University of Ghana http://ugspace.ug.edu.gh i ANATOMICAL, GERMINATION AND IN VITRO STUDIES ON SHEA TREE (Vitellaria paradoxa C.F.Gaertn.) SEED A Thesis presented to the Department of NUCLEAR AGRICULTURE AND RADIATION PROCESSING SCHOOL OF NUCLEAR AND ALLIED SCIENCES UNIVERSITY OF GHANA By Iddrisu Abdulai, 10358232 BSc. AGRICULTURE (University of Cape Coast), 2007 In partial fulfilment of the requirements for the degree of MASTER OF PHILOSOPHY in NUCLEAR AGRICULTURE (Plant Biotechnology and Mutation Breeding) July, 2013 University of Ghana http://ugspace.ug.edu.gh ii DECLARATION ‘‘This thesis is the result of research work undertaken by Iddrisu Abdulai in the Department of Nuclear Agriculture and Radiation Processing of the School of Nuclear and Allied Sciences, University of Ghana, under the supervision of Prof. George Y. P. Klu and Dr. Kenneth E. Danso.’’ Signed………………………………………………. Iddrisu Abdulai (Candidate) Signed………………………………………………… Prof. George Y. P. Klu (Supervisor) Signed………………………………………………… Dr. Kenneth E. Danso (Supervisor) University of Ghana http://ugspace.ug.edu.gh iii DEDICATION This work is dedicated to the Almighty Allah for giving me knowledge, wisdom and all other resources to complete it successfully, to my late friend Nategu Naah Mahama for his effort at collecting shea fruits for this work and to Franklin Otsyina for his support and encouragement. University of Ghana http://ugspace.ug.edu.gh iv ACKNOWLEDGEMENTS I wish to express my profound gratitude to my supervisors Prof. George Y. P. Klu and Dr. Kenneth E. Danso for their expert supervision and encouragement. I am particularly grateful to Dr. Kenneth E. Danso, Director of Biotechnology and Nuclear Agriculture Research Institute (BNARI), for allowing me unrestricted access to the laboratories and other facilities of the institute. My special thanks also go to Dr. Harry M. Amoatey (Head of Department, Nuclear Agriculture and Radiation Processing) for the fatherly care given to me throughout my study period at the School of Nuclear and Allied Sciences. I wish to thank all the technicians and scientists of the Biotechnology Centre, BNARI, especially Messrs. Alex Asumeng, Elegba Wilfered and Amposah Jonathan for their help. To all my colleagues Yayra, Gloria, Ebenezer, Ansah, Akwesi, Bransford and Antonio, Allah richly bless you for your support and encouragement. Special thanks go to the Management especially Mr. Nartey Baby Tei and all the staff of Cosmos Basic School for their support. I am also grateful to my friend Saaka Yakubu, who facilitated the transport of the shea fruits used for this study. Finally, I acknowledge the Government of Ghana and the School of Nuclear and Allied Sciences for the support during the research period. University of Ghana http://ugspace.ug.edu.gh v TABLE OF CONTENTS Title page……………………………………………………………………………….i Declaration ..................................................................................................................... ii Dedication .................................................................................................................... iii Acknowledgements ....................................................................................................... iv Table of contents ............................................................................................................ v List of tables .................................................................................................................. ix List of figures ................................................................................................................. x List of abbreviations .................................................................................................... xii ABSTRACT ................................................................................................................ xiv CHAPTER 1 .................................................................................................................. 1 1.0. Introduction ......................................................................................................... 1 CHAPTER 2 .................................................................................................................. 6 2.0. Literature review ..................................................................................................... 6 2.1. Origin and distribution ........................................................................................ 6 2.2. Taxonomy and classification ............................................................................... 8 2.3. Botany of shea tree .............................................................................................. 9 2.3.1. Canopy and branching .................................................................................. 9 2.3.2. Roots ........................................................................................................... 11 2.3.3. Leaves and flowers ..................................................................................... 12 2.3.4. Fruits ........................................................................................................... 12 2.3.5. Seed anatomy and morphology .................................................................. 13 2.4. Germination of Vitellaria paradoxa seeds ........................................................ 14 2.5. Socio-economic importance and uses of Vitellaria paradoxa .......................... 18 2.5.1. Domestic uses of Vitellaria paradoxa ........................................................ 18 University of Ghana http://ugspace.ug.edu.gh vi 2.5.2. International and industrial uses of sheanut and butter .............................. 19 2.6. Domestication of Vitellaria paradoxa .............................................................. 20 2.6.1. Domestication status of the species ............................................................ 20 2.6.2. Selection in agroforestry parklands ............................................................ 21 2.6.3. Breeding for earliness ................................................................................. 21 2.6.4. In vitro propagation of Vitellaria paradoxa ............................................... 22 REFERENCES ............................................................................................................ 24 CHAPTER 3 ................................................................................................................ 30 3.0. Anatomical and morphological studies on Vitellaria paradoxa seed ................... 30 3.1. Introduction ....................................................................................................... 30 3.2. Materials and methods ...................................................................................... 32 3.2.1. Shea fruit collection .................................................................................... 32 3.2.2. Studies on the morphology of Vitellaria paradoxa seed ............................. 33 3.3. Anatomical studies on the seed and embryo identification with topographical tetrazolium test ................................................................................................. 34 3.4. Results ............................................................................................................... 35 3.4.1. Morphology of Vitellaria paradoxa seed ................................................... 35 3.4.2. Anatomy of Vitellaria paradoxa seed ........................................................ 39 3.5. Discussion ......................................................................................................... 44 3.6. Conclusion ......................................................................................................... 49 REFERENCES ............................................................................................................ 50 CHAPTER 4 ................................................................................................................ 53 4.0. Germination studies on Vitellaria paradoxa seeds ............................................... 53 4.1. Introduction ....................................................................................................... 53 4.2. Materials and methods ...................................................................................... 54 University of Ghana http://ugspace.ug.edu.gh vii 4.2.1. Seed collection ............................................................................................ 54 4.2.2. Studies on Vitellaria paradoxa seedling development ............................... 55 4.2.3. Seed size and development of Vitellaria paradoxa seedlings .................... 56 4.2.4. Deshelling of seed and development of Vitellaria paradoxa seedlings ..... 58 4.2.5. Statistical analysis ....................................................................................... 58 4.3. Results ............................................................................................................... 59 4.3.1. Stages of the development of Vitellaria paradoxa seedlings ..................... 59 4.3.2. Effect of seed size on germination and emergence of V. paradoxa seedlings ............................................................................................................... 69 4.3.3. Effects of deshelling of seeds on the germination and growth of Vitellaria paradoxa seedlings .................................................................................... 74 4.4. Discussion ......................................................................................................... 80 4.4.1. Development of Vitellaria paradoxa seedlings .......................................... 80 4.4.2. Seed size and development of Vitellaria paradoxa seedlings .................... 84 4.4.3. Growth and morphology of Vitellaria paradoxa seedlings ........................ 86 4.5. Conclusion ......................................................................................................... 89 REFERENCES ............................................................................................................ 90 CHAPTER 5 ................................................................................................................ 93 5.0. In vitro propagation of Vitellaria paradoxa .......................................................... 93 5.1. Introduction ....................................................................................................... 93 5.2. Materials and methods ...................................................................................... 95 5.2.1. Collection of Vitellaria paradoxa fruits ..................................................... 95 5.2.2. In vitro culture of intact seeds .................................................................... 95 5.2.3. In vitro culture of deshelled seeds .............................................................. 96 5.2.4. Identification and culture of embryonic axes ............................................. 96 University of Ghana http://ugspace.ug.edu.gh viii 5.2.5. In vitro culture of rudimentary shoots ........................................................ 97 5.2.6. Data analysis ............................................................................................... 98 5.3. Results ............................................................................................................... 98 5.3.1. In vitro germination of intact seeds ............................................................ 98 5.3.2. In vitro germination of deshelled seeds ...................................................... 99 5.3.3. Response of embryonic axes to in vitro culture........................................ 100 5.3.4. In vitro regeneration of rudimentary shoots ............................................. 102 5.4. Discussion ....................................................................................................... 106 5.5. Conclusion ....................................................................................................... 110 REFERENCES .......................................................................................................... 111 CHAPTER 6 .............................................................................................................. 113 6.0. General conclusions and recommendations ........................................................ 113 6.1. Conclusions ..................................................................................................... 113 6.2. Recommendations ........................................................................................... 114 APPENDICES ........................................................................................................... 115 University of Ghana http://ugspace.ug.edu.gh ix LIST OF TABLES Table 4.1 Protrusion of pseudoradicles from different sides of germinating Vitellaria paradoxa seeds…………………………………………………………...61 Table 4.2 Number of pseudoradicles produced per germinating Vitellaria paradoxa seed……………………………………………………………………......62 Table 4.3 Effect of seed size on germination, emergence and emergence rate index of Vitellaria paradoxa seedlings………………………………...….…....71 Table 4.4 Effect of seed size on development of Vitellaria paradoxa seedlings…..71 Table 4.5 Effects of seed size on morphological features of Vitellaria paradoxa seedlings at bulging and at emergence.…………………………………..72 Table 4.6 Effects of deshelling of seeds on germination and emergence parameters of Vitellaria paradoxa seedlings………………………………………...74 Table 4.7 Effect of deshelling of seeds on the growth of Vitellaria paradoxa seedlings at 150 days after sowing……………………………………….76 Table 4.8 Effect of deshelling of seeds on the growth of Vitellaria paradoxa seedlings at 240 days after sowing……………………………………….77 Table 5.1 Effect of BAP and NAA on the response to culture, height and leaf production of rudimentary shoot explants 15 and 45 days after culture..103 University of Ghana http://ugspace.ug.edu.gh x LIST OF FIGURES Fig. 2.1 Maps showing the Shea Belt……………………………………………..…7 Fig. 2.2 Vitellaria paradoxa tree in agroforestry parkland…………………………10 Fig. 2.3 Bole of a Vitellaria paradoxa tree ………………………………………...11 Fig. 2.4 Fruits of Vitellaria paradoxa…………………………………..……….….13 Fig. 2.5 Cryptogeal seedling of Vitellaria paradoxa…………………………….....15 Fig. 2.6 Sheanuts and butter……………………………………………………..…18 Fig. 3.1 Map of the Upper West Region of Ghana showing Ga and Tanina………33 Fig. 3.2 Cartographic drawing of a Vitellaria paradoxa seed showing the different sides………………………………………………………………………..35 Fig. 3.3 Vitellaria paradoxa seeds…………………………………………………36 Fig. 3.4 Cotyledon morphology of Vitellaria paradoxa seeds…………………….38 Fig. 3.5 Transverse sections through partially dry V. paradoxa seeds………….....39 Fig. 3.6 Vitellaria paradoxa seeds stained by tetrazolium chloride …………...….40 Fig. 3.7 Location of the embryo in Vitellaria paradoxa seeds ……………………41 Fig. 3.8 Polyembryonic Vitellaria paradoxa seed ……………………………..….42 Fig. 3.9 Split cotyledons of a Vitellaria paradoxa seed…………………………… 43 Fig. 3.10 Exudation of latex from fresh Vitellaria paradoxa seed…………………44 Fig. 4.1 Sprouted Vitellaria paradoxa seeds……………………………………….60 Fig. 4.2 Pseudoradicles of Vitellaria paradoxa seedlings at sprouting stage………61 Fig. 4.3 Protrusion of pseudoradicles from different sides of germinating Vitellaria paradoxa seeds…………………………………………………….……....62 Fig. 4.4 Vitellaria paradoxa seedlings at the second and third developmental stages ………………………………………………………………………….......63 Fig. 4.5 Morphological features of the pseudoradicle………………………………64 . Fig. 4.6 Anatomical and morphological features of the pseudoradicle at bulging stage…………………………………………………………………….......65 Fig. 4.7 Development of the rudimentary shoot at the bulging stage……………....66 University of Ghana http://ugspace.ug.edu.gh xi Fig. 4.8 Vitellaria paradoxa seedlings at the fourth and fifth stages of development ……………………………………………………………………………...67 Fig. 4.9 Production of multiple shoots and seedlings in Vitellaria paradoxa….…...68 Fig. 4.10 Types of seedlings produced by Vitellaria paradoxa based on cotyledon exposition…………………………………………….……...……………..69 Fig. 4.11 Morphological features of Vitellaria paradoxa seedlings………………...73 Fig. 4.12 Emergence of a trapped Vitellaria paradoxa seedling……......………...…75 Fig. 4.13 Tuberous root crown of Vitellaria paradoxa seedlings…………………....78 Fig. 4.14 Vitellaria paradoxa seedlings showing monopodial growth………………79 Fig. 4.15 Vitellaria paradoxa seedlings with two (A) and three (B) apical growing points……………………………………………………………………....80 Fig. 5.1 Intact Vitellaria paradoxa seed cultured in vitro on MS basal medium supplemented with 2.0 mg/l BAP 10 days after culture……………………98 Fig. 5.2 Sprouted Vitellaria paradoxa seed cultured on MS basal medium amended with 1.0 mg/l BAP after 35 days of culture…………………………………99 Fig. 5.3 Days to sprouting and percentage sprouting of deshelled Vitellaria paradoxa seeds cultured on MS basal medium supplemented with 1.0–4.0 mg/l BAP ………………………………………………………....100 Fig. 5.4 Embryonic axis culture of Vitellaria paradoxa………………………......101 Fig. 5.5 Days to sprouting and percentage sprouting of Vitellaria paradoxa embryonic axes cultured on MS basal medium supplemented with 1.0–4.0 mg/l BAP …..…………………………………………………….101 Fig. 5.6 In vitro regeneration of Vitellaria paradoxa using rudimentary shoots..…105 Fig. 5.7 Regenerated Vitellaria paradoxa shoots cultured on MS basal medium amended with 2.0 mg/l BAP and 0.2 mg/l NAA at 30 days after culture....106 University of Ghana http://ugspace.ug.edu.gh xii LIST OF ABBREVIATIONS ANOVA - Analysis of variance BA - 6-benzyladenine BAP - 6-Benzylaminopurine CBE - Cocoa Butter Equivalent CER - Cryptocotylar epigeal reserve CHR - Cryptocotylar hypogeal reserve CRIG - Cocoa Research Institute of Ghana CuSO4 - Copper sulphate DAS - Days after sowing DE - Distal end DS - Dorsal side EI - Emergence index EP - Emergence percentage ERI - Emergence rate index EST - Establishment IPGRI - International Plant Genetic Resources Institute GAEC - Ghana Atomic Energy Commission GP - Germination percentage GS - Germinated seeds HgCl2 - Mercuric(II) chloride HTML - HyperText Markup Language LA - Leaf area LW - Length and width LSD - Least significant difference MGT - Mean germination time MS - Murashige and Skoog University of Ghana http://ugspace.ug.edu.gh xiii N - North NAA - Naphthaleneacetic acid PE - Proximal end PEF - Phanerocotylar epigeal foliaceous PER - Phanerocotylar epigeal reserve PHF - Phanerocotylar hypogeal foliaceous PHR - Phanerocotylar hypogeal reserve PRE - Pseudoradicle elongation psi - pound per square inch SA - Shoot appearance SAM - Shoot apical meristem SE - Shoot elongation SSA - sub-Saharan Africa TS - Total number of sown seeds TTC - Tetrazolium chloride TTZ - Topographical tetrazolium UV - Ultraviolet VS - Ventral side 2,4-D - Dichlorophenoxyacetic acid University of Ghana http://ugspace.ug.edu.gh xiv ABSTRACT In vivo and in vitro germination and regeneration studies were conducted on the development of Vitellaria paradoxa seedlings as an initial effort towards its domestication. However, to achieve this objective, the morphology and anatomy of the seeds were first studied because they influence germination. Although a smooth, brown coat encloses a V. paradoxa seed, it did not impose dormancy on the embryo. Transverse and longitudinal sections through the seed showed that the embryo is surrounded by latex- and fat-containing tissues which made its identification difficult. Thus, the embryo was identified by immersing transversely cut seeds in 1.0 % tetrazolium chloride (TTC) solution for 24 hours which stained it red. When V. paradoxa seeds of similar size were sown on nursery beds, the resulting seedlings developed through seven stages namely sprouting, pseudoradicle elongation, bulging, appearance of the shoot on the pseudoradicle, shoot elongation, emergence and seedling establishment. The pseudoradicle is the fused petioles of the two cotyledons and a transverse section through it revealed an outer sheath and lactiferous vessels surrounding a central hollow tube. Longitudinal section also showed the lactiferous vessels surrounding the central hollow tube in which the plumule moves through until it reaches the bulge of the pseudoradicle where it develops into a rudimentary shoot. The rudimentary shoot then protrudes from the pseudoradicle and grows upwards. Classifying seeds into three groups based on sizes and sowing them on nursery beds showed that seed size significantly affected days to germination and the morphology of the resulting seedlings. Large seeds germinated within one week after sowing with vigorous growth compared to small and medium seeds. Although the seedcoat of V. paradoxa never imposed dormancy, deshelling (removal of the seedcoat) significantly led to early germination and synchronous seedling emergence compared to those for University of Ghana http://ugspace.ug.edu.gh xv intact seeds (control). In vitro culture of intact and deshelled seeds on Murashige and Skoog (1962) basal salts modified with 6-benzylaminopurine (BAP) produced no plantlets although 80 % of the deshelled seeds developed long pseudoradicles on a medium supplemented with lower concentration of BAP (1.0 or 2.0 mg/l). Similarly, the culture of TTC identified embryonic axes did not produce plantlets, but rather significantly long pseudoradicles were produced with BAP having significant effect on pseudoradicle development. Contrastingly, in vitro culture of excised rudimentary shoots on the same MS medium modified with BAP and naphthaleneacetic acid (NAA) produced plantlets with distinct shoots and leaves. Significant reduction in days to emergence of seedlings from deshelled seeds and successful in vitro plantlet development using rudimentary shoot explants will enhance nursery establishment of this economically important tree species for domestication and reafforestation programmes in sub-Saharan Africa. University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER 1 1.0. Introduction The shea tree (Vitellaria paradoxa C.F.Gaertn.) belongs to the family Sapotaceae. It is indigenous to sub-Saharan Africa (SSA) and typically occurs in the interior savannas where it is the major oil crop (Nikiema and Umali, 2007). The plant still remains an indigen confined to its native 19 countries located in SSA. In Ghana, shea trees are commonly found in the northern sector with sparse populations in Brong- Ahafo, Ashanti, Eastern and Volta regions of southern Ghana (Fobil, 2007). Now, the importance of shea tree to the local inhabitants of the Shea Belt chiefly depends on the time when its products (shea fruit, nut and butter) are available, late March to September. The early part of this period represents the time when energy demands for farm labour are highest, whilst food supply is at its lowest. Hence, the shea fruit is most frequently consumed as a staple food. The fruit pulp is nutritious containing large quantities of vitamin C, proteins and minerals (Ugese et al., 2008a,b; Maranz et al., 2004). In rural areas, the nuts are either bartered for starchy foodstuffs or sold immediately and the money used to buy them (Yidana, 2004). The product of international trade which is extracted from the nuts is shea butter. Shea butter has characteristics similar to those of cocoa butter; hence, it is used as a cocoa butter equivalent (CBE) to manufacture confectionery (Shea Matters, 2011). It is also used as a base for medicines and lotions in the pharmaceutical and cosmetic industries respectively. The fat is used locally as cooking oil, for soap making and as fuel for lighting lamps. Medicinally, almost every part of the tree is used, including the epiphytes Tapinanthus spp. which normally parasitize the species. The protein- University of Ghana http://ugspace.ug.edu.gh 2 rich defoliatory caterpillar Cirina butyrospermi, associated with this species, is widely consumed in Nigeria and other parts of the Shea Belt (Ande, 2004). Vitellaria paradoxa is an iconic and unique tree of the Sudano-Sahelian savanna landscape. Nutritionally, shea fruits are available during the lean season (Sanou and Lamien, 2011). Morphoagronomically, shea trees grow abundantly on marginal soils and have high longevity. Ecologically, they proliferate in arid and semiarid savannas and this proliferation manifests the species’ capabilities to combat desertification, to ameliorate the microclimate and to recycle nutrients through annual leaf shedding (Dianda et al., 2009). Socioeconomically, shea provides employment and income to women and children who are the most vulnerable in society (Elias and Carney, 2007). With both the kernels and butter as export commodities, V. paradoxa thus has the potential of bridging the economic gap between Northern and Southern Ghana. Despite all these benefits and potentials, V. paradoxa still remains wild (Nyarko et al., 2012; Okao et al., 2012) and lacks a tradition of being planted (Sanou and Lamien, 2011). Traditionally, farmers rarely plant shea tree because the seedlings grow extremely slowly (Ugese et al., 2010a) leading to the long gestation period of about 15–20 years (Masters et al., 2004). Seasonally, shea trees bear fruits erratically and this unpredictable yielding pattern negatively affects the agro-industrial development of the species and the food security of rural dwellers whose livelihood depends on it (Yidana, 2004). Prices of shea products are also disappointingly low. Of all the factors, the long juvenile growth period of the tree is main disincentive to the food-insecure farmers of the Shea Belt and has been primarily blamed for the non- University of Ghana http://ugspace.ug.edu.gh 3 domestication of the plant (Shea Matters, 2011; Moore, 2008). Thus, a substation was established by Cocoa Research Institute of Ghana (CRIG) at Bole in 1976 to research into the ecology of V. paradoxa and to develop early-bearing varieties and techniques for propagating them. In spite of CRIG’s efforts which resulted in some successes on vegetative propagation, no major breakthrough in the reduction of the gestation period has been reported (Yeboah et al., 2011). Consequently, the tree still remains wild. Wild Vitellaria trees now face more threats than ever, with natural stands being extensively cleared for establishing other high-income earning cash crops such as mango (Mangifera indica L.) and for producing fuelwood and charcoal. Osei Agyeman et al. (2012) reported that mature shea trees are the most preferred woody species for charcoal burning in the Upper West Region of Ghana with an average of 4 trees felled per month per charcoal burner. As a result, the Sudan savanna zone which in the 1940s had the densest population (230 trees ha−1) now has as few as 5 trees ha−1 (Djossa et al., 2008). Thus, developing efficient propagation techniques to produce seedlings or plantlets for planting, or reafforestation is highly recommended. Prospects and evidences of crop improvement by intensively growing the genetically unmodified V. paradoxa seedlings abound. Yidana (2004) reported that trees on agroforestry parklands produce higher yields and bear fruit more consistently than those in bushlands. Also, fruit yield of individual trees vary extensively, with some trees even bearing fruits bi-annually. These characteristics of V. paradoxa trees suggest that selection for yield improvement is a real possibility (Nyarko et al., 2012). Therefore, higher yields are obtainable when seedlings or plantlets produced directly from plus trees are purposefully planted to establish plantations. University of Ghana http://ugspace.ug.edu.gh 4 Intentionally planting Vitellaria seedlings to establish plantations should be considered as a viable initial effort towards domesticating this species because the shea industry still relies on fruits collected from the wild. The major challenge to this approach of establishing shea plantations is how to produce uniform and vigorous seedlings in large quantities. First, the seed is recalcitrant and thus germinates readily after harvest but loses viability rapidly (Orwa et al., 2009). Second, seedlings produce long taproots which adapt them well to their savanna habitat (Jackson, 1974), but the long taproots make raising them in the nursery and subsequent transplanting difficult. Therefore, germination studies are necessary to circumvent some of these problems, especially by developing seedlings with shorter taproots. In vitro propagation techniques may offer a feasible alternative for producing planting materials abundantly. However, all the vegetative parts of V. paradoxa including shoot tips which are possible explants exude latex copiously. Latex hinders vegetative propagation (Masters, 2002), often contaminating in vitro plant cultures massively. Latex-related contamination may be minimized when explants with limited amount of latex are identified and excised for culture. For example, the immature cotyledons and the pink-coloured juvenile shoot located in the bulged portion of the pseudoradicle of the germinated seeds may contain little or no latex. The immature cotyledon explants were cultured on Murashige and Skoog (1962) basal medium supplemented with 2,4- Dichlorophenoxyacetic acid (2,4-D) and somatic embryos were induced and successfully transformed into embryogenic calli (CRIG, 2012). Although Ugese et al. (2010a) and Jackson (1968) have already described the type of germination and the stages associated with Vitellaria seedling establishment, the University of Ghana http://ugspace.ug.edu.gh 5 morphological and anatomical features of the seeds responsible for these events are still unknown. For instance, the exact location of the embryo can hardly be tracked with a reasonable degree of accuracy. Thus, morphological and anatomical studies on the seed may be a necessary prerequisite to further investigate the sequence of activities involved in its germination. Knowledge about the morphological and anatomical features of the seed may help to determine appropriate methods of seed pre-treatment that may lead to quicker germination and seedling emergence. Intentional and commercial planting of V. paradoxa depends on deploying an efficient system for developing its seedlings or plantlets on large-scale basis. However, success rate of the ex vitro vegetative propagation methods has been low (Yeboah et al., 2011) because the latex sap inhibits contact between the cambial cells of graft unions and also quickly blocks transpiration vessels of cut surfaces (Masters, 2002). Thus, sexual propagation still remains a reliable method of producing seedlings. Also, in vitro propagation techniques using juvenile plant parts with little or no latex such as the immature shoots may offer better chances of producing plantlets both in large quantities and on timely basis. Thus, the major objective of this study was to develop efficient techniques for propagating V. paradoxa as part of initial efforts for domesticating the plant. The specific objectives were to i. study the morphological and anatomical features of the seed ii. investigate the stages of seedling development iii. develop appropriate nursery protocol for producing seedlings abundantly iv. develop an in vitro protocol for regenerating V. paradoxa plantlets using rudimentary shoots as explants. University of Ghana http://ugspace.ug.edu.gh 6 CHAPTER 2 2.0. Literature review 2.1. Origin and distribution Shea tree (Vitellaria paradoxa C.F.Gaertn.) is of African origin growing wild in West and Central Africa (IPGRI, 2006). It is a major component of the woody flora of the Sudan and Guinea savanna vegetation zones of sub-Saharan Africa (Byakagaba et al., 2011). Vitellaria paradoxa occurs in 19 African countries namely: Benin, Burkina Faso, Cameroon, Central African Republic, Chad, Côte d’Ivoire, Ethiopia, Ghana, Guinea, Guinea Bissau, Mali, Niger, Nigeria, Senegal, Sierra Leone, Sudan, Togo, Uganda and Democratic Republic of Congo (Hall et al., 1996). Vitellaria paradoxa was first described by Mungo Park, the Scottish explorer, in 1796 in the Ségou region of Mali (Sanou and Lamien, 2011). According to Allal et al. (2011), West Africa has the highest genetic diversity and the densest tree populations, thereby implicating the subregion as the most probable centre of origin of the species. The Moroccan traveller Ibn Battuta had documented shea butter as a high-value commodity in regional trade across West Africa as early as 1354 (Masters, 2002). Unknown precisely is the exact location from where Vitellaria germplasm spread to the other African countries (IPGRI, 2006), including Ghana. In Ghana, V. paradoxa grows abundantly in the northern sector and particularly thrives well in the Northern Region in Eastern Gonja, Western Dagomba, Southern Mamprusi, Western Gonja and Nanumba with Eastern Gonja having the densest stands (Lovett and Haq, 2000a) (Fig. 2.1A). In the Upper West Region, it occurs in Lawra, Tumu, Wa and Wechiau where pure stands of the trees are commonly found. University of Ghana http://ugspace.ug.edu.gh 7 V. paradoxa grows extensively in the Guinea savanna and less abundantly in the Sudan savanna; thus, tree population in the Upper East Region is scarcely as dense as those in Northern and Upper West regions (Fobil, 2007). Sparse shea tree cover exists in Brong-Ahafo, Ashanti, Eastern and Volta regions of Southern Ghana (Fobil, 2007). The Shea Belt, the geographical region in Africa where Vitellaria grows, is approximately 5000 km long and 500 km wide and ranges from western Senegal to northwestern Uganda (Shea Matters, 2011). Within the Shea Belt, the species occurs in areas with 400–1800 mm annual rainfall: its distribution area spreads from West- East Africa and up to the Adamaoua Province in Cameroon (North–South Africa) (IPGRI, 2006). Vitellaria paradoxa is localized between the latitudes 9° and 14° N in West Africa, 7° and 12° N in Central Africa and 2° and 8° in East Africa (Fig. 2.1B). The species is thus absent from humid forest, coastal areas and highlands at altitudes above 1600 m (Bonkoungou, 2002). It thrives well on various soils but avoids alluvial hollows and those prone to flooding (Tropical Advisory Service, 2002). Shea Belt Fig. 2.1A Fig. 2.1B Fig. 2.1. Maps showing the Shea Belt; A, Ghana’s Shea Belt (Re-drawn from Quainoo et al., 2012); B, Native range of shea tree (Re-drawn from Hatskevich et al., 2011) University of Ghana http://ugspace.ug.edu.gh 8 2.2. Taxonomy and classification Shea tree is a bacciferous fruit tree belonging to the sapodilla family, Sapotaceae, which contain flowering plants of the order Ericales. The Sapotaceae include both evergreen and deciduous trees, shrubs and lianas in approximately 65 genera with pantropical distribution (Jessup and Short, 2011). Some sapotaceous species such as Synsepalum dulcificum, Chrysophyllum albidum, C. giganteum, Tieghemella heckelii and Vitellaria paradoxa all of which are indigenous to Ghana produce edible fruits, oils and timber (Hawthorne and Jongkind, 2008). Botanically, shea tree was validly named Vitellaria paradoxa in 1807 by Carl von Friedrich Gaertner. Later, the butter-producing tree of West African origin was renamed Butyrospermum parkii (G.Don) Kotschy in 1865 in which the genus name Butyrospermum translates from Latin as butter (butyros) and spermum (seed) and the specific epithet parkii honours Mungo Park. Butyrospermum parkii remained as the most popular name of the species throughout the 20th century. However, V. paradoxa has priority and is, therefore, the botanically valid name whilst B. parkii, and B. paradoxum are homotypic synonyms (McNeill and Turland, 2011). The genus Vitellaria is monotypic but has 2 subspecies known as Vitellaria paradoxa subsp. paradoxa [synonym: Butyrospermum parkii (G.Don) Kotschy] and subsp. nilotica (Kotschy) A.N.Henry, Chithra and N.C.Nair (synonym: Butyrospermum niloticum Kotschy) (Henry et al., 1983). The 2 subspecies are simply named V. paradoxa and V. nilotica respectively. Subspecies paradoxa has a less dense and shorter indumentum and slightly smaller flowers than those of nilotica (Nikiema and Umali, 2007). It occurs in West Africa whilst nilotica is found in East Africa with no University of Ghana http://ugspace.ug.edu.gh 9 overlap in their ranges. However, Hall et al. (1996) recognize no clear-cut distinction between the subspecies’ morphology and thus concluded that the difference is purely clinal. Further studies to clarify such differences would be useful for breeding. With the wide phenotypic diversity among shea tree populations, Diarrassouba et al. (2009, 2008) classified the species morphologically on the basis of shape of fruit and of tree canopies, which shows some amount of discrete variation. Diarrassouba et al. (2009) identified 5 morphotypes based on fruit shape which are fusiform, round, ovoid, reverse pear and ellipsoid fruits, while the morphotypes reported based on canopy shape are ball or spherical, broadly pyramidal, broom and oblong-shaped trees (Fig. 2.2). Similarly, Yidana (2004) described 4 major fruit types with differences that are consistent enough to serve as basis for varietal classification and development. These phenotypic differences suggest that it is possible to select for improved performance both in fruit production and in time to fruit production. 2.3. Botany of shea tree 2.3.1. Canopy and branching Vitellaria paradoxa is a small to medium-sized deciduous tree growing up to 7–25 m tall (Sanou and Lamien, 2011). Parkland and fallow trees are generally bigger than bushland trees (Fig. 2.2). The bole of the tree is short, usually 3–4 m long, up to 0.5– 1.5 m in diameter with the bark being blackish, greyish, rough, deeply fissured and splitting regularly into corky square or rectangular scales. It copiously produces white latex which coagulates on the corky barks when cut (Hall et al., 1996). University of Ghana http://ugspace.ug.edu.gh 10 Fig. 2.2. Vitellaria paradoxa tree in agroforestry parkland; Adapted from Diarrassouba et al. (2009) Boughs and twigs of V. paradoxa grow plagiotropically (Diarrassouba et al., 2009; Hall et al., 1996). Due to the plagiotropic branching, V. paradoxa trees produce epicormic shoots (Fig. 2.3) which readily sprout from disturbed main and secondary branches. Main and secondary branches produce a large number of twigs giving rise to canopy morphotypes of varied shapes and sizes which have been described as round to spindle-, umbrella- or broom-shaped (IPGRI, 2006) (see also Section 2.2). Morphotypes based on canopy shape play an important role in tree selection and management on parklands. For example, broom-shaped trees which usually overshade annual food crops face intensive, selective thinning in agroforestry parklands (Boffa, 2000). Also, trees with round canopies have been identified as higher yielding than those with erect or oblong shapes (Schreckenberg, 1996). University of Ghana http://ugspace.ug.edu.gh 11 1 2 3 Fig. 2.3. Bole of a Vitellaria paradoxa tree showing fissured corky bark (1), epicormic branch (2) and leaf (3); Adapted from Yidana (2004) 2.3.2. Roots Vitellaria paradoxa has a taproot system with the taproot growing up to 1.0 to 2.0 m long. It produces shallow lateral roots that are concentrated at a depth of 0.1 m extending up to 20.0 m outwards from the tree. Secondary lateral roots are also produced, which grow downwards to the same depth as the taproot. Due to the shallow root development, mature trees are easily toppled over by strong winds occurring in the rainy season (Moore, 2008). The shallow roots also contribute to early leaf abscission; a feature that is implicated by Soro et al. (2012) to promote precocity (early flowering and fruiting in mature trees). University of Ghana http://ugspace.ug.edu.gh 12 2.3.3. Leaves and flowers The leaves of V. paradoxa are simple and entire with craspedodromous venation and prominent marginal veins (Fig. 2.3). They are spirally arranged in dense clusters and at the tips of the branches (Nikiema and Umali, 2007). Leaf blade measures 10–25 cm long and 4–14 cm wide with cuneate to rounded or slightly cordate base and rounded to acute apex. Leaves are both stipulate and petiolate with petioles 3–10 cm long. The greenish or cream-yellowish, fragrant flowers develop between December and March. Leaves are shed and the inflorescence appears in clusters approximately 10–40 at the shoot apex in the axils of scale leaves (Maranz and Wiesman, 2003). Flowers are complete and allogamous (Okullo et al., 2004). Pollination is largely entomophilous and insect pollinators include bees, wasps and ants (Cardi et al., 2005). 2.3.4. Fruits The fruit of V. paradoxa is a berry containing 1–5 seeds (Fig. 2.4). Single-seeded fruits are the commonest and among the multiple-seeded fruits, double-seeded ones are those most frequently produced (Diarrassouba et al., 2009). The oval-shaped fruit is about 3–8 cm long and 2–4 cm wide and weighs between 10 and 60 g. Both fruit shape and size differ greatly among trees (Nyarko et al., 2012: Yidana, 2004). The fruit is initially green and pubescent, but turns yellowish green and smells tartly strong on maturity (Nikiema and Umali, 2007). The edible pulp of the fruit comprises the epicarp and the mesocarp (Fig. 2.4C). The thicker mesocarp overlying the hilum is lined with a fibrous and bitter-tasting funiculus which is scarcely consumed. Thus, the frugivorous dispersers, mainly flying foxes (Chiroptera: Pteropodidae), pick ripe fruits from the trees, carry them in their mouths to their roosts where they eat the entire pulp (Djossa et al., 2008) and then disperse the seed with the funiculus intact. University of Ghana http://ugspace.ug.edu.gh 13 2 1 h A B C Fig. 2.4. Fruits of Vitellaria paradoxa; A, Fusiform-shaped shea fruit; B, Multiple- seeded shea fruit; C, Single-seeded shea fruit showing the edible pulp (1), seed (2) with its hilum (h); Adapted from Diarrassouba et al. (2009) 2.3.5. Seed anatomy and morphology The seed of V. paradoxa has a fragile testa that encloses an oleaginous kernel (Orwa et al., 2009). The fat content of the shea seed on average is 50 % (Shea Matters, 2011). The seed or the kernel is often incorrectly called a nut (Nikiema and Umali, 2007). The incorrect naming arises from culinary sense in which any oily kernel found within a shell is termed a nut. The seed is globose or broadly ellipsoid, 3–5 cm long and 2.0–3.5 cm wide, and weighs between 5–16 g. The coat of the seed, commonly called shell, is shiny and has a conspicuous hilum, large and pale, that covers nearly one side (Jøker, 2000) (Fig. 2.4C). Colour of the shell of mature seeds is homogeneous within a given tree and may be dark brown, clear brown, greyish brown or blackish brown (Diarrassouba et al., 2008). The fresh kernel comprises 2 thick, fleshy cotyledons and an unexserted radicle (Nikiema and Umali, 2007). The seeds of sapotaceous trees may be put into 2 main groups based on their morphology and the amount of latex they contain. On the basis of morphologies, the seeds are either laterally compressed or broadly ellipsoid to globoid. Examples of the laterally compressed or flattened seeds are those produced by Chrysophyllum spp. and University of Ghana http://ugspace.ug.edu.gh 14 Micropholis guyanensis (Jessup and Short, 2011). The broadly ellipsoid to ovoid seeds are produced by V. paradoxa, C. giganteum and Tieghemella heckelii. The laterally compressed seeds, which usually contain little amount of latex, have appressed cotyledons and therefore exhibit schizocotyly. In contrast, the broadly globoid seeds which are particularly laticiferous have fused cotyledons (Watson and Dallwitz, 2012). The seeds of V. paradoxa have fused cotyledons and also produce latex copiously (Nikiema and Umali, 2007). Schmidt (2000) reported that seeds are asymmetrically shaped to ensure that they scarcely fall and lie vertically (radicle end up) during dispersal because this orientation reduces the rate of seedling emergence. The morphology of sapotaceous seeds clearly suggests the orientations in which they are more likely to fall when they are naturally dispersed. The laterally compressed seeds usually fall and lie with the hilum laterally exposed whilst the ellipsoid to globoid ones usually fall and lie on the hilum. The earliest description of the germination of Vitellaria by Jackson (1968) mentions the hilum as the flatter side on which seeds usually fall and lie to germinate. 2.4. Germination of Vitellaria paradoxa seeds Germination commences with the uptake of water by a seed and terminates when the radicle appears, or becomes visible. Subsequent events, including the mobilization of the major storage reserves, are associated with growth of the seedling. Seedlings become established when they exhaust the seed reserves. Therefore, germination, seedling growth and establishment are distinct phases with establishment marking the stage when a seedling dependent on seed reserves is transformed into a fully autotrophic plant (Bewley and Black, 1994). University of Ghana http://ugspace.ug.edu.gh 15 Germination of V. paradoxa seed is described as cryptogeal (Fig. 2.5), a terminology outside those associated with the traditional scheme of germination and seedling types based on cotyledonar traits. Cryptogeal germination is the germination in which the plumule is initially pushed into the soil where it develops into a shoot which then emerges above the soil (Burrows and Stockey, 1994). Jackson (1968) observed that germination of a V. paradoxa seed involves the cracking of the testa at the broader end, followed by the appearance of a pseudoradicle which pushes the plumule and true radicle into the soil. In the soil, the pseudoradicle forms a bulge or swelling about 5–7 cm along its length and above the swelling, a pink-coloured shoot with scale leaves appears and grows upwards. Below the swelling is the true radicle which continuous growing downwards, becoming severalfold longer than the shoot (Ugese et al., 2010a). The shoot also continues its upward growth and ultimately emerges above the soil in about 2–3 months after sowing. Fig. 2.5. Cryptogeal seedling of Vitellaria paradoxa showing 1, pseudoradicle; 2, bulge; 3, true root; 4, shoot; Adapted from Yidana (2004) University of Ghana http://ugspace.ug.edu.gh 16 Vitellaria paradoxa seedlings delay to emerge above the soil even though germination of the seeds occurs within a week after sowing (Ugese et al., 2010a). The long period of seedling growth through the soil is what has been incorrectly described as dormancy. Consequently, much of the research work on germination of V. paradoxa seeds has been focused on breaking dormancy or achieving quicker emergence instead of identifying the processes involved in seedling development. The seed of V. paradoxa is recalcitrant (Sanou and Lamien, 2011; Pritchard et al., 2004) and germinates rapidly after shedding to avoid desiccation-related mortality (Orwa et al., 2009); thus, it has no dormancy period. Despite being non-dormant, V. paradoxa seeds germinate and emerge non-uniformly. According to Ugese et al. (2010a), difference in seed provenances is one of the causes of variability in seedling emergence and growth. Sowing depth and seed size have also been proved to affect Vitellaria seed germination, emergence and growth of the resulting seedlings (Ugese et al., 2010b; Ugese et al., 2009). Although seedcoat has been implicated as the major cause of morphological dormancy in testaceous seeds (Msanga, 1998), Ugese et al. (2005) observed non-significant effect of the shell on the germination of V. paradoxa seeds. Accordingly, the causes of the long period of below-ground seedling growth still remain precisely unknown (Ugese et al., 2005). Nonetheless, identification of the factors responsible for this slow growth may help explain how dryland habitats are populated by recalcitrant species. Desiccation-sensitive seeds frequently occur in aseasonal tropical forests (Tweddle et al., 2003). It thus remained to be explained why V. paradoxa that characterizes the woody flora of arid and semiarid savannah landscapes of SSA evolved desiccation- University of Ghana http://ugspace.ug.edu.gh 17 sensitivity. This recalcitrance to storage may be explained by anatomical and morphological studies of the seed which are thus far lacking. Recalcitrant species do occur naturally in tropical drylands (Tweddle et al., 2003), yet little is known about their regeneration strategies (Pritchard et al., 2004). Some of the ecological adaptations of dryland species to desiccation-intolerance may be elucidated by detailed study of the germination of V. paradoxa seeds. Farmers hardly plant V. paradoxa; therefore, very little information exists on its germination and on the morphology of the seedlings (Sanou and Lamien, 2011; Ugese et al., 2010a). This scanty information is also highly conflicting. For example, Jøker (2000) reported that V. paradoxa seedlings emerge as late as 150 days after sowing whilst Yidana (2004) recorded seedling emergence in just 28 days after sowing. However, detailed information about germination and seedling morphology is crucial in determining why the shoot delays to emerge above the soil and in developing appropriate techniques for achieving quicker emergence (Ugese et al., 2005). Vitellaria paradoxa seedlings are slow growing (Asante et al., 2012). The trade-off between survival and rapid growth usually favours survival in which the seedlings preferentially allocate more growth resources to develop long taproots and large root crowns (Jackson, 1974). According to Dillenburg et al. (2010), this growth pattern enables seedlings to persist in the soil seedling bank from year to year. Morris and Doak (1998) reported that species persisting in agro-ecological zones with high interannual climate variation possess high longevity. Thus, on attainment of maturity, longevity up to 200 or 300 years has been reported for Vitellaria trees (Jøker, 2000). University of Ghana http://ugspace.ug.edu.gh 18 2.5. Socio-economic importance and uses of Vitellaria paradoxa 2.5.1. Domestic uses of Vitellaria paradoxa The commercialization of shea products represents a perpetual source of income at different parts of the community chain, beginning with rural children and women who gather and process nuts to town dwellers as well as entire countries (Shea Matters, 2011). The shea fruit is consumed by people of all ages whilst the nut and butter (Fig. 2.6) are both export commodities. The pulp is also used to make beverages and jam, which are much appreciated in Mali and Burkina Faso (Sanou and Lamien, 2011). A B Fig. 2.6. Sheanuts (A) and butter (B); Adapted from Moore (2008) In the Sudano-Sahelian savanna, shea butter is the major cooking oil being the most important source of fatty acids and glycerol in the diet (Hall et al., 1996). The butter is an unguent with anti-microbial properties and is widely used to prepare herbal ointments. Accordingly, it is used as an anodyne to treat sprains and relieve pains, and to heal wounds quickly. As a cosmetic, it is used as a moisturizer to protect the skin against the windy and sunny weather especially during harmattan, which is usually University of Ghana http://ugspace.ug.edu.gh 19 severe in Northern Ghana. The healing properties of shea butter are partly attributable to the presence of allantoin, a substance known to stimulate the growth of healthy tissue in ulcerous wounds (Wallace-Bruce, 1995). Furthermore, every other part of the Vitellaria tree has several medicinal uses depending on the locality. For example, leaves are used to treat stomach ache in children. Extract of stem bark possesses broad spectrum antibacterial activity against clinical isolates of some gram positive pathogenic bacteria and thus demonstrates its potential to provide lead molecules for the production of novel antibiotics (Ayankunle et al., 2012). Roots and root bark are ground to a paste and taken orally to cure jaundice and are also used to treat diarrhoea and stomach ache. Latex tapped from the bole is heated and mixed with palm oil to make glue which is used as a domestic adhesive (Hall et al., 1996). It is chewed as a gum especially by children and thus has the potential to be used industrially for making chewing gum just as chicle, latex obtained from sapotaceous species such as Manilkara zapota, M. chicle, M. staminodella and M. bidentata (Mathews, 2009). 2.5.2. International and industrial uses of sheanut and butter Approximately 95 % of sheanuts provide an important raw material for cocoa butter equivalents (CBEs) that are used in the confectionery industries to manufacture chocolate (Masters et al., 2004). Shea butter is used as a CBE because its melting point (32–45 °C) is similar to that of cocoa butter. It has high amounts of di-stearin (30 %) and some stearo-palmitine (6.5 %) which make it blend homogeneously with cocoa butter without altering its flow properties (Sanou and Lamien, 2011). University of Ghana http://ugspace.ug.edu.gh 20 The butter also has numerous uses in the cosmetic industry. Cosmetically, the high proportion of unsaponifiable matter, consisting of 60–70 % triterpene alcohols, gives shea butter creams good penetrative, moisturizing, regenerative and anti-wrinkle properties. These properties enable the butter or creams containing it to protect the body from ultraviolet (UV) radiation. Having properties similar to those of sebum, it is used to produce lipsticks, soaps and other skincare products (Shea Matters, 2011). The melting point of shea butter which is close to average body temperature of a healthy person (37 ℃) primarily makes it a suitable base for ointments and medicines (Hall et al., 1996). Thus, the butter is used in pharmaceutical industries and in herbal medicines. In Ghana, it is the most widely used butter for making herbal ointments and balms. Clinical tests with patients suffering from rhinitis and having moderate to severe nasal congestion showed that shea butter may relieve nasal congestion better than conventional nasal drops (Sanou and Lamien, 2011). 2.6. Domestication of Vitellaria paradoxa 2.6.1. Domestication status of the species Despite its importance and potentials, V. paradoxa still remains wild and is considered threatened by the World Conservation Union (Byakagaba et al., 2011). The major undesirable trait of the species in its wild form is probably the long gestation period, which according to Yeboah et al. (2011) and Yidana (2004) discourages farmers from domesticating it. Domestication of crops was not accomplished merely by gathering and even the most intensive harvesting of cereals never applied sufficient selection pressure to domesticate them fully. In contrast, deliberate planting exerts a strong selection pressure that fixes desirable alleles (Zohary, 2004). Lovett and Haq (2000b) described Vitellaria as a semidomesticate, University of Ghana http://ugspace.ug.edu.gh 21 but Shea Matters (2011) considers it wild because a tradition of deliberately planting the trees scarcely exists. In addition to the long gestation period, low and unstable prices of shea products have also been strongly opined to militate against domesticating the plant. Sheanuts have low prices which fluctuate widely both within and between seasons (Shea Matters, 2011). Thus, Vitellaria trees are lately being felled ruthlessly for short-term economic gains (Osei Agyeman et al., 2012; Masters, 2002) with a lot of nuts especially of trees far afield remaining uncollected (Shea Matters, 2011). Because farmer-domestication of plants is market-driven, remunerative and stable-pricing policy may tremendously accelerate domestication of the species. 2.6.2. Selection in agroforestry parklands Agroforestry parkland is a mixture of naturally established trees and shrubs that farmers select for certain functions and cultivate together with staple food crops such as millet (Paul, 2012; Boffa, 2000). It is the principal agricultural system used by subsistence farmers in the Sudano-Sahelian savannas where they select for Vitellaria trees that compete minimally with their annual crops and for those whose fruits are both large and sweet. This anthropic selection made Vitellaria the dominant woody species in most West African savanna parklands (Hall et al., 1996) with its genetic make-up reflecting generations of unconscious selection (Lovett and Haq, 2000a). 2.6.3. Breeding for earliness The early 1970s saw Vitellaria vegetable fat be announced as a CBE followed by a marked increase in interest from the pharmaceutical and cosmetics industries (Masters University of Ghana http://ugspace.ug.edu.gh 22 et al., 2004). Consequently, a substation was established by Cocoa Research Institute of Ghana (CRIG) at Bole in 1976 to research into the botany and ecology of shea tree and to develop early bearing varieties. The institute has thus far developed improved agroforestry practices such as pruning mistletoe-infested trees, and identification and selective thinning of unproductive trees and has been disseminating them to farmers. However, no early bearing varieties have been developed (Yeboah et al., 2011). Some of the methods employed to propagate Vitellaria vegetatively are grafting, stem and root cutting, and layering with grafting being the most promising method (Yidana, 2004). Using stem cuttings with retained petioles and Seradix 3 powder (rooting hormone), Yeboah et al. (2011) readily induced faster rooting when the cuttings were set in a propagating bin. However, the major limitations of these methods are the low success (20–25 %) and reproducibility rates caused mainly by the latex sap (Masters, 2002). More importantly, the few successfully grafted and rooted cuttings grow slowly. For example, Sanou et al. (2004) recorded an annual growth rate of 12.6 cm for grafted plants. Thus, the seed seems to be the most reliable propagule, albeit non-uniform germination and rapid loss of viability. 2.6.4. In vitro propagation of Vitellaria paradoxa As a result of the difficulties in conventionally propagating V. paradoxa, several researchers have attempted propagating the species in vitro. For instance, immature cotyledon explants were cultured on Murashige and Skoog (1962) basal medium amended with 2,4-Dichlorophenoxyacetic acid (2,4-D) at 0.01–1.0 mg/l in darkness. Maximum somatic embryo induction was obtained at 0.1 mg/l 2,4-D. However, the resulting plantlets were lost during weaning (CRIG, 2012). University of Ghana http://ugspace.ug.edu.gh 23 Using leaf disc explants of V. paradoxa, Fotso et al. (2008) also successfully induced callus on a half-strength MS basal medium supplemented with 0.6 % agar, 4.5 % sucrose, 0.5 mg/l BAP and 0.5–5.0 mg/l 2,4-D. At 28 days after culture, a BAP/2,4-D combination of 0.5/3.0 mg/l respectively yielded 87.3 % callogenesis. Further, a BAP/2,4-D ratio of 0.5/2.0 mg/l produced 62.1 % embryonic calli with an average of 27 embryos per callus in 97 days after culture. Again, the resulting bipolar embryos were never successfully transformed into viable plantlets. Consequently, the exact causes of the loss of somatic embryos or the regenerated plantlets remain unknown. Work done on miracle fruit (S. dulcificum), an African Sapotaceae, by Ogunsola and Ilori (2008) involved excising the micropylar end of the seeds containing the embryos. The embryos were successfully regenerated on MS medium supplemented with 0.1 mg/l Naphthaleneacetic acid (NAA) and 0.2 mg/l 6-Benylaminopurine (BAP). It may thus be possible to identify, locate and excise the embryo of V. paradoxa which is also a sapotaceous species for in vitro culture as well. Sapotaceous species are particularly laticiferous (Watson and Dallwitz, 2012). Bhore and Preveena (2011) attributed a 100 % contamination of nodal sector and leaf disc explants in Mimusops elengi (an Asian Sapotaceae) to the latex produced by the species. Consequently, the high amount of latex sap and saponins in V. paradoxa could also be contributing factors to its recalcitrance to in vitro propagation. Despite these challenges, in vitro propagation with its potential of regenerating plantlets all- year round using explants from any part of the plant still offers prospects of producing uniform plantlets to domesticate V. paradoxa fully. University of Ghana http://ugspace.ug.edu.gh 24 REFERENCES Allal, F. Sanou, H., Millet, L., Vaillant, A., Camus-Kulandaivelu, L., Logossa, Z. A., Lefe`vre, F. and Bouvet, J.-M. (2011). Past climate changes explain the phylogeography of Vitellaria paradoxa over Africa. Heredity, 107:174–186. Ande, A. T. (2004). The pupal habits of Cirina forda, a leading food insect in Kwara State, Nigeria. Journal of Sustainable Tropical Agricultural Research, 9:97– 100. Asante, W. J., Banidiyia, M. A. and Tom-Dery, D. (2012). Effect of planting depth on the germination and initial growth and development of shea (Vitellaria paradoxa C.F. Gaertn.). International Journal of Biosciences, 2(12):146–152. Ayankunle, A. A., Kolawole, O. T., Adesokan, A. A. and Akiibinu, M. O. (2012). Antibacterial activity and subchronic toxicity studies of Vitellaria paradoxa stem bark extract. Journal of Pharmacology and Toxicology, 7(6):2986–304. Bewley, J. D. and Black, M. (1994). Seeds: Physiology of development and germination. 2nd ed. New York: Plenum Press. 445 pp. Bhore, S. J. and Preveena, J. (2011). Micropropagation of Mimusops elengi Linn.: Identification of suitable explant and comparative analysis of immature zygotic embryos response on three basal media. American-Eurasian Journal of Agricultural and Environmental Sciences, 10(2):216–222. Boffa, J. M. (2000). West African agroforestry parklands: keys to conservation and sustainable management. Unasylva, 200(51):11–17. Bonkoungou, E. G. (2002). The Shea tree (Vitellaria paradoxa) and the African shea parklands. Paper presented at the international workshop on processing and marketing of shea products in Africa. 4–6 March, 2002. pp. 51–59. Dakar, Senegal. Burrows, G. E. and Stockey, R. A. (1994). The developmental anatomy of cryptogeal germination in bunya pine (Araucaria bidwillii). International Journal of Plant Sciences, 155(5):519–537. Byakagaba, P., Eilu, G., Okullo, J. B. L., Tumwebaze, S. B. and Mwavu, E. N. (2011). Population structure and regeneration status of Vitellaria paradoxa C.F.Gaertn. under different land management regimes in Uganda. Agricultural Journal, 6(1):14–22. Cardi, C. A., Vaillant, A., Sanou, H., Kelly, B. A. and Bouvet, J.-M. (2005). Characterization of microsatellite markers in the shea tree (Vitellaria paradoxa C.F.Gaertn.) in Mali. Molecular Ecology Notes, 5:524–526. CRIG (2012). Shea/cashew development. Progress report for 2011/12 and workplan for 2012/13. May, 2012, pp. 318–373. Tafo, Ghana. University of Ghana http://ugspace.ug.edu.gh 25 Dianda, M., Bayala, J., Diop, T. and Ouédraogo, S. J. (2009). Improving growth of shea butter tree (Vitellaria paradoxa C.F.Gaertn.) seedlings using mineral N, P and arbuscular mycorrhizal (AM) fungi. Biotechnology, Agronomy, Society and Environment, 13(1):93–102. Diarrassouba, N., Bup, N. D., Kapseu, C., Kouame, C. and et Sangaré, A. (2008). Phenotypic diversity of shea (Vitellaria paradoxa C.F.Gaertn.) populations across four agro-ecological zones of Cameroon. Journal of Crop Science and Biotechnology, 10(4):211–218. Diarrassouba, N., Fofana, I. J., Issali, A. E., Bup, N. D. and et Sangaré, A. (2009). Typology of shea trees (Vitellaria paradoxa) using qualitative morphological traits in Côte d’Ivoire. Gene Conserve, 8(33):752–780. Dillenburg, L. R., Rosa, L. M. G. and Mósena, M. (2010). Hypocotyl of seedlings of the large-seeded species Araucaria angustifolia: an important underground sink of the seed reserves. Trees, 24:705–711. Elias, M. and Carney, J. (2007). African shea butter: A feminized subsidy from nature. Africa, 77:37–62. Djossa, B. A., Fahr, J., Kalko, E. K. V. and Sinsin, B. A. (2008). Fruit selection and effects of seed handling by flying foxes on germination rates of shea trees, a key resource in northern Benin, West Africa. Ecotropica, 14:37–48. Fobil, J. N. (2007). Bole, Ghana: Research and development of the shea tree and its products. New Haven. CT: HORIZON Solutions International. Available at: http://www.solutions-site.org/node/110. Accessed on 31st December 2012. Fotso, S., Néhémie, D. T. and Denis, O. N. (2008). Comparison of first stages of somatic embryogenesis in Baillonella toxisperma and Vitellaria paradoxa. Biotechnology, Agronomy, Society and Environment, 12(2):131 (abs). Hall, J. B., Aebischer, D. P., Tomlinson, H. F., Amaning, E. O. and Hindle, J. R. (1996). Vitellaria paradoxa, a monograph, Publication Number 8, School of Agricultural and Forest Sciences, University of Wales, Bangor, UK. 105 pp. Hatskevich, A., Jeníček, V. and Antwi, D. S. (2011). Shea industry - a means of poverty reduction in Northern Ghana. Agricultura Tropica et Subtropica, 44(4):223–228. Hawthorne, W. D. and Jongkind, C. C. H. (2008). Woody plants of Western African forests: A guide to the forest trees, shrubs and lianes from Senegal to Ghana. Royal Botanic Gardens, Kew. 1040 pp. Henry, A. N., Chithra, V. and Nair, N. C. (1983). Vitellaria vs. Butryospermum (Sapotaceae). Taxon, 32:282–286. University of Ghana http://ugspace.ug.edu.gh 26 IPGRI, INIA (2006). Descriptors for Shea tree (Vitellaria paradoxa). International Plant Genetic Resources Institute, Rome, Italy; Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria, Madrid, Spain. 63 pp. Jackson, G. (1968). Notes on West African Vegetation – III: the seedling morphology of Butyrospermum paradoxum. Journal of West African Science Association, 13:215–222. Jackson, G. (1974). Cryptogeal germination and other seedling adaptations to the burning of vegetation in the savanna regions: the origin of the pyrophytic habit. New Phytologist, 73:771–780. Jessup, L. W. and Short, P. S. (2011). Sapotaceae. In: Flora of the Darwin Region (Short, P. S. and Cowie, I. D. eds). Volume 1, pp. 1–6. Northern Territory Herbarium, Department of Natural Resources, Environment, the Arts and Sport. Palmerston, Australia. Jøker, D. (2000). Vitellaria paradoxa Gaertn. F. Danida Forest Seed Centre, Seed Leaflet No.50, December, 2000. Lovett, P. N. and Haq, N. (2000a). Evidence for anthropic selection of the sheanut tree (Vitellaria paradoxa). Agroforestry Systems, 48:273–278. Lovett, P. N. and Haq, N. (2000b). Diversity of the sheanut tree (Vitellaria paradoxa C.F.Gaertn.) in Ghana. Genetic Resources and Crop Evolution, 47:293–304. Maranz, S. and Wiesman, Z. (2003). Evidence for indigenous selection and distribution of the shea tree, Vitellaria paradoxa, and its potential significance to prevailing parkland savanna tree patterns in sub-Saharan Africa north of the equator. Journal of Biogeography, 30:1505–1516. Maranz, S., Kpikpi, W., Wiesman, Z., De Saint Sauveur, A. and Chapagain, B. (2004). Nutritional values and indigenous preferences for Shea fruits (Vitellaria paradoxa C.F.Gaertn.) in African agroforestry parklands. Economic Botany, 58:588–600. Masters, E. T. (2002). The Shea Resource: Overview of research and development across Africa. Paper presented at the international workshop on processing and marketing of shea products in Africa, 4–6 March, 2002. pp. 13–29. Dakar, Senegal. Masters, E. T., Yidana, J. A. and Lovett, P. N. (2004). Reinforcing sound management through trade: shea tree products in Africa. Unasylva, 219(55):46–52. Mathews, J. P. (2009). Chicle: The chewing gum of the Americas, from the Ancient Maya to William Wrigley. Tucson: University of Arizona Press. pp. 19–21. University of Ghana http://ugspace.ug.edu.gh 27 McNeill, J. and Turland, N. J. (2011). Synopsis of proposals on botanical nomenclature – Melbourne 2011: A review of the proposals concerning the International Code of Botanical Nomenclature submitted to the XVIII International Botanical Congress. Taxon, 60 (1):243–286. Morris, W. F. and Doak, D. F. (1998). Life history of a long-lived gynodioecious plant, Silene acaulis (Caryophyllaceae), inferred from size based population projection matrices. American Journal of Botany, 85:784–793. Moore, S. (2008). The role of Vitellaria paradoxa in poverty reduction and food security in the Upper East Region of Ghana. Earth & E-nvironment, 3:209– 245. Msanga, H. P. (1998). Dormancy and germination. In: Tropical tree seed manual (Vozzo, J. A. ed.). pp. 149–176. United States Department of Agriculture, Forest Service. Murashige, T. and Skoog, F. (1962). A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiologia Plantarum, 15(3):473–497. Nikiema, A. and Umali, B. E. (2007). Vitellaria paradoxa C.F.Gaertn. http://database.prota.org/PROTAhtml/Vitellaria paradoxa_En.htm. Accessed on 22nd May 2013. Nyarko, G., Mahunu, G. K., Chimsah, F. A., Yidana, J. A., Abubakari, A-H., Abagale, F. K., Quainoo, A. and Poudyal, M. (2012). Leaf and fruit characteristics of Shea (Vitellaria paradoxa) in Northern Ghana. Research in Plant Biology, 2(3):38–45. Ogunsola, K. E. and Ilori, C. O. (2008). In vitro propagation of miracle berry (Synsepalum dulcificum Daniel) through embryo and nodal cultures. African Journal of Biotechnology, 7(3):244–248. Okao, M., Malinga, M., Okia, C. A. and Okullo, J. B. L. (2012). Vegetative propagation of Vitellaria paradoxa by stem cuttings: Effects of rooting substrate and planting technique. Paper presented at the third RUFORUM biennial meeting. 24–28 September 2012. pp. 289–293. Entebbe, Uganda. Okullo, J. B. L., Hall, J. B. and Obua, J. (2004). Leafing, flowering and fruiting of Vitellaria paradoxa subsp. nilotica in savanna parklands in Uganda. Agroforestry Systems, 60:77–91. Orwa, C., Mutua, A., Kindt, R., Jamnadass, R. and Simons, A. (2009). Agroforestree Database: a tree reference and selection guide version 4.0 http://www.worldagroforestry.org/af/treedb/. Accessed on 23rd November 2012. Osei Agyeman, K., Amposah, O., Imoro, B. and Lurumuah, S. (2012). Commercial charcoal production and sustainable community development of the Upper West Region, Ghana. Journal of Sustainable Development, 5(4):149–164. University of Ghana http://ugspace.ug.edu.gh 28 Paul, O., Jacob, G. A., Clemen, A. O. and John, B. L. O. (2012). On-farm management of Vitellaria paradoxa C.F.Gaertn. in Amuria District, Eastern Uganda. http://www.hindawi.com/journals/ijfr/2012/768946/. Accessed on 12th January 2013. Pritchard, H. W., Daws, M. I., Fletcher, B. J., Christiane, S., Gaméné, C. S., Msanga, H. P. and Omondi, W. (2004). Ecological correlates of seed desiccation tolerance in tropical African dryland trees1. American Journal of Botany, 91(6):863–870. Quainoo, A. K., Nyarko, G., Davrieux, F., Piombo, G., Bouvet, J.-M., Yidana, J. A., Abubakari, A. H., Mahunu, G. K., Abagale, F. K. and Chimsah, F. A. (2012). Determination of biochemical composition of shea (Vitellaria paradoxa) nut using near infrared spectroscopy (nirs) and gas chromatography. International Journal of Biology, Pharmacy and Allied Sciences, 1(2):84–98. Sanou, H., Kambou, S., Teklehaimanot, Z., Demb´el´e, M., Yossi, H., Sina, S., Djingdia, L. and Bouvet, J.-M. (2004). Vegetative propagation of Vitellaria paradoxa by grafting. Agroforestry Systems, 60:93–99. Sanou, H. and Lamien, N. (2011). Vitellaria paradoxa, shea butter tree. Conservation and sustainable use of genetic resources of priority food tree species in sub- Saharan Africa. Bioversity International (Rome, Italy). Available at www.bioversityinternational.org. Accessed on 2nd January 2013. Schmidt, L. (2000). Germination and seedling establishment. Available at http://curis.ku.dk/portal-life/files/20712899/Chapter10. Accessed on 12th May 2012. 24 pp. Schreckenberg, K. (1996). Forests, fields and markets: A study of indigenous tree products in the woody savannas of the Bassila Region, Benin. Ph.D. Thesis, University of London. 326 pp. Shea Matters (2011). Life in the Shea Belt. Publication of IOI Loders Croklaan, October, 2011. 36 pp. Available at: http://europe.croklaan.com. Accessed on 21st May 2012. Soro, D., Traore, K. and Ouattara, D. (2012). Precocity and lateness of shea tree fruit production. Journal of Applied Biosciences, 52:3676–3684. Tropical Advisory Service (2002). Butryospermum parkii, Family Sapotaceae. Leaflet: HYDRA No. TTS9. Ryton Organic Gardens, Coventry, UK. 2 pp. Tweddle, J. C., Dickie, J. B., Baskin, C. C. and Baskin, J. M. (2003). Ecological aspects of seed desiccation sensitivity. Journal of Ecology, 91:294–304. Ugese, F. D., Ojo, A. A. and Bello, L. L. (2005). Effect of presowing treatment and nut orientation on emergence and seedling growth of seeds of shea butter tree (Vitellaria paradoxa). Nigerian Journal of Botany, 18:294–304. University of Ghana http://ugspace.ug.edu.gh 29 Ugese, F. D., Baiyeri, K. P. and Mbah, B. N. (2008a). Nutritional composition of shea (Vitellaria paradoxa) fruit pulp across its major distribution zones in Nigeria. Fruits, 63:163–169. Ugese, F. D., Baiyeri, K. P. and Mbah, B. N. (2008b). Mineral content of the pulp of shea butter fruit (Vitellaria paradoxa C.F.Gaertn.) sourced from seven locations in the savanna ecology of Nigeria. Tree and Forestry Science and Biotechnology, 2(1):40–42. Ugese, F. D., Baiyeri, K. P. and Mbah, B. N. (2009). Intra- and inter-correlative responses among fruits physical traits, seedling growth parameters and fruit and nut proximate qualities of the Nigerian shea nut tree (Vitellaria paradoxa C.F.Gaertn.). Journal of Tropical Agriculture, Food, Environment and Extension, 8(2):110–115. Ugese, F. D., Baiyeri, K. P. and Mbah, B. N. (2010a). Determination of growth stages and seedling structures associated with slow emergence of shea butter tree (Vitellaria paradoxa C.F.Gaertn.) seedlings. Journal of Animal and Plant Sciences, 8(2):993–998. Ugese, F. D., Baiyeri, K. P. and Mbah, B. N. (2010b). Effect of sowing depth and mulch application on emergence and growth of shea butter tree seedlings (Vitellaria paradoxa C.F.Gaertn.). African Journal Biotechnology, 9(10):1443–1449. Wallace-Bruce, Y. (1995). Shea butter extraction in Ghana. In: Do It Herself: Women and technical innovation (Appleton, H. ed.). pp. 157–161. Intermediate Technology Publications, London. Watson, L. and Dallwitz, M. J. (2012). The families of flowering plants: Descriptions, illustrations, identification and information retrieval. http://delta-intkey.com. Accessed on 19th May 2012. Yeboah, J., Lowor, S. T. and Amoah, F. M. (2009). The rooting performance of shea (Vitellaria paradoxa Gaertn.) stem cuttings as influenced by wood type, sucrose and rooting hormone. Scientific Research and Essays, 4(5):521–525. Yeboah, J., Lowor, S. T., Amoah, F. M. and Owusu-Ansah, F. (2011). Propagating structures and some factors that affect the rooting performance of Shea (Vitellaria paradoxa C.F.Gaertn.) stem cuttings. Agriculture and Biology Journal of North America, 2(2):258–269. Yidana, J. A. (2004). Progress in developing technologies to domesticate the cultivation of shea tree (Vitellaria paradoxa) in Ghana. Agricultural and Food Science Journal of Ghana, 3:249–267. Zohary, D. (2004). Unconscious selection and the evolution of domesticated plants. Economic Botany, 58(1):5–10. University of Ghana http://ugspace.ug.edu.gh 30 CHAPTER 3 3.0. Anatomical and morphological studies on Vitellaria paradoxa seed 3.1. Introduction The anatomical and morphological features of seeds influence their germination and seedling development (Flores, 2002). A typical dicotyledonous seed comprises seedcoat, embryo and food reserves stored in either the cotyledons or the endosperms. The exact locations and nature of these features vary from plant to plant, but they evolve to ensure the survival of the seed. A seed is enclosed in a fruit which protects it and also guarantees efficient dispersal of the diaspore (Leubner, 2012). The term diaspore refers to the dispersal unit, which in some plants is only the seed, whereas in others such as V. paradoxa, it is the fruit containing seed. The seed usually has one embryo which occasionally divides into 2 or several giving rise to zygotic polyembryony. Polyembryony may also arise from the nucellus of the embryo sac in which case it is termed adventitious polyembryony (Filovona et al., 2002). Both types of polyembryony produce 2 or more seedlings from 1 seed as opposed to monoembryony which results in a single seedling. Another important feature of dicotyledonous seeds is cotyledon morphology based on which seeds are described as either schizocotylous or syncotylous. It determines the type of germination (Corby, 2008) and the optimal environmental conditions that the resulting seedlings require to develop properly (Kitajima and Fenner, 2000). Schizocotylous seeds, which have appressed cotyledons, germinate epigeally and do not tolerate shady conditions. Due to the presence of little food reserves in their cotyledons, the resulting seedlings develop photosynthetic ability rapidly, which explains their preference for well-lighted habitats. Schizocotylous seeds include the University of Ghana http://ugspace.ug.edu.gh 31 majority of dicotyledonous seeds that store their food reserves in the endosperm. Endospermic seeds have thin and papery cotyledons (Kitajima and Fenner, 2000). In contrast to schizocotylous seeds, syncotylous seeds have fused cotyledons and germinate crypto-hypogeally because the fusing of the cotyledons impedes their emergence from the seedcoat (Flores, 2002). These seeds usually store their food reserves in the cotyledons which are termed reserve cotyledons (Maia et al., 2005). The reserve cotyledons of V. paradoxa are fused to each other and remain so throughout germination and seedling establishment (Sanou and Lamien (2011). Also, all through these early growth and developmental stages, the thick cotyledons of V. paradoxa seedlings remain enclosed in the seedcoat until they are shed. The seedcoat may be thick especially in some orthodox seeds, but it is generally thin and fragile in most recalcitrant seeds. Thicker seedcoats impose exogenous dormancy on the embryo whilst thinner ones scarcely do so (Pritchard et al., 2004). A thin coat (shell) surrounds a V. paradoxa seed similar to those reportedly found in other sapotaceous species (Roosmalen and Garcia, 2000). The Sapotaceae contain species which display both schizocotyly and syncotyly. For example, African star apple (Chrysophyllum albidum) has schizocotylous seeds and consequently germinates epigeally (Ehiagbonare et al., 2008). Seeds germinate when they absorb moisture from the surrounding medium through their coats. The seedcoat has a micropyle through which the seed imbibes water and the radicle protrudes during germination (Gama-Arachchige et al., 2011). However, the radicles of some seeds protrude the seedcoat from a much larger opening called operculum University of Ghana http://ugspace.ug.edu.gh 32 (Pérez, 2009; Flores, 2002). The micropyle or operculum may be anatropous (located on the hilar side of seed) or synatropous (located on a different side). The locations of the micropyle or operculum and the hilum on seeds strongly influence how to orient the seeds in planting holes (Schmidt, 2000; Swaminathan et al., 1992). The seed of V. paradoxa is laticiferous and the latex and fat deposits surround the embryo making it difficult to be identified and may also explain its slow growth. This project was therefore aimed at studying the anatomical and morphological features of Vitellaria paradoxa seed. The specific objectives of the study were to i. identify the location of the embryo in the seed ii. examine the effects of the internal and external features of the seed responsible for the cryptogeal germination of the species. 3.2. Materials and methods 3.2.1. Shea fruit collection Mature fruits were collected from agroforestry parklands at Ga and Tanina in the Wa West District of the Upper West Region of Ghana (Fig. 3.1) and transported by road to the Biotechnology Centre of the Ghana Atomic Energy Commission (GAEC). Three hundred (300) fruits were depulped manually to obtain fresh seeds which were washed using tap water. During depulping, the number of seeds per fruit was recorded. Seeds were spread under a shade for 6 hours to dry after which their anatomical and morphological features were studied using magnifying lenses and a stereomicroscope (Leica ZOOM 2000, Cole-Parmer, Wetzlar, Germany). University of Ghana http://ugspace.ug.edu.gh 33 Fig. 3.1. Map of the Upper West Region of Ghana showing Tanina (T) and Ga (G) from where the shea fruits were collected 3.2.2. Studies on the morphology of Vitellaria paradoxa seed The dimensions of 50 seeds were measured using vernier callipers. All dimensions were measured linearly with the length taken form the proximal to the distal end whilst the breadth and thickness were measured at the broadest part of the seed. Shape of the seed was described according to the method of Diarrassouba et al. (2009) whilst colour of seedcoat was described by using HTML Colour Chart (http://www.html- color-names.com/color-chart.php). Any differences in shape and other features of seeds in single- and multiple-seeded fruits were observed. The location of the micropyle in relation to the hilum was also observed. Cotyledon morphology was studied by using 100 fresh seeds of which 50 seeds were deshelled and air-dried for 3–5 days whilst the remaining 50 were sown for 7–10 days to imbibe moisture for sprouting. Deshelling was done by using pliers to gently press the seed in the middle to rupture the seedcoat at the dorsal side. A knife was then used to remove the seedcoat (shell). Seeds were sown in polyethylene pots filled with a soil mix consisting of topsoil and well-decomposed sawdust in the ratio 5:1. The sprouted University of Ghana http://ugspace.ug.edu.gh 34 seeds were carefully uprooted and washed using tap water, and the thickness of their pseudoradicles were measured using vernier callipers after which they were trimmed off. Both the partially dry or sprouted seeds were then split open manually by pulling apart the cotyledons beginning from the distal end. Thickness of the seedcoat was measured by cutting 40 fresh seeds longitudinally into equal parts. The kernels in the sectioned parts were gently scooped out and the thickness of the shell at the dorsal and ventral or hilar sides was measured using vernier callipers. Photographs of the observed structures were taken using a 16.1-megapexil digital still camera (Sony Corporation, China) and then translated into cartographic drawing with Microsoft Coral Drawing (Version XIII) where necessary. 3.3. Anatomical studies on the seed and embryo identification with topographical tetrazolium test The location of the embryo was determined by using topographical tetrazolium (TTZ) test as described by Yu and Wang (1996). Six (6) fresh and 6 partially dry seeds (seeds air-dried for 72 hours) were deshelled and a quarter of their distal ends were transversely cut off. The remaining proximal portions of the seeds were washed and soaked in distilled water for 6 hours before immersion in 1.0 % tetrazolium chloride (TTC) solution in a Petri dish. The dish with its contents was wrapped tightly using parafilm and placed in a closed cabinet for 24 hours. Thereafter, the seeds were observed for staining of the embryos by the TTC solutions and photographs were taken using a Sony digital still camera (see Section 3.2.2). Twenty (20) fresh and 20 partially dry seeds were either transversely or longitudinally sectioned and the parts from where latex exuded were observed and photographed. University of Ghana http://ugspace.ug.edu.gh 35 3.4. Results 3.4.1. Morphology of Vitellaria paradoxa seed The number of seeds in a fruit varied from 1 to 5. The seed consists of 4 distinct sides termed proximal end, dorsal side, distal end and ventral or hilar side (Fig. 3.2). The distal end of the seed is comparatively smaller than the proximal end. Size of the seed is highly variable even among seeds from the same tree, with length ranging from 1.2 to 4.9 cm whilst the breadth and thickness were between 1.0–3.3 cm and 1.0–2.7 cm respectively. Seeds in single-seeded fruits were bigger than those in multiple-seeded fruits in which seed size also decreases with increasing number of seeds. Fig. 3.2. Cartographic drawing of a Vitellaria paradoxa seed showing the different sides; DE, Distal end; PE, Proximal end; DS, Dorsal side; VS, Ventral or hilar side; h, hilum; hc, hilar cap; m, micropyle Structurally, the seed of V. paradoxa comprises a coat that is flatter and slightly rough at the ventral side but smooth and convex at the dorsal side (Fig. 3.3A and B). The coat is the endocarp of the fruit wall. The ventral side comprises mainly the hilum which begins from the distal end where it is capped, broadens in the middle and tapers at the proximal end of the seed (Fig. 3.3B and D). The hilum is the flatter part of the seed with the hilar cap being completely woody (Fig. 3.3B). The shape of the hilum University of Ghana http://ugspace.ug.edu.gh 36 varies from seed to seed, but seeds from the same parent tree have comparatively similar-shaped hila. The hila of seeds of single-seeded fruits are centrally located whilst those of multiple-seeded fruits are almost laterally located (Fig. 3.3E). The convex side of the seed is shiny whilst the hilum appears dull. co hc h A B m m 1.0 cm C 1.0 cm D ms 1.0 cm E Fig. 3.3. Vitellaria paradoxa seeds showing A, differences in coat colour in seeds from the same parent tree; B, hilum (h) with the hilar cap (hc); C, micropyle (m); D, location of the micropyle (m); E, hila of different shapes from seeds of different parent trees; co, convex part of the seed; ms, seed from a multiple-seeded fruit University of Ghana http://ugspace.ug.edu.gh 37 The size of the Z-shaped micropyle ranges from 0.04 to 1.00 mm (Fig. 3.3C). It is located at the proximal end of the seed where it is separated from the hilum which is at the ventral side (Fig. 3.3D). The thickness of the seedcoat varies from seed to seed, but it ranges between 0.04 and 0.07 mm at the convex part to 0.08 and 1.20 mm at the hilar side. The seedcoat is thinnest at the dorsal side where it ruptures easily with moisture loss when little pressure is exerted on it. The colour of the seedcoat of V. paradoxa varies considerably even among seeds from the same tree whilst shape of the seed is comparatively the same. The colour of seedcoat is light to dark brown on the convex sides whilst that of the hilum ranges from beige to sandybrown or tan. Shape of the seed varies from spherical to globular or obovoid (Fig. 3.3A and B). Seeds in multiple-seeded fruits have flatter sides at the points of contact with each other in the fruit. Deshelled seeds show raphes at the distal ends which are oriented either parallel or perpendicular to the embryo (Fig. 3.4A and B). Seeds whose cotyledonary raphes run parallel to the embryo were named Type 1 seeds, whilst those with their raphes perpendicular to the embryo were termed Type 2 seeds. The size of the cotyledonary raphe increases as the moisture content of the seed decreases and this desiccation creates a depression in between the cotyledons. A fully split-open seed whether air dried or sprouted shows 2 distinct sections on its cotyledons (Fig. 3.4C). One of the sections broadens at the distal part but narrows sharply and ends bluntly just close to the proximal end. This section represents where the seed is schizocotylous (the cotyledons are appressed or adpressed to each other). The other section, located at the University of Ghana http://ugspace.ug.edu.gh 38 proximal sides, is where the seed is syncotylous (cotyledons are fused to each other). Thickness of the pseudoradicles in sprouted seeds ranged from 4 to 7 mm. A B x y x y x x C Fig. 3.4. Cotyledon morphology of Vitellaria paradoxa seed; A, Type 1 seed with a raphe (arrowed) parallel to the embryo; B, Type 2 seed with a raphe perpendicular to the embryo; C, Sprouted seed which is split showing where cotyledons are appressed (y) and fused (x) to each other University of Ghana http://ugspace.ug.edu.gh 39 3.4.2. Anatomy of Vitellaria paradoxa seed In a partially dry seed, the kernel shrank and pulled away from the seedcoat creating a space at the dorsal side whilst it remained appressed to the seedcoat at the hilar side (Fig. 3.5A). The cut surfaces of such partially dry kernels looked slightly pinkish or light brown depending on the seeds and their desiccation status with few drops of water-like fluids appearing on them (Fig. 3.5B). The colour of the kernels of fresh seeds is white, but it changes to light brown as the kernels lose moisture. Kernels are smooth all round with slight depressions at the raphes but appear slightly rough as they lose moisture. The raphe, visible at the distal end of the seed (Fig. 3.4A and B), widens inside the seed due to moisture loss (Fig. 3.5B). A B Fig. 3.5. Transverse sections through partially dry V. paradoxa seeds; A, Space created between the kernel and convex parts of the testa after moisture loss; B, Microscopic view (10) of the surface of a partially dry seed showing a wider cotyledonary raphe (arrowed) All the fresh seeds immersed in the TTC solution were stained by the tetrazolium chloride (Fig. 3.6A–C). They showed a red coloration in the slit along the 2 cotyledons with the staining being deeper towards the proximal end and visible at the University of Ghana http://ugspace.ug.edu.gh 40 exserted spot (Fig. 3.6A and B). A longitudinal section through a stained seed showed 2 differentially stained regions (Fig. 3.6C); the lighter portion is the radicle whilst the deeper part is the plumule. Embryos, the only metabolically active parts of seeds, contain dehydrogenase enzymes which release hydrogen that reduces tetrazolium chloride (with a pale yellow colour) to a bright-red formazan (Yu and Wang, 1996). Therefore, the deeply red-stained region was considered as the embryo. Partially dry seeds did not show any red coloration indicating that they were dead (Fig. 3.6D). A B r p C D Fig. 3.6. Vitellaria paradoxa seeds stained by tetrazolium chloride; A, Deep red stain (arrowed) indicating the presence of a live embryo; B, Transverse section through a stained seed showing the raphe (arrowed); C, Longitudinal section through a stained seed showing two differentially stained regions: r, radicle; p, plumule; D, Dark embryo spot (arrowed) showing a dead embryo University of Ghana http://ugspace.ug.edu.gh 41 When fresh seeds were deshelled, the spot indicated by the TTC staining as the embryo is normally light yellow (Fig. 3.7A). The light yellow spot (embryo) is either exserted or unexserted and in this study, the exserted embryos were bigger than the unexserted ones. The embryos are usually located at the proximal end, but in some seeds they were observed in other parts (Fig. 3.7B). Seeds having one embryo were considered monoembryonic. In sprouted seeds, the cotyledons usually swelled and distended at the point where the embryo is located (Fig. 3.7B). A B Fig. 3.7. Location of the embryo in Vitellaria paradoxa seeds; A, Fresh seed showing exserted embryo (arrowed) at the proximal end; B, Sprouted seed showing cotyledons distended into a 6 mm thick pseudoradicle (arrowed) at a lateral side Some seeds had 2 yellow spots (embryos) located at their proximal ends (Fig. 3.8E) and were considered polyembryonic. The polyembryonic seeds were large with their linear dimensions ranging from 3.7–4.2 × 3.0–3.5 cm. These seeds were flat with their dorsal parts scarcely convexed. In contrast to monoembryonic seeds, polyembryonic seeds have 3 cotyledonary raphes (1 central and 2 lateral) and could be split open easily into 2 parts along the central raphe with 1 embryo located in each of them (Fig. 3.8B). Along the 2 lateral raphes, a polyembryonic seed could be split into University of Ghana http://ugspace.ug.edu.gh 42 4 parts suggesting that each of these seeds had 4 cotyledons (Fig. 3.8B). As the moisture content of the seeds decreased, the embryos turned light brown to dark. 1 2 2 A B Fig. 3.8. Polyembryonic Vitellaria paradoxa seed; A, Proximal end of the seed showing two embryos (arrowed); B, Distal end of the seed showing the central (1) and lateral (2) raphes Fresh cotyledons of the seeds used for this study were usually unequal with the embryo appearing as a yellow thrusted spot at the proximal end of the smaller one and its notch on the bigger one (Fig. 3.9A). In contrast to the seeds used for the TTZ test, the freshly split open seeds did not reveal embryos being well differentiated into plumules and radicles (Fig. 3.9A). For instance, when these freshly split open seeds were observed under a stereomicroscope, their plumules were never clearly distinct from their radicles. Instead, the entire embryo was seen as small and linear, and surrounded by 2 large cotyledons (Fig. 3.9B). University of Ghana http://ugspace.ug.edu.gh 43 e A Radicle Embryonic e notch Plumule x x x x y y B Fig. 3.9. Split cotyledons of a Vitellaria paradoxa seed; A, Fresh cotyledons showing the embryo notch (arrowed) and embryo (e); B, Cartographic drawing showing the embryo (e) and where cotyledons are free (y) and fused (x) When fresh seeds were longitudinally sectioned, they copiously exuded latex only from the spot where the embryo is located (Fig. 3.10A). Conversely, transversely sectioned seeds exuded latex from the edges as well as from the slit in between the cotyledons (Fig. 3.10B). Partially dry seeds never yielded any latex on their cut surfaces when sectioned in either direction. University of Ghana http://ugspace.ug.edu.gh 44 A B Fig. 3.10. Exudation of latex from fresh Vitellaria paradoxa seed; A, Longitudinal section through a seed showing latex oozing at the embryo site; B, Transverse section showing latex oozing around the edges and slit (arrowed) in between the cotyledons 3.5. Discussion Knowledge of the anatomy and morphology of seeds is very important because these features influence germination and seedling establishment (Aguado et al., 2011). The convex part of a Vitellaria seed is smooth and shiny. The convex shape of the dorsal side confers some physiological advantage to the seed. According to Tompsett (1994), recalcitrant seeds have evolved convex shape to minimize desiccation by reducing the surface area exposed to insolation. The shiny, smooth seedcoat reflects radiant heat and thus minimizes seed desiccation. The variation in colour of coat of mature seeds from light brown to dark brown is probably due to genotypic effect of the mother tree as well as environmental conditions. In this study, variation in colour of seedcoat was observed even among seeds of the same tree contrary to the observation by Diarrassouba et al. (2009) who reported homogeneous coloration of seeds from the same tree at shea parkland of Tengrela Department in Côte d’Ivoire. University of Ghana http://ugspace.ug.edu.gh 45 The micropyle of a Vitellaria seed is located at the proximal end, whilst the hilum is at the ventral side. This morphology most probably precludes imbibition of moisture via the micropyle. In contrast, the micropyle of many dicotyledonous seeds is located on hilar side (Schmidt, 2000). Thus, when the hilum of a Vitellaria seed is in contact with the soil, the micropyle is raised slightly above the soil surface (Fig. 3.3D). Because moisture uptake depends on physical contact between soil moisture and the seed, the slightly airborne micropyle can hardly imbibe moisture. Naturally, Vitellaria seeds are oriented hilum down during germination (Hall et al., 1996) suggesting that some part of the hilum exclusively absorbs moisture. The smaller size of the micropyle (0.04 to 1.00 mm) implies that the thicker pseudoradicle (4.00 to 7.00 mm) of germinating seed cannot protrude the seedcoat through it either. Thus, the Z-shaped micropyle of the V. paradoxa seed might be the opening on the seedcoat for only gaseous exchange. The seedcoat is thin but the size (thickness) varied from the convex part to the hilar or ventral side. Wada et al. (2011) pointed out that differential thickness of the seedcoat implies differences in the sensitivities of various parts of the seedcoat to factors necessary for germination. A thin seedcoat promotes rapid germination of the seeds because it allows quick imbibition of water. According to Yidana (2004), V. paradoxa seeds germinate within 2–6 days after sowing. However, the thin dorsal side, which is always exposed, might also enhance moisture loss from the seed and might thus explain why extracted seed lose their germinability rapidly. Pritchard et al. (2004) reported that recalcitrant seeds have little physical defences such as thick seedcoat and therefore evolve rapid germination as a reproductive strategy to minimize desiccation- related mortality. Desiccation-related mortality of the Vitellaria seeds may also be University of Ghana http://ugspace.ug.edu.gh 46 caused by either the kernel shrinking away from the seedcoat at the dorsal side or the widening of the cotyledonary raphes inside the seeds. Either or both of the resulting spaces may decrease the conduction of imbibed water to the embryo. The hilum of V. paradoxa covers the entire ventral side running from the distal to the proximal end of the seed. The distal part of the hilum termed hilar cap is woody and appears porous and is thus the part that most likely absorbs moisture during germination. Although, many seeds imbibe water through the micropyle, a number of them absorb moisture mainly through the hilum (Xia et al., 2012; Maekawa, 1991) and this phenomenon may be true for Vitellaria seeds. All sapotaceous species thus far identified possess prominent hila (Jessup and Short, 2011; Graveson, 2009). The presence of a conspicuous hilum may be a useful taxonomic feature for identifying other members of the family particularly in tropical Africa where many plants remain undescribed and uncharacterized. According to McDonald (2013), both the texture and the area of the seed that contact the growth medium influence the rate at which water is imbibed. Thus, the flat, broad and rough hilum which is usually in contact with the soil during natural dispersal allows large soil–seed interface area which enhances rapid imbibition of water for germination and subsequent seedling growth. The orientation of the raphes of V. paradoxa seeds as observed in this study revealed 2 seed types based on cotyledon morphology referred to as Type 1 and Type 2 seeds. Type 1 seeds are those in which the raphes run parallel to the embryos while Type 2 seeds, which were more common, are those in which the raphes run perpendicular to the embryos. The 2 seed types are only distinguishable when their seedcoats are removed. Cotyledon morphology is a very useful taxonomic feature in classifying University of Ghana http://ugspace.ug.edu.gh 47 plants (Chandler, 2008) and has been used to identify different taxa in the Convulvulaceae family (Ogunwenmo, 2003). The genus Vitellaria is widely reported to have 2 subspecies namely paradoxa and nilotica (McNeill and Turland, 2011). However, Hall et al. (1996) observed no clear distinctions between the 2 subspecies. The 2 seed types with different cotyledon morphologies observed in this study may be 2 subspecies in the genus. However, this claim needs to be investigated further. A V. paradoxa seed comprises 2 large cotyledons which are distally free (schizocotylous) but proximally fused (syncotylous). Whilst Nikiema and Umali (2007) reported that the seeds of V. paradoxa are fully syncotylous, all the seeds observed in this study were only partially syncotylous. Partial or full syncotyly influences seed germination in 2 ways. First, it impedes the emergence of cotyledons from the seedcoat during germination (Flores, 2002). Second, in syncotylous seeds, some intercalary growth at the base of the cotyledons produces petioles from which the epicotyls appear (Finneseth et al., 1998). Thus, the proximally syncotylous morphology of V. paradoxa seeds explains why they germinate crypto-hypogeally producing long and fused cotyledonary petioles which are erroneously described as pseudoradicles by Ugese et al. (2010) and Jackson (1974). Contrastingly, Ehiagbonare et al. (2008) reported epigeal germination in Chrysophyllum albidum (a Sapotaceae) because its seeds are fully schizocotylous. All the fresh seeds immersed in the TTC solution stained red with the red stain more visible at the proximal ends whilst all the partially dry seeds showed no staining. Fresh or live embryos are light yellow in colour; thus, changes in the colour of the embryo to dark brown which indicated seed death was associated with a decrease in University of Ghana http://ugspace.ug.edu.gh 48 seed moisture content confirming the desiccation sensitivity of the seeds. Embryos of seeds contain dehydrogenase enzymes whose metabolic activities release hydrogen which on contacting 2,3,5-triphenyl tetrazolium chloride reduces it to a stable, bright red triphenylformazan. According to Yu and Wang (1996), the part of the seed where red formazan is produced is the embryo. Thus, TTC staining was also used to identify the embryos for in vitro culture (Section 5.2.4). In some seeds, the embryos were found in different parts besides the proximal end. Therefore, in the seeds of V. paradoxa, the embryos could be located at different parts. The embryo is small relative to the massive cotyledons that surround it. The TTZ test showed 2 differently stained parts identified as the radicle and plumule. Such differentiation was never observed in fresh seeds not used for the test suggesting that the seeds of Vitellaria most likely possess immature and rudimentary embryos. Immature embryos are commonly found in recalcitrant seeds (Berjak and Pammenter, 2008). Vitellaria paradoxa, therefore, might have evolved cryptogeal germination to allow the physiologically immature embryos to be pushed into the bulge of the pseudoradicle to mature before germination (producing true radicles). Some of the V. paradoxa seeds used for this study had 2 embryos both located at their proximal ends (polyembryony). In contrast to monoembryonic seeds, the polyembryonic seeds could easily be separated into 2 parts along a central raphe with 1 embryo located on each of the parts. Polyembryony has been reported in many plant families and it occurs in 2 main forms namely zygotic or nucellar polyembryony (Aleza et al., 2010). The location of the embryos in the 2 readily separable parts of the University of Ghana http://ugspace.ug.edu.gh 49 seeds as observed in this study suggests that the polyembryony is more likely zygotic and may result in the production of 2 seedlings per seed. A longitudinal section through the embryo of a fresh seed resulted in copious exudation of latex only at the embryo whilst a transverse section showed latex oozing from the edges and around the cotyledonary raphe. This pattern of latex exudation suggests the presence of laticiferous vessels running parallel to one another. The laticiferous nature of the seed most likely accounts for the difficulty in isolating the embryo as well as its slow growth. In contrast, seeds of other sapotaceous species such as Synsepalum dulcificum scarcely show any visible latex when cut, explaining why their embryos are easily isolated for in vitro culture (Ogunsola and Ilori, 2008). 3.6. Conclusion The seed of Vitellaria paradoxa has a shiny, brown seedcoat which is very fragile at the convex part whilst thicker at the flatter, hilar side. The shrinking of the kernel away from the seedcoat and the widening of the cotyledonary raphe reflect the recalcitrance of the seed to storage. The seed is proximally syncotylous which causes the cotyledons to swell and to elongate producing fused cotyledonary petioles during germination. It has a rudimentary embryo which is located in the small, light yellow, exserted or unexserted spot at the proximal end. The entire embryo is embedded in copious amount of latex making it difficult to be identified and isolated. University of Ghana http://ugspace.ug.edu.gh 50 REFERENCES Aguado, M., Martínez-Sánchez, J., Reig-Arminana, J., Garcia-Breijo, F. J., France, J. A. and Vicente, M. J. (2011). Morphology, anatomy and germination response of heteromorphic achenes of Anthemis chrysantha J. Gay (Asteraceae), a critically endangered species. Seed Science Research, 21:283–299. Aleza, P., Juárez, J., Ollitrault, P. and Navarro, L. (2010). Polyembryony in non- apomictic citrus genotypes. Annals of Botany, 106(4):533 (abs). Berjak, P. and Pammenter, N. W. (2008). From Avicennia to Zizania: Seed Recalcitrance in Perspective. Annals of Botany, 101: 213–228. Chandler, J. W. (2008). Cotyledon organogenesis. Journal of Experimental Botany, 59(11):2917–2931. Corby, H. D. L. (2008). Seed germination among the Leguminosae. 22 pp. Available at http://www.hdlcorby.com. Accessed on 21st June 2013. Diarrassouba, N., Fofana, I. J., Issali, A. E., Bup, N. D. and et Sangaré, A. (2009). Typology of shea trees (Vitellaria paradoxa) using qualitative morphological traits in Côte d’Ivoire. Gene Conserve, 8(33):752–780. Ehiagbonare, J. E., Onyibe, H. I. and Okoegwale, E. E. (2008). Studies on the isolation of normal and abnormal seedlings of Chrysophyllum albidum: A step towards sustainable management of the taxon in the 21st century. Scientific Research and Essay, 3(12):567–570. Filonova, L. H., von Arnold, S., Daniel, G. and Bozhkov, P. V. (2002). Programmed cell death eliminates all but one embryo in a polyembryonic plant seed. Cell Death and Differentiation, 9:1057–1062. Finneseth, C. H., Layne, D. R. and Geneve, R. L. (1998). Morphological development of the North American pawpaw during germination and seedling emergence. HORTSCIENCE, 33(5):802–805. Flores, E. M. (2002). Tropical tree seed biology. In: Agricultural Handbook 721 (Vozzo, J. A. ed.). pp. 13–118. USDA Forest Service, Washington DC. Gama-Arachchige, N. S., Baskin, J. M., Geneve, R. L. and Baskin, C. C. (2011). Acquisition of physical dormancy and ontogeny of the micropyle–water-gap complex in developing seeds of Geranium carolinianum (Geraniaceae). Annals of Botany, 108:51–64. Graveson, R. (2009). Plant taxonomy of Saint Lucia: Botanical descriptions of important species, species checklist and herbarium development. Technical Report No. 4 to the National Forest Demarcation and Bio-Physical Resource Inventory Project, FCG International Ltd, Helsinki, Finland. p. 26. University of Ghana http://ugspace.ug.edu.gh 51 Hall, J. B., Aebischer, D. P., Tomlinson, H. F., Amaning, E. O. and Hindle, J. R. (1996). Vitellaria paradoxa, a monograph, Publication Number 8, School of Agricultural and Forest Sciences, University of Wales, Bangor, UK. 105 pp. http://www.html-color-names.com/color-chart.php. Accessed on 6th June 2012. Jackson, G. (1974). Cryptogeal germination and other seedling adaptations to the burning of vegetation in the savanna regions: the origin of the pyrophytic habit. New Phytologist, 73:771–780. Jessup, L. W. and Short, P. S. (2011). Sapotaceae. In: Flora of the Darwin Region (Short, P. S. and Cowie, I. D. eds). Volume 1, pp. 1–6. Northern Territory Herbarium, Department of Natural Resources, Environment, the Arts and Sport. Palmerston, Australia. Kitajima, K. and Fenner, M. (2000). Seedling regeneration ecology. In: Seeds: Ecology of regeneration in plant communities (Fenner, M. ed.). 2nd ed. pp. 331–360. Wallingford, UK: CAB International. Leubner, G. (2012). The seed biology place. http://www.seedbiology.de. Accessed on 13th March 2013. Maekawa, S. (1991). Verbena seed hilum morphology contributes to irregular germination. HORTSCIENCE, 26(2):129–132. Maia, L. A., Maia, S. and Parolin, P. (2005). Seedling morphology of non-pioneer trees in Central Amazonian Várzea floodplain forests. Ecotropica, 11:1–8. McDonald, M. B. (2013). Physiology of seed germination. Seed Biology Program, The Ohio State University, Columbus, Publication No. OH 43210–1086. 8 pp. http://seedbiology.osu.edu/HCS631. Accessed on 20th June 2013. McNeill, J. and Turland, N. J. (2011). Synopsis of proposals on botanical nomenclature – Melbourne 2011: A review of the proposals concerning the International Code of Botanical Nomenclature submitted to the XVIII International Botanical Congress. Taxon, 60 (1):243–286. Nikiema, A. and Umali, B. E. (2007). Vitellaria paradoxa C.F.Gaertn. http://database.prota.org/PROTAhtml/Vitellaria paradoxa_En.htm. Accessed on 22nd May 2013. Ogunsola, K. E. and Ilori, C. O. (2008). In vitro propagation of miracle berry (Synsepalum dulcificum Daniel) through embryo and nodal cultures. African Journal of Biotechnology, 7(3):244–248. Ogunwenmo, K. O. (2003). Cotyledon morphology: an aid in identification of Ipomoea taxa (Convolvulaceae). Feddes Repertorium, 114:198–203. Pérez, H. E. (2009). Promoting germination in ornamental palm seeds through dormancy alleviation. HortTechnology, 19(4):882–685. University of Ghana http://ugspace.ug.edu.gh 52 Pritchard, H. W., Daws, M. I., Fletcher, B. J., Christiane, S., Gaméné, C. S., Msanga, H. P. and Omondi, W. (2004). Ecological correlates of seed desiccation tolerance in tropical African dryland trees1. American Journal of Botany, 91(6):863–870. Roosmalen, G. M. and Garcia, O. M. C. G. (2000). Fruits of the Amazonian forest, Part II: Sapotaceae. Acta Amazonica, 30(2):187–290. Sanou, H. and Lamien, N. (2011). Vitellaria paradoxa, shea butter tree. Conservation and sustainable use of genetic resources of priority food tree species in sub- Saharan Africa. Bioversity International (Rome, Italy). Available at www.bioversityinternational.org. Accessed on 2nd January 2013. Schmidt, L. (2000). Seed development, biology and ecology. Available at http://curis.ku.dk/portal-life/files/20712850/Chapter2. Accessed on 10th May 2012. 35 pp. Swaminathan, C., Vinaya Rai, R. S., Suresh, K. K. and Sivaganam, K. (1992). Improving seed germination of Derris indica by vertical sowing. Journal of Tropical Forest Science, 6(2):152–158. Tompsett, P. B. (1994). Capture of genetic resources by collection and storage of seed: a physiological approach. In: Tropical trees: the potential for domestication and the rebuilding of forest resources (Leakey, R. R. B. and Newton, A. C. eds). pp. 61–71. ITE Symposium No. 29, ECTF Symposium No. 1. London: HMSO. Ugese, F. D., Baiyeri, K. P. and Mbah, B. N. (2010). Determination of growth stages and seedling structures associated with slow emergence of shea butter tree (Vitellaria paradoxa C.F.Gaertn.) seedlings. Journal of Animal and Plant Sciences, 8(2):993–998. Wada, S., Kennedy, J. A. and Reed, B. M. (2011). Seed-coat anatomy and proanthocyanidins contribute to the dormancy of Rubus seed. Scientia Horticulturae, 130:762–768. Xia, K., Daws, M. I., Stuppy, W., Zhou, Z-K. and Pritchard, H. W. (2012). Rates of water loss and uptake in recalcitrant fruits of Quercus species are determined by pericarp anatomy. PLoS One, 7(10):1–11. Yidana, J. A. (2004). Progress in developing technologies to domesticate the cultivation of shea tree (Vitellaria paradoxa) in Ghana. Agricultural and Food Science Journal of Ghana, 3:249–267. Yu, S. L. and Wang, B. S. P. (1996). Tetrazolium testing for viability of tree seed. In: Rapid viability testing of tropical tree seed (Bhodthipuks, J. ed.). Training course proceedings No. 4:33–58. ASEAN Forest Tree Seed Centre Project. Muak Lek, Thailand. University of Ghana http://ugspace.ug.edu.gh 53 CHAPTER 4 4.0. Germination studies on Vitellaria paradoxa seeds 4.1. Introduction Germination is traditionally classified based on 3 cotyledonar traits namely position, exposition and function. On the basis of cotyledon position, germination is described as hypogeal if the cotyledons remain below the soil or epigeal if they are lifted above the soil. With reference to the exposition of cotyledons, Amritphale and Sharma (2008) classified germination as either cryptocotylar for which the cotyledons remain in the seed coat or phanerocotylar where the seed coat is shed. Using cotyledon function, Baraloto and Forget (2007) described cotyledons as reserve if they are non- photosynthetic or foliaceous if they are photosynthetic. Considering the wide diversity of germination, Maia et al. (2005) described a classification system that integrates all the cotyledonar traits giving rise to 5 or 6 seedling types known as cryptocotylar hypogeal reserve (CHR), cryptocotylar epigeal reserve (CER), phanerocotylar hypogeal reserve (PHR), phanerocotylar epigeal reserve (PER), phanerocotylar epigeal foliaceous (PEF) or phanerocotylar hypogeal foliaceous (PHF). Phanerocotylar hypogeal foliaceous seedlings remain unreported or are considered as non-existent. Usually, only 1 of these seedlings types is found within a genus. Essig (1987) reported that epigeal and hypogeal germination may be used to distinguish between different subgenera within some genera. Germination types based on cotyledonar traits lead to the understanding of seedling morphologies and culture conditions most appropriate for seedling development. Seedling types based on cotyledon morphology are used to study phylogenetic University of Ghana http://ugspace.ug.edu.gh 54 relationships because seedling traits are highly conservative (Ibarra-Manríquez et al., 2001). However, no attempt has been made to describe the germination and morphology of V. paradoxa seedlings with reference to cotyledonar traits. Such studies may therefore be useful for both taxonomic and agronomic purposes. Natural stands of shea trees are increasingly being cleared for farming despite the low recruitment rate of the species. Seed mortality is also rising because fruit shedding has now been coinciding with drought or dry season. These threats make it necessary to examine the germination of this economically important tree. In-depth knowledge of the germination of V. paradoxa will enhance nursery establishment which may eventually lead to its domestication. Thus, the major objective of this study was to examine the germination of V. paradoxa seeds critically as part of initial efforts towards the domestication of the plant. The specific objectives of this study were to i. investigate the stages of development of Vitellaria seedlings ii. evaluate the effect of seed size on germination and seedling establishment iii. determine the effects of deshelling on germination and seedling growth. 4.2. Materials and methods 4.2.1. Seed collection Shea fruits were collected as described in the previous chapter (Section 3.2.1). They were depulped manually to obtain fresh seeds which were then washed with tap water and air-dried for 6 hours. The extracted seeds were used for the various experiments unless otherwise stated. University of Ghana http://ugspace.ug.edu.gh 55 4.2.2. Studies on Vitellaria paradoxa seedling development Two hundred and forty (240) seeds of similar size were nursed in polyethylene bags filled with soil mix comprising topsoil and well-decomposed sawdust in the ratio 5:1. Seeds were treated with HerculeR 50 SC (IPROCHEM Co. Ltd, Shenzhen, China) against termites and then sown 2 cm deep with the hilar side down. They and the resulting seedlings were watered as and when necessary. On sprouting, the points on the seedcoat through which the pseudoradicles protruded were observed, while the number of pseudoradicles produced per seed was counted. Five (5) seedlings were then sampled destructively at 3-day intervals beginning 5 days after sowing (DAS) to observe plumule burial in the soil, its descent to the base of the pseudoradicle and its morphogenesis into a shoot. The pseudoradicles of another set of 15 seedlings which were also sampled at the same interval and air-dried for 24, 48 or 72 hours were dissected either transversely or longitudinally to observe their anatomical features using a magnifying lens and a stereomicroscope (Leica ZOOM 2000, Cole-Parmer, Wetzlar, Germany). Both sets of seedlings were sampled 7 times each. The observed anatomical features were photographed using a 16.1-megapexil digital still camera (Sony Corporation, China). The outer covering of the pseudoradicle was manually removed to observe the core anatomical structures. The remaining seedlings which were not sampled were allowed to develop to emergence. These seedlings were observed until 3 weeks after emergence to describe the stages through which they developed. They were further classified as shrubby or multiple seedlings. A shrubby seedling developed from a single pseudoradicle but produced one or more lateral shoots, whilst a multiple seedling developed from one of the several pseudoradicles produced by a seed. Seedling morphology was described using cotyledonar traits (degree of exposition of cotyledons and their function) as proposed by Flores (2002). University of Ghana http://ugspace.ug.edu.gh 56 4.2.3. Seed size and development of Vitellaria paradoxa seedlings The linear dimensions (length, breadth and width) of 180 seeds were measured with vernier callipers. The products of these dimensions (in cm3) were used to categorize seeds as small (13–15 cm3), medium (19–21 cm3) or large (25–27 cm3). The categorized seeds were then sown on seedbeds constructed using soil mix as described earlier (Section 4.2.2). The experimental design was randomized complete block with 60 seeds in each of the 3 replicates. All seeds were sown 2 cm deep with the hilar side down as it happens in the wild. Watering and hoe-weeding were done as and when necessary. Germinating seeds and seedlings were sprayed with Cydem Super and Akape 20 SC (IPROCHEM Co. Ltd, Shenzhen, China) against insects notably termites and leaf-eating pests. Seeds were observed at a 5-day interval for signs of germination for 30 days and thereafter for emergence beginning 30 DAS. Seeds were considered as germinated when their pseudoradicles had become visible. Germination percentage and mean germination time (MGT) were calculated from the data obtained. Germination percentage was calculated as GP = (GS*100)/ TS, where GP = germination percentage, GS = number of germinated seeds and TS = total number of sown seeds, whilst mean germination time (MGT) was computed as Σ(t × n ) MGT (days) =   Σn where ti is the number of days beginning from the date of sowing and ni is the number of germinated seeds at each day (Bewley and Black, 1994). Emergence percentage (EP), emergence index (EI) and emergence rate index (ERI) were computed using the formulae described by Adetimirin et al. (2006) as follows: University of Ghana http://ugspace.ug.edu.gh 57       EP = × 100 %            () EI =        ERI =  (  ) Mean germination time measures the duration to sprouting, emergence index measures the rate of seedling emergence and emergence rate index estimates the duration to the emergence of all seedlings in the absence of other limiting conditions. At bulging and emergence stages, length of pseudoradicles and taproots was measured using a metre rule. Length of pseudoradicle was measured from the seed to the cotyledonary node, whilst that of the taproot was measured from the cotyledonary node to the tip of the primary root. At emergence, root crown and shoot diameters were also measured using vernier callipers. Total shoot height (from the cotyledonary node to the tip of the shoot) was measured 3 weeks after emergence. Growth stages of V. paradoxa seedlings as described by Ugese et al. (2010) were modified and the duration of each stage was measured (in days). The 5 growth stages identified by Ugese et al. (2010) are sprouting, swelling of the pseudoradicle, appearance of a pink- coloured shoot on the pseudoradicle, elongation of the shoot and seedling emergence. Duration to shoot elongation was computed as the difference between the time the shoot appeared on the pseudoradicle and when the seedling emerged above the soil. Time to seedling establishment (exhaustion or transfer of seed reserve into the seedling structures) was estimated. Seed reserves were considered exhausted when the cotyledons turned dark brown both externally and internally. University of Ghana http://ugspace.ug.edu.gh 58 4.2.4. Deshelling of seed and development of Vitellaria paradoxa seedlings Freshly extracted seeds numbering 120 were used for this study. Sixty (60) seeds were deshelled whilst the remaining seeds were left intact as the control. Seeds were deshelled as described in Section 3.1.2. Both the deshelled and intact seeds were then sown 2 cm deep at a spacing of 50 × 25 cm on 3 seedbeds constructed using soil mix described in Section 4.2.2. The experimental design was randomized complete block with each of the beds as a replicate. All cultural practices performed were similar to those described in Section 4.2.3. Data were taken on germination and emergence as outlined in Section 4.2.3. Seedling growth parameters namely number of leaves, height, length and width of leaves, and stem and root crown diameters were measured at 150 and 240 DAS. Seedling height (taken from soil level) and leaf dimensions were measured with a metre rule, whilst stem diameter was measured at soil level with vernier callipers. Root crown diameters of seedlings were also measured at the widest point using vernier callipers. The linear dimensions of leaves were used to compute leaf area based on the model developed by Ugese et al. (2008a): LA = 4.41 + 1.14LW, where LA is leaf area and LW is the product of linear dimensions of the length and width at the broadest part of the leaf. Growth pattern of seedlings was described according to the terminologies of Wu and Hinckley (2010) and Tomlinson (1987). 4.2.5. Statistical analysis Data collected were subjected to analysis of variance (ANOVA) using the Genstat statistical package (9th Edition). Percentage data on germination and on emergence were transformed using square-root transformation before analysis. Means were University of Ghana http://ugspace.ug.edu.gh 59 separated where appropriate at 5 % significance level using least significant difference (LSD) test. 4.3. Results 4.3.1. Stages of the development of Vitellaria paradoxa seedlings The development of V. paradoxa seedlings occurred in 7 distinct stages. These were sprouting, elongation of the pseudoradicle, bulging, shoot appearance, shoot elongation, emergence and establishment. The first 5 stages represented the skotomorphogenic growth stages because they occurred below ground whilst stages 6 and 7 were the photomorphorgenic stages because they took place above ground. Sprouting or germination is the first stage and was marked by the protrusion of the pseudoradicle through the seedcoat. Prior to protrusion, the cotyledons swelled at the embryo side producing a thick pseudoradicle which then pushed against and ruptured the overlying seedcoat (Fig. 4.1A and B). Pseudoradicles which appeared at the proximal ends pushed against well-defined parts called opercula rupturing and sloughing the caps off (Fig. 4.1B). The operculum caps detached first at the dorsal side, but they remained attached to the seedcoat at the hilar side and were eventually pushed away by the extending pseudoradicles (Fig. 4.1C and D). The opercula and their caps were nearly circular except towards the hilar side where the boundaries were straight (Fig. 4.1C). The micropyles were still intact in the portions of the seedcoat (operculum caps) that were thrown off (Fig. 4.1D). University of Ghana http://ugspace.ug.edu.gh 60 r m A B m 2 1 C D Fig. 4.1. Sprouted Vitellaria paradoxa seeds showing A, the margin (r) of the operculum and the micropyle (m) on the operculum; B, a pseudoradicle pushing against an operculum cap; C, ruptured operculum cap (arrowed) still attached to the seedcoat; D, micropyle (m), operculum cap (1) and operculum (2) The protruding, blunt-ended pseudoradicle pushed the embryo out of the seed where it remained visible as a small light brown spot with a yellow background at the tip of the pseudoradicle (Fig. 4.2A). When the pseudoradicle became streamlined in shape, the embryo was no longer visible at the tip suggesting that it retracted into the base (Fig. 4.2B). The brittle pseudoradicle bruises easily and thereafter exudes latex profusely. University of Ghana http://ugspace.ug.edu.gh 61 A B Fig. 4.2. Pseudoradicles of Vitellaria paradoxa seedlings at the sprouting stage; A, Embryo at the tip of a pseudoradicle; B, Streamlined shaped pseudoradicle just beginning to elongate Ninety-two percent (92 %) of the pseudoradicles protruded through the proximal end of the seed whilst less than 4 % protruded through the other sides of the seedcoat (Table 4.1). Whereas the seedcoat ruptured along well-defined margins at the proximal end of the seed, it ruptured irregularly in distal, lateral, dorsal or ventral protrusions and often resulted in the shedding of a large part of the shell (Fig. 4.3A– E). Pseudoradicles protruding the seedcoat through the hilar side were usually thicker than those that protruded through the other sides. Table 4.1. Protrusion of pseudoradicles from different sides of germinating Vitellaria paradoxa seeds Side of pseudoradicle protrusion Percentage Proximal protrusion 92.0 Distal protrusion 1.2 Ventral protrusion 1.3 Dorsal protrusion 2.5 Lateral protrusion 3.0 University of Ghana http://ugspace.ug.edu.gh 62 A B C D E Fig. 4.3. Protrusion of pseudoradicles from different sides of germinating Vitellaria paradoxa seeds; A, Proximal protrusion; B, Distal protrusion; C, Ventral protrusion; D, Dorsal protrusion; E, Lateral protrusion About 93 % of the germinated seeds produced 1 pseudoradicle each whilst the remaining seeds produced either 2 or between 6 and 10 pseudoradicles each (Table 4.2). Where 2 pseudoradicles were produced from one seed, they were similar in size but were either separated from or appressed to each other (Fig. 4.4A and B). However, closer observation of appressed pseudoradicles showed a clear line of division between them. All 6 to 10 pseudoradicles produced from individual germinated seeds were unequal in size and separated from one another (Fig. 4.4C). Table 4.2. Number of pseudoradicles produced per germinating Vitellaria paradoxa seed Number of pseudoradicles produced Percentage per seed 1 92.5 2 0.03 > 6 7.47 University of Ghana http://ugspace.ug.edu.gh 63 ps b tr A B C Fig. 4.4. Vitellaria paradoxa seedlings at the second and third developmental stages; A, Sprouted Vitellaria paradoxa seed showing a pseudoradicle (ps), bulge (b) and true root (tr); B, Two pseudoradicles produced from one seed; C, Germinated seed with 6 pseudoradicles At the second stage of seedling development, the blunt-ended pseudoradicle became terete with the radicle visible at the tip. The pseudoradicle then elongated rapidly deep into the soil and formed a bulge or swelling at 2–8 cm along its length (Fig. 4.4A). Below the bulge, the true radicle continued its positive geotropic growth (Fig. 4.4A). Morphologically, the pseudoradicle is smooth and cylindrical and can be split into 2 equal parts along a defined raphe with each of the parts being attached to 1 of the cotyledons (Fig. 4.5A). Removing the outer sheath of the pseudoradicle reveals laticiferous or latex-containing vessels (Fig. 4.5B). In seeds that produced either 1 or 2 pseudoradicles, each of the pseudoradicles had a central hollow tube in addition to the outer sheath and laticiferous vessels (Fig. 4.5B and C). On the contrary, all the 6 University of Ghana http://ugspace.ug.edu.gh 64 to 10 pseudoradicles produced from individual seeds had only an outer covering and a laticiferous vessel each making them solid. co l pt A B C Fig. 4.5. Morphological features of the pseudoradicle; A, Cotyledons (co) showing the raphe (arrowed) along which their petioles were fused into a pseudoradicle; B, Pseudoradicle showing a laticiferous vessel (l) and the central hollow tube (pt); C, Seed showing two central hollow tubes (arrowed) Transverse and longitudinal sections through the pseudoradicles showed the 2 or 3 main parts described earlier. The outer sheath enveloped the vessels into a tubular geotropic structure (Fig. 4.6A). In seeds that produced 1 or 2 pseudoradicles, the laticiferous vessels varied from 6 to 8 (Fig. 4.6A). Conversely, in seeds that produced 6 or more pseudoradicles, each of the pseudoradicles had only 1 laticiferous vessel (Fig. 4.6B). The latex vessels which extend from the seed to the base of the bulge surround the central hollow tube (Fig. 4.6A). The plumule of the embryo moves through the central hollow tube during germination until it reaches the bulge suggesting that the hollow tube may be a plumule tube (Fig. 4.7A). University of Ghana http://ugspace.ug.edu.gh 65 pt os os l l 1 mm A 1 mm B Fig. 4.6. Anatomical and morphological features of the pseudoradicle at bulging stage; A, Transverse sections showing the outer sheath (os), plumule tube (pt) and laticiferous vessel (l); B, Outer sheath and laticiferous vessel The plumule which may be single or branched is surrounded by rhizoid- or hair-like structures. It is usually white but turns pink as it develops into a rudimentary shoot (Fig. 4.7B and C). Occasionally, some of the plumules never turn pink but remain white throughout the period when they develop into shoots. In this study, plumules that turned pink were more common than those than remained white. Plumules that branched developed 2 or more shoots in the bulge. The bulge, formed at the base of the pseudoradicle, is a swelling of 0.5–0.7 cm long that is produced when the descended plumule develops into the rudimentary shoot (Fig. 4.7A and C). The rudimentary shoot has internodes and nodes with scale leaves numbering 5 to 7 (Fig. 4.7D). The formation of the bulge marks the third stage of seedling development. University of Ghana http://ugspace.ug.edu.gh 66 pt p b 2 mm A 1 mm B p 4 b 3 2 1 2 mm C D Fig. 4.7. Development of the rudimentary shoot at the bulging stage; A, Longitudinal section through the pseudoradicle showing the plumule tube (pt), descended plumule (p) and bulge (b); B, Branched plumule; C, Plumule (p) developing into a rudimentary shoot; D, Rudimentary shoot showing an internode (1); epicotyl (2), node (3) and scale leaf (4); Bar: D  1.0 mm At the fourth stage of seedling development, the rudimentary shoot then protrudes from the pseudoradicle via a cotyledonary slit (Fig 4.8A). The shoots appeared ventral, lateral (Fig. 4.8B) or dorsal (Fig. 4.8D). Occasionally, 2 or more shoots University of Ghana http://ugspace.ug.edu.gh 67 appeared via the cotyledonary slit (Fig. 4.8C). The fifth stage is elongation of the shoot towards the soil surface (Fig. 4.8D). The shoots either elongated freely or were trapped within the pseudoradicles (Fig. 4.8B) or in-between the pseudoradicles and the seed (Fig. 4.8E). Trapped shoots could easily be freed by gently turning the pseudoradicle with the seed into another direction (Fig. 4.8F). 1 2 cs A B C D E F Fig. 4.8. Vitellaria paradoxa seedlings at the fourth and fifth stages of development; A, Pseudoradicle showing the cotyledonary slit (cs); B, Shoots that appeared lateral (1) and ventral (2); C, Three shoots that appeared ventral; D, Shoot elongating freely; E, Shoot that is trapped (arrowed); F, Trapped shoot freed by turning away the seed The elongated shoots then emerged above the soil level marking the sixth stage of seedling development. Emerged seedlings appeared pink or light green with the pinked-coloured seedlings outnumbering those that were light green. When the cotyledons of the seedlings turned dark brown, the seedlings were considered established. At this stage, both the leaves and stems of the pink-coloured seedlings University of Ghana http://ugspace.ug.edu.gh 68 had turned green except the growing points that still remained pink or light brown. Conversely, the light-green seedlings became fully green with their growing points still remaining light green. At the establishment stage, therefore, the seedlings had developed fully functional leaves for photosynthesis and were completely autotrophic (Appendix 7.1). Establishment is the final stage of seedling development. Any germinated seed that produced 1 pseudoradicle also produced 1 seedling usually with a single shoot emerging above the soil (Fig. 4.9A). Occasionally, some of those germinated seeds with 1 pseudoradicle produced 2 or more shoots (Figs. 4.9B and 10B). A seedling that produced 2 or more shoots was classified as a shrubby seedling. Germinated seeds that produced 2 pseudoradicles each also produced 2 clearly identifiable and easily separable seedlings (Fig. 4.9C). Two (2) seedlings produced from 1 seed were classified as multiple seedlings. Contrastingly, sprouted seeds that produced 6 or more pseudoradicles never produced any seedlings at all. A B C Fig. 4.9. Production of multiple shoots and seedlings in Vitellaria paradoxa; A, Seedling with one main axis; B, Seedling with two shoots; C, Two seedlings produced from one seed University of Ghana http://ugspace.ug.edu.gh 69 Seedlings were further classified as either cryptocotylar or phanerocotylar depending on the degree of cotyledon exposure. In those seeds whose cotyledons were lateral to each other when sown in the hilum down orientation, the resulting seedlings were the phanerocotylar type (Fig. 4.10A). In these seeds, the cotyledons usually split-open distally, but remained slightly proximally fused. For those seeds whose cotyledons lie on top of each other in the hilum down orientation, seedling morphology was cryptocotylar or semicryptocotylar (Fig. 4.10B and C). However, all the seedlings had reserve cotyledons irrespective of cotyledon arrangement. 1 2 A B C Fig. 4.10. Types of seedlings produced by Vitellaria paradoxa based on cotyledon exposition; A, Phanerocotylar seedling showing cotyledons with their raphe (arrowed); B, Cryptocotylar seedling with two shoots (1 and 2); C, Semicryptocotylar seedling 4.3.2. Effect of seed size on germination and emergence of V. paradoxa seedlings Seed size had a significant influence on the duration to germination, pseudoradicle elongation, bulging, shoot appearance, shoot elongation, emergence and establishment of Vitellaria seedlings (Tables 4.3 and 4.4). All the small and medium seeds sprouted or germinated whilst 95 % of the large seeds sprouted indicating that seed size did not University of Ghana http://ugspace.ug.edu.gh 70 have significant (P > 0.05) effect on germination. Similarly, emergence percentage, varying non-significantly from 93.15 % for medium seeds to 95.79 % for large seeds, was high among all the seedlings produced (Table 4.3). Duration to sprouting varied significantly (P < 0.05) from 7 days in large seeds to 12 days in the small seeds (Table 4.4.). Just as sprouting, the rate of pseudoradicle elongation was also statistically different among the seeds. The pseudoradicles elongated fully among the large seeds in 12 days which was significantly (P < 0.05) faster than those of both medium seeds (16 days) and small seeds (22 days). Days to shoot elongation (SE) which is the difference between days to emergence of seedlings above the soil and days to shoot appearance on the pseudoradicle (SA) had the longest duration of all the skotomorphogenic, or below-ground growth stages of Vitellaria seedlings (Table 4.4). It took 38 days in the seedlings produced from small seeds but followed no clearly defined trend because the shoots of the seedlings of medium seeds elongated earlier than those of large seeds. Seedlings produced by large seeds had the shortest duration to emergence (61 days), followed by seedlings produced by medium seeds (65 days) and finally those produced by small seeds (75 days) (Table 4.4). Seed size, therefore, had a significant effect (P < 0.05) on days to emergence of V. paradoxa seedlings. The corresponding emergence rate indices of seedlings produced by large and medium seeds (65 and 70 days respectively) were statistically the same, but both were significantly shorter than that of seedlings of small seeds (78 days) (Table 4.3). Time to seedling establishment University of Ghana http://ugspace.ug.edu.gh 71 varied significantly from 97 days for seedlings of small seeds to 99 and 114 days for seedlings of medium and large seeds respectively (Table 4.4). Table 4.3. Effect of seed size on germination, emergence percentage and emergence rate index of Vitellaria paradoxa seedlings Size class Germination Emergence Emergence rate percentage percentage index Small 100a 95.79a 78.15b Medium 100a 93.15a 70.25a Large 95a 93.30a 65.07a Means in the same column followed by the same letters are not significantly different (P < 0.05) Table 4.4. Effect of seed size on development of Vitellaria paradoxa seedlings Days to Seed size Sprouting PRE Bulging SA *SE Emergence EST Small 11.52c 21.52c 27.86c 37.99c 37.03 75.02b 97.42a Medium 9.15b 15.89b 25.69b 33.31b 32.16 65.47a 99.29a Large 6.98a 11.55a 20.76a 25.36a 35.34 60.70a 114.26b Mean 9.22 16.32 24.75 32.2 34.84 67.06 103.66 Means in the same column followed by the same letters are not significantly different (P < 0.05); PRE, Pseudoradicle elongation; SA, shoot appearance; SE, Shoot elongation; EST, Establishment and *SE = Emergence – SA The morphological features of V. paradoxa seedlings (Fig. 4.11) varied significantly as seed sized increased. The mean length of the pseudoradicles produced by small seeds was longer (6.75 cm) than those produced by medium seeds (5.12 cm) and large seeds (3.64 cm). These differences indicated that seed size had a significant (P < 0.05) influence on the length of the pseudoradicle (Table 4.5). Similarly, at the bulging stage, the difference between the mean length of taproot produced by seedlings of small seeds (10.29 cm) and that of those produced by medium seeds (7.65 cm) was University of Ghana http://ugspace.ug.edu.gh 72 significant. Although length of the taproot decreased as seed size increased, no significant difference existed between the length of taproot of seedlings produced by medium seeds (7.65 cm) and that of the seedlings produced by large seeds (6.73 cm). At the emergence stage, length of taproot also decreased as seed size increased with seedlings of small seeds producing significantly the longest taproots (31.61 cm). Table 4.5. Effects of seed size on morphological features of Vitellaria paradoxa seedlings at bulging and at emergence Length of Shoot Diameter of Size class taproot at taproot at height root pseudoradicle shoot bulging emergence collar Small 6.75c 10.29b 31.61c 8.45b 0.42c 0.15b Medium 5.12b 7.65a 24.15b 9.20ab 0.56b 0.19b Large 3.64a 6.37a 18.05a 10.05a 0.68a 0.25a Mean 5.17 8.10 24.60 9.23 0.55 0.20 Means in the same column followed by the same letters are not significantly different (P < 0.05) Contrary to length of pseudoradicle and taproot at bulging and at emergence, total shoot height increased significantly as seed size increased. The mean total shoot height of seedlings (Fig. 4.10C) produced by large seeds (10.05 cm) was significantly higher than that produced by seedlings of small seeds (8.45 cm) (Table 4.5). Similarly, at emergence stage, seed size had a significant (P < 0.05) effect on mean diameters of root collar and shoot because both growth parameters increased as seed size increased. Root collar diameter of seedlings of large seeds (0.68 cm) was wider than those of seedlings of medium seeds (0.56 cm) and small seeds (0.42 cm). Shoot diameter of seedlings produced by large seeds (0.25 cm) was significantly wider than that of seedlings of medium seeds (0.19 cm). However, no significant difference existed between shoot diameters of seedlings of medium and small seeds. University of Ghana http://ugspace.ug.edu.gh 73 tp 1.0 mm A B s cn rc C 15 mm D Fig. 4.11. Morphological features of Vitellaria paradoxa seedlings; A, Emerged seedling showing shoot above soil level; B, Seedling showing a 29 cm long taproot (tp); C, Seedling showing cotyledonary node (cn) and entire shoot (s); D, Seedling showing root crown (rc) that is buried 6.5 cm deep University of Ghana http://ugspace.ug.edu.gh 74 4.3.3. Effects of deshelling of seeds on the germination and growth of Vitellaria paradoxa seedlings Germination was observed in both the intact and deshelled seeds at the first 5 days after sowing. The germination percentage of the seeds varied non-significantly from 89.87 % for the intact, or control seeds to 94.76 % for the deshelled seeds (Table 4.6). Also, the mean germination time (11 days) for the intact seeds was not significantly different from that of the deshelled seeds (10 days). The last seedling, produced from an intact seed, emerged at 145 DAS with 2 shoots (Fig. 4.12). However, the percentage of seedlings that emerged from the deshelled seeds was significantly higher (94.67 %) than that for seedlings produced from the intact seeds (82.99 %). The emergence index of seedlings produced from the deshelled seeds (57 days) was significantly (P < 0.05) shorter than that of the intact seeds (75 days). Similarly, the corresponding emergence rate index of seedlings of the deshelled seeds (60 days) was significantly shorter than that of the seedlings of the control seeds (90 days). Table 4.6. Effects of deshelling of seeds on germination and emergence parameters of Vitellaria paradoxa seedlings Seed type Germination Mean Emergence Emergence Emergence percentage germ. Time percentage index (days) rate index (days) (days) Intact 89.87a 10.92a 82.99b 74.68b 90.00b Deshelled 94.67a 9.57a 94.67a 56.54a 59.70a Means in the same column followed by different letters are significantly different (P < 0.05) University of Ghana http://ugspace.ug.edu.gh 75 A B Fig. 4.12. Emergence of a trapped Vitellaria paradoxa seedling; A, Lateral shoot (ls) developing from the main shoot (ms) which was trapped by the pseudoradicle (ps); B, Cotyledonary node (cn) with two shoots; op, operculum; sc, seedcoat At 150 days after sowing, seedlings produced by the deshelled seeds produced a mean stem diameter of 0.41 cm which was non-significantly larger than that produced by seedlings of the intact seeds (Table 4.7). The mean height of seedlings produced by the deshelled seeds was 5.79 cm whilst that of the seedlings of the intact seeds was 5.04 cm. A significant difference in the height of the seedlings, therefore, occurred at 150 DAS. In contrast to seedling height, the mean numbers of leaves produced by the seedlings (6.62 and 5.55 for the seedlings from deshelled and intact seeds respectively) were statistically similar to each other. Similarly, no significant difference existed between the leaf areas of the seedlings. However, mean root crown University of Ghana http://ugspace.ug.edu.gh 76 diameter of the seedlings produced from deshelled seeds (0.78 cm) was significantly wider than that of seedlings of the intact seeds (0.42 cm). Table 4.7. Effect of deshelling of seeds on the growth of Vitellaria paradoxa seedlings at 150 days after sowing Stem Height Number of Leaf area Root collar Seed diameter (cm) leaves (cm2) diameter (cm) (cm) Intact 0.35a 5.04b 5.55a 38.04a 0.42b Deshelled 0.41a 5.79a 6.62a 51.91a 0.78a Means in the same column followed by the same letters are not significantly different (P < 0.05) Stem diameter of seedlings produced from the deshelled seeds increased more than two times at 240 DAS as compared with that recorded at 150 DAS. Consequently, the mean stem diameter of seedlings of deshelled seeds (1.12 cm) was significantly wider than that of seedlings of intact seeds (0.54 cm) (Table 4.8). Seedlings produced by the deshelled seeds had a mean height of 11.10 cm but it was non-significantly different from that recorded for seedlings of the control seeds (10.35 cm). Similarly, the mean numbers of the leaves produced by the seedlings were non-significantly different from each other (Table 4.8). In contrast, both leaf area and root collar diameter were significantly different. The leaf area of seedlings produced by the deshelled seeds (194.51 cm2) was significantly wider than that of seedlings of the control seeds (138.15 cm2). Also, the root collar diameter of seedlings of deshelled seeds (2.64 cm) was significantly greater than that of the seedlings of the intact seeds (1.58 cm). University of Ghana http://ugspace.ug.edu.gh 77 Table 4.8. Effect of deshelling of seeds on the growth of Vitellaria paradoxa seedlings at 240 days after sowing Stem diameter Height Number of Leaf area Root collar Seed (cm) (cm) leaves (cm2) diameter (cm) Intact 0.54b 10.35a 13.01a 138.15b 1.58b Deshelled 1.12a 11.10a 14.65a 194.51a 2.64a Means in the same column followed by different letters are significantly different (P < 0.05) Morphologically, some seedlings produced by the deshelled seeds developed 2 or more shoots above the soil level (Fig. 4.13). These seedlings still had their main growing axes (leaders) intact. Lateral shoots were produced only by the seedlings with much swollen root crowns (Fig. 4.13B). These lateral shoots were distinguished from axillary shoots by their upright growth pattern. Digging up the soil around the seedlings and examining them closely revealed that the lateral shoots were produced from two different sides, one of which was located belowground. The belowground lateral shoots were produced directly from the top of the root collar (Fig. 4.13B). These shoots appeared 2 to 4 weeks after the main axes had emerged. The second group of lateral shoots arose from the main growing axes on top of the root collar just at or above the soil level (Fig. 4.13C and D). Such growing axes from which they arose also appear swollen right from the root collar up to the soil level (Fig. 4.13C). These shoots appeared 6 to 8 weeks after the emergence of the leaders. These lateral shoots arising directing from the leaders, and not from the root crowns, grew so quickly that irrespective of their time of appearance, they soon became the tallest shoots in just 2–3 weeks suggesting that they may be an excellent source of scions for grafting or shoot tips for in vitro culture (Fig. 4.13C and D). University of Ghana http://ugspace.ug.edu.gh 78 A B C D Fig.4.13. Tuberous root crown of Vitellaria seedlings; A, Seedling produced by a deshelled seed (s) showing a tuberous root crown (rc); B, Lateral shoot (ls) growing from the root crown; C, Lateral shoot growing from the main axis below the soil level; D, A 15 day-old lateral shoot developing from a 240-day old main shoot above the soil level University of Ghana http://ugspace.ug.edu.gh 79 Apical growth of the seedlings occurred in 2 contrasting forms. In the commoner form, which was observed in 99 % of seedlings of the intact seeds, the shoot apical meristem (SAM) continuously grew upwards with or without producing axillary shoots or branches (monopodial growth). Those seedlings without branches were upright (Fig 4.14A) whilst those with one or more branches had their SAMs bent towards the direction of the branches (Fig. 4.14B). m A B Fig. 4.14. Vitellaria paradoxa seedlings showing monopodial growth; A, Seedling without branches; B, Seedling with 3 branches and a bent shoot apical meristem (m) In the second growth or branching pattern, which occurred in only seedlings produced by the deshelled seeds and accounted for 30 % of their population, a lateral shoot initially emerged from the tip of the shoot apical meristem and grew upwards at an angle less than 30° (Fig. 4.15A). The lateral shoot increased in length up to 12–19 cm before another one emerged from it and also grew upwards. At the time the other shoot emerged from it, the lower SAM from which it had emerged had also started producing new growth (Fig. 4.15 B). University of Ghana http://ugspace.ug.edu.gh 80 A B Fig. 4.15. Vitellaria paradoxa seedlings with two (A) and three (B) apical growing points; Lateral shoot (ls) produced on top of the shoot apical meristem (m); 1, 2 and 3 are three apical growing points 4.4. Discussion 4.4.1. Development of Vitellaria paradoxa seedlings Germination or sprouting of Vitellaria seeds commenced when the pseudoradicles protruded through the seedcoats via the opercula by pushing against and rupturing the overlying operculum caps. This observation is in contrast to germination in most dicotyledonous plants in which the radicle protrudes the seedcoat through the micropyle. Pérez, (2009) and Gong et al. (2005) observed that radicles or pseudoradicles of monocotyledonous plants commonly protrude from the seed via the opercula; thus, operculum protrusion in V. paradoxa (a eudicot) makes its University of Ghana http://ugspace.ug.edu.gh 81 germination unique. The large size of the pseudoradicle relative to that of the micropyle most likely precludes micropylar protrusion. Proximal protrusion (pseudoradicles appearing at the proximal or micropylar end of the seeds) represented 92 % of the protrusions with the remainder representing pseudoradicles protruding from other parts of the seedcoat. The embryos which swell to form the pseudoradicles are occasionally located in any other parts of seed besides the micropylar ends. This seed morphology explains why the pseudoradicles protruded from different parts of the seedcoat as observed in this study. This observation supports the finding of Msanga (1998) who recorded radicles of some tropical plants such as Hopea ferrea and Markhamia lutea protruding from the seedcoat via different parts other than the micropylar end in contrast to temperate plants whose radicles exclusively protrude the seedcoat from the micropyle. The development of V. paradoxa seedlings occurred in 7 distinct stages namely: sprouting, pseudoradicle elongation, bulging, shoot appearance, shoot elongation, emergence and establishment. Jackson (1968) described the germination of V. paradoxa but never identified the distinct stages as done in this study. Ugese et al. (2010) outlined only 5 stages because shoot appearance and shoot elongation were considered as a single stage whilst establishment was excluded. The pseudoradicle of a sprouted Vitellaria seed elongates to push the plumule and the radicle deeper into the soil. The morphology of the pseudoradicle clearly indicates that it is the fused petioles of the 2 cotyledons of the seeds and may therefore be described as the cotyledonary tube. Alabarce and Dillenburg (2012) and Burrows and University of Ghana http://ugspace.ug.edu.gh 82 Stockey (1994) used the term cotyledonary tube to describe a similar positively geotropic structure produced by cryptogeally germinating Araucaria angustifolia and A. bidwillii respectively. Another term that is suggested based on observations made in this study is cotyledonary axis, which is analogous with embryonic axis. A transverse section through the cotyledonary tube or pseudoradicle showed an outer sheath, laticiferous vessels and an inner hollow tube. Functionally, the outer sheath protects both the shoot and root apical meristems in its hollow tube. The laticiferous vessels translocate both latex sap and food reserves from the seed to the base of the bulge where the embryo develops. Germinating seeds that developed 6 laticiferous vessels produced 1 shoot each. However, germinating seeds with 7 or 8 lactiferous vessels typically produced 2 or more shoots (multiple shoots) above the soil level. The increased number of laticiferous vessels might have facilitated rapid translocation of seed reserves to the base of the pseudoradicle for use by the developing plumules or shoots which in turn caused them to develop branches. The number of pseudoradicles per germinating seeds also varied resulting in 1 seedling (monoembryony) or 2 seedlings (polyembryony). Monoembryonic seeds that produced 1 pseudoradicle characteristically produced 1 seedling. The single seedling may emerge above the soil with 1 or more shoots (multiple shoots). Multiple shoots were most likely produced by precocious branching of the plumules which resulted in 2 or more shoots appearing via the cotyledonary slits. Polyembryony which results in the production of 2 or more seedlings from 1 seed has already been reported in an Indian Sapotaceae Madhuca indica (Verma et al., 2009). It may, thus, account for the University of Ghana http://ugspace.ug.edu.gh 83 production of 2 seedlings from 1 seed as observed in this study. Multiple seedlings can readily be separated and replanted just as monoembryonic seedlings. Vitellaria paradoxa produces cryptogeal seedlings (Benti, 2009) but this seedling morphology remains unreported in any other sapotaceous species. Ibarra-Manríquez et al. (2001) reported that seedling traits are evolutionary conservative reflecting phylogenetic relationships. Therefore, Vitellaria might have evolved cryptogeal germination in response to bushfires occurring widely in the Shea Belt. Jackson (1968) observed that when Vitellaria seedlings push their plumules and radicles into the soil via the elongating fused cotyledonary petioles, they safely bury their root crowns in soil where they are protected against both bushfires and heat during the long dry season. Accordingly, Jackson (1974) described cryptogeal germination as plumule burying. Another benefit of cryptogeal germination is that it enables the seedlings to resprout quickly after grazing and fires (Clarkson and Clifford, 1987). In this study, seedlings of V. paradoxa were either cryptocotylar or phanerocotylar reserve types depending on cotyledon morphology. Seeds whose cotyledonary raphes run parallel to their embryos produced phanerocotylar seedlings, whereas those whose cotyledonary raphes were perpendicular to their embryos produced cryptocotylar seedlings. Sapotaceous species produce both cryptocotylar and phanerocotylar seedlings. For example, Cruz (2005) observed phanerocotylar seedlings in Pouteria pachycarpa whilst Mundhra and Paria (2009) described cryptocotylar seedlings in Madhuca indica. Essig (1987) reported that within any given genus, different subgenera could be distinguished by seedling morphologies. Thus, the production of University of Ghana http://ugspace.ug.edu.gh 84 both phanerocotylar and cryptocotylar seedlings may be useful for taxonomic classification within the genus Vitellaria. 4.4.2. Seed size and development of Vitellaria paradoxa seedlings The germination of Vitellaria seeds and the subsequent morphology of the seedlings were significantly influenced by sizes of the seeds sown. In contrast, Ugese et al. (2008b) observed lower emergence percentage when small-sized seeds were sown as compared with large ones. The high germination percentage of all the seeds used in this study could probably be due to their freshness. Jøker (2000) reported that Vitellaria seeds have to be sown fresh because they have a very high germination percentage at that state but it decreases with loss of moisture. The time to sprouting or germination of large seeds was significantly (P < 0.05) earlier (7 days) than those of medium and small seeds. Ugese et al. (2010) had earlier reported that V. paradoxa seeds often sprout within 7 days when sown. Large seeds germinated earlier most probably because their developing embryos have more seed reserves, and their widest area could have also enabled them to imbibe more moisture rapidly. Subsequently, the pseudoradicles of the seedlings produced by large seeds elongated faster producing bulges quicker than those of medium and small seeds. On the contrary, germination was delayed in small seeds, an observation which had also been reported by Ugese et al. (2007). The delayed germination may be attributed to little food reserves in the small seeds. The interval between the appearance of the shoot on the pseudoradicle and seedling emergence denoted by SE (shoot elongation) represented the single longest period in University of Ghana http://ugspace.ug.edu.gh 85 the skotomorphogenic growth of Vitellaria seedlings. Shoots of seedlings produced from small seeds took the longest period to elongate fully. This delayed shoot elongation was most likely caused by obstruction of the shoots by the pseudoradicles or by the seeds or by both. Due to the presence of more reserves, both medium and large seeds produced seedlings with more vigorous shoots, which were less obstructed. Without giving any methods, Ugese et al. (2010) suggested that V. paradoxa seedlings could emerge quickly by manipulating them at any of the stages between bulging and emergence. One of such methods is turning the seed into a different direction any time the shoots appear ventral on the pseudoradicle. The emergence index of seedlings produced by large seeds was significantly shorter than that of seedlings of small seeds. The faster emergence could be due to the presence of more food reserves, which were then mobilized for growth. According to Ugese et al. (2010), emergence of V. paradoxa seedlings ranged from 51 to 79 days after sowing. In this study, seedling emergence ranged from 61 to 75 days depending on the seed size. Contrastingly, Yidana (2004) reported that V. paradoxa seedlings emerge as early as 28 days after sowing. The differences in seedling emergence indices may be due to differences in their genetic make-up. Large seeds produced seedlings which exhibited vigorous growth in terms of height of shoots and diameters of the root collars and shoots. The vigorous growth observed in shoots developing from these seedlings may again be attributed to the large amount of food reserves available for the developing shoots. These seedlings also developed shorter taproots suggesting that they could easily be raised in nursery containers. Contrastingly, National Research Council (2006) reported that the containerization University of Ghana http://ugspace.ug.edu.gh 86 and subsequent transplanting of V. paradoxa seedlings are difficult because they develop long taproots which quickly outgrow the containers. However, seedlings produced from small and medium seeds developed longer taproots which usually adapt them well to their ecological conditions. Optimal partitioning theory predicts that plants allocate biomass preferentially to harness resources that are most limiting to growth (Kobe et al., 2010). In the Shea Belt, the most-limiting factor to plant growth is insufficient moisture which is usually accompanied by widespread bushfires (Jackson, 1974). The longer taproots of seedlings of small and medium seeds may enable them to reach moisture deep into the soil than those produced by large seeds. Also, the longer pseudoradicles of these seedlings pushed their shoot meristems located in the root crown deeper into the soil. Thus, seedlings produced from small and medium seeds are better equipped with the ability to resprout than those of large seeds. 4.4.3. Growth and morphology of Vitellaria paradoxa seedlings Although seedcoat or testa plays a protective role in seeds, it imposes dormancy, thereby inhibiting germination in most species (Debeaujon et al., 2000). Thus, the effect of seedcoat on germination and seedling growth of V. paradoxa was studied by removal of the testa. Deshelling did not have any significant (P > 0.05) effect on mean germination time and germination percentage. However, it significantly (P < 0.05) influenced emergence index, emergence percentage and emergence rate index. The non-significant difference in germination between the deshelled and intact seeds suggests that in V. paradoxa, the seedcoat does not impose dormancy on the embryo. University of Ghana http://ugspace.ug.edu.gh 87 Ugese et al. (2005) earlier concluded that any treatment aimed at breaking dormancy or achieving quicker seedling emergence should not be targeted at the seedcoat. Jøker (2000) emphasized that Vitellaria seeds require no pre-treatment besides extraction from the fruit because they are non-dormant. On the contrary, the removal of the shell allowed the cotyledons to swell quickly into a pseudoradicle which in turn elongated rapidly. This rapid pseudoradicle elongation might have accounted for the significantly higher emergence percentage, and the smaller emergence index and emergence rate index of seedlings produced by the deshelled seeds than those produced by the intact seeds. Smaller emergence and emergence rate indices indicate faster and synchronous seedling emergence (Ugese et al., 2011). The emergence of seedlings from the deshelled seeds suggests a highly synchronized seedling development. Therefore, removal of the shell of V. paradoxa seed has the potential of producing more uniform seedlings in the nursery. In this study, the last seedling which emerged 145 days after sowing (DAS) had its main axis obstructed by the pseudoradicle and this obstruction caused the seedling to produce 2 shoots. According to Jøker (2000), seedlings of the subspecies V. nilotica emerged faster than those of V. paradoxa in which emergence could even occur 150 DAS. The delayed emergence of V. paradoxa seedlings is caused by the shoots being obstructed from elongating freely through the soil, usually by the pseudoradicles and the seeds or by both. In non-obstructed seedlings, emergence never exceeded 70 DAS. One of the most likely factors explaining why seedlings produced by the deshelled seeds were less obstructed was that their pseudoradicles and seeds withered and shrank more rapidly. This rapid withering and shrinking of seed remnants might also University of Ghana http://ugspace.ug.edu.gh 88 explain why the emergence percentage of these seedlings was significantly higher than that of the seedlings produced by the intact seeds. Deshelling of the seeds did not produce any significant effect on height and number of leaves produced by the seedlings. In contrast, however, the mean stem and root crown diameters as well as the leaf area were significantly (P < 0.05) influenced by the removal of the seedcoat. The rate of leaf production is always proportional to apical growth in seedlings because leaves are produced from nodes which result in increased seedling height. Thus, the non-significant difference in seedling height also translated into the non-statistical difference in mean number of leaves produced by both seedling types. Asante et al. (2012) who earlier recorded a mean highest number of 4.6 leaves produced by 24-week-old V. paradoxa seedlings attributed that observation to their cryptogeal morphology. Ugese (2010) observed a positive, significant correlation between leaf size and stem diameter of tamarind (Tamarindus indica L.) seedlings. The significantly wider leaves of the seedlings of deshelled seeds produced more photosynthates which they eventually used to develop bigger stems. Cryptogeally germinating seeds have evolved a means of transferring the reserves in their surface-borne seeds and other aboveground parts onto underground sinks which appear swollen throughout their sapling stage and enable them to persist in the seedling bank of the soil (Dillenburg et al., 2010). Therefore, photosynthates mobilized by the seedlings were subsequently translocated below ground. The wider root crowns of seedlings produced by the deshelled seeds could have been due to the presence of more photosynthates which were produced using their wider leaves. University of Ghana http://ugspace.ug.edu.gh 89 Only seedlings produced by deshelled seeds which developed the larger root crown diameters produced both below- and aboveground lateral shoots. Diarrassouba et al. (2009) and Nikiema and Umali (2007) reported that V. paradoxa produces plagiotropic branches 4–7 years after sowing. By their orthotropic growth, lateral shoots observed in this study were clearly distinguished from branches. The production of 2 or more shoots which gives rise to several growing axes and branches numbering as many as 5 per seedling was considered phenomenal. Lovett and Haq (2000) explained that the current genetic make-up of Vitellaria reflects a millennium of anthropic selection due to its association with human habitation. Thus, semi- domestication is most likely being manifested by V. paradoxa seedlings which are less than one year old and yet produced 2 or more growing points or axillary shoots. 4.5. Conclusion The development of a Vitellaria paradoxa seedling occurs in 7 distinct stages termed: sprouting, elongation of the pseudoradicle, bulging, shoot appearance, shoot elongation, emergence and establishment with shoot elongation below ground being the longest. At shoot appearance stage, turning the pseudoradicles with the seeds into another direction may shorten the duration to emergence. Large seeds produced seedlings with vigorous aboveground growth and with shorter taproots. Medium and small seeds produced seedlings with longer pseudoradicles and taproots. Where seeds are to be planted in situ, medium seeds may be used whilst large seeds may be preferable for nursery establishment. Deshelling of seeds prevented seedling trapping and therefore resulted in higher germination percentage, shorter emergence index and a more synchronous seedling development. University of Ghana http://ugspace.ug.edu.gh 90 REFERENCES Adetimirin, V. O., Kim, S. K. and Szczech, M. (2006). Factors associated with emergence of shrunken-2-maize in Korea. Journal of Agricultural Science, 144:63–68. Alabarce, F. S. and Dillenburg, L. R. (2012). Resprouting ability and intensity after damage in seedlings of the large-seeded species Araucaria angustifolia, Brazilian Journal of Plant Physiology, 24(2):131–140. Amritphale, D. and Sharma, S. K. (2008). Seedlings of dicots: Form and function. Resonance, pp. 468–474. http://www.ias.ac.in/resonance/May2008/. Accessed on 23rd May 2013. Asante, W. J., Banidiyia, M. A. and Tom-Dery, D. (2012). Effect of planting depth on the germination and initial growth and development of shea (Vitellaria paradoxa C.F.Gaertn.). International Journal of Biosciences, 2(12):146–152. Baraloto, C. and Forget, P.-M. (2007). Seed size, seedling morphology, and response to deep shade and damage in neotropical rain forest trees. American Journal of Botany 94(6): 901–911. Benti, D. G. (2009). Ecology, composition and population structure of Vitellaria paradoxa subspecies nilotica (Kotschy) A.N.Henry, Chithra & N.C.Nair in Ethiopia: Implication for sustainability of production. Paper presented on naturally African: In support of African natural products development and marketing sub-regional consultative workshop for Eastern Africa, Nairobi, Kenya, p. 9. 18–20 May 2009. Bewley, J. D. and Black, M. (1994). Seeds: Physiology of development and germination. 2nd ed. New York: Plenum Press. 445 pp. Burrows, G. E. and Stockey, R. A. (1994). The developmental anatomy of cryptogeal germination in bunya pine (Araucaria bidwillii). International Journal of Plant Sciences, 155(5):519–537. Clarkson, J. R. and Clifford, H. T. (1987). Germination of Jedda multicaulis Clarkson,J.r. (Thymelaeaceae). An example of cryptogeal germination in the Australian flora. Australian Journal of Botany, 35(6):715(abs). Cruz, E. D. (2005). Quantitative characteristics of fruits and seeds of Pouteria pachycarpa Pires – Sapotaceae1. Revista Brasileira de Sementes, 27:159–164. Debeaujon, I., Le´on-Kloosterzie, K. M. and Koornneef, M. (2000). Influence of the testa on seed dormancy, germination, and longevity in Arabidopsis1. Plant Physiology, 122:403–413. Diarrassouba, N., Fofana, I. J., Issali, A. E., Bup, N. D. and et Sangaré, A. (2009). Typology of shea trees (Vitellaria paradoxa) using qualitative morphological traits in Côte d’Ivoire. Gene Conserve, 8(33):752–780. University of Ghana http://ugspace.ug.edu.gh 91 Dillenburg, L. R., Rosa, L. M. G. and Mósena, M. (2010). Hypocotyl of seedlings of the large-seeded species Araucaria angustifolia: an important underground sink of the seed reserves. Trees, 24:705–711. Essig, F. B. (1987). Seedling morphology and subgeneric classification in Clematis (Ranunculaceae). American Journal of Botany, 74:733–733. Flores, E. M. (2002). Tropical tree seed biology. In: Agricultural Handbook 721 (Vozzo, J. A. ed.). pp. 13–118. USDA Forest Service, Washington DC. Gong, X., Bassel, G. W., Wang, A., Greenwood, J. S. and Bewley, J. D. (2005). The emergence of embryos from hard seeds is related to the structure of the cell walls of the micropylar endosperm, and not to endo-b-mannanase activity. Annals of Botany, 96:1165–1173. Ibarra-Manríquez, G., Martinez, R. M. and Oyama, K. (2001). Seedling functional types in a lowland rain forest in Mexico. American Journal of Botany, 88: 1801–1812. Jackson, G. (1968). Notes on West African Vegetation – III: the seedling morphology of Butyrospermum paradoxum. Journal of West African Science Association, 13:215–222. Jackson, G. (1974). Cryptogeal germination and other seedling adaptations to the burning of vegetation in the savanna regions: the origin of the pyrophytic habit. New Phytologist, 73:771–780. Jøker, D. (2000). Vitellaria paradoxa Gaertn. F. Danida Forest Seed Centre, Seed Leaflet No.50, December, 2000. Kobe, R. K., Iyer, M. and Walters, M. B. (2010). Optimal partitioning theory revisited: Non-structural carbohydrates dominate root mass responses to nitrogen. Ecology, 91:166–179. Lovett, P. N. and Haq, N. (2000). Evidence for anthropic selection of the sheanut tree (Vitellaria paradoxa). Agroforestry Systems, 48:273–278. Maia, L. A., Maia, S. and Parolin, P. (2005). Seedling morphology of non-pioneer trees in Central Amazonian Várzea floodplain forests. Ecotropica, 11:1–8. Msanga, H. P. (1998). Dormancy and germination. In: Tropical tree seed manual (Vozzo, J. A. ed.). pp. 149–176. United States Department of Agriculture, Forest Service. Mundhra, N. and Paria, N. D. (2009). Epigeal cryptocotyly in Madhuca indica J.F. Gmel (Sapotaceae). International Journal of Botany, 5(2):200–202. National Research Council (2006). Shea (Vitellaria paradoxa). In: Lost Crops of Africa (Vietmeyer, N. D. ed.). Volume II: Vegetables. pp. 303–322. The National Academies Press, Washington, DC. University of Ghana http://ugspace.ug.edu.gh 92 Nikiema, A. and Umali, B. E., 2007. Vitellaria paradoxa C.F.Gaertn. http://database.prota.org/PROTAhtml/Vitellaria. Accessed on 22nd May 2013. Pérez, H. E. (2009). Promoting germination in ornamental palm seeds through dormancy alleviation. HortTechnology, 19(4):882–685. Tomlinson, P. B. (1987). Architecture of tropical plants. Annual Reviews of Ecology and Systematics, 18:1–21. Ugese, F. D., Ojo, A. A. and Bello, L. L. (2005). Effect of presowing treatment and nut orientation on emergence and seedling growth of seeds of shea butter tree (Vitellaria paradoxa). Nigerian Journal of Botany, 18:294–304. Ugese, F. D., Ojo, A. A. and Adedzwa, D. K. (2007). Effect of sowing depth and seed size on emergence and seedling growth of seeds of shea butter tree (Vitellaria paradoxa). Nigerian Journal of Botany, 20:93–102. Ugese, F. D., Baiyeri, K. P. and Mbah, B. N. (2008a). Leaf area determination of shea butter tree (Vitellaria Paradoxa C.F.Gaertn.). International Agrophysics, 22:167–170. Ugese, F. D., Baiyeri, K. P. and Mbah, B. N. (2008b). Effect of seed source and watering intervals on growth and dry matter yield of shea butter tree (Vitellaria paradoxa Gaertn. F.) seedlings. Bio-Research, 6:303–307. Ugese, F. D. (2010). Effect of nursery media on emergence and growth of tamarind (Tamarindus indica L.) seedlings. Journal of Animal and Plant Sciences, 8:999–1005. Ugese, F. D., Baiyeri, K. P. and Mbah, B. N. (2010). Determination of growth stages and seedling structures associated with slow emergence of shea butter tree (Vitellaria paradoxa C.F.Gaertn.) seedlings. Journal of Animal and Plant Sciences, 8(2):993–998. Ugese, F. D., Baiyeri, K. P. and Mbah, B. N. (2011). Variability in seedling growth of seeds of shea butter tree (Vitellaria Paradoxa C.F.Gaertn.) sourced from nine locations in Nigeria. Tree and Forestry Science and Biotech., 5(1):72–77. Verma, S. K., Rana, B. S. and Singh, B. P. (2009). Occurrence of cojointed twin seedlings in Madhuca latifolia Roxb. Indian Forester, 135: 571–573. Wu, R. and Hinckley, T. M. (2010). Phenotypic plasticity of sylleptic branching: Genetic design of tree architecture. Critical Reviews in Plant Sciences, 20(5): 467 (abs). Yidana, J. A. (2004). Progress in developing technologies to domesticate the cultivation of shea tree (Vitellaria paradoxa) in Ghana. Agricultural and Food Science Journal of Ghana, 3:249–267. University of Ghana http://ugspace.ug.edu.gh 93 CHAPTER 5 5.0 In vitro propagation of Vitellaria paradoxa 5.1. Introduction Although V. paradoxa is amenable to both sexual and asexual methods of reproduction, the long juvenile growth period of saplings, stem and root cuttings, and grafted seedlings makes ex vitro propagation methods agronomically unattractive (Yeboah et al., 2009). Moreover, the recalcitrant seeds are only available during April to September and lose viability easily through desiccation (Danthu et al., 2000). In addition, seeds germinate non-synchronously making it difficult to produce uniform seedlings on large-scale basis. As has already been reported for other sapotaceous species such as Argania spinosa (Nouaim et al., 2002) and Pouteria lacuma (Padilla et al., 2006), in vitro propagation may be an effective option for propagating this economically important species Several researchers including CRIG (2012) and Fotso et al. (2008) have attempted in vitro regeneration of V. paradoxa because every living cell is totipotent, but success rate has thus far been low. The low success rate may be attributed to saponins and latex sap which are abundantly exuded from any living part of the plant. These exudates reduce the success rate of grafting (Masters, 2002) and could be a source of in vitro contamination. For other reasons precisely unknown, somatic embryos of Vitellaria are difficult to transform into plantlets (Fotso et al., 2008) or the resulting plantlets are easily lost during post-flask management (CRIG, 2012). Despite the losses of in vitro plantlets, Adu-Gyamfi et al. (2012) successfully regenerated V. paradoxa plantlets by somatic embryogenesis using immature cotyledon explants. The cotyledon explants were cultured on Murashige and Skoog University of Ghana http://ugspace.ug.edu.gh 94 (MS) (1962) basal medium supplemented with 30.0 g/l sucrose, 2.4 g/l phytagel and 5 different concentrations of 2,4-Dichlorophenoxyacetic acid (2,4-D) (0–0.5 mg/l) to induce embryogenic calli. The embryogenic calli developed into somatic embryos when transferred onto a hormone-free MS medium amended as described earlier. In addition to non-synchronous development, the somatic embryos germinated poorly (15 %). Lovett and Haq (2013) also successfully regenerated Vitellaria using shoot tips explanted from 1 to 2-month-old seedlings and cultured on half- and quarter- strength MS medium supplemented with 6-benzyladenine (BA) and NAA. However, transfer of the regenerated shoots into a rooting medium resulted in less than 30 % successful root induction. Therefore, other protocols that may result in high success rate leading to the production of plantlets in large quantities are yet to be developed. Latex-related contaminations reduce the success of many in vitro techniques used for propagating sapotaceous species. Being laticiferous species, the Sapotaceae produce the latex sap as a secondary metabolite when they are developing making its total elimination from the growth media difficult (Bhore and Preveena, 2011). Repeated subculturing to minimize latex contamination not only increases cost of production but also leads to losses of plantlets. Thus, one of the most feasible options for successful in vitro regeneration of Vitellaria would be to identify and to excise any part of the seed or seedling that contains little or no latex for culture. The major objective of this study was therefore to develop an in vitro protocol for V. paradoxa. The specific objectives were to i. propagate V. paradoxa using whole seeds as explants ii. regenerate V. paradoxa plantlets using embryonic axis explants iii. regenerate V. paradoxa using rudimentary shoots of in vivo seedlings. University of Ghana http://ugspace.ug.edu.gh 95 5.2. Materials and methods 5.2.1. Collection of Vitellaria paradoxa fruits Mature fruits of V. paradoxa collected from farmlands and fallows at Tanina and Ga in the Wa West District of the Upper West Region of Ghana were used for this study (Section 3.2.1). 5.2.2. In vitro culture of intact seeds Fresh fruits were depulped manually to obtain seeds. Eighty (80) seeds were selected and thoroughly washed in distilled water and air-dried for 24 hours. They were then sterilized by immersing 0.2 % mercuric chloride (HgCl2) for 2 minutes followed by rinsing with 3 changes of sterile distilled water and cultured on 60 ml Murashige and Skoog (1962) basal medium in honey jars. The MS medium was prepared from stock solutions and amended with 30.0 g/l sucrose, 100.0 mg/l myo-inositol, 100.0 mg/l copper sulphate (CuSO4), 100.0 mg/l activated charcoal and 1.0, 2.0, 3.0 or 4.0 mg/l BAP. The pH of the medium was adjusted to 5.8 using 1.0 M KOH before the addition of 3.0 g/l phytagel and autoclaving at 121 ℃ for 15 minutes at 15 psi. The cultured explants were incubated in the growth room at a temperature of 25 ± 2 ℃ under 16-hour photoperiod with light provided by fluorescent tubes at an intensity of 3000 lux. Completely randomized design was used with 20 seeds per each of the 4 different concentrations of BAP. The number of days to sprouting, sprouting percentage, days to seedling emergence, seedling height and number of leaves were recorded. Seeds were considered sprouted when their pseudoradicles became visible. University of Ghana http://ugspace.ug.edu.gh 96 5.2.3. In vitro culture of deshelled seeds Eighty (80) freshly extracted seeds were deshelled by pressing the seeds in between pliers to rupture the shell (testa) at the dorsal side. A knife was then used to remove the remaining parts of the shell. The deshelled seeds were thoroughly washed and rinsed with 4 changes in distilled water and air-dried for 12 hours. They were soaked in 100.0 mg/l ascorbic acid solution for 3 hours to remove phenolic compounds in them and then sterilized using 0.2 % mercuric chloride (HgCl2) for 90 seconds and cultured on MS medium prepared as described in Section 5.2.2. The growth room conditions in which the cultured seeds were incubated, the experimental design and data collection were also the same as outlined in Section 5.2.2. 5.2.4. Identification and culture of embryonic axes Eighty (80) freshly extracted seeds prepared as already described (Section 5.2.3) were soaked in distilled water for 6 hours. The deshelled seeds were again soaked in 1.0 % TTC solution to identify their embryos (see Section 3.3). The stained embryo spots or axes were excised using a 1.0 cm diameter cork borer, trimmed to 0.8 cm long and then soaked in a 100.0 mg/l solution of ascorbic acid for 3 hours to remove any phenolic compounds. They were immersed in 0.2 % mercuric chloride (HgCl2) for 60 seconds and then rinsed with 4 changes of sterile distilled water. After the sterilization, the embryonic axes were cultured in honey jars containing 60 ml MS medium supplemented with 30.0 g/l sucrose, 100.0 mg/l myo-inositol, 100.0 mg/l copper sulphate (CuSO4), 100.0 mg/l activated charcoal, 3.0 g/l phytagel and 1.0, 2.0, 3.0 or 4.0 mg/l BAP prepared as described earlier and incubated under the same growth room conditions (see Section 5.2.2). The experimental design and data collection were the same as described in Section 5.2.2. University of Ghana http://ugspace.ug.edu.gh 97 5.2.5. In vitro culture of rudimentary shoots Freshly extracted seeds were washed using tap water and air-dried for 6 hours. The seeds were treated with HerculeR 50 SC (IPROCHEM Co. Ltd, Shenzhen, China) against termites and then sown 2 cm deep with the hilum down in polyethylene pots filled with a mixture of compost and sawdust in the ratio 5:1. The polyethylene pots with the sown seeds were placed in a plant barn and watered every other day. The germinated seeds were uprooted after 2–3 weeks and the bulged portions of the pseudoradicles containing the rudimentary shoots were excised using forceps. The excised portions measuring 1.5 cm each were sterilized by washing thoroughly under tap water and thereafter soaked in distilled water for 1 hour to remove any latex. They were immersed in 0.2 % mercuric chloride (HgCl2) for 90 seconds and rinsed with 3 changes of sterile distilled water. They were then dissected to remove the rudimentary shoots which were cultured in honey jars containing 40 ml MS basal medium supplemented with 30.0 g/l sucrose, 100.0 mg/l myo-inositol, 100.0 mg/l copper sulphate and 1.0, 2.0, 3.0 or 4.0 mg/l BAP and 0.0, 0.1, 0.2 or 0.4 mg/l NAA. The culture medium was prepared as described in Section 5.2.2. The cultured explants were incubated in the growth room under the same conditions as described in Section 5.2.2. The factorial experimental design comprising 2 factors was used. The duration to response of explants to culture, height of developing shoots, and the number of leaves, root and shoots produced per explant were recorded. Response of explants to culture was determined by change in colour (pink to light green). Changes in colour of the cultured explants were described using HTML Colour Chart (http://www.html- color-names.com/color-chart.php). Observations were made at 5-day intervals beginning from the day of inoculation. University of Ghana http://ugspace.ug.edu.gh 98 5.2.6. Data analysis Data were subjected to analysis of variance using Genstat statistical package (9th Edition). Means were separated where appropriate using the least significant difference (LSD) test. 5.3. Results 5.3.1. In vitro germination of intact seeds Between 5 and 10 days after culture, almost all the seeds showed signs of germination because their seedcoats had ruptured visibly. Thereafter, they exuded phenolics excessively into the culture medium, which subsequently hindered the germination process. Consequently, all the cultured seeds died after 50 days of culture (Fig. 5.1). Fig. 5.1. Intact Vitellaria paradoxa seed cultured on MS basal medium supplemented with 2.0 mg/l BAP showing visible seedcoat rupture (arrowed) 10 days after culture University of Ghana http://ugspace.ug.edu.gh 99 5.3.2. In vitro germination of deshelled seeds In contrast to intact seeds, deshelled seeds sprouted within 7 days of culture on BAP amended medium. The seeds sprouted when their cotyledons swelled and elongated into pseudoradicles at the proximal ends (Fig. 5.2). However, the number of seeds that sprouted varied depending on the concentration of BAP in the culture medium. The duration to sprouting significantly delayed as the concentration of the BAP increased suggesting that higher BAP concentration may be detrimental to sprouting (Fig. 5.3A). Also, the percentage of seeds that sprouted decreased as the concentration of the BAP in the medium increased. Eighty percent (80 %) of the deshelled seeds cultured on 1.0 mg/l BAP supplemented medium sprouted, but the sprouting decreased to 50 % at 3.0 or 4.0 mg/l BAP. However, no significant difference existed in the percentage sprouting (Fig. 5.3B). Due to intense phenolic exudation, the sprouted seeds could not develop into plantlets after 150 days of incubation. Fig. 5.2. Sprouted Vitellaria paradoxa seed cultured on MS basal medium amended with 1.0 mg/l BAP after 35 days of culture; Bar = 1.0 cm University of Ghana http://ugspace.ug.edu.gh 100 90 12 b b 80 a 10 70 a 60 8 a a a a50 6 40 4 30 20 2 10 0 0 1.0 2.0 3.0 4.0 1.0 2.0 3.0 4.0 Concentration of BAP (mg/l) Concentration of BAP (mg/l) Fig. 5.3A Fig. 5.3B Fig. 5.3. Days to sprouting (A) and percentage sprouting (B) of deshelled Vitellaria paradoxa seeds cultured on MS basal medium supplemented with 1.0–4.0 mg/l BAP; Bars with the same letters are not significantly different (P < 0.05) 5.3.3. Response of embryonic axes to in vitro culture Excised TTC stained embryonic axes cultured on MS medium supplemented with BAP swelled and developed pseudoradicles between 7 and 12 days after culture (Fig. 5.4). Sprouting was significantly earlier (6 days) on the medium supplemented with 1.0 mg/l BAP than on the other concentrations. On the medium with 4.0 mg/l BAP, embryonic axis explants took approximately twice the time used by those cultured on 1.0 mg/l BAP to sprout (Fig. 5.5A). Latex or phenolics were not exuded in the culture medium as compared with the culture of deshelled seeds where whitish exudates accumulated on the surface of the growth medium. The percentage sprouting also varied depending on the concentration of BAP in the culture medium. It was significantly higher (90 %) on the medium with 2.0 mg/l BAP and lower (50 %) on the medium with 4.0 mg/l BAP (Fig. 5.5B). After 70 days of culture, the explants became brown and eventually wilted without any sign of emergence of plantlets. Days to sprouting Percentage sprouting University of Ghana http://ugspace.ug.edu.gh 101 ps A B Fig. 5.4. Embryonic axis culture of Vitellaria paradoxa; A, Embryonic axis explants showing the embryo (arrowed); B, Sprouted embryonic axis cultured on MS basal medium amended with 1.0 mg/l BAP 10 days after culture; ps, pseudoradicle 12 100 c a 10 bc 90 ab 80 8 ab 70 bc a 606 c50 40 4 30 2 20 10 0 0 1.0 2.0 3.0 4.0 1.0 2.0 3.0 4.0 Concentration of BAP (mg/l) Concentration of BAP (mg/l) Fig. 5.5A Fig. 5.5B Fig. 5.5. Days to sprouting (A) and percentage sprouting (B) of Vitellaria paradoxa embryonic axes cultured on MS basal medium supplemented with 1.0–4.0 mg/l BAP; Bars with the same letters are not significantly different (P < 0.05) Days to sprouting Percentage sprouting University of Ghana http://ugspace.ug.edu.gh 102 5.3.4. In vitro regeneration of rudimentary shoots Although the cultured deshelled seeds of V. paradoxa sprouted by producing long pseudoradicles, they never developed shoots throughout the culture period. Thus, rudimentary shoots in the bulges of the pseudoradicles from in vivo seedlings were excised and cultured on MS medium supplemented with BAP and NAA at different concentrations. The rudimentary shoot explants responded to culture when their scale leaves turned from pink to green within 5 to 15 days after culture (Table 5.1). Days to response to culture were influenced by the presence of both BAP and NAA in the culture medium. An increase in BAP concentration from 1.0 mg/l enhanced the rate of response to culture conditions. All explants cultured on 1.0 mg/l BAP responded within 2 weeks independent of the NAA concentration in the culture medium. But on the medium with the highest NAA concentration (0.4 mg/l), response was delayed to 15 days compared with 8 and 10 days when NAA concentration was decreased to 0.2 or 0.1 mg/l respectively (Table 5.1). At 2.0, 3.0 or 4.0 mg/l BAP, days to response was significantly reduced to 5 or 6 days (within a week) except when explants were cultured on a medium supplemented with 0.4 mg/l NAA. At this highest NAA concentration, response was delayed to 13 days for 2.0 mg/l BAP, 8 days for 3.0 mg/l BAP and 7 days for 4.0 mg/l BAP. Statistical analysis showed significant interactions between BAP and NAA on duration to response of explants to culture (Appendix 7.38). University of Ghana http://ugspace.ug.edu.gh 103 Table 5.1. Effects of BAP and NAA on the response to culture, height and leaf production of rudimentary shoot explants 15 and 45 days after culture Mean shoot height Mean number of Mean number of Days to response Mean shoot height Conc. of BAP Conc. of NAA 15 days after leaves 15 days after leaves 45 days after to culture 45 days after culture (mg/l) (mg/l) culture culture culture 1.0 0.0 11.67 ± 1.56de 1.33 ± 0.19cde 1.33 ± 0.41bc 1.93 ± 0.20f 2.67 ± 0.61hi 0.1 10.00 ± 1.56cd 1.47 ± 0.19cd 1.67 ± 0.41bc 2.50 ± 0.20f 4.00 ± 0.61efg 0.2 8.33 ± 1.56bc 1.20 ± 0.19def 2.00 ± 0.41b 1.90. ± 0.20f 3.33 ± 0.61fgh 0.4 15.00 ± 1.56ef 0.87 ± 0.19f 1.00 ± 0.41c 1.27 ± 0.20g 2.00 ± 0.61i 2.0 0.0 5.00 ± 1.56a 1.27 ± 0.19cde 2.00 ± 0.41b 3.10 ± 0.20de 5.00 ± 0.61bcde 0.1 5.00 ± 1.56a 2.60 ± 0.19a 3.00 ± 0.41a 4.60 ± 0.20a 5.67 ± 0.61abc 0.2 6.67 ± 1.56ab 2.13 ± 0.19b 2.00 ± 0.41b 3.77 ± 0.20bc 6.67 ± 0.61a 0.4 13.33 ± 1.56ef 1.00 ± 0.19ef 1.33 ± 0.41bc 3.00 ± 0.20e 3.00 ± 0.61ghi 3.0 0.0 5.00 ± 1.56a 1.23 ± 0.19cde 1.67 ± 0.41bc 3.53 ± 0.20bc 5.33 ± 0.61bcd 0.1 5.00 ± 1.56a 1.43 ± 0.19cd 1.67 ± 0.41bc 3.73 ± 0.20bc 4.67 ± 0.61cde 0.2 5.00 ± 1.56a 1.57 ± 0.19c 2.00 ± 0.41b 3.43 ± 0.20cd 5.00 ± 0.61bcde 0.4 8.33 ± 1.56bc 0.87 ± 0.19f 1.33 ± 0.41bc 2.53 ± 0.20f 2.67 ± 0.61hi 4.0 0.0 5.00 ± 1.56a 1.43 ± 0.19cd 3.00 ± 0.41a 3.40 ± 0.20cde 5.00 ± 0.61bcde 0.1 5.00 ± 1.56a 1.53 ± 0.19cd 2.00 ± 0.41b 3.90 ± 0.20b 6.00 ± 0.61ab 0.2 5.00 ± 1.56a 1.03 ± 0.19ef 2.00 ± 0.41b 3.00 ± 0.20e 4.33 ± 0.61def 0.4 6.67 ± 1.56ab 1.03 ± 0.19ef 1.00 ± 0.41c 2.53 ± 0.20f 3.33 ± 0.61fgh Means in the same column followed by the same letters are not significantly different (P < 0.05) University of Ghana http://ugspace.ug.edu.gh 104 As shown by the height of shoots and number of leaves, subsequent growth of the plantlets was also strongly influenced by the growth regulators (BAP and NAA) in the culture medium. After 15 days of culture, the height of the shoots decreased as the concentration of both BAP and NAA in the culture medium increased (Table 5.1). The optimal concentration of the growth regulators for rapid shoot development was 2.0 mg/l BAP when NAA concentration was 0.1 mg/l (Fig. 5.6). At this concentration, height of the plantlets was significantly higher (2.60 cm) than those for all the remaining treatments where the height did not exceed 1.5 cm. The number of leaves produced by the plantlets at 15 days of culture also decreased as the concentration of both growth regulators increased. Similar to shoot height, the optimal concentration for leaf production was 2.0 mg/l BAP and 0.1 mg/l NAA. This concentration of the growth regulators resulted in significantly higher number of leaves (3), albeit being the same as the number of leaves produced at BAP/NAA combination of 4.0 and 0.0 mg/l respectively (Table 5.1). At this concentration of BAP (4.0 mg/l), leaf production clearly decreased as NAA concentration increased resulting in the least number of leaves (1). At 45 days of culture, height of plantlets almost doubled at the optimal concentration of 2.0 mg/l BAP and 0.1 mg/l NAA. For leaf production, a BAP/NAA combination of 2.0 and 0.2 mg/l respectively resulted in the highest number of leaves (7). However, no significant difference occurred between this highest number of leaves and that produced by a BAP and NAA combination of 4.0 mg/l and 0.1 mg/l respectively. University of Ghana http://ugspace.ug.edu.gh 105 1 1 A B 2 2 10.0 mm 5.0 mm C D Fig. 5.6. In vitro regeneration of V. paradoxa using rudimentary shoots; A, Pseudoradicle showing the bulge (1); B, Excised bulges of the seedlings containing rudimentary shoots (1); C, Dissected bulge showing rudimentary shoot (2) used as explant; D, Regenerated plantlet from a rudimentary shoot 15 days after culture on MS medium supplemented with 2.0 mg/l BAP and 0.1 mg/l NAA Morphologically, the plantlets produced showed distinct shoots with leaves similar to those produced by in vivo seedlings except that they had no roots. Some of the plantlets even showed signs of producing multiple shoots (Fig. 5.7A). In contrast to in vivo seedlings, rudimentary shoot explants produced scale leaves all of which expanded upon culture becoming green and fully functional leaves. University of Ghana http://ugspace.ug.edu.gh 106 A B Fig. 5.7. Regenerated Vitellaria paradoxa shoots cultured on MS basal medium supplemented with 2.0 mg/l BAP and 0.2 mg/l NAA at 30 days after culture; A, Shoot with a lateral shoot (arrowed) just beginning to develop; B, Shoot with expanded leaves 5.4. Discussion In vitro regeneration of V. paradoxa will immensely accelerate the domestication of this economically important species and thereby improve the socioeconomic life of the majority of people especially women whose livelihood depends on it. However, to-date, little effort has been made to domesticate, or to establish plantations of shea because the tree grows in the wild and has largely been regarded as such (Gwali et al., 2012). Vitellaria paradoxa has the ability to grow abundantly on marginal soils and to improve soil quality due to annual leaf shedding (Dianda et al., 2009). With these potentials, V. paradoxa should be urgently considered for use for both afforestation and reafforestation programmes because the vegetative cover is being rapidly destroyed by bushfires. This forestation programme may, however, require the use of alternative modes of propagating this species. This study was thus aimed at developing efficient in vitro regeneration techniques for V. paradoxa. University of Ghana http://ugspace.ug.edu.gh 107 Plantlets were never produced, albeit signs of sprouting and development of pseudoradicles, when intact and deshelled seeds were cultured respectively. For deshelled seeds, pseudoradicle development and elongation were significantly (P < 0.05) influenced by the presence of BAP in the culture medium because the days to sprouting were reduced from 2 weeks on high BAP (3.0 or 4.0 mg/l) amended medium to 1 week on low concentration (1.0 or 2.0 mg/l) of the growth regulator. Similarly, the percentage number of seeds that developed pseudoradicles was also significantly reduced as the BAP concentration increased. At higher concentrations, phytotoxicity of BAP on plant tissues has already been observed in plants such as Gladiolus (Shaheenuzzaman et al., 2011). Thus, the delay in sprouting and decrease in sprouting percentage as BAP concentration increased might also be due to the phytotoxic effects of the growth regulator on V. paradoxa. Also, the failure of sprouted seeds to develop shoots may be attributed to the phenolic compounds they exuded, which eventually caused them to turn brown. Exudates of phenolic compounds cause excessive browning or necrosis and subsequent death of explants (Abdelwahd et al., 2008; Dibax et al., 2005). The phenolics are usually produced as metabolites during the growth of the plant (Arnaldos et al., 2001) and in the case of V. paradoxa, their production rate was probably high because some of the sprouted deshelled seeds turned brown as early as 28 days after culture despite pre- soaking in antioxidant solutions for 3 hours. Another contributing factor to the poor development of shoot in vitro may be the presence of latex. Bhore and Preveena (2011) observed 100 % contamination of explants during in vitro propagation of Mimusops elengi (an Asian Sapotaceae) University of Ghana http://ugspace.ug.edu.gh 108 because of the latex excreted by the explants. They concluded that the latex could hardly be taken out from the culture medium completely because the plant produced it during growth. Consequently, the contamination of the medium with white sap visible on the exposed parts of the pseudoradicles could have been caused by the latex produced by the explants as they were developing. The culture of embryonic axes stained with TTC also resulted in a higher production of pseudoradicles than those observed in the deshelled seeds. The response of the embryonic axes to BAP in the culture medium was also different from those of the intact seeds. The optimal concentration of BAP that enhanced sprouting was 2.0 mg/l, but sprouting decreased to almost half (50 %) at the highest concentration of BAP (4.0 mg/l) again suggesting the phytotoxic effects of the growth regulator. Yet again, full plantlets were never regenerated in embryonic axis culture although latex-related contamination was eliminated. In contrast, whole seedlings were successfully regenerated in vitro from zygotic embryos of M. elengi (Bhore and Preveena, 2011). One of the factors that most likely accounted for the inability to regenerate whole plantlets could be the cryptogeal morphology of Vitellaria seedlings. The cryptogeal morphology which indicates embryo development taking place in the pseudoradicle as germination progresses could place a genetic barrier on embryo development outside the pseudoradicle. Cryptogeal seedling development remains unreported for any other Sapotaceae (Jackson, 1974). Although embryonic axes failed to produce full plantlets, rudimentary shoots excised from the pseudoradicles and cultured on the same MS medium developed plantlets University of Ghana http://ugspace.ug.edu.gh 109 with well-distinct shoots and leaves. The development of the plantlets was influenced by the presence of BAP and NAA in the regeneration medium. The growth regulators had a significant effect (P < 0.05) on the response of rudimentary shoot explants to culture conditions and subsequent development of both shoots and leaves. Increasing the concentration of BAP produced a significantly faster response and subsequently resulted in faster growth in terms of both shoot height and number of leaves produced. Effects of NAA at lower concentrations (0.1 or 0.2 mg/l) were stimulating on both growth parameters, whilst the highest dose (0.4 mg/l) inhibited growth. A BAP concentration of 2.0 mg/l was optimal giving significantly taller shoots and a higher number of leaves. A combination of 2.0 mg/l BAP with 0.1 or 0.2 mg/l NAA giving a cytokinin/auxin ratio ranging from 10:1 to 20:1 was the optimum concentration of the growth regulators for both shoot and leaf production. Lovett and Haq (2013) had earlier observed maximum shoot regeneration of V. paradoxa at a high BA/NAA ratio between 5:1 and 50:1. According to Kalidass and Mohan (2009), cytokinins have phytotoxic effects on shoot production and on growth of many plants, and this phytotoxicity may also be true for V. paradoxa because shoot height and leaf production were significantly reduced at higher BAP concentrations (3.0 or 4.0 mg/l). On the medium with BAP/NAA combination of 2.0 mg/l and 0.2 mg/l respectively, the shoots grew up to 6.67 cm tall 45 days after culture. Asante et al. (2012) recorded a mean seedling height of 4.66 cm at 168 days after sowing suggesting that above- ground growth of V. paradoxa seedlings under ex vitro nursery conditions is slow. Thus, the production of plantlets as tall as 7.0 cm in just 45 days of culture may University of Ghana http://ugspace.ug.edu.gh 110 indicate accelerated growth of the plants in culture. In vitro propagation thus has the potential to speed up regeneration of V. paradoxa for domestication. Despite the rapid growth of shoots, the regenerated shoot plantlets produced no roots after 45 days of culture. Lovett and Haq (2013) also observed no root development when axillary shoot tips of V. paradoxa were cultured for 42 days on a medium that contains a high combination of BA/NAA. The failure of plantlets to produce roots as observed in this study might probably be due to the high cytokinin concentration in the medium, but this claim requires further investigation. In vitro regeneration of S. dulcificum by Ogunsola and Ilori (2008) also demonstrates the need for high auxin/cytokinin ratio for root production, at least in the Sapotaceae. 5.5. Conclusion Rapid in vitro regeneration of V. paradoxa was achieved by using rudimentary shoots as explant. A BAP concentration of 2.0 mg/l in combination with 0.1 or 0.2 mg/l NAA was optimal for shoot and leaf production. Rudimentary shoots may, therefore, be ideal explants for in vitro regeneration of V. paradoxa via organogenesis and somatic embryogenesis. This protocol is both simple and cost-effective because it significantly minimizes latex-associated contamination. The rudimentary shoot explants also have the potential of producing multiple shoots and may thus enable superior genotypes to be rapidly multiplied on large-scale basis for farmers. University of Ghana http://ugspace.ug.edu.gh 111 REFERENCES Abdelwahd, R., Najat, H., Mustapha, L. and Sripada, M. U. (2008). Use of an adsorbent and antioxidants to reduce the effects of leached phenolics in in vitro plantlet regeneration of faba bean. African Journal of Biotechnology, 7 (8):997–1002. Adu-Gyamfi, P. K. K., Barnor, M. T., Dadzie, A. M., Lowor, S., Opoku, S. T., Opoku-Ameyaw, K., Bissah, M. and Padi, F. K. (2012). Preliminary investigation on somatic embryogenesis from immature cotyledon explants of Shea (Vitellaria paradoxa C.F.Gaertn.). Journal of Agricultural Science and Technology B, 2:1171–1176. Arnaldos, T. L., Munoz, R., Ferrer, M. A. and Calderon, A. A. (2001). Changes in phenol content during strawberry (Fragaria × ananasa, cv. Chandler) callus culture. Physiologia Plantarum, 113:315–322. Asante, W. J., Banidiyia, M. A. and Tom-Dery, D. (2012). Effect of planting depth on the germination and initial growth and development of shea (Vitellaria paradoxa C.F.Gaertn.). International Journal of Biosciences, 2(12):146–152. Bhore, S. J. and Preveena, J. (2011). Micropropagation of Mimusops elengi Linn.: Identification of suitable explant and comparative analysis of immature zygotic embryos response on three basal media. American-Eurasian Journal of Agricultural and Environmental Sciences, 10(2):216–222. CRIG (2012). Shea/cashew development. Progress report for 2011/12 and workplan for 2012/13. May, 2012, pp. 318–373. Tafo, Ghana. Danthu, P., Gueye, A., Boye, A., Bauwens, D. and Sarr, A. (2000). Seed storage behaviour of four Sahelian and Sudanian tree species (Boscia senegalensis, Butyrospermum parkii, Cordyla pinnata and Saba senegalensis). Seed Science Research, 10:183–187. Dianda, M., Bayala, J., Diop, T. and Ouédraogo, S. J. (2009). Improving growth of shea butter tree (Vitellaria paradoxa C.F.Gaertn.) seedlings using mineral N, P and arbuscular mycorrhizal (AM) fungi. Biotechnology, Agronomy, Society and Environment, 13(1):93–102. Dibax, R., Eisfeld, C. L., Cuquel, F. L., Koehler, H. and Quoirin, M. (2005). Plant regeneration from cotyledonary explants of Eucalyptus camaldulensis. Scientia Agricola, (Piracicaba, Brazil), 62(4):406–412. Fotso, S., Tchinda, N. D. and Ndoumou, D. O. (2008). Comparison of first stages of somatic embryogenesis in Baillonella toxisperma and Vitellaria paradoxa. Biotechnology, Agronomy, Society and Environment, 12(2):131(abs). University of Ghana http://ugspace.ug.edu.gh 112 Gwali, S., Nakabonge, G., Okullo, J. B. L., Eilu, G., Nyeko, P. and Vuzi, P. (2012). Morphological variation among shea tree (Vitellaria paradoxa subsp. nilotica) ‘ethnovarieties’ in Uganda. Genetic Resources and Crop Evolution, 59 (8):1883–1898. http://www.html-color-names.com/color-chart.php. Accessed on 12th June 2012. Kalidass, C. and Mohan, V. R. (2009). In vitro rapid clonal propagation of Phyllanthus urinaria L. (Euphorbiaceae): A medicinal plant. Researcher, 1(14):56–61. Jackson, G. (1974). Cryptogeal germination and other seedling adaptations to the burning of vegetation in the savanna regions: the origin of the pyrophytic habit. New Phytologist, 73:771–780. Lovett, P. N. and Haq, N. (2013). Progress in developing in vitro systems for shea tree (Vitellaria paradoxa C.F.Gaertn.) propagation. Forests, Trees and Livelihoods, 22(1):60–69. Masters, E. T. (2002). The Shea Resource: Overview of research and development across Africa. Paper presented at the international workshop on processing and marketing of shea products in Africa, 4–6 March, 2002. pp 13–29. Dakar, Senegal. Murashige, T. and Skoog, F. (1962). A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiologia Plantarum, 15(3):473–497. Nouaim, R., Mangin, G., Breuil, M. C. and Chaussod, R. (2002). The argan tree (Argania spinosa) in Morocco: propagation by seeds, cuttings and in vitro techniques. Agroforestry Systems, 54(1):71–81. Ogunsola, K. E. and Ilori, C. O. (2008). In vitro propagation of miracle berry (Synsepalum dulcificum Daniel) through embryo and nodal cultures. African Journal of Biotechnology, 7(3):244–248. Padilla, I. M. G., Carmona, E., Westendorp, N. and Encina, C. L. (2006). Micropropagation and effects of mycorrhiza and soil bacteria on acclimatization and development of lucumo (Pouteria lucuma R. and Pav.) var. La Molina. In vitro Cellular and Developmental Biology-Plant, 42(2):193(abs). Shaheenuzzaman, M., Haque, M. S., Karim, M. M. and Noor, Z. U. (2011). In vitro shoot proliferation and development of micropropagation protocol from leaf disc of Gladiolus. Journal of the Bangladesh Agricultural University, 9(1):21– 26. Yeboah, J., Lowor, S. T. and Amoah, F. M. (2009). The rooting performance of shea (Vitellaria paradoxa C.F.Gaertn) stem cuttings as influenced by wood type, sucrose and rooting hormone. Scientific Research and Essay, 4(5):521–525. University of Ghana http://ugspace.ug.edu.gh 113 CHAPTER 6 6.0. General conclusions and recommendations 6.1. Conclusions Vitellaria paradoxa is an indigen of sub-Saharan Africa and became a plant of international trade in the early 1950s when the butter extracted from its kernel was discovered as one of the best cocoa butter substitutes. Much of the research work on V. paradoxa has been primarily focused on exploiting it. Despite its socioeconomic value of sustaining the livelihood of women and children, it still remains undomesticated largely due to the non-availability of reliable methods of producing planting materials commercially and its long gestation period. This study sought to develop appropriate techniques for propagating V. paradoxa as part of initial efforts towards hastening its domestication. The following conclusions have been made from the study: 1. The embryo of Vitellaria paradoxa seed is located at the proximal end where it is embedded in copious amount of latex and fatty tissues making its identification and isolation difficult. 2. The seed is proximally syncotylous which accounts for the production of long and fused cotyledonary petioles (pseudoradicles) during germination. 3. Vitellaria paradoxa seedlings developed through 7 distinct stages namely: sprouting of the seed, elongation of the pseudoradicle, formation of a bulge in which the plumule develops into a rudimentary shoot, appearance of the rudimentary shoot on the pseudoradicle, elongation of the shoot, emergence and establishment. University of Ghana http://ugspace.ug.edu.gh 114 4. Seed size had significant effect on seedling development because large seeds produced seedlings with shorter taproots and vigorous above-ground growth, whilst small and medium seeds produced seedlings with longer taproots. 5. Seedlings produced from deshelled seeds emerged faster and more synchronously than those produced by intact seeds. 6. Seedlings produced by the deshelled seeds developed large tuberous root crowns from which lateral shoots developed. These seedlings also produced several axillary shoots making them shrubby. 7. Amongst the 3 different types of explants cultured, only in vivo rudimentary shoots successfully and rapidly developed into plantlets. 8. Murashige and Skoog basal medium supplemented with 2.0 mg/l BAP and 0.1 or 0.2 mg/l NAA was the optimum concentration of the growth regulators for shoot regeneration because 86 % of the cultured rudimentary shoot explants developed into plantlets. 6.2. Recommendations Based on the findings of this study, the following recommendations would be very useful for developing appropriate techniques for propagating Vitellaria paradoxa. 1. The effect of seed storage on viability should be investigated. This study and its findings will facilitate international germplasm exchange. 2. Further investigations are required to achieve rooting of regenerated rudimentary shoots in vitro. 3. Rudimentary shoots should be used as explants for somatic embryogenesis. 4. The induction of multiple shoots from rudimentary shoot explants should be further investigated. University of Ghana http://ugspace.ug.edu.gh 115 APPENDICES 7.0. Appendices Appendix 7.1A. Seeds of Vitellaria paradoxa f hc A B C A, Seed showing the bitter-tasting funiculus (f); B, Seed showing the woody hilar cup (hc) eaten by xylophagus insects; C, Naturally dispersed seed lying hilum down Appendix 7.1B. Stages of the development of Vitellaria paradoxa seedlings A B C D E F G A, Sprouting; B, Pseudoradicle elongation; C, Bulging; D, Shoot appearance; E, Shoot elongation; F, Emergence; G, Establishment; Stages A–E are the skotomorphogenic growth stages; F and G are the photomorphorgenic growth stages; oc, operculum cap; ps, pseudoradicle; b, bulge; tr, true root; s, shoot; sl, soil level; es, exhausted seed University of Ghana http://ugspace.ug.edu.gh 116 Appendix 7.2. Production of multiple shoots from monoembryonic seeds A B C D A. Monoembryonic seed showing an unexserted embryo (arrowed) at the proximal end B. Sprouted seed showing a pseudoradicle (arrowed) C. Three shoots appearing from the cotyledonary slit D. Three shoots that have emerged above the soil; cotyledonar raphe (arrowed) University of Ghana http://ugspace.ug.edu.gh 117 Appendix 7.3. Production of multiple seedlings from polyembryonic seeds A B C D E A. Polyembryonic seed showing two embryos (arrowed) B. Sprouted polyembryonic seed showing two appressed pseudoradicles C. Sprouted polyembryonic seed showing two free pseudoradicles D. Two seedlings conjoined to each other at the root–shoot junction (arrowed) E. Two free seedlings still attached to the seed by their pseudoradicles (arrowed) University of Ghana http://ugspace.ug.edu.gh 118 Appendix 7.4. ANOVA for effect of seed size on germination percentage Source of variation df SS MS v.r. F Seed size 2 0.14222 0.07111 2.81 0.173 Replication 2 0.05056 0.02528 1.00 Residual 4 0.10111 0.02528 Total 8 0.29389 Appendix 7.5. ANOVA for effect of seed size on emergence percentage Source of variation df SS MS v.r. F Seed size 2 0.03909 0.01954 1.07 0.424 Replication 2 0.11209 0.05604 3.08 Residual 4 0.07284 0.01821 Total 8 0.22402 Appendix 7.6. ANOVA for effect of seed size on emergence rate index Source of variation df SS MS v.r. F Seed size 2 260.39 130.20 11.29 0.023 Replication 2 79.89 39.94 3.46 Residual 4 46.14 11.54 Total 8 386.42 Appendix 7.7. ANOVA for effect of seed size on days to sprouting Source of variation df SS MS v.r. F Seed size 2 30.9809 15.4904 23.95 0.006 Replication 2 3.1505 1.5752 2.44 Residual 4 2.5867 0.6467 Total 8 36.7180 Appendix 7.8. ANOVA for effect of seed size on days to pseudoradicle elongation Source of variation df SS MS v.r. F Seed size 2 149.959 74.980 8.94 0.033 Replication 2 9.534 4.767 0.57 Residual 4 33.547 8.387 Total 8 193.040 University of Ghana http://ugspace.ug.edu.gh 119 Appendix 7.9. ANOVA for effect of seed size on days to formation of bulge on the pseudoradicle Source of variation df SS MS v.r. F Seed size 2 79.5475 39.7737 48.15 0.002 Replication 2 1.8224 0.9112 1.10 Residual 4 3.3040 0.8260 Total 8 84.6739 Appendix 7.10. ANOVA for effect of seed size on days to appearance of the shoot on the pseudoradicle Source of variation df SS MS v.r. F Seed size 2 244.485 122.242 88.84 <.001 Replication 2 12.825 6.413 4.66 Residual 4 5.504 1.376 Total 8 262.814 Appendix 7.11. ANOVA for effect of seed size on days to seedling emergence Source of variation df SS MS v.r. F Seed size 2 318.891 159.445 17.02 0.011 Replication 2 9.174 4.587 0.49 Residual 4 37.464 9.366 Total 8 365.529 Appendix 7.12. ANOVA for effect of seed size on days to seedling establishment Source of variation df SS MS v.r. F Seed size 2 511.15 255.57 20.17 0.008 Replication 2 13.74 6.87 0.54 Residual 4 50.68 12.67 Total 8 575.57 University of Ghana http://ugspace.ug.edu.gh 120 Appendix 7.13. ANOVA for effect of seed size on the length of the pseudoradicle at bulge formation Source of variation df SS MS v.r. F Seed size 2 14.5204 7.2602 17.76 0.010 Replication 2 1.8074 0.9037 2.21 Residual 4 1.6349 0.4087 Total 8 17.9627 Appendix 7.14. ANOVA for effect of seed size on the length of the taproot at bulge formation Source of variation df SS MS v.r. F Seed size 2 23.9654 11.9827 13.02 0.018 Replication 2 1.6177 0.8088 0.88 Residual 4 3.6816 0.9204 Total 8 29.2647 Appendix 7.15. ANOVA for effect of seed size on the length of the taproot at emergence Source of variation df SS MS v.r. F Seed size 2 276.604 138.302 72.74 <.001 Replication 2 4.107 2.054 1.08 Residual 4 7.605 1.901 Total 8 288.316 Appendix 7.16. ANOVA for effect of seed size on the total shoot height three weeks after emergence Source of variation df SS MS v.r. F Seed size 2 3.8138 1.9069 7.20 0.047 Replication 2 1.3524 0.6762 2.55 Residual 4 1.0590 0.2647 Total 8 6.2252 University of Ghana http://ugspace.ug.edu.gh 121 Appendix 7.17. ANOVA for effect of seed size on diameter of the root crown three weeks after emergence Source of variation df SS MS v.r. F Seed size 2 0.1017556 0.0508778 194.85 <.001 Replication 2 0.0046889 0.0023444 8.98 Residual 4 0.0010444 0.0002611 Total 8 0.1074889 Appendix 7.18. ANOVA for effect of seed size on diameter of the shoots 3 weeks after emergence Source of variation df SS MS v.r. F Seed size 2 0.0150889 0.0075444 21.90 0.007 Replication 2 0.0021556 0.0010778 3.13 Residual 4 0.0013778 0.0003444 Total 8 0.0186222 Appendix 7.19. ANOVA for effect of deshelling of the seeds on germination percentage Source of variation df SS MS v.r. F Deshelling of seed 1 0.096 0.096 0.63 0.512 Replication 2 0.281 0.141 0.91 Residual 2 0.308 0.154 Total 5 0.685 Appendix 7.20. ANOVA for effect of deshelling of the seeds on mean germination time Source of variation df SS MS v.r. F Deshelling of seed 1 2.734 2.734 2.61 0.247 Replication 2 3.101 1.550 1.48 Residual 2 2.092 1.046 Total 5 7.927 University of Ghana http://ugspace.ug.edu.gh 122 Appendix 7.21. ANOVA for effect of deshelling of the seed on emergence percentage Source of variation df SS MS v.r. F Deshelling of seed 1 0.582 0.582 43.93 0.022 Replication 2 0.299 0.149 11.25 Residual 2 0.027 0.013 Total 5 0.908 Appendix 7.22. ANOVA for effect of deshelling of seeds size on emergence index Source of variation df SS MS v.r. F Deshelling of seeds 1 493.41 493.41 297.18 0.003 Replication 2 29.24 14.62 8.81 Residual 2 3.32 1.66 Total 5 525.97 Appendix 7.23. ANOVA for effect of deshelling of the seeds on emergence rate index Source of variation df SS MS v.r. F Deshelling of seed 1 1378.95 1378.95 92.95 0.011 Replication 2 185.07 92.53 6.24 Residual 2 29.67 14.84 Total 5 1593.70 Appendix 7.24. ANOVA for effect of deshelling of seeds on mean stem diameter of seedlings 150 days after sowing Source of variation df SS MS v.r. F Deshelling of seed 1 0.0054 0.00540 15.43 0.059 Replication 2 0.0019 0.00095 2.71 Residual 2 0.0007 0.00035 Total 5 0.0080 University of Ghana http://ugspace.ug.edu.gh 123 Appendix 7.25. ANOVA for effect of deshelling of seeds on seedling height 150 days after sowing Source of variation df SS MS v.r. F Deshelling of seed 1 0.84375 0.84375 1875.00 <.001 Replication 2 0.15790 0.07895 175.44 Residual 2 0.00090 0.00045 Total 5 1.00255 Appendix 7.26. ANOVA for effect of deshelling of seeds on number of leaves produced by seedlings 150 days after sowing Source of variation df SS MS v.r. F Deshelling of seed 1 1.7281 1.7281 3.81 0.190 Replication 2 0.0506 0.0253 0.06 Residual 2 0.9076 0.4538 Total 5 2.6863 Appendix 7.27. ANOVA for effect of deshelling of seeds on mean leaf area of seedlings 150 days after sowing Source of variation df SS MS v.r. F Deshelling of seed 1 288.57 288.57 3.44 0.205 Replication 2 231.16 115.58 1.38 Residual 2 167.87 83.93 Total 5 687.59 Appendix 7.28. ANOVA for effect of deshelling of seeds on mean root crown diameter of seedlings 150 days after sowing Source of variation df SS MS v.r. F Deshelling of seed 1 0.201667 0.201667 22.74 0.041 Replication 2 0.030400 0.015200 1.71 Residual 2 0.017733 0.008867 Total 5 0.249800 University of Ghana http://ugspace.ug.edu.gh 124 Appendix 7.29. ANOVA for effect of deshelling of seeds on mean stem diameter of seedlings 240 days after sowing Source of variation df SS MS v.r. F Deshelling of seed 1 0.49882 0.49882 49.55 0.020 Replication 2 0.12413 0.06207 6.17 Residual 2 0.02013 0.01007 Total 5 0.64308 Appendix 7.30. ANOVA for effect of deshelling of seeds on mean height of seedlings 240 days after sowing Source of variation df SS MS v.r. F Deshelling of seed 1 0.8588 0.8588 2.84 0.234 Replication 2 1.8097 0.9048 2.99 Residual 2 0.6050 0.3025 Total 5 3.2735 Appendix 7.31. ANOVA for effect of deshelling of seeds on mean number of leaves produced by seedlings 240 days after sowing Source of variation df SS MS v.r. F Deshelling of seed 1 4.034 4.034 0.93 0.437 Replication 2 6.546 3.273 0.75 Residual 2 8.689 4.345 Total 5 19.270 Appendix 7.32. ANOVA for effect of deshelling of seeds on mean leaf area of seedlings 240 days after sowing Source of variation df SS MS v.r. F Deshelling of seed 1 4763.5 4763.5 24.98 0.038 Replication 2 307.9 154.0 0.81 Residual 2 381.3 190.7 Total 5 5452.8 University of Ghana http://ugspace.ug.edu.gh 125 Appendix 7.33. ANOVA for effect of deshelling of seeds on root crown diameter of seedlings 240 days after sowing Source of variation df SS MS v.r. F Deshelling of seed 1 1.70667 1.70667 32.28 0.030 Replication 2 0.21280 0.10640 2.01 Residual 2 0.10573 0.05287 Total 5 2.02520 Appendix 7.34. ANOVA for the effect of concentration of BAP in culture medium on duration to sprouting of deshelled seeds Source of variation df SS MS v.r. F Concentration of BAP 3 212.500 70.833 7.56 <.001 Residual 44 412.500 9.375 Total 47 625.000 Appendix 7.35. ANOVA for the effect of concentration of BAP in culture medium on percentage number of sprouted deshelled seeds Source of variation df SS MS v.r. F Concentration of BAP 3 12000 4000 1.69 0.176 Residual 76 180000 2368 Total 79 192000 Appendix 7.36. ANOVA for the effect of concentration of BAP in culture medium on duration to sprouting of embryonic axes Source of variation df SS MS v.r. F Concentration of BAP 3 175.00 58.33 4.69 0.006 Residual 52 646.43 12.43 Total 55 821.43 Appendix 7.37. ANOVA for the effect of concentration of BAP in culture medium on GP embryonic axes Source of variation df SS MS v.r. F Concentration of BAP 3 20000 6667 3.42 0.021 Residual 76 148000 1947 Total 79 168000 University of Ghana http://ugspace.ug.edu.gh 126 Appendix 7.38. ANOVA for factorial analysis of the effect of concentration of BAP and NAA in the culture medium on the duration to response of rudimentary shoots to culture conditions Source of variation df SS MS v.r. F Concentration of BAP 3 239.062 79.688 21.86 <.001 Concentration of NAA 3 168.229 56.076 15.38 <.001 Conc. of BAP × Conc. of NAA 9 75.521 8.391 2.30 0.040 Residual 32 116.667 3.646 Total 47 599.479 Appendix 7.39. ANOVA for factorial analysis of the effect of concentration of BAP and NAA in the culture medium on height of regenerated rudimentary shoots 15 days after culture Source of variation df SS MS v.r. F Concentration of BAP 3 2.27167 0.75722 15.87 <.001 Concentration of NAA 3 4.19833 1.39944 29.33 <.001 Conc. of BAP × Conc. of NAA 9 2.85333 0.31704 6.65 <.001 Residual 32 1.52667 0.04771 Total 47 10.85000 Appendix 7.40. ANOVA for factorial analysis of the effect of concentration of BAP and NAA in the culture medium on number of leaves produced by regenerated rudimentary shoots 15 days after culture Source of variation df SS MS v.r. F Concentration of BAP 3 2.7292 0.9097 3.64 0.023 Concentration of NAA 3 6.7292 2.2431 8.97 <.001 Conc. of BAP × Conc. of NAA 9 5.8542 0.6505 2.60 0.022 Residual 32 8.0000 0.2500 Total 47 23.3125 University of Ghana http://ugspace.ug.edu.gh 127 Appendix 7.41. ANOVA for factorial analysis of the effect of concentration of BAP and NAA in the culture medium on height of regenerated rudimentary shoots 45 days after culture Source of variation df SS MS v.r. F Concentration of BAP 3 20.74167 6.91389 119.38 <.001 Concentration of NAA 3 10.94167 3.64722 62.97 <.001 Conc. of BAP × Conc. of NAA 9 1.84000 0.20444 3.53 0.004 Residual 32 1.85333 0.05792 Total 47 35.37667 Appendix 7.42. ANOVA for factorial analysis of the effect of concentration of BAP and NAA in the culture medium on number of leaves produced by regenerated shoots 45 days after culture Source of variation df SS MS v.r. F Concentration of BAP 3 29.4167 9.8056 17.43 <.001 Concentration of NAA 3 40.0833 13.3611 23.75 <.001 Conc. of BAP × Conc. of NAA 9 5.8542 1.3796 2.45 0.030 Residual 32 18.0000 0.5625 Total 47 99.9167