Evaluation and phenotypic plasticity of taro [Colocasia esculenta (l.) Schott.] genotypes for nutrient and anti-nutrient composition

The study was carried out to determine the nutritional and anti-nutritional composition of taro genotypes and also determine the phenotypic plasticity of the genotypes in two agro ecological zones in Ghana. The towns and zones were Bunso in the semi deciduous forest (an upland) and Tano Dumasi in the forest savannah transition agro-ecological (a waterlogged area) zone in the Eastern and Ashanti regions respectively.Two (2) freshly harvested corms of each genotype from each location were assessed for their nutritional (moisture, protein, carbohydrate, ash and fat) and anti-nutritional (phytate, oxalate and tannin) composition Data collected were subjected to analysis of variance and AMMI analysis using GenStat 12 edition to assess the effect of genotype, environment and their interaction on the traits studied. Phenotypic plasticity for the genotypes and the traits studied was also calculated. Pearson correlation was also conducted to assess the relationship between the traits studied. There were significant differences among the genotypes for nutrient and anti-nutrient composition except for percentage fat, indicating enough genetic variability among the genotypes, giving room for good selection progress for development of taro varieties. A higher magnitude of the environment over genotype and genotype by environment interaction observed indicates the influence of environment in the expression of the nutritional and anti-nutritional traits. Observed varied phenotypic plasticity among the genotypes for the nutrient and anti-nutrients composition also indicates varied adaptation of the genotypes to the environment. Genotypes BL/SM/115, CE/MAL/32 and CE/IND/16 and hybrids KAO19 × CE/MAL/32 and CE/IND/16×KAO19, CE/IND/16 × BL/SM/10, and CE/IND/16 × BL/SM/115 which recorded high nutrients and low anti-nutrients content and were stable across the environments can be released to farmers for cultivation. They could also be included in breeding programs for the development of enhanced nutritional quality of taro in Ghana.


Introduction
Root crops such as taro [Colocasia esculenta (l.) Schott.]play a major role in the food security of many developing countries as it serves as an important food or subsistence crop for millions of people in these developing countries [1] (Misra et al., 2008).Taro is ranked the fifth most utilized root tuber after cassava, potato, sweet potato and yam in the world [2,3] (Bamidele et al., 2014;Igbabul et al., 2014).It also has the potential of improving the livelihood of many poor people who depend on it for food [4] (Akalu and Geleta, 2017).
Taro has a lot of food and medicinal properties.All the plant parts that is the young leaves (petiole and blade), corms and cormels are used as vegetables in sauces and soups [5][6][7] ( Owuamanam et al., 2010;Angami et al., 2015;Matthews, 2014).The corms can be consumed after baking, roasting, steaming or boiling.It can also be fried as chips, dried and made into flour for preparation of pastries [8] (Krishnapriya and Suganthi, 2017).The corms contain anthocyanins, cyanidin 3-glucoside and flavonoids which improve blood circulation by strengthening the capillaries of the heart [9] (Wagner, 1985).It also acts as potent antioxidants and anti-inflammatory agent and inhibit human cancer cell growth [10] (Youdim et al., 2000).
The oxalate compounds in raw taro are responsible for theacridity in taro therefore, it is advisable to process taro before consumption [17] (Bradbury et al., 1995).In Ghana, taro varieties with higher nutritional content and low oxalate, phytate and tannin content are unavailable.There is also limited genotype by environment interaction (GEI) analysis on nutritional components of taro limiting selection and recommendation of suitable genotypes which possess desirable nutritional traits for food, feed and industrial use [18,19] (Andrade et al., 2016;Tumwegamire et al., 2016).Breeding for improved nutrient content in taro will go a long way to improve the nutritional status of most rural poor in Ghana.Nutritional composition of crops is highly influenced by genotype by environment interaction and therefore the genotypes generated should be evaluated under different environmental conditions to assess their phenotypic plasticity of environment.Phenotypic plasticity of a genotype is the extent to which their performance varies under different environmental conditions [20,21] (Bradshaw, 2006;Lande, 2009).Phenotypic plasticity evaluations are also important in the face of climatic change [22] (Hidayatullah et al., 2020).When the plasticity of a genotype for a particular trait near zero, the genotype is deemed stable and when it is far from zero, the genotype is plastic that is unstable [23] (Tripodi et al., 2020).Understanding the phenotypic plasticity of genotypes is crucial in breeding programs as it aids in selecting genotypes for varietal improvement and for predicting theirperformances in different environments [24] (Tchokponhoue et al., 2019).Phenotypic plasticity helps the breeder to understand the crops adaptation and the links between phenotype and genotype [25,26] (Sadras and Trentacoste, 2011; Mohammadi, 2014).
The objective of the study therefore was to evaluate and determine the phenotypic plasticity of parents and single cross hybrids of taro for nutritional and anti-nutritional composition under two agro-ecological environments and select stable genotypes of taro that are low in anti-nutrients and rich in beneficial nutrients and health related functional traits for varietal development of taro in Ghana.Recently, [51] Oue ´draogo et al (2023) have reported on the nutritional composition in Bukina Faso, [33] Boampong et al (2019) also reported on nutritional composition of taro accessions in Ghana, however this paper assesses the nutritional and anti-nutritional composition of parents and hybrids of taro in two contrasting environments.

Research materials
Harvested corms of equal weight from parents and single cross hybrids were used in the chemical analysis.The genotypes used for the chemical analysis were involved in the evaluation of yield and yield components of taro.Table 1 shows the names of parents, hybrids and details of crosses used in the research.The genotypes used as parents except for KAO 19 which is a local accession from Ghana were introduced in-vitro into Ghana as part of the germplasm distributed during 2004-2014 by Centre for Pacific Crops and Trees (CePaCT) as part of their "Fighting TLB Disease in Samoa and at Global Level Through Networking and Sharing Genetic

Details Name
Resources" program due to the narrow genetic base of taro in Ghana.All the exotic genotypes were conserved at Plant Genetic Resource Research Institute (PGRRI) of the Council for Scientific and Industrial Research (CSIR), Bunso in the Eastern region of Ghana.

Research sites
2.2.1 Environment 1 -Bunso.The research was conducted at Plant Genetic Resources Research Institute of the Council for Scientific and Industrial Research (PGRRI-CSIR), research field at Bunso in the Eastern region of Ghana.Bunso is located in the semi deciduous forest agro-ecological zone at longitude 0˚27.634´W,latitude 6˚17.715´N and altitude 208.4m.The region is known for the large-scale cultivation and consumption of taro.The Bunso research site was an upland.

Environment 2 -Tano Dumasi.
The research was also conducted at the Ministry of Fisheries and Aqua Culture research site at Tano Dumasi in the Sekyere South District of Ashanti Region.Tano Dumasi lies in the forest savannah transition agro-ecological zone at longitude13 0.2360´W, latitude 6˚53.2530´N and altitude 277.1m.Tano Dumasi was previously known for the cultivation of taro but currently most of the taro fields are used for rice cultivation due to the incidence of taro leaf blight disease.The field was a waterlogged area or a lowland.
The two fields were selected for the research to also assess the effect of moisture on the nutrient and anti-nutrient composition of taro, since taro is believed to be cultivated mostly in waterlogged areas in Ghana.
Table 2 shows the monthly average temperature and rainfall at the research sites for the period of the research.Table 3 shows the routine soil analysis and the soil type of the research sites.

Nutrient and anti-nutrient analysis
Two corms were randomly selected from the harvested corms from each genotype and sent to the Food Science Laboratory of the Food Science and Technology Department of Kwame Nkrumah University of Science and Technology (KNUST) [27].The AOAC (1990) protocol for chemical analysis was used for nutrient and anti-nutrient analysis.The analysis was done in triplicates for all the nutritional and anti-nutritional traits.

Protein determination.
The protein content was determined using the Kejdahl method as follows: Digestion.Two (2) g of sample and a half of selenium-based catalyst tablets were put in a digestion flask with a few anti-bumping agents.Twenty-five (25) ml of concentrated H 2 SO 4 was then added and shaken until the entire sample was thoroughly wet.The flask was put on digestion burner and heated slowly until boiling ceased and the resulting solution was clear.The mixture was then cooled to room temperature.The digested sample solution was transferred into a 100 ml volumetric flask and made up to the mark.
Distillation.Twenty-five (25) ml of 2% boric acid was pipetted into a 250 ml conical flask and two drops of mixed indicator added.The conical flask and its contents were placed under a condenser in such a position that the tip of the condenser is completely immersed in solution.Ten (10) ml of the digested sample solution was measured into the decomposition flask of the Kejdahl unit and fixed.Excess of 40% NaOH (about 15-20 ml) was added to it.The ammonia produced was distilled into the collection flask with the condenser tip immersed in the receiving flask till a volume of about 150 ml-200 ml was collected.
Titration.The distillate was titrated with 0.1N HCl solution.Acid was added until the solution was colorless.The nitrogen content was determined in triplicate, and a blank determination was run using the same amount of all reagents as used for the sample.
The total nitrogen was then calculated as follows: Va-volume in ml of standard acid used in titration Vb-volume in ml of standard acid used in blank NA-normality of acid W-Weight of sample taken 2.3.2Ash content.Five (5) g sample was weighed into a tarred crucible.The crucible was placed in a cool muffle furnace and ignited for 2 hours at about 600 degrees.The muffle furnace was turned off and opened until the temperature dropped to at least 250 degrees preferably lower.The door was opened carefully to avoid losing ash that may be fluffy.Safety tongs were used to quickly transfer crucibles to a desiccator with a porcelain plate and desiccant.The desiccator was closed and crucibles allowed to cool prior to weighing.The ash content was calculated as follows:

Fat content. The fat content was extracted using the soxhlet extraction as follows:
A previously dried (air oven at 100˚C) sample was accurately weighed into 250 ml round bottom flask.Five (5) grams of dried sample was weighed unto a 22 ×80mm paper thimble or a folded filter paper.A bit of cotton or glass wool was placed into the thimble to prevent loss of the sample.One hundred and fifty (150) ml of petroleum spirit B.P 40-60˚C was added to the round bottom flask and the apparatus assembled.Condenser was connected to the soxhlet extractor and reflux for 4-6 hours on the heating mantle.After extraction, the thimble was removed and the solvent recovered by distillation.The flask and fat/oil were heated in an oven at about 103˚C to evaporate the solvent.The flask and its content were cooled to room temperature in a desiccator.
The flask was weighed and weight of fat/oil collected determined as follows: 2.3.4Moisture content.The moisture content was analyzed using the oven drying method as follows: Five (5) g of sample was transferred to previously dried and weighed dish.The dish was placed in an oven thermostatically controlled at 105 degrees for 5 hours.The dish was removed and placed in a desiccator to cool to room temperature and weighed.The dish was dried again for 30 minutes, cooled down and weighed.The drying was repeated, cooled and weighed until a constant weight was reached.The moisture content was determined as follows: 2.3.5 Carbohydrate.The total carbohydrate was calculated as follows 2.3.6 Phytate determination.Four (4) g of taro sample was soaked in 100ml of 2% HCl for 3 hours and filtered through Whatman filter paper.Twenty-five (25) ml of the filtrate was weighed into 250 ml conical flask and 5ml of 0.3% ammonium thiocyanate solution added as an indicator and 53.5ml distilled water was added to give the desired acidity.The solution was titrated with standard iron III chloride solution which contained about 0.001905 g of iron per ml until a brownish yellow colour was attained and persisted for 5 minutes.The phytate content was calculated as: 2.3.7 Oxalate determination.The oxalate content in all the samples were analyzed by following titration method using KMnO 4 described in [27] AOAC (1990).In the determination of oxalate, 1 g of each selected samples were weighed and mixed with 20 ml of 0.1M HCl in a 50ml beaker to extract total oxalate.All beakers with samples and extracting solvents were kept in a water bath at 100˚C for 30 minutes, later filtrated using Wattman No 1 filter paper and 0.5 ml of 5% calcium chloride was added to the filtrate to precipitate out calcium oxalate, the precipitate was separated by centrifugation at 3500 rpm for 15 minutes, and supernatant was discarded.The calcium oxalate precipitate was washed with 2 ml of 0.35 M ammonium hydroxide and then dissolved in 0.5 M of sulphuric acid.The dissolved solution was titrated with 0.1 M of potassium permanganate at 60˚C till faint pink color persisted for at least 15 seconds.The oxalate content was calculated by using stoichiometric formula.The total oxalate contentwas expressed in mg/100 g of dry weight.

Oxalate ¼
titre � titrant � molar mass of oxalic acid � dilution � 100 wt of sample ð7Þ 2.3.8Tannin determination.0.2 g of the taro sample was soaked in 10 ml of 70% acetone and then placed in an ice bath (to prevent acetone from evaporation).The set up was shaken for 12-15 minutes to extract tannin.The solution was allowed to cool for about 30 minutes and then filtered to collect the supernatant.Then 0.5 ml of the supernatant was placed in a test tube and 0.5 ml of distilled water was added followed by 0.5 ml of Folins' reagent and 2.5 ml of 20% Na 2 CO 3 solution.The test tube was vortexed and incubated at room temperature for 40 minutes.The absorbance of the resulting solution was read at 725 nm with a calorimeter and standard tannic acid curve plotted.Concentration of the sample was extrapolated from the plot.
No permit was required for the research because it did not involve any animal life and the Lab protocols used were referenced.

Statistical analysis
The Analysis of Variance (ANOVA) was conducted for the nutrient and anti-nutrient composition using GenStat 12 th edition [28] (GenStat, 2009) to determine the significance of the genotypes across the environment.The means were separated using the LSD at 5% significant level.
AMMI analysis was conducted using GenStat 12 edition to assess the effect of genotype, environment and their interaction on the nutrient and anti-nutrient compositions.
The coefficient of relative phenotypic plasticity (CRP) for the genotypes used and each trait was calculated according to [29] Dingemanse et al. (2010) as follows: Where Vi is the variance of the individual (i) VPp is the overall phenotypic variance of the population.Genotypic means were used to compute Pearson correlation coefficients between the nutrient and anti-nutrient composition across the environments using Statistix version 9.1 [30] (Statistix, 2013).

Diversity among the genotypes for the nutrient and anti-nutrient compositions
The combined means of the parents and single cross hybrids for the nutrient and anti-nutrient compositions across the environment are presented in Table 4 and in the various environments as S1 and S2 Tables in S1 File.The analysis of variance (ANOVA) revealed highly significant (p<0.001)differences among the genotypes for the nutrient and anti-nutrient composition except for percentage fat content (Table 4). Among

Environmental and genotypic effects on the nutrient and anti-nutrient composition
The environmental variance analysis showed highly significant (p<0.01)difference between the environments for all the nutrient and anti-nutrient composition studied (Table 5).The genotype by environment interaction (GEI) was also significant (p<0.01) for all the traits studied except for percentage fat content.The analysis also revealed that the magnitude of the mean square for all the traits was higher for the environment than the GEI and the genotype (Table 5).
The environmental means for the nutrient and anti-nutrient composition of the parents and hybrids are presented in Table 6.The analysis revealed that the means of the anti-nutrient composition (Oxalate, Tannin and Phytate) were lower in environment 1 than environment 2. For the nutritional composition; percentage ash, percentage protein and percentage fat contents were higher in environment 1 while moisture content and total carbohydrate were also higher in environment 2. The AMMI analysis of variance revealed highly significant (p<0.001)genotype, environment and GEI effects for all the nutrient and anti-nutrient compositions for the parents and hybrids evaluated (Table 7).
Only one IPCA accounted for the total GEI for all the traits.The GEI effect was the highest contributor to the total variation for ash content (37.78%) and moisture content (46.76%).Environmental effect was the highest contributor to the total variation for protein content (78.11%), total carbohydrate (66.27%) and fat content (26.15%).Genotypic effect was the highest contributor to the total variation for oxalate, phytate and tannin content (Table 7).

Phenotypic plasticity of the parents and hybrids of taro for nutrient and anti-nutrient composition
The phenotypic plasticity index (PPI) (Figs 1 and 2) revealed stability for both parents and hybrids for the nutrient and anti-nutrient composition.Genotypes which recorded near zero PPI are more stable while those that recorded above zero PPI are more plastic.Parents BL/SM/ 10, CE/MAL/32 and CE/IND/16 were observed to be stable for combined nutrient and antinutrient composition studied as they recorded near zero PPI.KAO19 was however more plastic, recording above zero PP1 (Fig 1).For the hybrids, KAO19 x CE/IND/16 was observed to be highly stable for the combined nutrient and anti-nutrient composition followed by BL/SM/ 115 × BL/SM/10, CE/MAL/32 × CE/IND/16 and CE/IND/16 × CE/MAL/32 respectively.Hybrid BL/SM/115 × CE/IND/16 was however observed to be highly plastic for the combines nutrient and anti-nutrient composition (Fig 2).
The nutrient composition showed variable degree of stability (Tables 8 and 9).The average PPI for the traits ranged from 0.23 for oxalate content to 1.14 for tannin content.High plasticity was observed for the parents than the hybrids for the nutrient composition except for ash, oxalate and tannins contents with an average PPI of 0.60, 0.11 and 0.88 respectively for parents and 0.794, 0.29 and 1.23 respectively for hybrids.The average PPI for the parents ranged from 0.11 for oxalate content to 1.55 for protein content.Among the parents, BL/SM/10 was stable with PPI of 0.1 while CE/MAL/32 was plastic with PPI of 1.2 for ash content.BL/SM/10 recorded the highest PPI of 3.08 while BL/ SM/115 had the least PPI of 0.6.For total carbohydrates, BL/SM/115 was more stable while

Association among nutrient and anti-nutrient composition for taro genotypes
The correlation analysis for nutrient and anti-nutrient composition is presented in Table 10.The analysis revealed positive and highly significant (p<0.001)correlation between phytate and oxalate content (r = 0.49).Tannin (r = 0.89) and moisture (r = 0.68) contents showed positive and highly significant (p<0.001)correlation with oxalate contents.Percentage ash was negatively correlated with all the traits studied except for protein content (r = 0.33) although not significant (p>0.05).There was non-significant (p>0.05)negative correlation between protein content and oxalate content (r = -0.05)and non-significant positive correlation between phytate (r = 0.09) content.Total carbohydrate was also positively correlated with phytate (r = 0.14), oxalate (r = 0.42) and tannin (r = 0.34) contents although not significant (p>0.05).Fat content was also positively associated with oxalate (r = 0.28) and tannin (r = 0.34) content while negatively associated with phytate (r = -0.05)content (Table 10).

Diversity among the genotypes for nutritional and anti-nutrient compositions
Developing taro genotypes with enhanced nutritional and low anti-nutritional composition is important for improved human nutrition and industrial purposes.Assessing the nutritional profile of accessions provides important information for crop improvements [31,32]

Environmental and genotypic effects of the genotypes for nutritional and anti-nutrient compositions
The significant differences between the environments indicate the diversity among the environments used for the research and this is evidence in the differences in the climatic conditions that existed in the two agro-ecological zones used for the research (Table 2).The higher magnitude of the environment over genotype and GEI indicates the influence of environment in the expression of the nutritional and anti-nutritional traits making environment an important factor in the breeding of taro genotypes for nutrient and anti-nutrients composition.The higher levels of the nutritional composition and the lower levels of the anti-nutritional composition among the genotypes cultivated at Bunso, suggests Bunso to be an ideal environment for cultivation of taro for improved nutritional composition.The higher levels of the anti-nutritional composition of the genotypes at the Tano Dumasi environment may be due to high soil moisture content as Tano Dumasi research site was a waterlogged area and acridity content of taro increases when it comes into contact with moisture [41] (Fufa et al., 2021).These findings are in agreement with the work of [42] Huang et al. (2007) who reported higher anti-nutrient composition among taro genotypes in paddy areas than in upland fields.
The higher genotypic effect on oxalate, phytate and tannin composition suggests the presence of high level of diversity among the genotypes for anti-nutrient traits studied.This gives opportunity for selecting superior genotypes among the genotypes used for the research for further evaluation and onward distribution to farmers for cultivation or to include them in breeding programs for varietal development of taro for low anti-nutrient content in Ghana [43].Nduwumuremyi (2017) also reported high genotypic effects on carotene content in cassava.[44] Oduro (2013) and [45] Gurmu (2015) also reported high genotype effect on nutritional traits in potatoes [46].Brown et al. (2010) however, reported non-significant (p>0.05)environment and genotype effect on mineral content in potatoes.
The higher environment effect on protein, carbohydrate and fat contents indicate that the variability among the genotypes for those traits are largely influenced by environment.Therefore, specific selection for protein, carbohydrate and fat is recommended to overcome environmental influence [36].Gurmu

Phenotypic plasticity and stability of the genotypes for nutritional and anti-nutrient composition
The varied phenotypic plasticity among the genotypes for the nutrient and anti-nutrients composition is an indication of varied adaptation of the genotypes to the environment.It also confirms the effect of GEI in the expression of the traits [48].(Des Marais et al., 2013).Genotypes that exhibited PPI of near zero for most of the nutrient and anti-nutrient compositions are more stable across the environments [49] (Tan et al., 2020).These genotypes can therefore be recommended for cultivation in wide range of environments.Therefore, genotypes KAO19 and BL/SM/115 × CE/IND/16 can be recommended for cultivation in a wide range of environments for cormquality.The genotypes which presented high PPI are more plastic and unstable and therefore can be selected for particular environments.Phenotypic plasticity among taro genotypes for nutritional traits has not been reported in literature, however, [22] Hidayatullah et al. (2020) reported phenotypic plasticity for eddoe and dasheen taro for yield and yield components [23].Tripodi et al. (2020) reported plasticity among chilli pepper genotypes for health-related compounds in two environments [49].Tan et al. (2020) also reported on phenotypic plasticity of the rice grain ionome.

Association among the nutritional and anti-nutritional compositions
Significant and positive correlation among traits indicates that the traits can be improved together while significant and negative correlations indicate that the traits will have to be improved separately.The inter-correlation between all the anti-nutrients indicates that they can be improved together.The negative correlation between the anti-nutrients studied and protein and ash contents which are important nutritional qualities suggest that these traits can be improved together since low values of the anti-nutrients in the taro cormis desirable [50].Mwenye et al. (2011) also observed positive and negative significant correlations among nutritional composition of taro [51].Oue ´draogo et al. (2023) also observed positive correlation between nutritional compositions in taro genotypes in Burkina Faso.

Conclusions
There were lots of variability among the genotypes of taro studied for nutritional and antinutritional compositions.Environment played an important role in the expression of the nutrients and anti-nutrients compositions as the genotypes showed varied levels of stability for the traits.Genotypes BL/SM/115, CE/MAL/32 and CE/IND/16 which recorded high nutrient and low anti-nutrient contents and were stable across the environments can be included in breeding programs for the development of enhanced nutritional quality taro varieties.Crosses KAO19 × CE/MAL/32 and CE/IND/16×KAO19, CE/IND/16 × BL/SM/10, CE/ IND/16 × BL/SM/115 which also showed stability with high nutrients and low anti-nutrients can be released to farmers for cultivation.

Table 7 . AMMI analysis of variance showing sum of squares and its significance test and % total variation of environment, genotype and GEI for nutrient and anti- nutrient compositions among taro genotypes across two environments.
https://doi.org/10.1371/journal.pone.0291358.t007

Table 10 . Correlation matrix for nutrient and anti-nutrient composition in taro.
[32][33][34]et al., 2019)emenla et al., 2019).Genotypes which contain less amount of anti-nutrients, high amount of nutrient and low moisture content are preferable by consumers since they are seem to be of high quality[32](Khatemenla et al., 2019).The significant differences among the accessions for the nutritional and anti-nutritional traits indicates genetic variability among the genotypes and also suggests that, good selection responsecan be made in the development of taro varieties with high nutritional and low anti-nutrient compositions[32][33][34].Boampong et al. (2019), Khatemenla et al., (2019) and Matikiti et al., 2017 reported diversity among taro accessions for nutritional traits [6].Angami et al. (2015) and [35] Azene and Molla (2017) have also reported similar results on oxalic and moisture content in taro.The finding of the research also agrees with the works of [36] Gurmu et al. (2020) and [19] Tumwegamire et al. (2016) who reported significant differences among newly developed sweet potato clones for nutritional composition [37].Polycarp et al. (2012) also reported significant difference in chemical factors and anti-nutritional composition in Ghanaian yams.Breeding for high nutritional composition such as high protein, ash and carbohydrate as well as low fat, oxalate, phytate, tannin and moisture contents in crops is important for improving the health of humans ([36, 38] Magwaza et al., 2016; Gurmu et al., 2020).Therefore, selecting of taro genotypes with high protein, ash and carbohydrate contents and low fat, oxalate, phytate, tannin and moisture contents will serve as useful genetic resource for improving human nutrition in Ghana.Hybrids from KAO 19 × CE/IND/16, CE/MAL/32 × KAO19, KAO 19 × BL/SM/115 and CE/IND/16 × BL/SM/115 and genotypes BL/SM/115, CE/MAL/32 and BL/SM/10 which recorded high protein, ash and carbohydrate content with low fat, oxalate, phytate, tannin and moisture contents below the daily recommended amount for human consumption ([39, 40] Massey et al., 2001; Reddy et al., 1982) can be included in breeding programs for the nutritional improvement of taro in Ghana.