http://ugspace.ug.edu.gh 87 early and late storage phases. There were, however, no interaction between the economic losses suffered by the maize varieties and sampling periods in any of the two phases. Analyses of variance of the contrast variables in the early phase revealed a significant difference (FI,9 =18.37, P=0.002) (see Appendix 6) between the value loss levels of second and fourth sampling occasions in the early phase. This observation was not surprising, considering that insect pests, especially S. zeamais density sharply increased in this storage phase (see Fig. 3), most probably due to colonization and reproduction (Markham, 1981). Similarly, during the late phase the value loss increased significantly (FI,9 =19.16, P=0.002) between the fifth and the eighth sampling occasions. However, this increase may have been gradual, as no differences were detected after the fifth sampling occasion except between the last two sampling occasions. This significant increase was possibly due to the surprisingly decreased value loss figures recorded for Abulia and Abeleehi varieties at the seventh sampling period which might itself be due to experimental error. University of Ghana http://ugspace.ug.edu.gh 5.0 SUMMARY AND CONCLUSIONS The influence of maize variety and season on the dynamics of the larger grain borer, Prostephanus truncatus and Sitophilus zeamais population was studied in the traditional faim-level storage bam. The weight losses experienced by the different varieties as a result of infestation by the insect pests were also studied. Moreover, the study included the distribution of the primary pests in the vertical layers, their ovipositional distribution on cobs as well as their association with other secondaiy insects especially with Teretriosoma nigrescens, a predator of P. truncatus recently introduced into Ghana. Relative economic value loss of the cobs were also studied with the aid visual of a scale of damage. From the study, the following results can be summarised: 1. The incidence of P. truncatus was generally low. However maize variety showed significance with Abutia consistently recording the highest pest densities throughout the storage season. P. truncatus densities also increased over time. 2. Densities of S. zeamais were similarly high on all varieties. S. zeamais population increased rapidly and reached its maximum after three months of storage. The densities on all varieties remained relatively stable after then, fluctuating slightly with an amplitude of 100 insects per kg grain. 3. Maize variety influenced weight losses early in the storage season but not in the late phase. Weight losses significantly increased with time. Dzolokpuita recorded the least weight loss at almost every sampling occasion 4. Husk number and grain hardness were thought to explain the low P. truncatus density and weight loss. But varietal characteristics did not influence S. zeamais population. No differences were observed among the distribution of either pest with University of Ghana http://ugspace.ug.edu.gh 89 respect to the vertical section of the bam compartment. However the successful rate of emergence of S. zeamais adults from reared-out cobs early in the storage season decreased as more P. truncatus adults emerged from the these cobs later in the season. 6. Both P truncatus and S. zeamais densities were significantly associated with the more abundant secondary pests, namely, Cathartus quadricollis, Palorus subdepressus and Cryptolestes ferrugineus. 7. P. truncatus density strongly and positively correlated with Teretriosoma nigrescens density. However, T. nigrescens did not have an overall restrictive effect on P. truncatus population growth. 8. P. truncatus density in the long season were not different from those of the short season. However S. zeamais densities were different for the first three sampling occasions. Weight losses sustained by the variety were similar for the two storage seasons 9. Cob damage development for Abeleehi and Dzolokpuita varieties appeared slower than for Abutia, presumably because of low P. truncatus incidence. However more cobs of Abutia escaped insect damage. 10. The economic value loss increased over time. Although Abutia recorded the highest value loss figures per sampling time, Abeleehi lost much of its economic value, about 16%, by the end of the storage season. From the results presented, the relative varietal susceptibility of maize cultivars proved to be important in influencing the population build-up of P truncatus and subsequent weight loss of stored maize. Husk numbers and grain hardness appeared to explain the relative resistance to P. truncatus damage. These and other varietal characteristics University of Ghana http://ugspace.ug.edu.gh 90 such as strength and wholeness of husk, phenolic content should be further explored in a wider spectrum of maize cultivars. Since P. truncatus density and loss increased over time, length of storage should be considered important. The performance of T. nigrescens on P. truncatus in the introduced areas should be assessed. It is not clear whether the introduced T. nigrescens is undergoing establishment difficulties here in Ghana or T. nigrescens was actually responsible for the collapse of P. truncatus in the Volta region during the 1994/1995 storage season. For a successful integrated pest management package for the small-scale, resource-poor rural farmers for P. truncatus control, these problematic issues should be considered carefully. University of Ghana http://ugspace.ug.edu.gh 91 6.0 LITERATURE CITED Adams, J. M. (1977) The evaluation of losses in maize stored on selection of small farms in Zambia, with particular reference to methodology. Tropical stored Products Information, 33, 19-24. Addo, S (1994) Tolerance of different maize varieties and wood species used in maize storage structures in the Volta Region, Ghana, to the Larger grain borer, Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae). M.Phil Thesis. Univ. of Ghana, Legon - Accra, Ghana. Almy, S. W. and Asanga, C. T. (1988). On-farm maize storage and seed preservation in Fako and Meme Divisions, Southwest Province, Cameroon. Report of testing and Liaison Unit, Ekona, Cameroun NCRE/USAID/IITA llpp Anon (1989) Post-harvest Food Losses in Developing countries National Academy of science, Washington, DC 206pp Apert, J. (1987). The storage of food grains and seeds CTA, Macmillan Publishers vi + 146pp. Ayertey, J. N. and Akibu, S. (1982) Notes on the relative incidence of Sitophilus zeamais Mots, and Sitophilus oryzae (L.) on stored maize from Funtua agricultural development project areas. Niger. J. PI Prot. 6:44-45. Ayertey, J. N. and Brempong - Yeboah, C. Y (1991). A survey of the Larger Grain Borer, Prostephanus truncatus (Horn) in some outbreak areas in Ghana. In: Ndoye, M. (ed.). Exotic pests in Africa, Prevention and Control. Symposium 9th Scientific. Conference. African Assoc. Ins. Scientists, Accra, Ghana, Sept. 1991. p.59-66. Badu-Apraku, B., Twumasi-Afiiyie, S. and Sallah, P. Y. K. (1992) Ghana. In J. M. Fajemisin, ed. Maize production in West and Central .Africa; trends and research orientation, p. 29-35 Ouagadougou, Burkina Faso, (Semi-Arid Food Grain Research and Development Project (SAFGRAD) IITA 71pp. Barnes, D. E, De las Casas and Ramirez Genel (1959) Fumigacion de granos almacenados. Agricultura tecnica en Mexico 8:7:-10. University of Ghana http://ugspace.ug.edu.gh 92 Beckett, S. J. Longstaff, B. C. and Evans, D. E. (1994) A comparison of the demography of four major stored grain coleopteran pest species and its implications for pest management. Proc. 6th Int. Wkg. conf. stored Prod. Prot 1:491-508. Bell, R. J. and Watters, F. L. (1982) Environmental factors influencing the development and rate of increase of Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) on stored maize. J. stored Prod. Res. 18:131 -142 Bhatia, S. K. (1976) Resistance to insects in stored grain. Trop. Stored. Prod. Inf. 31:21- 35. Biliwa, A. (1988) Les essais insecticides mene’s au Togo pour proteger le mais egrene, le mais sur lepi et le manioc seche contre l'infestation par le Prostephanus truncatus et le Sitophilus sp. In G. G. M. Schulten and A. J. Toet, (eds.), Proceedings of the Workshop on the containment and control of the larger grain borer, Arusha. Tanzania. Report II p 85-109. Rome. FAO 209pp Borgemeister, C, Adda, C, Djomamou, B, Degbey P, Agbaka A. Djossou, F, Meikle W. G. and Markham. R. H. (1994) The effect of maize cob selection and the impact of field infestation on stored maize losses by the larger grain borer, {Prostephanus truncatus (Horn) (Col. :Bostrichidae) and associated storage pest. Proc. 6th Int. Wkg. Conf. stored - prod. Proc. 2:906-909. Boxall, R. A. (1986) A critical review of the methodology for assessing form - level grain losses after harvest. Report of the tropical Development and Research Institute, G191, viii + 139pp. Boye, J. (1988) Autokologische Untersuchengen zum Velhatten des Grosen Kombohcers Prostephanus truncatus (Horn) (Col.: Bostrichidae) in Costa Rica. Disertation, Christian - Abrecht-Universitat, Kiel, 195pp. Boye, J., Burde, S., Keil, H., Laborius, G. A. and Schulz, F. A. (1988) The Possibilities of biological, integrated control of the larger grain borer {Prostephanus truncatus) (Horn) in Africa. Proc. Workshop on the containment and control of the larger grain borer, Arusha, Tanzania, 110-140 Chestnut, T. L. (1972) Flight habits of the maize weevil as related to field infestation of com. J. Econ. Entomol\ 65 (2): 434-435. University of Ghana http://ugspace.ug.edu.gh 93 Chittenden, F. H. (1911) Papers on insects affecting stored products. The larger grain borer. USD A, Division of Entomology, Bulletin No. 96. 48 - CIMMYT (1990)world maize facts and trends: realizing the potential of maize in sub- Saharan Africa, Mexico, D.F. 72pp. Compton, J. A. F. (1991). Survey of farm stored maize and dried cassava, central region, Togo; February - March 1991. Natural Resources Institute Report R1773, xi +90 pp. Chatham, UK Compton, J. A. F., Floyd, S. and Agbo, B. (1995). The modified count and weigh procedure: An improved method for assessing weight loss in stored maize cobs, (unpublished) Compton, J. A. F., Wright, M. A. P., Gay, C. and Stabrawa, A. (1992) A rapid method for loss assessment in stored maize and dried cassava. Tropical Pest management (in press). Cork, A., Hall, D. R., Hodges, R. J. and Pickett, J. A. (1991) Identification of male - produced aggregation pheromone of larger grain borer, Prostephanus truncatus (Horn) Col. Bostrichidae) Journal of Chemical Ecology 17(4): 789 - 803. De Lima, C. P. F. (1979) Assessment of losses due to insects and rodents in maize stored for subsistence in Kenya. Trop. stored prod. Inf 38:21-25. Dendy, J., Dobie, P., Saidi, J. A., Smith, J. and Urono, B. (1989) Trapping the larger grain borer, Prostephanus truncatus in maize fields using synthetic pheromones. Entomologia experimentalis et applicata 50:241-244. Detmers, H. B. (1990) Response of the larger grain borer Prostephanus truncatus (Horn) Coleoptera: Bostrichidae) to methyl bromide, p. 90. In 5th Int. Wkg. conf. stored Prod. Prot. Bordeaux France, Sept. 9-14, 1990. Abstract volume. 264pp Detmers, H. B., Helbig J., Laborius G. A., Richner J. and Schulz F.A. (1990) Investigation on the development capabilities of Prostephanus truncatus (Horn) Col.: Bostrichidae) on different types of wood, p.7. In 5th Int Wkg. Conf. Stored Prod. Prot. Bordeaux France, Sept. 9-14, 1990. Abstract volume 264pp University of Ghana http://ugspace.ug.edu.gh 94 Dick, K. (1988) A review of insect infestation of maize in farm storage in Africa with special reference to the ecology and control of Prostephanus truncatus. NRI Bulletin No. 18. NRI, Chatham, Kent, MEA 4TB, UK 42 pp. Dick, K. M. Rees, D. P. Lay, K. K. and Ofosu, A. (1989) Ghana: Occurrence of larger grain borer, Prostephanus truncatus (Horn), in Ghana. FAO Plant Prot. Bull, 37 (3): 32. Dix, D. E. and All, J. N. (1985) Invasion patterns and sex ratio dynamics of the maize weevil (Coleoptera: Curculionidae) infesting field com. Journal of Economic Entomology; 78: 1072-1075 Dobie, P. (1974) The laboratory assessment of the inherent susceptibility of maize varieties to post-harvest infestation by Sitophilus zeamais Motch. (Coleoptera: Curculionidae).,/. Stored Prod Res., 10:183-197. Dobie, P. (1977) The contribution of the Tropical stored products center to the study of insect resistance in stored maize Trop. stored Prod. Inf. 34:7-32. Dustan, W. R. and Magazini, I. A. (1981). The larger grain borer on stored products in Tanzania. FAO Plant Prot. Bull 29:80-81 Eden, W. G. (1952) Effect of husk cover of com on rice weevil damage in Alabama. J. Econ. Entomol} 45:543-544 Fadamiro, H. Y. and Wyatt, T. D. (1995) Flight initiation by Prostephanus truncatus in relation to time of day, temperature relative humidity and starvation. Entomologia Experimentalis etApplicata, 75: 273-277 FAO - PAL, (1990). The FAO Prevention of Food Losses Action. Programme: Moving towards a systems approach. GAGS Newsletter, 14:13-15. Fisher, H. U. (1987) Insects causing major losses in grain stores. Their identification and characterization with special reference to Prostephanus truncatus. Report of the seminar on post-harvest food losses. ECA/FAO 9-13 Feb., 1987. Fisher, W. S. (1950) A revision of the North America species of beetles belonging to the family Bostrichidae. USDA, Miscellaneous Publication No. 698. 157pp Floyd, E. H. and Newsom, L. D. (1959) Biological study of the rice weevil complex. Ann. Entomol. Soc. Am. 52: 687-695 University of Ghana http://ugspace.ug.edu.gh 95 GGDP (1991) Annual Report. Ghana Grains Development Project / CRI, Kumasi, Ghana. Giles P. H. and Leon, O. (1975) Infestation problems in farm stored maize in Nicaragua, pp. 68-76. InProc. 1st Int. Wkg. conf. Stored Prod. Entomol Savannah, Georgia, USA Oct. 7-11, 1974 Giles, P. H. (1977) Bean storage problems in Nicaragua Trop. Stored Prod Inf 34:63-67. Giles, P. H. and Ashman, F. (1971). A study of pre-harvest infestation of maize by S. zeamais Motschulsky (Coleoptera Curculionidae) in the Kenya highlands. J. stored Prod. Res. 7:69-83 Gilman, G. A (1984) Presence et importance economique de Prostephanus truncatus et efforts supra-regionaux visant a empecher sa proliferation en Afrique. In: Le control de lTnfestation des Denrees stockees par Trogodema granarium et Prostephanus truncatus. Compte rendu d'un seminaire international tenu a Lome 10-15 Dec. 1984. GTZ, Eschlom, Germany. 71-77. GLGBP (1993). Field studies in the Volta Region. MOFA/ODA (UK), Ghana LGB Project, Quarterly report 2, April - June 1993. Golob, P (1984) Prostephanus truncatus (Horn), the larger grain borer, in East Africa: the development of a control strategy. In: Proc. 3rd Int. Wkg. Conf on stored Prod Entomol, Kansas state University, Manhattan, Kansas, USA, Oct. 23-28, 1983 pp. 711 - 722. Golob, P. (1988). Current status of the larger grain borer Prostephanus truncatus (Horn) in Africa. Insect Science and its Applications 9: 737 - 745. Gutherie, W. D. and Russel, W. A. (1989) Breeding methodologies and the genetics basis of resistance to the European com borer. In towards insect Resistant maize for the third world. (Edited by Mihm, J. A. Wisenam, B. R. and Davis, F. M. pp. 192- 202. CIMMYT, Mexico. Haines, C. P. (1981). Insects and arachnids from stored products a report on specimens received by the Tropical Stored Product center 1973 - 1977. Report L54. London. UK TPI. 73pp University of Ghana http://ugspace.ug.edu.gh 96 Hall, D. W. (1970) Handling and storage of food grains in tropical and subtropical areas. FAO Agricultural Development Paper No. 90. Rome. 350 pp. Halstead, D. G. H.(1964) The separation of Sitophilus oryzae (L.) and S. zeamais Motschulsky (Col.: Curculionidae), with summary of their distribution. Eniomologisr s Mon. M a g 99: 72-74. Harris, K. L. and Lindblad, C. J.(1978) (Eds.) Post-Harvest Grain Loss Assessment methods. Published by the American Association of cereal chemists in cooperation with the league for International food education, the tropical products Institute, the FAO and the group for Assistance on systems relating the grain After-harvest. 193pp. Haubruge, E. and Verstraeten, C. (1987) Interaction entre Prostephanus truncatus (Horn) (Col.: Bostrichidae) et quatre especes de coleopteres des denrees, ravageuses du mais Medelingen Faculteit Landbouwwetens chappen Rijksuniverstaten Gent, 52: 241-246 Helbig, J., Detmers, H. B. Laborius, G.A. and Schulz, F.A. (1990) Investigations on the capability of Prostephanus truncatus (Horn) (Col.: Bostrichidae) to develop on different types of Wood. In Fleurat - Lessad, F. and Ducan P. eds., Proc. 5th Int. Wkg. Conf. stored - Prod. Prot, Bordeaux, France, September 9-14 1990, 1:89-97. Helbig, J., Laborius, G.-A and Schulz, F.A. (1992) Resherches our l'influence de Teretriosoma nigrescens Lewis (Col.: Histeridae) sur le developpement des populations de Prostephanus truncatus (Horn) (Col.: Bostrichidae) dans des condition semi practiques au Togo. Proc. FAO/GTZ coordinating meeting, Implementation of and further research biological control of the larger grain borer, Lome, Togo, 111 120 Henckes, C. (1992). Investigation into the insect population dynamics, damage and losses of stored maize an approach to IPM on small in Tanzania with special reference to Prostephanus truncatus (Horn). Ph.D. thesis University of Berlin. 124pp. Hinton, H. E. (1945) The Histeridae associated with stored products. Bull Entomol Res. 35:309-340 University of Ghana http://ugspace.ug.edu.gh 97 Hodges R. J., Meik, J. and Denton, H. (1985). Infestation of dried cassava (Manihot esculenta Crantz) by Prostephanus truncatus (Coleoptera: Bostrichidae). J. stored Prod. Res. 21 73-77. Hodges, R. J. (1984) Field ecology and monitoring of Prostephanus truncatus (Horn) pp. 32-48. In: The larger grain borer Prostephanus truncatus (Horn) GAGS Workshop, TPE storage Dept., Slough 24 - 25 Februaiy 1983 storage Dept., Slough 24 - 25 Februaiy 1983. Hodges, R. J. (1986) The biology and control of Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) - a destructive storage pest with an increasing range. J. stored Prod. Res. 22 (1) 1-14 Hodges, R. J. (1994) Recent advances in the biology and control of Prostephanus truncatus (Coleoptera: Bostrichidae) Proc. 6th Int. Wkg. conf stored - Prod. Prot. 2:929 - 934. Hodges, R. J. and Meik, J. (1984) Infestation of maize cobs by Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) - aspects of biology and control. J. Stored Prod. Res. 20 (4): 205 - 213. Hodges, R. J. Dustan, W. R. Magazini, I. and Golob, P. (1983a) An outbreak of Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) in East Africa. Protection Ecology 5:183-194. Hodges. R. J. (1982) A review of the biology and control of the greater grain borer Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae). Trop. stored Prod. Inf, 43: 3-9 Hoppe, T. (1986) Storage insects of basic food grains in Honduras. Tropical science 26 (1): 25-38 Horber, E. (1989) Methods to evaluate resistance in maize to grain insects in the field and in storage. In Mihm. J. A. Wiseman, B. R. and Davis, F. M (eds.) Towards Insect resistant maize for the third world pp. 140-150. CIMMYT, Mexico Horn, G. H. (1878) Revision of the Bostrichidae of the United States. Proceedings of the American Philosophical Society 17:540 - 555 University of Ghana http://ugspace.ug.edu.gh 98 Howard, D. C. (1983) The population biology of the greater grain borer Prostephanus truncatus (Horn). Ph.D. Thesis University of Reading, U.K. 180 pp. Jalloh, I. S. T. (1989) An assessment of the ability of Prostephanus truncatus (Horn) to infest and breed on some stored root and tuber crops, carob and copra. MSc. Thesis, University of Reacting, U.K. vi + 73 pp. Kalivogui, K. and Muck, O. (1991) Larger grain borer (.Prostephanus truncatus) in Guinea. FAO Plant Protection Bulletin 39:43 Kega, V. K. and Wurui, C. W. (1983) Prostephanus truncatus in Cape Province Kenya - Trop. stored Prod. Inf. 46:2. Keil, H. (1988). Losses caused by larger grain borer in farm stored maize, pp. 28-52. In G. G. M. Schulten and A. J. Toet (eds.) Report II Technical papers presented at the workshop, Workshop on the containment and control of the larger grain borer, Arusha, Tanzania, 16-21 May 1988. Kingsolver, J.M. (1971) Key to the general and species of Bostrichidae commonly intercepted in USDA plant quarantine inspections. USD A, AQIM Bi, 697, Kossou, D. K. ,Bosque - Perez, N.A. and Mareck, J. H. (1992) Effects of shelling maize cobs on the oviposition and development of Sitophilus zeamais Motschulsky. J. stored Prod. Res. 28 (2): 187 - 192. Kossou, D. K., Mareck, J. H. and Bosque - Perez, N.A. (1993) Comparison of improved and local maize varieties in the republic of Benin with emphasis on susceptibility to Sitophilus zeamais Motschulsky. J. stored Prod. Res. 29 (4) 333-343. Krall S. (1984) A new threat to farm level maize storage in West Africa: Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae). Trop. stored Prod. Inf. 50: 26-31 Krall, S. and Favi, F. (1986) Benin. Further distribution of the larger grain borer (Prostephanus truncatus) in West Africa. FAO Plant Prot. Bull. 34 (4): 213-214. Laborious, G.A. (1990) Biologically - integrated control of pp. 16-24. In: R. H. Markham and H.R. Heiren (eds.) Biological control of the larger grain borer. Proc. ETA/FAO coordination meeting, Cotonou, Rep. of Benin, 2-3 June 1989. Laborius, G.A. Leliveldt, B. and Schulz F.A. (1989). Evaluation of biological control methods to control the larger grain borer Prostephanus truncatus (Horn) University of Ghana http://ugspace.ug.edu.gh 99 Coleoptera: Bostrichidae) in Africa with special reference to the predator Teretriosoma nigrescens Lewis (Coleoptera: Histeridae). In: Integrated pest management in tropical and sub-tropical cropping systems. Proceedings of the International DLG symposium, Bad Durkheim, Germany, February 8-15 1989. Leliveldt, (1990) Antagoristen von Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) Doktorarbeit, Freie Universitat Berlin, ix + 215pp. Leliveldt,. B. and Laborius, G.A. (1990) Effectiveness and specificity of the antagonist Teretriosoma nigrescens Lewis (Col.: Bostrichidae)pp. 87-102. In: R4 Markham and H.R. Herren (eds.) Biological control of the larger grain borer. Prog DTA/FAO coordination meeting, Cotonou, Benin 2-3 June 1989. DTA, Ibadan. Nigeria viii + 171pp Lesne, P. (1898) Revision des coleopteres de la famille des Bostrychides. Ann. Soc. ent. fr. LXVI, 319-350 (April 1898) Lewis, G. (1891) On two new species of Mexican Histeridae. Ent. Monthly Mag. 2(2), 106-107. Li, L. (1988) Behavioural ecology and life history evolution in the larger grain borer, Prostephanus truncatus (Horn). Doctoral Thesis. University of Reading U.K. vii + 229pp. Longstaff, B. C. (1981) Biology of the grain pest species of the genus Sitophilus (Coleoptera: Curculionidae) A critical review. Prot Ecol. 3:83-100 Magrath, P. (1994) Cost-benefit analysis (CBA) of fanner storage . In Compton, J. (ed.) Ghana Larger Grain Borer Project, NRI Report. 6th Quarterly Report, April - June, 1994. Magrath, P. and Compton, J. A. F. (1994) Maize quality-value study. In Compton, J. (ed.) Ghana Larger Grain Borer Project, NRI Report. 8th Quarterly Report, October - December, 1994. pp. 14 - 22. Markham R. H., Wright, V. F and Rios Ibarra, R. M. (1991) A selective review of research on Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) with an annotated and updated bibliography. CEIBA 32 (1) 90pp. University of Ghana http://ugspace.ug.edu.gh 100 Markham R.H. (1981) The ecology of insect pest populations in maize storage in Nigeria. Ph.D. Thesis University of London. Markham, R.H., Borgemeister, C. and Meikle, W. G. (1994b) Can biological control resolve the larger grain borer crisis. In: E. Highley E. J. Wright, H. J. Banks and B. R. Champ (eds.) Proc 6th Int. Wkg. Conf. stored-Prod. Prot. 17-23 April, 1994, Canberra, Australia. 2:1087 - 1097. Markham, R.H., Borgemeister, C., Meikle, W. G. and Espinal, R. (1994c). Larger grain borer and the maize weevil. UTA Annual Report, 1995. p56 - 67. Markham, R.H., Bosque - Perez, N.A., Borgemeister, C. and Meikle, W. G. (1994a) Developing pest management strategies for Sitophilus zeamais and Prostephanus truncatus in the tropics. FAO Plant Prot. Bull. 42 (3) 97-116. McCullagh, P. (1980) Regression models for ordinal data (with discussion) J. Royal Stat. Soc. B42: 109-142. McFarlane, J. A. (1988) Pest management strategies for Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) as a pest of stored maize grain: present status and progress. Trop. Pest Manage 34 (2)121 - 132. MOA/WFED (1990) Recipes for Soybean, Maize and Cowpea in Ghanaian traditional dishes. MO A/Extension S4. Women Farmers Extension Division, Department of Agricultural Extension services .Ministry of Agriculture, Accra, Ghana, June, 1990. Mushi, A. M.(1984) The larger grain borer, Prostephanus truncatus (Horn) problem in Tanzania. In the larger grain borer Prostephanus truncatus (Horn). GASGA Workshop Tropical Product institute storage Department, Slough, 24-25 Feb., 1983 GTZ GmbH Eschbom. Federal republic of Germany Nang’ayo, F. L. O., Hill, M. G., Chandi, E. A., Chiro, C. T.5 Nzeve, D. N. and Obiero, J. (1993) The natural environment as a reservoir for the larger grain borer, Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) in Kenya. African Crop Science Journal, 1:39-47. NARP (1993) Cereals: A report by the commodity committee of NARP. Final revised Version. Appendix VI, 1: 1-37. University of Ghana http://ugspace.ug.edu.gh 101 Novillo Rameix, P. (1991) Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) en el ambiente de alamacenamiento Traditional de maiz en el Valle de El Zamorano, Honduras, C.A. Tesis de Ingeniero Agronomo, Escuela Agricola Panamericana Honduras viii + 12a pp. Nyakunga, Y.B. (1982) The biology of Prostephanus truncatus (Horn) (Col.: Bostrichidae) on cassava Master of Science thesis University of Reading, England. 76pp. Nyanteng, V.K. (1972) The storage of foodstuff in Ghana. Univ. of Ghana, Legon, Ghana. ISSER Technical Publication series 18. 92pp. Nyanteng, V.K. and van Apeldoom, G.J. (1971) The fanner and the marketing of foodcrops. Tech. Pub ISSER. Univ. of Ghana, Legon. Series No. 19. 84pp Okelana, F.A. and Osuji, F.N.C. (1984). Influence of relative humidity at 30°C on the oviposition, and development and mortality of Sitophilus zeamais Motsch. (Coleoptera: Curculionidae) in Maize Kernels. J. Stored Prod. Res. 21:13-19. Oussou, R. A., Meilde, W. G., Borgemeister, C., Markham, R.H., Degbey, P., Adda, C., Djomamou, B., Azoma, K., Bonou, C.A. (1995). Growth and development of Teretriosoma nigrescens Lewis (Col: Histeridae) under different humidity regimes. Conf. Association of African Insect Scientists (AAIS), on the theme Management of insects for better health and natural Resource Management. 6-11th Aug. 1995. Yamoussoukro, Cote d’Ivoire. Pantenius, C. U. (1988) Storage losses in traditional maize granaries in Togo. Insect Sci. Applic. 9 (6): 72-735. Pike, V., Akinnibagbe, JJ.A and Bosque - Perez, N.A. (1992) Nigeria: Larger grain borer CProstephanus truncatus) outbreak in Western Nigeria. FAO Plant Prot Bull. 40 (4): 170 - 173. Pike, V., Smith, J.L., White, R.D. and Hall, D.R. (1994) Studies of responses of stored - product pests Prostephanus truncatus (Horn) and Sitophilus zeamais Motsch., to food volatiles. Proc. 6th Int. Wkg. Conf. stored-Prod. Prot. 1:566-569. University of Ghana http://ugspace.ug.edu.gh 102 Poschko, M. (1994) Research into the biology and host - specificity of Teretriosoma nigrescens, a potential predator of Prostephanus truncatus. Ph.D. Thesis. Univ. of Berlin GTZ, Eschbom, Germany. PPMED (1993) Agriculture in Ghana: Facts and figures, issued by policy, Planning, Monitoring and Evaluation Division of the Ministry of Food and Agriculture, Nov. 1991 Proctor D. L. (1971) An additional aedeagal character for distinguishing Sitophilus zeamais Motsch. from Sitophilus oryzae (L) (Coleoptera: Curculionidae). J. stored Prod. Res. 6 (4): 351-352. Ramirez Genel, M. (1959) Conservacion de granos almacenados en Mexico. Chapingo 12:377-389. Ramirez Genel, M. (1960) Infestacion de campo por insectos de granos almacenaclos. Obsvaciones en maiz de la mesa central y det tropico en 1959. Agricultura Tecnica en Mexico 10: 32-35. Ratnadass, A. and Fleurat - Lessard, F. (1991) Improvements in the practical methods of assessment of losses caused by insects in grain stored at the village level in tropical Africa. Proc. 5th Int. Wkg. Conf. stored-Prod. Prot Bordeaux, Sept 9-14, 1990, 3:1681-1691. Rawnsley, J. (1969) crop storage. Food Research and Development unit Technical Report No.l FAO PL:SF/GHA 7 89pp. Rees, D. P. (1985) Life history of Teretriosoma nigrescens Lewis (Coleoptera: Histeridae) and its ability to suppress populations of Prostephanus truncatus (Horn) Coleoptera: Bostrichidae). J. stored Prod. Res., 21:115-118. Rees, D. P. (1987) Laboratory studies on predation by Teretriosoma nigrescens Lewis (Col: Histeridae) on Prostephanus truncatus (Horn) (Col: Bostrichidae) infesting maize cobs in the presence of other maize pests. J. stored Prod. Res. 23:191-95. Rees, D. P. (1991) The effect of Teretriosoma nigrescens Lewis (Coleoptera: Histeridae) on three species of storage Bostrichidae infesting shelled maize. J. stored Prod. Res. 27:83-86. University of Ghana http://ugspace.ug.edu.gh 103 Rees, D. P. Rodriguez Rivera, R. and Herrera Rodriguez, F.J. (1990) Observation on the ecology of Teretriosoma nigrescens Lewis (Col: Histeridae) and its prey, Prostephanus truncatus (Horn) (Col: Bostrichidae) in the Yucatan peninsula, Mexico. Trop. Sci, 30:153-165. Riley, C.V. (1894) the insect occurring in the foreign exhibits of World’s Colombian Exposition. Insect life 6:213-227. Schoonhoven, A. V. Mills, R. B. and Horber, E. (1974). Development of Sitophilus zeamais Motsch in maize kernels and pellets made from maize kernel fractions. J. stored Prod. Res., 10:73-80. Schulten G. G. M (1976) Insects in stored maize ears. Abstr. Trop. A g r i c 2 (6): 9-17. Schulten, G. G. M (1975) Losses in stored maize in Malawi (C. Africa) and work undertaken to prevent them. Eppo Bull 5:113-120. Shade, R. E. Furgason, E.S. and Murdock, L. L. (1990) Detection of hidden infestations by feeding - generated ultrasonic signals. American Entomologist .36 (3):231-234. Sharifi, and Mills, R. B. (1971) Development activities and behaviour of the rice weevil inside wheat kernels. J. Econ. Entomol. 64:1114 - 1118. Shires, S. W. (1977) Ability of Prostephanus truncatus (Coleoptera: Bostrichidae) to damage and breed on several stored food commodities. J. stored Prod Res., 13: 205 - 208. Shires, S. W. (1979) Influence of temperature and humidity on survival, development period and adult sex ratio in Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae).,/. Stored Prod. Res. 15:5-10. Shires, S. W. (1980) Life history of Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) at optimum conditions of temperature and humidity. J. stored Prod. Res. 16:147- 150. Shires, SAW and McCarthy, S. (1976) A character for sexing live adults of Prostephanus, truncatus (Coleoptera: Bostrichidae). J. stored ProdI Res., 12:273-274. Spilman, T. J. (1984) Identification of larvae and pupae of the larger grain borer, Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae), and the larger black flour beetle, Cynaeus angustus (Coleoptera: Tenebrionidae), pp. 44 - 53. In Proc. University of Ghana http://ugspace.ug.edu.gh 104 3rd int. Wkg. conf. stored-Prod. Entomol, Kansas State Univ. Manhattan, Kansas, USA, Oct. 23-28, 1983. 726+7pp. Stabrawa, A. (1992) Study of the maize and cassava farming and storage systems in central Togo with reference to the impact of the larger grain borer. Report R.1844, xxi + 86pp. Natural Resources Institute, Chatham, UK Stem, V. M. (1973) Economic thresholds. Ann. Rev. Entomol .18:259-280. Subramanyam, Bh. and Hagstrum, D. W. (1991) Quantitative analysis of temperature, relative humidity, and diet influencing development of the larger grain borer, Prostephanus truncatus (Horn) Coleoptera: Bostrichidae) Trop. Pest Mgt. 37 (3): 195 -202. Subramanyam, Bh., Cutkomp, L. K. and Darveaux, B. A. (1985) A new character for identifying larval instars of Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae). J. Stored Prod. Res. 23 (3): 151 155. Subramanyam, Bh., Harein, P. K. Cutkomp, L. K. and Pegors, C (1988). Feeding by adults of Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) on shelled maize: Influence of damaged grain, adult crowding and grain density. Ins Sci Applic. 9 (2) : 263-266. Tigar, B. J., Osborne, P. E., Key, G. E. Flores - S, M. E. and Vazquez - A, M. (1994) Distribution and abundance of Prostephanus truncatus (Coleoptera: Bostrichidae) and its predator Teretriosoma nigrescens (Coleoptera: Histeridae) in Mexico. Bull. Ent. Res. 84: 555 -565. Tipping, P. W (1986). Mechanisms and inheritance in whole kernel com Zea mays L. to the maize weevil Sitophilus zeamais Motschulsky. Ph.D. Thesis. University of Kentucky. 128pp. Twumasi - Afiiyie, S. Badu-Apraku, B., Sallah, P. Y. K. and Dzah, B. D. (1992). Potential of maize as a major source of quality protein for Ghana. A paper presented at the 12 Annual Maize and legumes Workshop, Kumasi. March 24-26. Urrelo, R and Wright. V. F (1989) Development and behaviour of immature stages of the maize weevil (Coleoptera: Curculionidae) within kernels of resistant and susceptible maize. Ann. Entomol. Soc. 82 (6): 712-716. University of Ghana http://ugspace.ug.edu.gh 105 Vowotor, K. A. (1992). Effect of maize variety and storage form on opposition and development of the maize weevil Sitophilus zeamais Motschulsky (Coleoptera: Curculionidae). M. Phil. Thesis, University of Ghana.. Vowotor, K. A. Bosque - Perez, N.A. and Ayertey, J. N. (1994) Effect of maize variety and storage form on oviposition and development of maize weevil. In Proc. 6th Int. Wkg conf. stored - Prod. Prot. Canberra, Australia. 1:595-598. Vowotor, K. A., Boateng, B. A., Ayertey, J. N., Oussou, R.D., Meikle, W. G., Borgemeister, C. and Markham, R.H. (1995b) Controlling the larger grain borer, Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) in maize using host - plant resistance and biological control techniques. Paper presented at CTA/HBC Conference, Addis Ababa, 9-14th October, 1995. Vowotor, K. A., Bosque - Perez, N.A. and Ayertey, J. N (1995a). Effect of maize variety and storage form on the development of the maize weevil, Sitophilus zeamais Motschulsky. J. Stored Prod. Res. 31 (1): 29-36. Walgenbach, C. A. and Burkholder, W. E. (1987) Mating behaviour of the maize weevil, Sitophilus zeamais (Coleoptera: Curculionidae) Ann. Ent. Soc. Amer. 80: 578-583. Wheatly, P. E. (1973) Relative susceptibility of maize varieties. Trop. Stored Prod. Inf. 25:16-18. Wright, M. Akou - Edi, D. and Stabrawa, A. (1993). SPV, NRI research project on Prostephanus truncatus in cassava stored as cossettes: preliminary findings from Togo. SPV, NRI - Report 1992, 134pp. Yoshida, T. (1983). Damage and losses in stored products from insect pests in tropical countries. Proc. 3rd NFRI-UNU Workshop, Bio-loss of post-harvest food and its prevention technology, Jan. 21 1983. Tsukuba Science City Japan. Young, W. R. Canadia, D., Bames, Z. D. and Smith, D. L. (1961-62) Almacena miento de maiz en el tropics bajo grados differentes de humedad. Agricultura Tecnica en Mexico, 12: 13-16. University of Ghana http://ugspace.ug.edu.gh 106 Appendix 1 Aggggsmgnt r>fmai7e varieties from Ghana for resistance to Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) Summary of Materials and Methods Maize varieties were frozen and placed in a controlled temperature and humidity chamber (CTH) room at 27°C and 70% r.h. Moisture content of the grains was equilibrated between 13.5% -14.5%. Maize Hardness Testing A Glen Creston mill was used at a speed of 6000rpm. Three 20g samples of maize were ground to warm up the mill. The mill was left for 20 seconds after being turned on to obtain a constant speed. Two replicates per sample of 20g were milled for 20 seconds. The flour left in the mill after this time was brushed out. The entirety was sieved through a 710um sieve for four minutes using an Endecotts Octagon 200 test sieve shaker on maximum amplitude. Coefficient of hardness is the weight retained divided by the weight that passed through the sieve (i.e. the higher the coefficient obtained the harder the variety). VARIETY HARDNESS COEFFICIENT SAFITA-2 2.3 KAWANZIE 2.3 DZOLOKPUITA 1.5 PENYI 1 ABUROTIA 1.9 LEKLEBI 1.2 ABELEEHI 1.8 DORKE-SR 1.3 CULTURE MAIZE 0.9 ABUTIA 0.8 University of Ghana http://ugspace.ug.edu.gh 107 SUMMARY OF ANALYSIS OF VARIANCE PROCEDURE REPEATED MEASURES ANALYSIS OF VARIANCE REPEATED MEASURES LEVEL INFORMATION APPENDIX 2(Prostephanus truncatus} Dependent variable PT1 PT 2 PT 3 PT 4 Level o f PT 1 2 3 4 Source df Anova SS Mean square F Value Pr > F Variety 2 0.8607 0.4304 7.75 0.0110 Error 9 0.4995 0.0555 Test of Hvootheses for Within subiect effects Source df Anova SS Mean square F Value P r> F PT 3 0.1385 0.0461 3.17 0.0404 PT*Variety 6 0.2622 0.0435 3.0 0.0224 Error(PT) 27 0.3934 0.0146 Analysis of variance of contrast Variables PT.N represents the contrast between the nth level of PT and the las Contrast variable: PT.l Source df Anova SS Mean square F Value Pr > F Mean 1 0.2107 0.2107 5.38 0.0455 Variety 2 0.4074 0.2037 5.20 0.0315 Error 9 0.3524 0.0392 Contrast variable: PT.2 Source df Anova SS Mean square F Value Pr>F Mean 1 0.0660 0.0660 1.49 0.2536 Variety 2 0.3157 0.1579 3.56 0.0727 Error 9 0.3993 0.0444 Contrast variable: PT.3 Source df Anova SS Mean square F Value Pr > F Mean 1 0.2002 0.2002 10.38 0.0105 Variety 2 0.2372 0.1186 6.15 0.0207 Error 9 0.1736 0.0193 Dependent variable PT 5 PT 6 PT 7 PT 8 Level of PT 1 2 3 4 Test of Hypotheses for between subject effects Source______df Anova SS Mean square F Value_____ Pr> F Variety 2 9.4367 4.7183 4.3 0..0489 Error 9 9.877 1.0974 University of Ghana http://ugspace.ug.edu.gh 108 Test of Hypotheses for Within subject effects Source_____ df Anova SS Mean square F Value_____ Pr>F PT 3 6.1978 2.0659 25.11 0.0001 PT* Variety 6 0.2480 0.0413 0.5 0.8009 Error (PT) 27 2.2211 0.0823 Analysis of variance of contrast Variables PT.N represents the contrast between the nth level of PT and the last Contrast variable: PT. 1 Source df Anova SS Mean square F Value Pr >F Mean 1 10.6974 10.6974 37.04 0.0002 Variety 2 0.0533 0.0267 0.09 0.9127 Error 9 2.5660 0.2888 Contrast variable: PT.2 Source df Anova SS Mean square F Value Pr >F Mean 1 7.1302 7.1302 40.52 0.0001 Variety 2 0.2601 0.1301 0.74 0.5044 Error 9 1.5835 0.1760 Contrast variable: PT.3 Source df Anova SS Mean square F Value Pr > F Mean 1 2.4210 2.4210 22.34 0.0011 Variety 2 0.0753 0.0377 0.35 0.7155 Error 9 0.9751 0.1083 APPENDIX 3 (Sitophilus zeamais) Dependent variable SZ 1 SZ 2 SZ 3 SZ 4 Level of SZ 1 2 3 4 Test of Hypotheses for between subject effects Source df Anova SS Mean square F Value Pr > F Variety 2 0.0478 0.2339 0.87 0.4515 Error 9 0.2472 0.0275 Test of Hypotheses for Within subject effects Source df Anova SS Mean square F Value Pr>F SZ 3 1.8773 0.6258 51.92 0.0001 SZ*Variety 6 0.4003 0.0667 5.54 0.0008 Error(SZ) 27 0.3254 0.0121 Analysis of variance of contrast Variables SZ.N represents the contrast between the nth level of SZ and the last Contrast variable: SZ. 1 Source df Anova SS Mean square F Value Pr> F Mean 1 2.0336 2.0336 87.55 0.0001 Variety 2 0.5923 0.2962 12.75 0.0024 Error 9 0.2090 0.0232 University of Ghana http://ugspace.ug.edu.gh 109 Contrast variahle: SZ.2 Source df Anova SS Mean square F Value Pr > F Mean 1 0.0096 0.0096 0.31 0.5927 Variety 2 0.0244 0.0144 0.46 0.6454 Error 9 0.2820 0.0313 Contrast variable: SZ.3 Source df Anova SS Mean square F Value Pr> F Mean 1 0.1541 0.1541 5.92 0.0378 Variety 2 0.0136 0.0068 0.26 0.7757 Error 9 0.2345 0.0261 Dependent variable SZ 5 SZ 6 SZ 7 SZ 8 Level of SZ 1 2 3 4 Test of Hypotheses for between subject effects Source______df Anova SS Mean square F Value_____ Pr > F Variety 0.0035 0.0018 0.13 0..876 Error 0.1181 0.0131 Test of Hypotheses for Within subject effects Source df Anova SS Mean square F Value_____ Pr > F SZ 3 0.0547 0.0182 1.78 0.1740 SZ*Variety 6 0.0427 0.0071 0.70 0.6552 Error(SZ) 27 0.2762 0.0102 Analysis of variance of contrast Variables SZ.N represents the contrast between the nth level of SZ and the last Source df Anova SS Mean square F Value Pr > F Mean 1 0.0675 0675 5.09 0.0505 Variety 2 0.0616 0.0308 2.32 0.1539 Error 9 0.1194 0.0133 Contrast variable: SZ.2 Source df Anova SS Mean square F Value Pr > F Mean 1 0.0936 0.0934 2.78 0.1299 Variety 2 0.0223 0.0112 0.33 0.7265 Error 9 0.3032 0.0337 Contrast variable: SZ.3 Source df Anova SS Mean square F Value Pr>F Mean 1 0.0280 2.028 1.4 0.2666 Variety 2 0.0405 0.0217 1.09 0.3776 Error 9 0.1799 0.02 University of Ghana http://ugspace.ug.edu.gh 110 APPENDIX 4 (Percent weight loss) Dependent variable Loss 1 Loss 2 Loss 3 Loss 4 Level of Loss 1 2 3 4 Test of Hypotheses for between subject effects Source______df Anova SS Mean square F Value_____ Pr > F Variety 2 0.2842 0.1421 17.5 0.0008 Error 9 0.0731 0.0081 Test of Hypotheses for Within subject effects Source_____ df Anova SS Mean square F Value_____ Pr > F Loss 3 0.3232 0.1077 10.06 0.0001 Loss*Variety 6 0.2034 0.0339 3.17 0.0175 Error(Loss) 27 0.0731 0.0081 Analysis of variance of contrast Variables Loss.N represents the contrast between the nth level of Loss and the last Contrast variable: Loss. 1 Source df Anova SS Mean square F Value Pr > F Mean 1 0.5677 0.5677 81.65 0.0001 Variety 2 0.0599 0.0299 4.3 0.0488 Error 9 0.0626 0.0069 Contrast variable: Loss.2 Source df Anova SS Mean square F Value Pr > F Mean 1 0.1408 0.1408 4.46 0.0638 Variety 2 0.1623 0.0826 2.62 0.1270 Error 9 0.2841 0.0316 Contrast variable: Loss.3 Source df Anova SS Mean square F Value P r> F Mean 1 0.0217 0.0217 2.02 0.1886 Variety 2 0.1262 0.0631 5.89 0.0232 Error 9 0.0964 0.0107 Dependent variable Loss 5 Loss 6 Loss 7 Loss 8 Level of Loss 1 2 3 4 Test of Hypotheses for between subject effects Source df Anova SS Mean square F Value Pr > F Variety 2 0.1164 0.0832 4.25 0..0501 Error 9 0.1460 0.0196 Test of Hypotheses for Within subject effects Source df Anova SS Mean square F Value Pr > F Loss 3 0.3363 0.1121 9.72 0.0002 Loss* Variety 6 0.1495 0.0249 2.16 0.0786 Error(Loss) 27 0.3112 0.0115 University of Ghana http://ugspace.ug.edu.gh I ll Analysis of variance of contrast Variables Loss.N represents the contrast between the nth level of Loss and the last Contrast variable: Loss. 1 Source df Anova SS Mean square F Value Pr > F Mean 1 0.4219 0.4219 38.79 0.0002 Variety 2 0.0236 0.0118 1.08 0.3790 Error 9 0.0978 0.0109 Contrast variable: Loss.2 Source df Anova SS Mean square F Value Pr>F Mean 1 0.5720 0.5720 29.1 0.0004 Variety 2 0.1209 0.0604 3.07 0.0960 Error 9 0.1769 0.0197 Contrast variable: Loss.3 Source df Anova SS Mean square F Value Pr > F Mean 1 0.2581 0.0258 9.99 0.1886 Variety 2 0.0477 0.0239 0.92 0.4318 Error 9 0.2326 0.0258 APPENDIX 5 (Visual scale loss') Dependent variable vsloss 1 vsloss 2 vsloss 3 Level of vsloss 1 2 3 Test of Hvootheses for between subject effects Source df Anova SS Mean square F Value Pr> F Variety 2 29.15 14.57 5.06 0.0336 Error 9 25.91 2.88 Test of Hvootheses for Within subject effects Source df Anova SS Mean sauare F Value Pr > F vsloss 2 42.96 21.48 6.47 0.0076 vsloss*Varie1y 4 11.94 2.99 0.9 0.4851 Error(vsloss) 18 59.78 3.32 Analysis of variance of contrast Variables vsloss.N represents the contrast between the nth level of vsloss and the Contrast variable: vsloss. 1 Source df Anova SS Mean square F Value Pr > F Mean 1 84.64 84.64 18.14 0.0021 Variety 2 23.4 11.7 2.51 0.1362 Error 9 41.99 4.67 Contrast variable: vsloss.2 Source df Anova SS Mean square F Value Pr > F Mean 1 13.13 13.13 1.11 0.3188 Variety 2 3.42 1.71 0.15 0.8670 Error 9 106.06 11.78 University of Ghana http://ugspace.ug.edu.gh 112 Dependent variable vsloss 4 vsloss 5 vsloss 6 vsloss 7 Level of vsloss 1 2 3 4 Test of Hypotheses for between subject effects Source______df Anova SS Mean square F Value_____ Pr > F Variety 2 19.44 9.72 3.61 0.0708 Error 9 24.26 2.7 Test of Hypotheses for Within subject effects Source______df Anova SS Mean square F Value_____ Pr>F vsloss 3 105.45 35.18 7.25 0.001 vsloss*Variety 6 15.42 2.57 0.53 0.7806 Error(vsloss) 27 130.94 4.85 Analysis of variance of contrast Variables vsloss.N represents the contrast between the nth level of vsloss and the last Contrast variable: vsloss. 1 Source df Anova SS Mean square F Value Pr>F Mean 1 200.17 200.17 17.3 0.0025 Variety 2 11.03 5.51 0.48 0.6357 Error 9 104.14 11.57 Contrast variable: vsloss.2 Source df Anova SS Mean square F Value Pr> F Mean 1 39.04 39.04 3.39 0.0986 Variety 2 0.86 0.43 0.04 0.9636 Error 9 103.55 11.51 Contrast variable: vsloss.3 Source df Anova SS Mean square F Value Pr > F Mean 1 87.53 87.53 6.58 0.0305 Variety 2 5.13 2.57 0.19 0.8278 Error 9 119.8 13.31 APPENDIX 6 rvalue loss) Dependent variable vloss 1 vloss 2 vloss 3 Level of vloss 1 2 3 Test of Hypotheses for between subject effects Source df Anova SS Mean square F Value Pr>F Variety 2 114.8716 57.4435 7.24 0.0133 Error 9 71.356 7.9284 Test of Hypotheses for Within subject effects Source df Anova SS Mean square F Value Pr > F vloss 2 130.4717 65.2358 6.92 0.0059 vk>ss*Variety 4 39.0867 9.7717 1.04 0.4158 Error(vloss) 18 169.7147 9.43 University of Ghana http://ugspace.ug.edu.gh 113 Analysis of variance of contrast Variables vloss.N represents the contrast between the nth level of vloss and the last Contrast variable: vloss.l Source df Anova SS Mean square F Value Pr> F Mean 1 258.54 258.54 18.37 0.002 Variety 2 77.32 38.66 12.75 0.1171 Error 9 126.64 14.07 Contrast variable: vloss.2 Source df Anova SS Mean square F Value Pr>F Mean 1 44.85 44.85 1.39 0.2689 Variety 2 13.13 6.56 0.2 0.8189 Error 9 290.79 32.31 Dependent variable vloss 4 vloss 5 vloss 6 vloss 7 Level of vloss 1 2 3 4 Test of Hvootheses for between subiect effects Source df Anova SS Mean square F Value Pr> F Variety 2 77.81 38.91 5.22 0...0312 Error 9 67.12 7.46 Test of Hvootheses for Within subiect effects Source df Anova SS Mean square F Value Pr > F vloss 3 319.86 106.62 8.15 0.0005 vloss*Variety 6 45.71 7.62 0.58 0.7414 Error(vloss) 27 353.4 13.09 Analysis of variance of contrast Variables vloss.N represents the contrast between the nth level of vloss and the last Contrast variable: vloss. 1 Source df Anova SS Mean sauare F Value Pr > F Mean 1 613.47 613.47 19.16 0.0018 Variety 2 37.81 18.9 0.59 0.5743 Error 9 228.19 32.02 Contrast variable: vloss.2 Source df Anova SS Mean square F Value Pr>F Mean 1 130.02 130.02 4.22 0.07 Variety 2 2.11 1.06 0.03 0.9664 Error 9 276.99 30.78 Contrast variable: vloss.3 Source df Anova SS Mean square F Value Pr>F Mean 1 260.4 260.4 6.65 0.0297 Variety 2 12.75 6.38 0.16 0.8522 Error 9 352.38 39.15