Towards an Enhanced Breeding in Cocoyam: A Review of Past and Future Research Perspectives

Abstract

Cocoyams (Taro and Tannia) are important food crop in the tropical world. Beyond their food and nutritious values, they have cultural, religious and social meanings which vary within cultures. As with most tropical underutilized crops, cocoyam is affected by biotic and abiotic stresses resulting to low yield. In addition, limited genetic advancement in post-harvest and organoleptic properties is visible. While significant breeding gains have been made in the past, knowledge gaps exist in respect to flowering, genetics, cytogenetics and options for hybridization that could accelerate crop improvement. This review presents a concise but timely effort to explore these breeding limitations while highlighting solutions to overcome these challenges. The wider implication of this review is accelerated breeding and crop improvement.

Introduction

The most important food aroids belong to the genera of Colocasia and Xanthosoma with Colocasia esculenta (L.) Schott and Xanthosoma sagittifolium (L.) Schott highly popular [1]. The former is an emergent, perennial, aquatic and semi-aquatic herbaceous species native to Asia while the later is native to tropical America [2,3]. They are important root crops mainly cultivated by small-scale farmers [4]. in Asia, Africa and Latin America. The corm, cormels, and leaves are important source of carbohydrates for human nutrition, animal feed and supplemental food [3,5]. Colocasia esculenta is commonly known as Taro, true cocoyam, old cocoyam and several common names while Xanthosoma sagittifolium is referred to as tannia, yatua, malanga, callalo, coco or new cocoyam [6]. Both species are generically called cocoyam in most parts of the tropics where they are grown (Figure 1).

Figure 1: Global distribution of cocoyam (taro and tannia) production.

Cocoyam is an important food crop for more than 400 million people worldwide, especially in the tropics and subtropics [7,8]. Cocoyams ensure food security; have rich economic and socio-cultural connotations, serves as a cash crop and foreign exchange earner [8]. To underscore the importance of cocoyams it features prominently in the folklore and old traditions of many cultures in Oceania and south east Asia [8]. Cocoyams are well adapted food crops in many agro-ecological zones of sub- Saharan Africa and ranks third in importance after cassava and yam amongst the root and tuber crops [9]. The underutilized crops like cocoyam give poor inhabitants of the growing region an alternative source of income-paths out of poverty. In spite of its importance as a staple food in many countries, cocoyam remains a neglected crop mainly grown for subsistence agriculture. Their potential is seldom realized, mainly because of knowledge gaps in physiological and biological processes that influence breeding advances. The removal of the crop from the focus of Consultative Group for International Agricultural Research (CGIAR) centres in the past contributed to the limited research investment by the international community. International, regional and national efforts under the auspices of The International Network for Edible Aroids (INEA), Taro Network for Southeast Asia and Oceania (TANSOA), Taro genetic resources: conservation and utilization project (TaroGen), root and tuber research project (ROTREP) in Cameroon and Cocoyam rebirth initiative in Nigeria have sought to develop adaptations to climate and commercial challenges. Part focus of these initiatives is breeding improvement and international germplasm exchange. The future of cocoyam depends on selection of high yielding, good quality genotypes as well as development of low cost technologies that will enhance its sustainable production. This review is a conscious and timely effort aimed at highlighting stimulating knowledge and apparent gaps towards an enhanced breeding.

Economic Importance

Cocoyam (Taro and Tannia) corms and cormels are edible and are usually cooked by boiling, roasted, baked, steamed or fried and used as a starchy vegetable and supplemental food [3,5]. In addition to sustaining food security in domestic market, it also brings import earnings [10]. Cocoyam features prominently in the folklore and oral traditions of many cultures in Oceania and Southeast Asia. Samoa and Tonga have prominent depictions of cocoyam on their currencies [8]. In Hawaii, images of cocoyam farmers can be found throughout the islands, in murals, posters, original arts and other visuals, where its symbolic importance reflects its continuing role as a common food and common element in the agricultural landscape Cocoyam represents an excellent source of carbohydrate, the majority being starch of which 17% to 28% is amylase, and the remainder is amylopectin [11,12]. Cocoyam starch is one of the most nutritious and 98.8% digestible, a quality attributed to its granule size making it ideal for people with digestive difficulties [13]. The leaves and petioles are used as green vegetables after thorough removal of the acrid elements through special processing [3]. The leaves contain significant levels of protein and are also excellent source of carotene, potassium, calcium, phosphorous, iron, riboflavin, thiamine, niacin, vitamin A, vitamin C and dietary fibre [14]. In addition, they also contain greater amounts of vitamin B-complex than whole milk and can be useful to people allergic to cereals and can be consumed by children who are sensitive to milk thus its use in infant food formulae [15]. Cocoyams are used for soup thickening while the inflorescence is commonly used as a local food spice [16,17]. The leaf extracts have been implicated as expectorant, decongestant, contraceptive, anti-cancerous anti-oxidant, anti-inflammatory and anti-bacterial in action [18-20]. Cocoyams have been reported to have the potential of serving as a health supplement for the treatment of bone diseases such as osteoporosis [21].

Origin, Domestication and Dispersion

Taro cocoyam

Cytological and archaeological studies indicate that Taro probably originated in the Indo-Malaysian Peninsula over 50,000 years ago [22]. Evidence equally indicate human use of the plants 28,000 years ago in the Solomon Islands [23]. A general consensus in modern times is that its origin and domestication started from eastern India to Southeast Asia, from where it dispersed to other parts of the world [24,25]. With the aid of advances in marker technology, the possibility of domestication in different regions of Southeast Asia and Melanesia has further been strengthened [26]. The dispersion of Taro likely began in about 1600 to 1200 BC, when long-distance voyaging canoes were developed and the crop was taken further east into Fiji and western Polynesia and then into eastern Polynesia with the movement of migrating voyagers around 800 to 900 AD [27]. The geographical distribution of colocasia species indicates that the genus is naturally distributed from South Asia to Southeast Asia (including China and Indonesia), in lowland tropics areas as well as in the cooler conditions of the Himalayan mountains [28]. The crop probably arrived the island of Madagascar in Africa through the migrating Indonesians as early as 500 AD from where it spread across the Africa to the Guinea coast [3]. In the post-Columbian period, it was introduced to Caribbean and tropical American regions [29]. Presently Taro is cultivated throughout the tropics, subtropics and warm temperate regions of Asia, Oceania, Africa and America [5] (Figure 2).

Figure 2: Approved breeding scheme for cocoyam (taro and tannia) in National Root Crops Research Institute Umudike, Nigeria.

Tannia cocoyam

The native niche of tannia is hypothesized to South America including Colombia, Peru, Ecuador and Venezuela. When the European migrants arrived to America, tannia was said to have gained cultivation prominence from Central America to Bolivia, with greater intensity of cultivation observed in the West Indies [30]. To buttress this fact, the year 1881 saw tannia being enlisted as a common species in Puerto Rico [31]. Portuguese missionaries introduced tannia into Africa only in 1840 in the Gold Coast (now Ghana) [32]. Tannia reached West Africa between the 16th and 17th centuries and was spread further by traders, missionaries and other travellers [33]. It however became more popular in African cultivation and diet than Taro which was earlier introduced from Southeast Asia [34]. Tannia was rapidly adopted in West Africa because of its resemblance to the more familiar Taro and thus became known as new cocoyam in some locations [29]. Disease and pest resistance greatly enhanced tannia rapid adaptation and spread in Africa, Asia and some Pacific islands. They are now grown practically in all regions of the tropic [16].

Breeding and Crop Improvement Challenges

Poor flowering

Taro and Tannia hardly flower under natural environmental conditions and inflorescence develops only when the basic environmental, physiological, genetic and developmental conditions are fulfilled [35]. Cultivated or semi-cultivated genotypes are not stable with flowering and there is reported evidence of reduced flowering [36]. The main factor limiting classical interspecific hybridization is the irregularity of flowering and the abnormalities of the inflorescence structure [35]. Flowering of wild cultivars is common in its natural environment but when they are transferred from this optimal natural environment to field conditions they often continue to flower, sometimes even more intensively but the inflorescences are smaller with little or no pollen. It is reported that cultivars of African and Malaysian origin flower more than those of Japanese origin [37]. The mechanism for this remains unclear but there is a likelihood it is linked with photo periodic factors. Report shows a series of abnormal and unusual inflorescences in a number of genotypes. The phenotypic expression of abnormal floral structures is strongly influenced by nongenetic factors [35].

Sexual hybridization and seed sett

Sexual hybridization of Taro and Tannia is labor intensive and takes time in terms of field preparation, planting of parents, induction of flowering, pollination, development and maturation of fruit heads and seed harvesting [38]. In addition, the germination and planting of seedlings and screening processes takes up to 10 years or more from the time you make pollination, until the new, improved cultivar finally reaches a large number of farmers [39]. Taro and tannia have been propagated through the centuries by asexual method while seed production has been considered rare and of little significance [40]. Due to the challenges of seed propagation fertile seeds rarely develop [41]. Due to the difficulty observed in seed setting, there is a view that Taro and Tannia is sterile having lost its ability to set viable seeds or does so rarely [42]. Propagation by seed is rare and may be more frequent in the wild if environmental conditions are optimal [35].

Cytogenetics

Cytological studies on Taro and Tannia indicate confusion concerning basic chromosome number of the genus as shown in Table 1. The utilization of karyotypic data in Taro has produced a hypothesis for two separate lineages of the plant within contemporary populations [43-52]. In addition, various cytotypes have been observed within 2n=28 and 42 forms [27,53-55]. Reports have highlighted that the chromosomes are prone to unpredictable behaviour during cell divisions, thus the chromosome number per cell lack uniformity within the crop [55-64]. In tannia, thirteen bivalents (n=13) could be assigned to 12 chromosomal types of which 8 of them could be regarded as homologous, reducing the basic number to n=8 [65]. The somatic chromosome number of tannia is 26 but the haploid chromosome complement at the pachytene stage of meiosis could be resolved into 12 types based on morphology and staining pattern [66,62]. The argument of basic chromosome number of 26 was further strengthened [27,64]. From a breeding perspective, polyploidy as observed in Taro and Tannia can result to changes in cellular structures which thus lead to difficulties in mitosis as a result of spindle irregularities as well as irregular meiosis due to the formation of multivalent at meiotic metaphase I. In addition, the viability of gametes and zygotes arising from autotriploids is low while epigenetic instability is very common [67].

Table 1: List of x and 2n status of taro and tannia with corresponding references.

Limited genetic/genomic resources for accelerated breeding

The breeding of cultivars is a complex process which requires experience, adequate genetic resources, and reliable data about inheritance of crucial agronomic traits. The genetic diversity as revealed by morphological, cytological and DNA based studies suggest that diversity is low with the existence of two distinct gene families with no distinct allelic difference between wild and cultivated types [25,27,68]. The limited genetic diversity within cultivated gene pools threatens sustainable production [69-71]. Randomly amplified polymorphic DNA (RAPD), amplified fragment length polymorphisms (AFLP) and microsatellites (SSR) revealed low genetic variation when used for fingerprinting [72-76]. Mitochondrial and chloroplast-specific restriction fragment polymorphisms revealed limited species-level variability as well [77]. This is evidenced through in-breeding depression, a significant phenomenon that results to low yield and susceptibility to biotic and abiotic stress conditions. Selection of economic important traits can become much more efficient when the desired expression of the most important plant characteristics such as yield, eating quality and disease resistance are associated with DNA based markers [35].

History of Varietal Development and Priorities in Breeding

History of varietal development

Breeding programs of Taro were initiated in late 1970s with varietal releases recorded in Fiji in 1978, Samoa in 1982 and 1996, Solomon Islands in 1978 and 1992, Papua New Guinea in 1993 and India in 1995 [78,79]. Successful controlled hybridization was recorded by in Cameroun resulting to the production of 12 F1 hybrid combinations and 3 self-pollinations [80]. Unfortunately, these attempts to combine different genotypes through sexual crossings did not result to new released varieties. Taro breeding programme by Central Tuber Crops Research Institute (CTCRI) in India has succeeded in developing novel genotypes with characteristics such as erect lines, early maturity and resistance to TLB. The first successful controlled hybridization of Taro in Nigeria was reported in year 2015 with seeds presently undergoing clonal evaluation [81]. Earlier successful crosses in tannia have been reported in USA, Africa and India [7,82-84]. Unfortunately, most of this recombination did not result in varietal release.

Breeding priorities in Taro and Tannia

Progeny selection in a breeding scheme depends on the objectives of the breeding programme. Yield increase, disease and pest resistance, regular attractive corm shape, non-acrid tuber with relatively high dry matter content is desired. Apart from yield and eating quality, an ideotype is determined by its maturity period, corm shape, the number of suckers, the absence of stolons, the number of leaves, and the verticality of petioles [79]. Breeding against disease focus on Taro leaf blight (TLB) caused by Phytophthora colocasiae, and five viruses namely Dasheen mosaic virus (DsMV), Colocasia bobone disease virus (CBDV), Taro bacilliform virus (TaBV), Taro vein chlorosis virus (TaVCV), and Taro reovirus. When TaBV combines with CBDV it causes Alomae- Bobone disease, and is the most damaging taro viral disease, largely spread in the Solomon Islands and Papua New Guinea [79,85]. The main disease in the Solomon Islands and Papua New Guinea remains TLB, Alomae-Bobone virus complex and infections of nematode Hirschmanniella miticausa [35]. The most important disease in Africa is the cocoyam root rot disease (CRRD) observed in Tannia [86]. CRRD is caused by the oomycete Phthium myriotylum. Resistant varieties to CRRD is not yet available despite advances like generation of hybrids, mutants, polyploidy induction, and a better understanding of host pathogen interaction [87,88].

Advances in breeding

Floral induction in Taro and Tannia is possible with the help of Gibberelic acid (GA) [89,90]. GA aid the morphogenesis of flowering followed by heavy fused corms that exhibit many apical buds. This phenomenon is not observed under non-GA treated plants. The GA concentration required can be predicted from the propensity for natural flowering; 500 ppm GA is adequate in clones which often flower naturally whereas those which rarely flower require 1500 ppm GA [37]. GA floral induction in tannia shows dependence on GA concentration and plant age with 500 mg/L concentration giving the best result [91]. The authors observed that geographical location had no effect on flowering. They hold the view that flowering is only affected by physiological stage of corms meristem. The highest number of inflorescences as well as pollen quantity in Tannia was obtained at GA concentration at 750 and 1000 ppm [92]. GA applied at 1500 ppm, as foliar spray or pre-plant soak, effectively promoted flowering in Nigerian clones [37]. Lower concentration of 250 ppm to 500 ppm has been recommended, due to differential response of cultivars to GA treatment [93, 94]. Regardless of concentration level, all treatments promote abundant flowering in reasonable time and would be satisfactory for breeding purposes [37]. Floral promotion in cocoyam is dependent on the genotype and the method of GA application. Early flower emergence was found in plants soaked in 500 ppm of GA for 30 mins immediately prior to planting [95]. The Pro-Gibb Plus foliar spray proved to be more effective than pre-plant soak with the same material in reference to number and vigour of the inflorescences produced, duration of flowering and the percentage of plants that flowered [37]. Tannia have no obligated day length requirement for flowering although there may be a quantitative interaction between flowering induced by GA and day length [90]. The likelihood of other factors other than GA to be involved in flowering in aroids has been suggested [96]. Flower induction in tissue culture derived Tannia plants occurred 20 to 30 days earlier than that reported for non-tissue culture derived plants [92]. The tissue culture derived plants may have a lower endogenous level of GA, and spraying them with an artificial source may have provided an optimum level of GA for vegetative and reproductive growth [92]. Similar trend was observed in bananas and plantains [97,98].

From a cytological perspective, the majority of cultivated wild genotypes of Taro are diploids and are found throughout Asia and Oceania [49,59]. Triploids have equally been reported in the Asian continent [24,27,48,52]. Triploids in Taro are assumed to be autotriploids by cytogenetical, biochemical and morphological studies [52,68,26]. Allopolyploid origin for triploids is also possible [99]. From a morphological perspective, triploid and diploid plants referred to as ‘alowane’ and ‘alokine’ by Solomon Island farmers differ in plant height with the former taller than the later [100]. Isozyme analysis at the Aat-1 of aspartate aminotransferase of six hybrid triploids revealed that the seed parent was the double genome donor in five, whereas the pollen parent served as the donor in one. Both seed and pollen parents are able to serve as a double genome donor of the triploids [52]. The pollen mother cells (PMCs) of the diploid hybrids formed univalents during meiosis leading to complete sterility of pollen grains while the PMCs of the triploid hybrid behaved differently in the course of meiosis with homologous chromosomes of Taro forming bivalents while those of C. gigantea formed univalent as well as improved pollen grains at the rate of 21%. Tetraploid induction, using colchicines showed that treated tissues cultures resulted to diploids (2n=26), tetraploids (2n=52) and aneuploids (2n-2=24) respectively [101,102]. Dominating mixoploid plantlets of 4x were observed following treatment with colchicine [102]. The continuous exposure of tannia tissue cultures to colchicine is an effective method for the production of synthetic tetraploids [101]. An alternative approach in the induction of tetraploids and mixoploid with reduced mortality is the use of oryzallin at 0.05% for 3 days [103]. Two sets of plants exhibiting basic chromosome numbers of x=12 (characterised with somatic chromosome counts of 24 and 48) and x=14 (characterised with somatic chromosome counts of 28 and 42) had their origin from India and Japan respectively [46]. Interestingly, studies of chromosome numbers in Indian genotypes of Taro suggest that they have more genetic diversity than those from any other geographic area [24]. These observations tend to support the hypothesis of center of origin of Taro. Studies from cytogenetics report show that chromosome numbers can vary due to fission, fusion or genome doubling, and there is ample evidence that such changes can contribute to speciation [104]. The evolution of Taro and Tannia may have arisen from common progenitor, which has a basic chromosome number of n=x=6 with the former undergoing polyploidization while the later evolved through one-step chromosome doubling [57,58]. The argument that basic numbers x=14 or x=12 must have been the starting points for the derivation of all the modern chromosome numbers in the Araceae is buttressed [105]. A model-based approach aided by phylogenetic analysis was used to reconstruct ancestral haploid chromosome numbers in Araceae [106]. Findings from this study disagrees with previously inferred basic numbers x=14 and x=7. Maximum likelihood optimization from their study revealed an ancestral haploid chromosome number of n=16 with a Bayesian inference of n=18. Chromosome fusion (loss) is the predominant inferred event, whereas polyploidization events occurred less frequently [106]. A genome size of 13789.8 and 7863.12 Mbp (C) for Taro and Tannia respectively have been reported [107]. The epidermal variations in stomatal index found in Taro and Tannia reflect their ecological adaptation to variation in response to soil moisture [58]. The size and shape of the chromosomes are quite variable differing in size and shape thus buttressing differences within the karyotypes of certain species with Taro ranging 2.1 μ to 4.8 μ and Tannia 2.7 μ to 6.0 μ [54,108]. Positive and significant correlation coefficients were observed between stomata pore lengths, guard cell lengths and number of chromosomes.

Prolonging the life span of cocoyam pollen through storage can assist plant breeders to effect artificial pollination. Viable cocoyam pollen could be obtained after 28 days in storage with the best storage condition at 5°C and 30% relative humidity [109]. Pollens stored at 8°C and 80% relative humidity remained viable for about 8 days; however storage period is strongly dependent on pathogen interaction with pollen grain [80]. Five percent aqueous solution of sucrose was found to promote pollen germination up to 76% while distilled water supported pollen germination at a low rate and with little tuber growth [110]. Seed can be germinated between layers of moist filter paper, in soil and in agar culture [40]. Germination rates of 80% were observed in seeds grown in a greenhouse potting mix or its extract or distilled water with filter paper as a support [111]. Seeds that developed from pollination between cultivars appeared to have better germination than those coming from pollination within varieties [112]. Seed weight do not affect seed germination and seedling growth, while the reduction of moisture content of seed to 3% to 12% do not affect germination negatively [113]. Seeds planted in soil emerged within 14-24 days depending on depth of planting. Due its small size Taro and Tannia seed is not thought to contain much stored food and the period of viability seems to be very short [29]. The seeds can be conserved for at least 2 years at constant 5°C to -20°C when seed moisture is reduced to 10% to 12% and at room temperature when seed moisture content is reduced to 7.3% [114]. An 80% germination failure was reported when seeds are stored in air tight containers for more than 30 days. This is in contrast of seed germination rates exceeding 85% after storage of more than 7 months at 8°C and 80% relative humidity. Germinating seeds can be colonised by pathogens like Curvularia spp., Fusarium spp., Penicillium spp., and Rhizoctonia sp. The pathogens do not have any negative impact on germination [114]. Rescued embryo can be grown aseptically via in vitro seed rescue culture (SRC) technique [115]. The regenerated plantlets can be free from contaminants but has potential of expressing a wide range of morphological and physiological variability.

The breeding schemes adopted by the breeding programmes are bi-parental crossing and recurrent selection. Breeding efficiency is achieved by conducting the selection process at an early stage of development without waiting for the plants to be uprooted. This approach has been developed for determining traits like Taro corm flesh and corm fiber colors, which were correlated to the color of different petiole zones [116]. A vegetative growth index (VGI) that takes into account four vegetative traits has proved to be useful for the rapid assessment of genotypes with good yield potential [117]. One of the lessons of the Papua New Guinea breeding program was the difficulty of getting rid of wild deleterious traits. Under resistance breeding, a parent with wild and immune phenotypes is often associated with horizontal resistance (HR) is ideal. The polygenic nature of HR makes it difficult for population of pathogen to overcome resistance. This practice has been predominantly adopted in Papua New Guinea and Samoa. Breeders are favouring resistance sources within the cultivated genepool [118]. However, when using a collection with high genetic diversity, breeders conduct a high number of crosses in order to find the best heterotic combinations [79]. The creation of a small number of large full-sib families when working with a narrow genetic base and the creation of numerous small full-sib families when dealing with a broad genetic base is recommended [119]. Agronomic characters are heritable in Tannia but were quantitative in their inheritance and probably multigenic. Progenies of Tannia grown from true seed in the first filial generation showed considerable variation. Such variation indicates considerable heterozygosity in the parent cultivars. This variability may be attributed to sexual recombination, and perhaps somatic mutation associated with continuous vegetative propagation and the subsequent selection by farmers based on adaptability and culinary qualities from exotic and novel varieties [120]. Protogyny and self-incompatibility systems in the inflorescence facilitate cross-fertilisation, which usually results in variable progenies [121]. Combining genotypes from the two major genepools is strategic towards establishing a broad base for any breeding programme. The phenotypic expression of abnormal floral structures is strongly influenced by non-genetic factors. Breeders can efficiently use several types of abnormal inflorescences. Some types with a double, multiple or fascinated spadix can be very productive. The most undesirable types are those with a reduced female portion of the spadix, a reduced male portion, inflorescence with sterile female parts, inflorescence where the syandria do not produce pollen or pollen is not viable, false inflorescence, small lateral inflorescences at the base of the corm ‘head’ and multiple inflorescences with reduced individual part. It has been shown that root rot disease might be associated with an increased peroxidase activity in the roots. Plants with different levels of root rot disease tolerance have been obtained through the irradiation with gamma rays on in vitro grown apices [122]. Breeders can apply this test to screen large progenies [88]. Isolates of P. colocasiae from different countries showed a very high diversity suggesting a high capacity of the pathogen to evolve rapidly in isolated insular regions [123]. The implication is that resistant cultivars in some countries have the potential of being susceptible in another, thus breeding for TLB resistance should be conducted against local isolates in each country affected by TLB.

Future Perspective

One of the major challenges of sustainable production is genetic erosion occasioned by loss of Taro and Tannia germplasm mostly held in farmers’ fields and in the wild. This presents a serious risk to germplasm conservation and thus threatens the crops sustainability. Such a situation raises the need to explore, collect and safeguard the existing genetic diversity. There is need for global, regional and in country assessment of collections in production areas so as to ensure maximum diversity for the crop’s improvement as well as conservation. International germplasm exchange as envisioned in recent international projects play an important role in broadening genetic diversity. Detailed passport information under conservation should be readily available using standardised and commonly agreed descriptor list among stakeholders. Cocoyams show a diverse array of agro-morphological polymorphism. The first descriptor list for Taro, developed by international board for plant genetic resources was used in earlier attempts to characterize germplasm [124]. An updated version of this descriptor list developed by international plant genetic resources institute in collaboration with TaroGen is currently in use. Additionally, the TANSAO network developed a descriptor list using major agro-morphological markers in all of the partner countries with a view to select national core samples to form a regional core collection [125,126]. In a recent study, Indian national collections were morphologically characterized using the National Bureau of Plant Genetic Resources descriptors [127]. The onus lies within the scientific community to harmonise these descriptor lists.

Collection of germplasm representing the genetic diversity is a prerequisite for its effective conservation and utilization for crop improvement. The use of complementary conservation strategies, especially in vitro techniques, for efficient conservation and utilization of germplasm is thus necessary. Collection missions in the 1980s were maintained ex situ [128]. A lot of the accessions were lost before being characterized due to inadequate level of husbandry in terms of weed control, insect pest and disease management. Natural calamities such as floods and prolonged dry periods also had impacts on the field GenBank [120]. Such errors have proved expensive and unsustainable [129]. The success of breeding programmes is anchored on availability and utilization of genetic resources. The enormous diversity in south east Asia, Pacific, India-China, South America and Africa can be efficiently utilized through inter regional collaboration thus strengthening the capacity of NARs breeding programmes. Early breeding programs used narrow genetic base and heterosis was not evident as most hybrids produced were susceptible to diseases. Different genepools can be effectively combined to develop heterotic populations. An interesting approach would be to combine genotypes from Asia, Pacific, Africa, and Latin America. Pacific cultivars are the result of intense local selection; they produce corms of good quality, but are susceptible to pests and diseases. Conversely, in Asia, co-evolution with numerous and diverse strains of P. colocasiae has produced resistant genotypes but because Taro is not as important as in the Pacific, most cultivars produce numerous suckers and stolons and have irregular corm shapes. For this breeding approach to be successful there is need for international exchange of germplasm [130]. This approach has been effective in cassava breeding in Africa where Latin American germplasm has been successfully introgressed into African germplasm [131]. There is need to assess the genetic diversity in the wild population as this offers significant potentials for breeding programmes [132]. Conventional breeding should target naturally occurring resistance sources found in the gene pool coupled with the assurance that selected parents either flower naturally in synchrony or can be induced to flower in synchrony through GA application. Repeated backcrossing offers a platform to eliminate deleterious genes from wild populations. To prevent inbreeding depressions, a better understanding of the genetic diversity in germplasm is essential when selecting parents. To circumvent the likelihood of inbreeding depression in the course of breeding, the recurrent parent should be replaced in each new backcross by a different genotype that is phenotypically similar. Taro and Tannia are mostly grown in marginal field environments with soils exhibiting varying levels of nutrient depletion. Breeding advancements within these environments have not been as successful as anticipated due to Gx E interaction and the prevailing farming systems adapted by locals. Under on station trials (researcher managed) high yielding individuals are products of optimal agronomic growth conditions. Thus, these cultivars produce low yield and respond poorly in non-optimal growth conditions. It is imperative to grow and promote individuals that are widely adapted over large expanse of marginal areas which is the more prevalent farming conditions. The participatory breeding approach can aid identification of superior cultivars, increase access of farmers to these cultivars while broadening the genetic base towards sustainable food production. A good starting point will be to engage diversity of farmers with more genotypes from which they can make selections across diverse agro environments. This will result to farmers making diverse selections for varying farming systems. The development of innovative strategies towards increasing multiplication rate of planting material, safe germplasm transfer and enhanced distribution of improved varieties will ensure the sustainability of participatory breeding.

It will be very useful to develop haploids, such that with the duplication of the chromosome number it will be possible to obtain homozygous diploids which are essential in Mendellian studies of qualitative inheritance. More data are needed to validate ploidy as numerous explanations need stronger empirical support to be accepted. Good mitotic indexing of cultivars would generate information required for proper chromosomal characterisation. Good knowledge of the crop genome is very necessary in order to establish a sound approach to its improvement. Induction of polyploids can be used to bridge the ploidy levels in intraspecific crosses and to move genes across interspecific breeding barrier [133]. Phenotypic and genotypic variations within each group of abnormal and unusual inflorescences will be excellent source of characters for selection and improvement, while further embryological studies can unravel the exact nature of incompatibility mechanism. Improvement in the techniques for storing pollen could help in germplasm transfer through introduction of pollen rather than corms and cormels from countries possessing better varieties [109]. The seed rescue culture technique may be used for safer and convenient international exchange of germplasm and to widen genetic diversity in breeding.

Genetic markers are very useful in breeding. Selection can become much more efficient when the desired expression of the most important plant characteristics such as yield, eating quality and disease resistance are associated with such markers [35]. Only a few genes or their coding proteins that are related to yield, quality or disease resistance have been isolated and identified. Genetic markers associated with desired traits will increase the selection efficiency of conventional breeding. There is need to saturate the currently existing genetic map [134]. The future efforts should include a larger number of co-dominant and informative markers such as SNPs. The current high-throughput sequencing techniques are quite promising for gene mapping. Optimization of QTL detection through the use of large populations of progenies as well as trial replications is important and critical when low heritability traits are considered. Genetically distant cultivars are most desirable in hybridization schemes and development of populations [135]. This will aid the advancement of basic knowledge on the genetics of most quantitative traits which is presently lacking [79]. Factorial or diallel designs should be implemented in order to provide precise estimates of additive and dominance variances in agronomic traits [79]. Before any new hybrids are released, extensive multi-location trials have to be conducted in order to test their adaptability and stability.

The limited success achieved with breeding can be linked to the weak institutional capacity of most NARs engaged in cocoyam breeding. This trend is not sustainable for the future of the crop and needs an urgent reappraisal. In optimizing breeding methodologies, the consolidation and development of already existing regional and international network is very critical. The international network for edible aroids (INEA) is presently leading this step but more efforts are needed within the breeding community to translate these into practical field breeding that will lead to varietal release in participating countries. In addition, the CGIAR research program on roots, tubers and bananas (RTB) that has part mandate to develop the underutilized minor root and tuber crops should take a lead. The implementation of recommendations in respect to status of cocoyam production in West and Central Africa could serve as a working document for the region and should be given utmost priority.

Conclusion

One of the great challenges for sustainable Taro and Tannia production is to mitigate effect of biotic and abiotic stress while ensuring increased production levels with reduced cropping area. The future lies in adding value to its organoleptic characteristics and nutritional properties, widening of export markets, diversification of use and promotion of more intensive consumption. A combination of expertise including geneticists, breeders, pathologists, taxonomists, food scientists, chemists, ecologists and physiologists will be critical in sustaining production. Natural variation in wild and cultivated germplasm provides an excellent platform for the discovery of diagnostic markers for marker-assisted selection (MAS). Efforts should be channelled towards exploiting advances in “omics” technologies. The development of genome database towards an accelerated breeding effort will surely transform the face of cocoyam breeding research while maintaining food and income security while reducing farmers’ risk in vulnerable agricultural environments.

Acknowledgement

The authors are thankful to the management of National Root Crops Research Institute Umudike, for facilitation of funds to undertake this compilation. Special thanks go to Dr. O. Onabanjo and Dr. A. Manasboonphempool for critically reviewing the manuscript.

References

1. Erlinawati I. The diversity of terrestrial Araceae in Mt. Watuwila complex, Southeast of Sulawesi. Berk Penel. Hayati J Biol Res. 2010;15:131-137.
2. Plucknett DL. Edible aroids. Alocasia, Colocasia,Cyrtosperma, Xanthosoma. In:Simmonds,Evolution of crop plants. London.1976;7:10-12.
3. Yamaguchi M. World vegetables, principle, production and nutritive values. Chichester, England.1983.
4. Onwueme IC and Charles WB. Cultivation of cocoyam. In: Tropical root and tuber crops. Production, perspectives and future prospects. FAO Plant Production and Protection. 1994;126:139-161.
5. Prakash PS and Nayar NM. Flowering and pollination in taro. J Root Crops. 2000;261:15-19.
6. O'Hair SK. Tropical root and tuber crops. In: Janick J and Simon JE (Eds.) Advances in new crops. Timber Press, Portland.1990;12:424-428.
7. Onokpise OU, et al. Evaluation of macabo cocoyam germplasm in Cameroon. In: Janick J (Ed.) Perspective on new crops and new uses. ASHS Press, Alexandria. 1999;394-396.
8. Onwueme I. Taro Cultivation in Asia and the Pacific. RAP Publication. Food & Agriculture Organization of the United Nations, Regional Office for Asia and the Pacific, Bangkok. 1999.
9. Onyeka J. Status of Cocoyam (Colocasiaesculenta and Xanthosoma spp) in West and Central Africa: Production, Household Importance and the Threat from Leaf Blight. Lima (Peru). CGIAR Research Program on Roots, Tubers and Bananas (RTB).2014;Available online at: www.rtb.cgiar.org.
10. Revill PA, et al. Incidence and distribution of viruses of taro (Colocasiaesculenta) in Pacific Island countries. Australasian Plant Path. 2005;34:327-331.
11. Matthews PJ. Taro in Hawaii: present status and current research. Plant Genet Resour Newsletter 1998;116:26-29.
12. Oke OL. Roots, tubers, plantains and bananas in human nutrition. Rome: FAO Corporate Documentary Repository, Rome, Italy. 1990.
13. Langworthy CF and Duel HJ. Digestibility of raw rice, arrow-root, canna, cassava, taro, tree-fern, and potato starches. J Biol Chem. 1992;52:251-261.
14. Bradbury JH and Holloway WD. Chemistry of tropical root crops: significance for nutrition and agriculture in the Pacific. ACIAR Monograph.1988;6:1-201.
15. Lee W. Taro (Colocasiaesculenta) [Electronic Version]. Ethnobotanical Leaflets. 1999.
16. Okeke SE. The correct nomenclature of the Nigerian species of Xanthosoma Schott (Araceae). Bot J Linn Soc. 1992;110:267-275.
17. UkpongIJ, et al. Phytochemical screening and mineral elements compositionof Xanthosoma sagittifolium inflorescence. Asian J Plant Sci Res. 2014;46:32-35.
18. Brown AC, et al. The anti-cancer effects of poi (Colocasiaesculenta) on colonic adenocarcinoma cells in vitro. Phytotherapy Res.2005;19:767-771.
19. Cambie RC and Ferguson LR. Potential functional foods in the traditional maori diet. Mutat Res. 2003;109-117.
20. Nair RT, et al. Antibacterial activity of some selected Indian medicinal flora. Turk J Biol.2005;29:41-47.
21. de Oliviera GL, et al. Xanthosoma sagittifolium and Laporteaaestuans: species used to prevent osteoporosis in Brazilian traditional medicine. Pharm Biol.2012;50:930-932.
22. White JP and O’Connell JF. A prehistory of Australia, New Guinea and Sahul. Academic Press, Sydney. 1982;
23. Loy TH, et al.Direct evidence for human use of plants 28,000 years ago: Starch residues on stone artefacts from the northern Solomon Islands. Antiquity. 1992;66:898-912.
24. Yen DE and Wheeler JM. Introduction of taro into the Pacific: The indications of the chromosome numbers. Ethnology. 1968;7:259-267.
25. Kuruvilla KM and Singh A. Karyotypic and electrophoretic studies on taro and its origin. Euphytica.1981;30:405-413.
26. Lebot V and Aradhya KM. Isozyme variation in taro (Colocasiaesculenta (L.) Schott) from Asia and Oceania. Euphytica1991;56:55-66.
27. Coates DJ, et al. Chromosome variation in taro, Colocasiaesculenta: Implications for the origin in the Pacific. Cytologia. 1988;53:551-560.
28. Ahmed I. Evolutionary dynamics in taro (Colocasiaesculenta L.). PhD dissertation Massey University, Palmerston North, New Zealand. 2014;
29. Rubatzky VE and YamagushiM. Edible aroids. In: Rubatzky, VE and YamagushiM.(Eds.) World Vegetables. Principles, Production and Nutritive Value, Chapman and Hall: International Thomson Publishing, New York. 1997;2:183-96.
30. FAO, Food and Agricultural Organisation species profile: 2013 Xanthosoma sagittifolium. Available on http://www.fao.org/docrep/T0646E/T0646E0o.htm
31. Bello Espinosa D. Apuntes para la flora de Puerto Rico. Primera parte. Anal Soc Española de Hist. Nat.1881;10:231-304.
32. CourseyDG. The edible aroids. World Crops. 1968;20:25-30.
33. Aroids BD. Plants of the Arum Family (2ndedn). Timber Press, Portland. 2000.
34. Tsafack TJJ and Gilbert C. Improvement Strategies for Cocoyam (Xanthosoma sagittifolium (L.) Schott). Conventional Breeding to Tissue Culture Technology. In: Bulbous Plants Biotechnology, Edited by Mérillon JM, CRC Press. 2014;75-91.
35. Ivancic A. Abnormal and unusual inflorescence of taro, Colocasiaesculenta (Araceae). Am J Bot.1995;43:475-489.
36. Hay A. Aroids of Papua New Guinea. Madang, Papua New Guinea: CRI publication.10.1990.
37. Wilson JE. Effects of formulation and the method of applying gibberellic acid on flower promotion in cocoyam. Exp agr. 1981;17:317-322.
38. Deo PC, et al. Improving taro (Colocasiaesculenta var. esculenta) production using biotechnological approaches. South Pacific J Nat Sci.2009;27:6-13.
39. Wilson JE. Agro facts, taro breeding, IRETA Publication No: 3/89 W. Samoa.
40. Yang Z, et al. A cytological study on three species of Colocasia (Araceae) from Yunnan. Caryologia.2003;56(3):323-327.
41. Kikuta A, et al. Seeds and seedlings of the taro, Colocasiaesculenta. Am J Bot.1938;25:186-188.
42. MacCaughey V and Emerson JS. The kalo in Hawaii. Hawaiian Forester and Agriculturist.1913;10:225-231.
43. PlucknettDL.Colocasia, Xanthosoma, Alocasia, Cyrtosperma, and Amorphophallus. In:Plucknett DL (ed) Tropical Root and Tuber crops Tomorrow. Univof Hawaii. 1970;1:127-135.
44. Huang XF, et al. Chromosomal ploidy identification of taro (Colocasia)germplasm resources. China Vegetables.2012;1:42-46.
45. Simmonds NW.Chromosome behaviour in some tropical plants. Heredity.1954;8:139-146.
46. Jones GE. Chromosome numbers and phylogenetic relationships in the Araceae. Ph.D. thesis, UnivVirg, USA. 1957.
47. Darlington CD and Wylie AP. Chromosome atlas of flowering plants. Allen and Unwin, London. 1955.
48. Fukushima E, et al. Chromosom numbers of the taro varieties cultivated in Japan. Chromosome Info Serv. 1962;3:38-39.
49. Guman, et al. The relationship between geographic distribution and ploidy level of taro, Colocasiaesculenta. Euphytica.1990;47:25-27.
50. Matthews PJ. The origins, domestication and dispersal of taro. PhD Thesis, Austral Nat Univ, Canberra. Australia. 1990.
51. Sreekumari MT and Mathew PM. Karyomorphology of five morphotypes of taro (Colocasiaesculenta (L.) Schott.). Cytologia. 1991;56:215-218.
52. Kokubugata G and Konishi T. Implication of a basic chromosome number of x=14 in seven cultivars of two varieties of Colocasiaesculenta by fluorescent in situ hybridization using rDNA probe. Cytologia.1999;64:77-83.
53. Isshiki S, et al. A probable origin of triploids in taro. J Jap Soc Hort Sci. 1999;68:774-779.
54. Parvin S, et al. Karyotype analysis of seven varieties of taro Colocasiaesculenta (L.) Schott. from Bangladesh. J BioSci. 2009;16:15-18.
55. Ramachandran K, Cytological studies on South Indian Araceae. Cytologia. 1978;43:289-303.
56. Krishnan R. and Magoon ML. Edible aroids-new insights into phylogeny. In: Leakey CLA (Ed.), Proc. 3rd Int. Sump Trop. Root Crops, Int. Inst. Tropical Agric. Nigeria. 1977;58-60.
57. Sreekumari MT. Cytological studies in taro: A review. J Root Crops. 1997;23:1-7.
58. Ekanem AM and Osuji JO. Mitotic index studies on edible cocoyams (Xanthosoma and Colocasiaspp.) Afr J Biotech. 2006;5:846-849.
59. Osuji JO and Nwala PC. Epidermal and cytological studies on cultivars of Xanthosoma (L.) Schott. and Colocasia (L.) Schott. (Araceae ). Int J Plant Soil Sci. 2015;4:149-155.
60. Kawahara T. Chromosome numbers of taros in Nepal and India. Chrom Info Serv. 1978;24:4-5.
61. Jackson GVH, et al.Tissue culture of taro (L.) Schott). J Hort Sci 1977;52:373-382.
62. Pfitzer P, Chromosomenzahlen von Araceen. Chromosoma.1957;8:436- 446.
63. Krishnan R, et al.Cytological studies in Xanthosoma sagittifolium. Nucleus. 1970;13:142-148.
64. Marchant CJ. Chromosome variation in Araceae II. Richardieae to Colocasieae. Kew Bull 1971;25:47-56.
65. Kuruvilla KM, et al.Karyomorphological investigations on aroids of North-Eastern Hills. J Cytol Genet.1989;24:13-22.
66. OnwuemeI. The tropical tuber crops: Yams, cassava, sweet potato and cocoyams. John Wiley and Sons, 1978;199-225.
67. Lakhanpaul S, et al. Analysis of genetic diversity in Indian taro [Colocasiaesculenta (L.) Schott] using random amplified polymorphic DNA (RAPD) markers. Genet Resour Crop Evol. 2003;50:603-609.
68. Comai L. The advantages and disadvantages of being polyploidy. Nat Rev Genet. 2005;6:836-846.
69. Tanimoto T and Matsumoto T. Variations of morphological characters and isozyme patterns in Japanese cultivars of Colocasiaesculenta Schott and C. gigantean Hook. Jap J Breed. 1986;36:100-111.
70. Irwin SV, et al. Molecular characterization of taro (Colocasiaesculenta) using RAPD markers. Euphytica.1998;99:183-189.
71. Schnell RJ, et al. Genetic similarities among cocoyam cultivars based on randomly amplified polymorphic DNA (RAPD) analysis. Scientia Horticulturae. 1999;80:267-276.
72. Offei S, et al. Genetic structure of eventy cocoyam (Xanthosoma sagittifoliumL. Schott) accessions in Ghana based on RAPD. Hereditas.2004;140:123-128.
73. Caillon S, et al. Nature of taro (Colocasiaesculenta(L.) genetic diversity prevalent in a Pacific ocean island, Vanua lava, Vanuatu. GenetResour and Crop Evol. 2006;53:1273-1289.
74. Sharma K, et al. Analysis of AFLP variation of taro population and markers associated with leaf blight resistance gene. Acad J Plant Sci. 2008;1:42-48.
75. Singh D, et al. Assessment and rationalization of genetic diversity of Papua New Guinea taro (Colocasiaesculenta) using SSR DNA fingerprinting. Genet Resour Crop Evol. 2008;55:811-822.
76. Macharia MW, et al. Genetic structure and diversity of East African taro (Colocasiaesculenta (L.) Schott). Afr J Biotech. 2014;13:2950-2955.
77. Cathebras C, et al.Characterization of microsatellites in Xanthosoma sagittifolium (Araceae) and cross-amplification in relatessd species. Appl Plant Sci.2014;2:1400027.
78. Brown VM and AsemotaH. PCR-based characterization of dasheen (Colocasia sp.) and cocoyam (Xanthosoma sp.). J Biotech Res. 2009;1:28-40.
79. Jackson GVH and Pelomo PM. Breeding for resistance to diseases of taro, Colocasiaesculenta, in Solomon Islands. In: International Symposium on Taro and Cocoyam, Visayas State College of Agriculture, Baybay, Leyte. Provisional Report. 1980;5:287-298.
80. Quero- García J, et al. Taro and cocoyam. In: Bradshaw JE (Ed.) Root and Tuber Crops, Hand Book of Plant Breeding. 2010;7:149-172.
81. Goenaga R and Hepperly P. Flowering induction, pollen and seed viability and artificial hybridization of taniers (Xanthosoma spp.). J AgrUniv Puerto Rico. 1990;74:253-260.
82. Amadi CO, et al. Hybridization and seed germination of taro (Colocasiaesculenta) in Nigeria. J Crop Improv. 2015;29:106.
83. Volin RB and Zettler FW. Seed propagation of cocoyam, Xanthosoma caracu Koch and Bouche. HortSci.1976;11:459-460.
84. Volin RB and Beale AJ. Genetic variation in F1 cocoyam (Xanthosoma spp.) hybrids.Proc. Florida State Hort Soc. 1981;94:235-238.
85. Onokpise OU, et al.Hybridisation and fruit formation in macabo cocoyam (Xanthosoma sagittifolium (L.) Schott). Ann Appl Biol. 1992a;120:527-535.
86. Jos JS, et al. Female fertility in tannia. Ann Bot. 1980;46:123-125.
87. Nzietchueng S.Some physiological characteristics of Pythium myriotylum, a pathogen of macabo root rot (Xanthosoma sagittifolium) in Cameroon. In:Degras L (Ed.) Proceedings of the 7th Symposium of the ISTRC. INRA, Paris. 1988;117-131.
88. Boudjeko T, et al. Peroxidase and pectin methylesterase activities in cocoyam (Xanthosoma sagittifolium L. Schott) roots upon Pythium myriotylum inoculation. J Phytopathology. 2005;153:409-416.
89. Nyochembeng LM, et al.Peroxidase activity, isozyme patterns and electrolyte leakage in roots of cocoyam infected with Pythium myriotylum. J Phytopathol.2007;155:454-461.
90. McDavid CR and Alamu S. Promotion of flowering in tannia (Xanthosoma sagittifolium) by gibberellic acid. Trop Agr. 1976;53:373-374.
91. McDavid CR and AlamuS. Effect of day-lenght and gibberellic acid on the growth and promotion of flowering in tannia (Xanthosoma sagittifolium). Trop Agr. 1979;56:17-23.
92. Saborío F, et al.Induccion de floracionentiquisque (Xanthosoma spp) encinco regions de Costa Rica. AgronomíaCostarricense. 2000;24:37-45.
93. Onokpise OU, et al. Acclimatization and flower induction of tissue culture derived cocoyam, (Xanthosoma sagittifolium Schott) plants. Agronomie. 1992;12:193-199.
94. Alamu S and McDavid CR. Promotion of flowering in edible aroids by gibberellic acid. Trop Agr. 1978a;55: 81–86.
95. Alamu S and McDavid CR.Effect of time and method of application of gibberellic acid on the growth and promotion of flowering in tannia (Xanthosoma sagittifolium). Trop Agr. 1978b;55:235-241.
96. Ferraren DO, et al. Flowering in taro (Colocasiaesculenta (L.) Schott) as affected by genotype and method of gibberellic acid application. Philipine J Crop Sci. 1991;16(1).
97. DostalR.Gibberellic acid and growth correlations. Nature (London).1959;183:1338.
98. Drew RA and Smith MK. Field evaluation of tissue culture banana in South-Eastern Queensland. Aust J Exp Agric. 1990;30:569-574.
99. Robinson JC, et al. A field comparison of conventional suckers with tissue culture planting over three crop cycles. J Hort Sci.1993;68:831-836.
100. Okada H and Hambali GG. Chromosome behaviours in meiosis of the inter-specific hybrids between Colocasiaesculenta (L.) Schott and C. gigantean Hook. fCytologia.1989;54:389-393.
101. Wang JK. Taro: A review of Colocasiaesculenta and its potential. HonululuUniv of Hawaii. 1983.
102. Esnard J, et al. Induction of tetraploidy in the tanier genotype “Inglesa” (Xanthosoma sagittifolium (L.) Schott). Plant Breeding. 1993;111:335-338.
103. Tambong JT, et al.In vitro induction of tetraploids in colchicine-treated cocoyam plantlets. Euphytica. 1998;104:191-197.
104. Oumar D, et al. Determination of ploidy level by flow cytometry and autopolyploid induction in cocoyam (Xanthosoma sagittifolium). AfrJ Biotech. 2011;10:16491-16494.
105. Guerra M. Chromosome numbers in plant cytotaxonomy: concepts and implications. Cytogen. Gen Res.2008;120:339-350.
106. Petersen G. Cytology and systematics of Araceae. Nord J Bot. 1989;9:119-166.
107. Cusimano N, et al. Maximum likelihood inference implies a high, not a low, ancestral haploid chromosome number in Araceae, with a critique of the bias introduced by ‘x’. Ann Bot. 2012;109:681-692.
108. Saikia JP, et al. Genome size determination and RAPD analysis of four edible aroids of Northeast India: The IIOAB J. 2010;1:25-27.
109. Chauhan KPS and Brandham PE. Chromosome and DNA variation in Amorphophallus. Kew Bull. 1985;40:745-758.
110. Agueguia A and Fatokun CA. Pollen storage in cocoyam (Xanthosoma sagittifolium (L.) Schott). Euphytica. 1988;39:195-198.
111. Tambong JT and Meboka M. Cocoyam (Xanthosoma sagittifolium L.) hybridization studies: Pollen viability and seed germination. Acta Horticulturae. 1994;380:448- 459.
112. Strauss MS, et al.Seed storage and germination and seedling proliferation in taro. Colocasiaesculenta (L.) Schott. Ann Bot. 1979;43:603-612.
113. Pardales JR. Floral morphology and biology, fruit and seed set, seed germination and seedling development of taro [in the Philippines. Ann Trop Res.1981;3:169-176.
114. Tyagi AP, et al. Seed germination and seedling development in taro (Colocasiaesculenta). The South Pacific J Nat Sci. 2004;22:61-65.
115. Price TV, et al. Effect of seed storage under ambient and cold temperatures on germination of taro seed (Colocasiaesculenta). Seed Sci Techn. 2007;35(3): 674-687.
116. Wagih ME. Seed rescue culture: a technique for regeneration of taro (Colocasiaesculenta). Afr Crop Sci J. 1994;2:9-15.
117. Ivancic A, et al. Development of visual tools for selecting qualitative corm characteristics of taro (Colocasiaesculenta(L.) Schott). Aust J Agri Res.2003;54:581-587.
118. Lebot V, et al.Comparative performance of local and introduced cultivars of taro (Colocasiaesculenta (L.) Schott) in Vanuatu. In: Proceedings of the 14th Symposium of the ISTRC. Thiruvananthapuram, Kerala, India, 2006;
119. Lebot V. The tropical root and tuber crops: cassava, sweet potato, yams and aroids. Crop Prod Sci. Hort Ser.2009;17:432.
120. Quero-García J, et al.Heritability of the main agronomic traits of taro. Crop Sci. 2006a;46:2368-2375.
121. Okpul T, et al.Genetic resources of taro (Colocasiaesculenta L.) Schott) in Papua New Guinea: A review. PNG JAgr For Fish. 2002;45:33-45.
122. Okpul T and Ivancic A. Hybridization of taro (ColocasiaesculentaL. Schott): floral development and stigma receptivity. PNG. J Agr For Fish. 1996;39:12-18.
123. Saborío F, et al. Induction of genetic variation in Xanthosoma spp. In: Genetic improvement of under-utilized and neglected crops in low income food deficit countries through irradiation and related techniques. International Atomic Energy Agency). Vienna, Austria.2004;143-154.
124. Lebot V, et al. Isozyme and RAPD variation among Phytophthora colocasiae isolates from South East Asia and the Pacific. Plant Pathol. 2003;52:303-313.
125. IBPGR. Descriptors for taro (Colocasiaesculenta). IPGRI, Rome, Italy. 1980.
126. IPGRI. Descriptors for taro (Colocasiaesculenta). IPGRI, Rome, Italy. 1999.
127. TANSAO. Annual Report. Taro Network for South East Asia and Oceania: Evaluation and breeding for rain-fed cropping systems in South East Asia and Oceania, Port Vila. 1998.
128. Devi AA, et al. Analysis of genetic diversity in edible aroid accessions of India based on morphological characters. J Root Crops. 2013;39:51-56.
129. Akus WL, et al. Taro germplasm collection and maintenance. In:Bubia Agricultural Research Centre, Annual Report Department of Agriculture and Livestock, Lae. 1989.
130. Godden D. Economic rationale for cooperation in germplasm conservation. Paper presented at ACIAR/NARI Workshop on Regional Co-operation in Conservation and Management of Plant Genetic Resources in the Pacific. Lae. 1999;
131. Lebot V. Taro (Colocasiaesculenta) genetic improvement: A need for internationalcollaboration. Aroideana.2005;28:166-173.
132. Okogbenin E, et al. Marker-assisted introgression of resistance to cassava mosaic disease into Latin American germplasm for the genetic improvement of cassava in Africa. Crop Sci. 2007;47:1895-1904.
133. Kreike CM, et al. Genetic diversity of taro, Colocasiaesculenta (L.) Schott, in Southeast Asia and the Pacific. Theor Appl Genet. 2004;109:761-768.
134. Poehlman JM. Breeding Field Crops.AVI Publ. Co. Inc. Westport Connecticut. 1987;451- 507.
135. Quero- García J, et al. First genetic maps and QTL studies of yield traits of taro (Colocasiaesculenta (L.) Schott). Euphytica. 2006b;151:187-199.