Volume 6, Issue 2, April Issue - 2018, Pages:370-385
|Authors: Nicolas OUEDRAOGO, Renan Ernest TRAORE, Pauline BATIONO/KANDO, Mahamadou SAWADOGO, Jean-Didier ZONGO|
|Abstract: Aim of present study was to describe the agro-morphological characterization of fifty (50) exotic varieties of taro introduced in Burkina Faso. Further agronomic performances and genetic parameters were also evaluated in present study. During study, 50 varieties which were introduced from 7 South-east Asian and Pacific countries were characterized. Nine quantitative and seventeen qualitative characters were studied during the study. Headsets of 250-300 grams were planted according to a randomized complete block design. The results showed a great variability among the agro-morphological characters which are may be because of genetic diversity. Moreover, the agronomic performances are significant compared with those of the local accessions. Also, the study revealed the existence of eighteen (18) significant positive correlations at a dissimilarity limit of 0.6 among the nine (09) quantitative characters studied, and associations between some qualitative and quantitative characters. Structuring the diversity allowed us to get five (05) different groups of phenotypes. The estimate of the genetic parameters indicates that the variability is more influenced by the genotype than the environmental factors.|
|Full Text: |
Taro is a tuber crop, a member of the Araceae family that is largely cultivated for its edible corms. It includes two different botanical morphotypes (Purselove, 1972), which differ basically by the size and shape of the main corm and its cormels: C. esculenta var. esculenta (dasheen type) is characterized by the presence of large central mother corm and very few (or no) side cormels, whereas C. esculenta var. antiquorum (eddoe type), has a small central mother corm with several well-developed side cormels. However, the differences are not always obvious because of the high variation within each group.
This staple crop is grown in tropical and subtropical countries and is consumed by over 400 million people worldwide (Bown, 2000).Taro is mainly produced in Africa (75% of the global production), Asia (21.3%), and Oceania (4.1%).It also plays an important role in food security (Akwee et al., 2015), especially in Oceania, where its per-capita annual consumption is amongst the highest in the world. Taro corms are valuable sources of carbohydrates. They contain considerable amounts of starch, about 21-26 % on fresh weight basis (Lebot et al., 2011). Small size granules offer smooth textured starch gel and enables high level of digestion (Nip, 1997). Corms are also moderately good sources of dietary fibers. The main sugar is sucrose, followed by glucose, fructose, mannose and xylose. In some varieties, maltose and a small amount of raffinose were also detected (Mbofung et al., 2006). Other abundant minerals are also magnesium, phosphorus, zinc and sodium. From a nutritional standpoint it is indeed unfortunate that taro is rather low in iron and manganese (Huang et al., 2007; Lewu et al., 2010). Taros can also be considered as relatively good sources of some vitamins and phenolic compounds. Studies of Huang et al. (2007) and Champagne et al. (2010) showed large amounts of β-carotene, thiamin, riboflavin and ascorbic acid.Taro is used in modern and traditional medicine. According to Watt (2014) in India, the juice of the petioles can be used to stop the arterial hemorrhages and to treat earaches.
In Burkina Faso, taro generates substantial incomes for producers, but its production does not satisfy the demand of local and supra-local markets (Traoré et al., 2013). According to the same author, taro is consumed at the national level before the period of harvest of staple food plants which are the cereals.
Despite this socio-economic importance, little attention is paid to taro among the international agricultural research community. Taro is one of the neglected plants also called "orphan" plant of the international agronomic research system, which according to Lebot (2002), do not form part of the mandate of the major research institutes and are thus neglected by Western researchers who tend to focus on the staple food plants, especially cereals.
To remedy this situation, a study on taro (Colocasia esculenta (L.)Schott) has been undertaken since 2004 in Burkina Faso in order to know the existing genetic diversity of the plant and to set up a national collection so as to preserve its heritage and to evaluate its potentialities for the varietal improvement. The first results of this study show that the genetic base of taro is narrow (Traoré, 2014; Chaïr et al., 2016).
Moreover, the recent creation of international collaboration allows to set up an international network of researcher on the valorization of taro throughout the world. This network is INEA (International Network for Edible Aroids) and gathers researchers of 16 developing countries, including Burkina Faso. The objectives of this network are to assemble and share the genetic resources of the taro (Colocasia esculenta) and to create new varieties adapted to the environmental and economic changes.
Thus in 2011, thanks to the international project ACPCCCC (Adapting clonally propagated crops to climatic and commercial changes) of network INEA (International Network for Edible Aroids), the University of Ouagadougou received a collection of fifty (50) varieties from seven (07) countries of South-east Asia and the Pacific regions in order to broaden the genetic base of taro in Burkina Faso. It proved to be necessary to carry out an agro-morphological characterization of these varieties in their hosting environment. Thus, the general objective of the present study is to determine the agro-morphological variability of the exotic varieties of taro in order to broaden its genetic base in Burkina Faso. More specifically, it aims to: (i) describe the agro-morphological characters of the exotic varieties; (ii) determine the agronomic performances of the exotic varieties; (iii) and evaluate the genetic parameters.
2 Material and Methods
2.1 Vegetable material
Fifty (50) taro varieties which were collected from seven (07) countries of South-East Asia and the Pacific were used as experimental material. These varieties were received in the form of vitro seedlings within the framework of international project ACPCCCC (Adapting clonally propagated crops to climatic and commercial changes).These verities were selected from the gene bank of genetic resources of taro preserved by the South Pacific Community (SPC) located in the Fiji Islands. The varieties received in 2011 were acclimatized under greenhouse then planted in the garden of the University of Ouagadougou. Planting material consists of headsets of 250-300 grams (Figure1). These headsets were obtained after 12 months’ growth of the vitro-seedlings in station.
The origin countries of collected varieties have various climatic zones with varied pluviometries. Table 1 presents the list of the vegetable materials, their origin and the annual pluviometry which characterizes each zone.
2.2 Experimental site
The experimentation was carried out in the experimental station of the Institute of the Rural Development (IDR) located in Gampèla at 12°24',29'' N and 1°21',66'' W (province of Kadiogo). This station is located at 18 km of Ouagadougou, on the Ouagadougou-Niamey route. It covers a surface of 490 hectares and is crossed from West to East by an affluent of Massili River (Bureau National des Sols, 1988).
The experimental station has Sudanian climate type characterized by a short rainy season extending from June to October and a long dry season from November to May (Tiombiano & Kampmann, 2010). Precipitations are not regular during the rainy season. The cumulated pluviometry of the agricultural season of 2012 amounts to 962 mm. The study was carried out on a sandy soil and deep ground (> 60cm) (Bureau National des Sols, 1988).
2.3 Experimental design
The test was established according to a randomized complete block design. Headsets were planted on twenty-five (25) elementary plots, i.e. two (02) varieties in each elementary plot. The difference between elementary plot and that between headsets holes was 1 m.
2.4 Practical farming
Headsets were planted on 28 July 2012. Transplanting was made in holes dug on a ground plowed using a tractor after organic spreading of manure. The 30 cm edged cubic holes are filled up to the 2/3 of their volume with compost mixed with organic manure fifteen (15) days before transplanting. The artificial fertilizer (NPK) was applied 4 months after transplanting, at a dose of 50 kg/ha. The spreading of manure was followed by weeding and study was conducted for 12 months. During the first (03) three months, watering was made with rainwater andsupplemented by water from wells during dry season. From Novemberuntil harvest, watering was carried out by manually. The weeding was carried out when time needed.
2.5 Studied characters
Twenty-six (26) agro-morphological characters including 17 qualitative characters and 9 quantitative characters were studied. These characters were selected among the descriptors of taro (IPGRI, 1999).
2.5.1 Quantitative characters
The agro-morphological quantitative characters were related to aerial as well as underground part. The aerial part is represented by the leaves. A leaf is made up of limb, petiole and sheath. Measurements were made on the last entirely developed leaf at the plant maximum growth stage, that is to say 6 months after transplanting. Thus, the height of plant (PLH) corresponding to the maximum vertical distance reached by the leaves starting from the basewas measured. The stolons are the horizontal underground stems of taro. The parameters studied for stolons are the number and length. The number of stolons (NUS) per principal plant was counted. The measure of stolon length (STL) was done on longest of them. The suckers being direct shoots on the collet of the plant, the number of suckers (NSU) around the principal plant was counted. The petiole length (PEL) as sheath length (SHL) were measured starting from the collet. The lamina length (LAL) was raised in the longest part and the lamina width (LAW) in the broadest part, the petiole was excluded. The principal corm weight (PCW) was weighed with maturity. Table 2 gives the modalities of the studied quantitative characters.
2.5.2 Qualitative characters
The qualitative characters were observed on the aerial and the underground part of the plant (Table 3). Concerning aerial part, the observations were related to the two last entirely opened leaves at the maximum stage of growth of the plant, that is to say 6 months after transplanting. The presence or absence of flowering was also noted. The observations concerned the shape of leaf lamina (SLL) which represents the position of the lobes of the lamina; Orientation of leaf lamina (OLL) which is the space position of the lamina compared to the ground; Leaf blade margin (LBM) which was appreciated according to whether the edges of this one are whole, corrugated broad and corrugated narrow; Leaf blade colour (LBC); Leaf main vein colour (LVC); Petiole junction colour (PJC); Colour of petiole top third (CPT); Colour of petiole middle third (CPM); Colour of petiole lower third (CPL), Leaf sheath colour (LSC); Corm flesh colour (CFC); Corm flesh fibre colour (CFF); Vein pattern (VEP) which is the form of the pigmentation on the veins of the lower face of the leaf; it was presented either in "V", or in "Y"; Sine between lobes (SIL) which is the incision between the two lobes of the limb.This character was noted according to the degree of opening of the incision; Flowering (FLO) is observed only for some varieties while others did not have flowering stage.
The observations of the underground part concernedonly corms at maturity, collected twelve months after transplanting. These charactersare relatedtocorm shape (COS), which was observed on the principal corm; Degree of fibrousness of corm (DFC) which was appreciated according to the more or less significant presence of fibers on the surface of the principal corm.
2.6 Statistical analyses
The descriptive analysis of the characters was made. The calculation of the genetic parameters was done from the components of the
analysis of variance and according to formulas used by Assefa et al. (2001), Rex (2002) and Hosseini et al. (2012). Table 4 presents the estimated genetic parameters and their calculation formulas. EXCEL (2013) software was used for that.
The following analyses were made with the software XLSTAT-Pro 7.1. Thus the variance analysis (ANOVA) was made to determine the discriminating characters. Prior to this analysis, normality was tested. Three (03) of the nine (09) quantitative characters, namely number of stolons (NUS), stolon length (STL) and number of suckers (NSU) did not present a normal distribution and were thus transformed by the square root (Quero Garcia, 2004). Only the character number of suckers (NSU) then presented a normal distribution after the transformation by the square root. Thus ANOVA concerned seven of the nine studied quantitative characters. The matrix of correlation was produced to study the relations between the quantitative characters. The analysis of the multiple correspondences (ACM) carried out to determine associations between the characters, related to two quantitative characters (plant height, principal corm weight) and four qualitative characters (colour of petiole middle third, flowering, corm flesh colour, corm flesh fibre colour). The characters were selected according to their importance for the improvement of taro.
The ascending hierarchical clustering (CAH) was carried out with four quantitative characters (height of the plants, numbers of stolons, number of suckers, and weight of the principal corm) to define the structuring of the variability of the collection. The four characters were selected because of their importance in the criteria of selection. The five others were eliminated because of existing of
significant positive correlations with those selected. The groups obtained by the ascending hierarchical clustering (CAH) were characterized by the discriminating factorial analysis (AFD).
3.1 Variation of qualitative characters
The results presented in table 5 showed a variability in studied qualitative characters. Modalities concerning the colour of the various parts of the leaf are varied (Figures 2 and 3). Among the observed colours, prevalence of the green colour leaf blade (72%) with variants viz., dark green (14%), yellow-green (12%), green-purple (2%) was observed. The colour of the principal vein as that of the petiolar junction is often different from that of the leaf blade. Some varieties have a uniform petiole colour while others have petioles of which the various parts (top third, middle third and lowerthird of petiole, sheath foliar) are different colours. The very distinctive form of the corm presented six (06) modalities (Figure 4) of which most significant is the elliptic form (64%).The colour of the corm flesh is also varied (Figure 5). However the majority of studied varieties (78%) have white colour flesh of corms. Purple colour (8%), pink colour (6%) and yellow colour (8%) of corm flesh were also observed. The flesh of the corms contains fibres whose colour is different from the rest of the flesh. Some studied qualitative characters are relatively less varied. Only two modalities were observed for following characters such as leaf lamina shape (Cup-shaped: 80%; flat: 20%), vein pattern (Y: 76%; V: 24%) and sine between lobes (Acute: 90%; Broad: 10%). The results of this study revealed a low reproductive capacity by sexual way of the varieties of the collection. Flowering is absent in 88% of the varieties, rare in 8% and is present in only 4% of varieties.
3.2 Variation of quantitative characters
The results of the descriptive analysis have been represented in table 6, result of study revealed highly significant values of F to the threshold of 5% for all the studied quantitative characters which presented a normal distribution. Various characteristics such as plant height (PLH), number of suckers (NSU), petiole length (PEL), sheath length (SHL), lamina length (LAL), lamina width (LAW) and principal corm weight (PCW) were analyzed in present study. The coefficients of variation are very high (above 100%) for some characters such as number of stolons (NUS), stolon length (STL) and number of suckers (NSU). However, in addition to the last characters, the length of the sheath and the weight of the principal corm have a coefficient of variation higher than 30 %. The study revealed that the individuals of the varieties of the collection have an average height of 104.44 cm with a maximum of 182.72 cm and a minimal height equal to 34.35 cm. Stolons were either absent or present with an average number of 5.39 and the maxima up to 48.89. In varieties those are showing suckers, the maximum number suckers observed was 71.60. The values of the size of the various parts of the leaf also had significant amplitudes. The weight of the corms varied between 0.44 kg and 1.65 kg with an average of 0.5 kg among the varieties of the collection.
3.3 Relations between characters
3.3.1 Correlations between studied characters
The matrix of correlation of Pearson represented by table 7 revealed the existence of eighteen (18) significant correlations all positive to the threshold of dissimilarity of 0.6 between the
nine (09) studied quantitative characters. Plant height (PLH) is correlated with leaf dimensions (PEL, SHL, LAL, LAW) and principal corm weight (PCW). In the same way the various variables of the leaf are correlated among them and are correlated with the principal corm weight. Further, a correlation between the number of stolons (NUS), stolon length (STL) and number of suckers (NSU) were also reported. All the negative correlations are non significant.
3.3.2 Association of the characters
The results of multiple correspondences (ACM) analysis are displayed in table 8 revealing that axis 1 (27.44% of inertia) associates with positive values variables such as crimson colour of petiole middle third (CPM-crimson), green colour of petiole middle third (CPM-Green), and pink corm flesh colour (CFC-pink), pink colour of corm flesh (CFF-pink) and height of the dwarf plant (PLH-Dwarf) on the one hand; variable like purple colour of petiole middle third (CPM-purple), white corm flesh colour (CFC-White), great height of the plant (PLH-Large) and large principal corm weight (PCW-Large) are associated with negative values.
As for axis 2 (16.34% of inertia), it associates the positive values of the variables red colour of petiole middle third (CPM-Red), purple corm flesh colour (CFC-Purple), purple colour of corm flesh fibre (CFF-Purple) and the average principal corm weight (PCW-Average).This axis associates the negative values of the variables are yellow corm flesh colour (CFC-Yellow), yellow colour of corm flesh fibre (CFF-Yellow), dwarf plant height (PLH-Dwarf) and small principal corm weight (PCW-Small).
3.4 Structuring of diversity
The dendrogramme resulting from the ascending hierarchical clustering (CAH) helps to identify five groups of morphological diversity (figure 6).This regrouping of the varieties is not dependent on their origin country. Group A is composed of 3 varieties from 3 countries, the group B of 11 varieties from 6 countries, the group C of 4 varieties from 3 different countries, the group D consists of 9 varieties from 5 countries and the group E is made of 23 varieties from 6 countries. Table 9 displays the composition of the groups resulting from the CAH of the 50 varieties.
3.5 Characterization of the groups resulting from the CAH
Table 10 gives the percentage of inertia and the definition of the axis in the discriminating factorial analysis (AFD). Axis 1 cumulates 72.44% of inertia and is defined by the characters plant height (PLH) and principal corm weight (PCW); it is the yield axis. Axis 2 of inertia 26.01% is defined by the characters number
of stolons (NUS) and number of suckers (NSU); it is the axis of the wild characters of the varieties.
Figure 7 represents the various groups resulting from the CAH in the factorial design ½ (98. 45% of total inertia).The groups show the following characteristics: Group A is consisted of the varieties of high height giving large corms. Those do not have stolons and have an average number of suckers. These varieties are improved and productive. Group B consists of small height varieties producing small corms. These varieties do not develop stolons and have a low number of suckers. Group C comprises varieties of average height with smallest corms. This group is characterized especially by a remarkable presence from stolons and suckers which are with wild characters. Group D consists of varieties of high height and producing largest corms. We notice an absence of stolons and a low presence of suckers. Group E is made of varieties of average height producing average corms. These varieties do not present stolons and contain an average number of suckers. This group contains more individuals.
Table 11 displays the results of the estimate of the genetic parameters. Result of study revealed highly strong values of heritability for all studied quantitative characters. These values are higher than 40 % and vary from 93.00 % for the character weight of principal corm to 98.16 % for the length of the petiole. High genotypic and phenotypic variances for plant height was also reported (VG=682.712; VP=696.840); those of the weight of the principal corm are weakest (VG=0.055; VP=0.060). For all the studied characters, phenotypic variances are slightly higher than the genotypic variances. The genetic gain expected compared to the average of the character (GAx) is very high for all the studied characters. They vary from 2569.603 for the number of suckers to 6235.427 for the weight of the principal corm.
4.1 Agro-morphological diversity of exotic varieties
The results of this study revealed a significant morphological and agronomic variability within the collection. Variability was observed for both qualitative and quantitative characteristics. The qualitative characters presented several modalities while forthe quantitative characters, high coefficients of variation and the highly significant values of F were reported. The studied varieties are different in morphological and agronomical point of view. This phenotypic variability would be due to a genetic diversity. Indeed, colour of the various parts of the leaf is also strongly varied. According to Lebot et al., (2004) and Paul &Bari (2011) this variability of the colour of the leaf depends on the genotype.
As taro is reproduced mainly by vegetative propagation, low intraspecific variability is expected. According to Okpul et al. (2004), this variability can be ascribed to sexual recombination, migration and perhaps to the change followed by a later selection of the farmers on the basis of the various agro-ecological modes, the farming systems and culinary preferences. Moreover, the studied varieties are from an international center of conservation of the genetic resources of taro. This center is the South Pacific Community (SPC) which preserves many taro varieties from various countries of South-east Asia and the Pacific. These countries are well-known to have a tradition of taro research. Some of the varieties were already selected or improved during former taro cultivation programmes. The absence of stolons in certain varieties confirms the enhanced character of the latter. The introduction of the collection could contribute to broaden the genetic base of taro in Burkina Faso which according to Traoré et al. (2013) is narrow.
4.2 Agronomic potentialities of exotic varieties
Colocasia esculenta is a species cultivated mainly for its thickened rhizomes. Its leaves are also consumed and some time it is also used as decorative plant. It is grown in Burkina Faso especially for its edible corms. The varieties of the studied collection present in terms of performance a significant agronomic potentiality. The yield only evaluated on the principal corm gives weights ranging from 0.44 kg to 1.65 kg. Moreover, the size of the leaves has high values for this verity. The length of the limb varies between 19.55 cm and 66.50 cm, the width between 15.94 cm and 51.60 cm. Traoré (2006) found within a collection of local cultivars an average weight of the principal corm equal to 0.33 kg, a maximum weight equal to 0.52 kg, a length of the limb which varies between 17.4 cm and 23.19 cm; a width of the limb set between 13.49 cm and 18.88 cm. Similarly, Ivancic &Lebot (2000) reported limb length between 30 and 80 cm and width between 20 and 50 cm. Thus, the varieties of present collection are more powerful than the local accessions and are close to the maximum performances.
4.3 Description of exotic varieties
This study showed positive correlations between the quantitative characters of the leaves and the parameters of the yield. These results are similar to the findings of Traoré et al. (2013). The positive correlation between dimensions of the leaf particularly the length and the width of the limb with the weight of the principal corm, means that the wingspan of the plant at the maximum of vegetative development stage gives an idea of the expected yield at harvest.
The results of the ACM show that within the collection, we can distinguish two types of varieties of small height. There are varieties of small height having purple or green colour of petiole middle third and whose flesh as well as fibres of the flesh are of pink colour (BL/SM/13; BL/SM/136). The other type of varieties of small height produces corms whose flesh is of yellow colour (CE/THA/07). The varieties of high height produce large corms whose flesh colour is white and the colour of the petiole middle third is purple (BL/SM/143; BL/SM/148; BL/SM/08). As for the varieties of average height, they produce corms whose corm flesh colour is crimson and the petiole middle third is of red colour (BL/SM/80).The groups identified by the CAH and characterized by the AFD are composed of more or less homogeneous varieties. The characterization of the groups rested especially on criteria of selection that is agronomic performance. The groups A and D are composed of varieties which produce large corms and not have stolons. These varieties seem to be improved and will likely be of interest taro producers. Their popularization would contribute to improve the productivity of taro in Burkina Faso and thereby ensure food security. The group E contains the most varieties. The varieties of this group give average corms. These varieties have an average number of suckers and that constitutes a significant character for the availability of the material of propagation. The group C consists of varieties characterized by the strong presence of stolons. According to Lebot et al. (2004), the strong presence of the stolons is one of the characteristics of the wild forms of taro. The wild forms generally contain resistant genes. These varieties could be used as sources of genes necessary in varietal creation. The group B is composed of varieties of the eddoe botanical variety, less demanding in water and adapted to rain production. Corms of these varieties are also well liked by some consumers because of their high mucilage content which makes them sticky.
4.4 Significance of the results of the estimate of the genetic parameters
The possibility of hybridization between the varieties for varietal creation in order to obtain varieties adapted to the pedoclimatic conditions is the immediate prospect for this study. However, hybridization is useful only if observed variability is related to a genetic diversity. According to Lakshmana et al. (2009), a weak difference between the phenotypic and genotypic coefficients of variation of a character indicates environment has very little influenced on genetic diversity. Also, the joint estimate of the heritability and genetic progress (GAx) can bring more reliable information (Govindaraj et al.,2011). The results obtained from the estimate of the genetic parameters namely the values of the heritability in the broad sense very high combined with the weak differences between the phenotypic and genotypic coefficients of variation on the one hand and on the other hand the genetic profits expected compared to the average of the character (GAx) strongly high indicate that the studied characters are influenced very little by environmental factors. The expression of the characters is therefore due to the genotype. The variability observed is especially influenced by the expression of the genes. According to Visscher et al. (2008), the phenotype is a good predictor of the genotype. Hybridizations will then influence the studied characters.
The characterization of the 50 exotic varieties of taro under the climatic conditions of Burkina Faso allows to detect a significant agro-morphological diversity strongly influenced by the expression of the genes. Several groups with different interests were made up and should allow a rational use of the varieties of the collection. Some of them could be popularized, and others could be used as sources of genes in programmes of varietal improvement of the taro. Their introduction will contribute to increase the genetic diversity of taro in Burkina Faso and will offer more increased possibilities for the creation of new varieties adapted to the pedoclimatic conditions of the country.
To enhance knowledge on the varieties of the collection, further studies are needed. It is a comparative study with the local accessions to confirm the contribution of this collection to the productivity of taro in Burkina Faso, an evaluation of precocity and of resistance to water stress which will allow toselect the varieties adapted to the pedoclimatic conditions and a molecular characterization.
We would like to gratefully thank the South Pacific Community (SPC) in Fiji, all partners of the INEA (International Network for Edible Aroids) network and the ACPCCCC (Adapting clonally propagated crops to climatic and commercial changes) project for carrying out of this study.
Assefa K, Tefera H, Merker A, Kefyalew T,Hundera F (2001) Variability, heritability and genetic advance in Phenomorphic and agronomic traits of Tef [Eragrostistef (Zucc.)Trotter] germplasm from eight Regions of Ethiopia.Hereditas 134: 103–113.
Bown D (2000) Aroids Plants of the Arum family. Timber Press, Portland, OR.
Bureau National des Sols(1988) Etude pédologique de la station expérimentale de Gampela, échelle 1/5000. N° 59, Pp.279.
Chaïr H, Traoré RE, Duval MF, Rivallan R, Mukherjee A, Aboagye LM (2016) Genetic Diversification and Dispersal of Taro (Colocasiaesculenta (L.)Schott).Plose one 11: 0157712.
Champagne A, Bernillon S, Moing A, Rolin D, Legendre L,Lebot V (2010) Carotenoid profiling of tropical root crop chemo types from Vanuatu. Journal of Food Composition and Analysis 23: 763–771.
Govindaraj M, Selvi B, Rajarathinam S, Sumathi P (2011) Genetic Variability and Heritability of Grain Yield Components and Grain Mineral Concentration in India’s Pearl Millet (PennisetumGlaucum (L) R. Br.) accessions. African Journal of Food, Agriculture, Nutrition and Development11: 4758-4771.
Hosseini SJ, Sarvestani ZT, Pirdashti H, Afkhami A,Hazrati S (2012) Estimation of heritability and genetic advance for screening some rice genotypes at salt stress conditions.International journal of Agronomy and Plant Production3:475-482.
Huang CC, Chen WC, Wang CCR(2007) Comparison of Taiwan paddy- and upland- cultivated taro (Colocasiaesculenta L.) cultivars for nutritive values. Food Chemistry 102:250–256.International Plant Genetic Resources Institute (1999) Descripteurs du taro (Colocasiaesculenta).Via de Delle Settle Chiese 142, 00145 Rome, Italie, ISBN 92-9043-410-4, Pp.56 .
IPGRI. 1999. Descriptor list for Colocasia. International Plant Genetic Resources Institute, Rome, Italy.
Ivancic A,Lebot V (2000) The genetics and breeeding of taro. CIRAD, ISSN 1251-7224, ISBN 2-87614-414-X, 194 p.
Lakshmana D, Biradar BD,Ravikumar RL (2009) Genetic variability studies for quantitative traits in a pool of restorers and mainteners lines of pearl millet (Pennisetumglaucum (L.)). Karnataka Journal of Agricultural Sciences 22: 881-882.
Lebot V, Malapa R, Bourrieau M (2011) Rapid estimation of taro (Colocasiaesculenta) quality by near-infrared reflectance spectroscopy. Journal of Agricultural and Food Chemistry 59: 9327–9334.
Lebot V, Prana MS, Kreike N, Van HH, Pardales J, Okpul T, Gendua M, Thongjiem M, Hue H, Viet N, Yap TC (2004) Characterization of taro (Colocasia esculenta (L.)Schott) genetic resources in Southeast Asia and Oceania.Genetic Resources and Crop Evolution 51: 381-392.
Lebot V (2002) La domestication des plantes en Océanie et les contraintes de la voie asexuée. Journal de la Société des Océanistes 1-2 : 114-115.
Lewu MN, Adebola PO, Afolayan AJ(2010) Comparative assessment of the nutritional value of commercially available cocoyam and potato tubers in South Africa. Journal of Food Quality 33:461-467
Mbofung CMF,Njintang YN, AboubakarAbdouBouba A, Balaam F (2006) Physicochemical and functional properties of six varieties of taro (Colocasia esculenta (L.) Schott) flour. Journal of Food Technology 4: 135–142.
Nip WK (1997) Taro. In: Smith DS, Cash JN, Nip WK and Hui YH (Ed), Processing vegetable and technology. Technomic Publishing, Pensylvania, USA, pp. 355-387.
Okpul T, Singh D, Gunua T,Wagih ME (2004) Assessment of diversity using agro-morphological traits for selecting a core sample of Papua New Guinea taro (Colocasia esculenta (L.) Schott) collection. Genetic Resources and Crop Evolution 51: 671-678.
Paul KK, Bari MA (2011) Accession characterization and genetic variability studies in Colocasiaesculenta(L.) Schott. Bangladesh Journal of Scientific and Industrial Research 46: 235-242.
Akwee PE, Netondo G, Kataka JA, Palapala VA (2015) A criticalreview of the role of taro (Colocasia esculenta(L.) Schott) to food security: A comparative analysis of Kenya and Pacific Island taro germplasm. Scientia Agriculturae 9 : 101-108.
Purseglove JW (1972) Tropical crops, monocotyledons. Longman: London.
Quero Garcia J (2004) Diversité génétique et amélioration des taros du Vanuatu. PHD Thesis submitted to Ecole Nationale Supérieure Agronomique Montpellier, France, Pp. 197.
Rex B (2002) Breeding for quantitative traits in plants. Stemma press Woodbury MN55125, Hardback, Pp. 369, ISBN 0-9720724-0-3.
Tiombiano A, Kampmannn D (2010) Atlas de la biodiversité de l’Afrique de l’Ouest, Tome II : Burkina Faso. Ouagadougou et Frankfurt/Main.
Traoré ER (2006) Etude de la variabilité agromorphologique d’une collection de taro (Colocasiaesculenta (L.) Schott) originaire des domaines soudanien et soudano-guinéen du Burkina Faso. DEA, Université de Ouagadougou, Burkina Faso, Pp. 54.
Traoré ER, Nanema RK, Bationon/Kando P, Sawadogo M, Nébié B, Zongo JD (2013) Variation agromorphologique dans une collection de taro (Colocasia esulenta (L.) Schott) adapté aux conditions de culture pluviale au Burkina Faso. International Journal of Biological and Chemical Sciences 7 : 1490-1502.
Traoré ER (2014) Etude de la diversité du Taro (Colocasia esculenta (L.) Schott): cas d’une collection du Burkina Faso et d’une collection internationale. Thèse unique, Université de Ouagadougou, Burkina Faso, Pp.161.
Visscher PM, Hill WG, Wray NR (2008) Heritability in the genomicsera-concepts and misconceptions. Nature Reviews Genetics 9: 255-266.
Watt G (2014) A dictionary of the economic products of India, vol 2.Cambridge University Press, Cambridge, Pp 696.