HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Zanklan, A. S. Articles by Grüneberg, W. J. Search for Related Content PubMed Articles by Zanklan, A. S. Articles by Grüneberg, W. J. Agricola Articles by Zanklan, A. S. Articles by Grüneberg, W. J.

PLANT GENETIC RESOURCES

Evaluation of the Storage Root-Forming Legume Yam Bean (Pachyrhizus spp.) under West African Conditions

^a Dép. de Biologie Végétale, Faculté des Sciences et Techniques, Univ. d'Abomey-Calavi, 01 BP 526, Cotonou, Bénin
^b Institute of Agronomy and Plant Breeding, Georg-August-Univ. Göttingen, Von-Siebold-Str. 8, 37075 Göttingen, Germany
^c Institute of Agricultural Chemistry, Georg-August-Univ. Göttingen, Carl-Sprengel-Weg 1, 37075 Göttingen, Germany
^d Dep. of Genetic Resources and Crop Improvement, International Potato Center, P.B. 1558, Lima 12, Peru

^* Corresponding author (w.gruneberg{at}cgiar.org).

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The yam bean (Pachyrhizus spp.) contains three closely related cultivated species: P. tuberosus (Lam.) Sprengel, P. erosus (L.) Urban, and P. ahipa (Wedd.). Its storage root dry matter content (SRDM) is usually low, although genotypes with a high SRDM have been identified (‘Chuin’ accessions). Flowers are often removed through flower pruning (FP) to increase storage root fresh matter yield (SRFY). The main objectives of this study were to investigate the potential for use in Benin (West Africa), to estimate the effect of FP, and to test whether roots could be processed into gari. In total, 34 accessions were tested at one drought-stress and one irrigated location. Means and genetic variance components were estimated for 33 agronomic traits. Without FP, the SRFYs of P. tuberosus, P. erosus, and P. ahipa were 13.9, 23.4, and 12.4 t ha^–1, respectively, and the seed yields were 2.2, 5.2, and 2.1 t ha^–1, respectively. The FP caused SRFY to increase by 48, 91, and 61% in P. tuberosus, P. erosus, and P. ahipa, respectively. The storage root dry matter yield (SRDY) of P. erosus was only slightly higher (8.5 t ha^–1) than that of the Chuin accessions (8.0 t ha^–1). Under drought, the SRDY was least affected in P. erosus. Early-maturing P. ahipa accessions were identified. All species could be used to make gari, which contained, on average, 5.5% protein, 58.5% starch, and 23.8% total dietary fiber. The crop has the potential for use in West Africa and has a large genetic variation for genetic improvement.

Abbreviations: BCMV, bean common mosaic virus • BIOM, biomass production • DM, dry matter • FM, fresh matter • FP, flower pruning • G x E, genotype-by-environment • G x L, genotype-by-location • SRDM, storage root dry matter content • SRDY, storage root dry matter yield • SRFY, storage root fresh matter yield

Evaluation of the Storage Root-Forming Legume Yam Bean (Pachyrhizus spp.) under West African Conditions

^a Dép. de Biologie Végétale, Faculté des Sciences et Techniques, Univ. d'Abomey-Calavi, 01 BP 526, Cotonou, Bénin
^b Institute of Agronomy and Plant Breeding, Georg-August-Univ. Göttingen, Von-Siebold-Str. 8, 37075 Göttingen, Germany
^c Institute of Agricultural Chemistry, Georg-August-Univ. Göttingen, Carl-Sprengel-Weg 1, 37075 Göttingen, Germany
^d Dep. of Genetic Resources and Crop Improvement, International Potato Center, P.B. 1558, Lima 12, Peru

^* Corresponding author (w.gruneberg{at}cgiar.org).

The yam bean (Pachyrhizus spp.) contains three closely related cultivated species: P. tuberosus (Lam.) Sprengel, P. erosus (L.) Urban, and P. ahipa (Wedd.). Its storage root dry matter content (SRDM) is usually low, although genotypes with a high SRDM have been identified (‘Chuin’ accessions). Flowers are often removed through flower pruning (FP) to increase storage root fresh matter yield (SRFY). The main objectives of this study were to investigate the potential for use in Benin (West Africa), to estimate the effect of FP, and to test whether roots could be processed into gari. In total, 34 accessions were tested at one drought-stress and one irrigated location. Means and genetic variance components were estimated for 33 agronomic traits. Without FP, the SRFYs of P. tuberosus, P. erosus, and P. ahipa were 13.9, 23.4, and 12.4 t ha^–1, respectively, and the seed yields were 2.2, 5.2, and 2.1 t ha^–1, respectively. The FP caused SRFY to increase by 48, 91, and 61% in P. tuberosus, P. erosus, and P. ahipa, respectively. The storage root dry matter yield (SRDY) of P. erosus was only slightly higher (8.5 t ha^–1) than that of the Chuin accessions (8.0 t ha^–1). Under drought, the SRDY was least affected in P. erosus. Early-maturing P. ahipa accessions were identified. All species could be used to make gari, which contained, on average, 5.5% protein, 58.5% starch, and 23.8% total dietary fiber. The crop has the potential for use in West Africa and has a large genetic variation for genetic improvement.

Abbreviations: BCMV, bean common mosaic virus • BIOM, biomass production • DM, dry matter • FM, fresh matter • FP, flower pruning • G x E, genotype-by-environment • G x L, genotype-by-location • SRDM, storage root dry matter content • SRDY, storage root dry matter yield • SRFY, storage root fresh matter yield

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
THE ROOT AND tuber crops produced by legumes have long been recognized as important, and the FAO (1979) has recommended them as a source of human nutrition. Nonetheless, the crop group is still underused. One of the most important root- and tuber-producing legumes is the yam bean (Pachyrhizus spp.). The genus contains three closely related cultivated species, all of which can be successfully crossed: the Amazonian yam bean (P. tuberosus), the Mexican yam bean (P. erosus), and the Andean yam bean (P. ahipa). Unlike its close relative the soybean (Glycine max [L.]), the yam bean is grown exclusively for its storage roots (Sørensen, 1996; Sørensen et al., 1997).

The yam bean produces heavy storage roots that weigh between 200 and 1500 g and have a relatively high protein content. Up to 18% crude protein has been observed in yam bean storage root dry matter content (SRDM) (Velasco and Grüneberg 1999; Kale 2006), three to five times higher than that of commonly grown tropical root and tuber crops such as cassava (Manihot esculenta Crantz), sweetpotato [Ipomoea batatas (L.) Lam] and yam [Dioscorea spp. (L.)]. Because storage roots consist mainly of starch, there is the possibility of using them for starch production (Bergthaller et al., 2001). Moreover, Kale (2006) observed relatively high iron contents in storage roots (up to 130 mg kg^–1 SRDM), a micronutrient seriously lacking in human diets. However, the storage roots of almost all yam bean species are characterized by their low SRDM (less than 20%). For this reason, they are usually used as a vegetable. The only exception is a P. tuberosus cultivar group that was recently found in Amazonian Peru. Known as the ‘Chuin’ group (Sørensen et al., 1997; Nielsen et al., 1998a), this type of yam bean is, in the area in which it originated, prepared in a way similar to cassava. Sørensen et al. (1997) reported that this type of yam bean has an SRDM of 30 to 32%, Nielsen et al. (1998a) found an SRDM of 7.5 to 32.6%, and Grüneberg et al. (2003) found that its SRDM ranged from 27 to 30%.

The 1000-seed weight of yam bean is high (180–230 g). The seeds themselves have high protein (26–32% by weight) and oil content (22–26% by weight). However, they are never consumed by people because they contain high levels of the insecticide rotenone (about 1% by weight) (Santos et al., 1996; Grüneberg et al., 1999). In both small-scale and commercial production, all the flowers are removed, as flower pruning (FP) increases storage root production. Only a few plants are not flower pruned and left to produce seeds, since seed is the common method to propagate yam bean.

Yam bean can be grown without nitrogen fertilizer because the plants have an efficient symbiotic relationship with rhizobia (Castellanos et al., 1997; Nielsen et al., 1998b). Mycorrhizae are also associated with the crop, thus facilitating phosphorus uptake. For these reasons, yam bean is highly suited to the needs of small farmers, and from both an ecological and a socioeconomic perspective, it has the potential to become an integral part of sustainable land-use systems (Grum and Sørensen, 1998). Currently, however, the yam bean is only cultivated locally in Central and South America (P. tuberosus, P. erosus, and P. ahipa), and in South and East Asia and the Pacific (P. erosus). Usually, it is grown for use as a root vegetable, the exception being the P. tuberosus Chuin type from the Ucayali River of Peru, which has a high dry matter content and is processed to produce flour.

Root and tuber crops, mainly cassava, sweetpotato, and yam, are important sources of nourishment in many West African countries. For example, about 80% of the population of Nigeria, Togo, Benin, and Ghana eat gari on a daily basis. This staple food made from cassava consists of about 1% protein. Only a few studies have explored the possibility of cultivating yam bean under West African conditions. For example, high storage root yields and adaptation to semiarid conditions were observed in field trials of P. erosus in Senegal (Annerose and Diouf 1998), Benin (Adjahossou 1998), and Sierra Leone (Belford et al., 2001). The main objective of this study was to investigate whether yam bean has the potential to be used as a root crop in West Africa. Because of gari's importance as a staple food, our study included tests that involved processing storage roots to produce gari. To our knowledge, this is the first comparison of all three cultivated yam bean species, under field conditions, using a large number of genotypes. Special emphasis was placed on the evaluation of the Chuin accessions grown that have a high SRDM.

	MATERIALS AND METHODS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In total, 34 accessions were studied. These represented the better-performing agronomic types from various areas and consisted of 14 P. ahipa accessions and breeding lines, 14 P. erosus accessions, and 6 P. tuberosus accessions—5 Chuin type and, for the purposes of comparison, 1 representative of the ‘Ashipa’ type (Tapia and Sørensen 2003). The material used originated from Peru and the Caribbean (P. tuberosus), Mexico, Guatemala, Brazil, China, and Vietnam (P. erosus), and Bolivia (P. ahipa). A detailed description of the accessions used is given in Table 1 . Seed multiplication was conducted from June 2000 to January 2001 at the Centre Songhai in Porto-Novo, Benin, and involved four to eight seeds per accession or line, respectively.

The experiments were performed from 22 June 2001 to 20 Jan. 2002. The trials at Songhai were irrigated during August and September (about 800-mm irrigation), which can be considered an environment without drought stress (total water supply for the crop duration was about 1260 mm [rainfall and irrigation]). At Niaouli, no irrigation was applied, and plants suffered drought stress. Separate experiments were conducted for each species (P. tuberosus, P. erosus, and P. ahipa), and data were classified relative to the experimental factors genotype, FP treatment, locations, and replications. The experimental design was a randomized block design with four blocks and two replications. Genotypes were completely randomized within blocks, and the two FP treatments ("with FP" and "without FP") were applied randomly to blocks. The experimental plots consisted of four rows, each containing 24 plants. The distance between plots was 1 m. In the experimental plots, planting distances were 0.75 m between rows and 0.25 m within rows. Two seeds were sown per hole at a depth of about 2 cm. Five weeks after sowing, extra plants were removed so that only one plant was left per hole. Weeds were removed every 2 wk. Two stakes were used to ensure that the P. tuberosus and P. erosus plants remained upright. No fertilizer, pesticide, or rhizobial inoculum was applied. In the FP treatment, all flowers were removed once a week.

In total, 33 agronomic characteristics were measured for all genotypes without FP treatment at both locations for all plot replications (Table 2 ), except starch, sucrose, glucose, and fructose content of storage roots, which were recorded without plot replications for all genotypes without FP treatment at both locations. The descriptors for Phaseolus spp., Vigna spp., and Ipomoea batatas (International Board for Plant Genetic Resources 1985, 1987, 1991) were used with minor modifications. Data for single plants were recorded using six randomly selected individuals within each plot's two central rows. Storage root crude protein content was determined by measuring the N content using the Dumas method (Sweeney and Rexroad, 1987) and multiplying the N content by 6.25; starch content was determined using the polarimetric standard analysis No. 123/1 (ICC, 1999), while sucrose, glucose, and fructose contents were analyzed enzymatically as described by Boehringer Mannheim (1983). The protein, starch, sucrose, glucose, and fructose contents of storage roots were determined twice in the laboratory for each plot sample; the values obtained were then averaged to obtain one plot value. When the FP treatment was applied, only the following 10 agronomic characteristics were measured: (i) storage root fresh matter yield (SRFY), (ii) storage root dry matter yield (SRDY), (iii) weight of vines and leaves, (iv) biomass production (BIOM), (v) SRDM, (vi) harvest index for storage root yield, (vii) damage to stem and leaves by insects, (viii) damage to storage roots by nematodes, (ix) number of storage roots per plant, and (x) protein content of storage roots (Table 2).

Agronomic data were analyzed for each species separately using an ANOVA and the PLABSTAT program (Utz, 1997). The data obtained for the gari samples were only described using mean values. The parameters estimated were means () for species, treatments, and locations; and variance components due to genotype (_G²), genotype-by-location (G x L) interactions (_GxL²), and the error term (²). For all traits recorded in the without FP treatment (except the starch, sucrose, glucose, and fructose content of storage roots), PLABSTAT was used with the model statement x_i = G + L + GL + R:L + RGL, which corresponds to the statistical model

where G = genotype, L = location, and R = replications; Y_ijln is the plot value of the ith trait of the jth genotype, for the lth location and nth block; µ_i is the trial mean of the ith trait; g_ij, l_il, and gl_ijl are, respectively, the effects of genotype, location, and genotype-location interactions; bl(l)_in(l) is the effect of blocks within locations; and _ijln is the plot error.

For the starch, sucrose, glucose, and fructose content of storage roots, PLABSTAT was used with the model statement x_i = G + L + GL, which corresponds to the statistical model

where '_ijl is an error term comprising gl_ijl, the G x L interactions, and _ijln, the plot error ('_ijl = gl_ijl + _ijln); other effects are as given in the statistical model described above.

For all those traits recorded with and without FP, PLABSTAT was used with the model statement x_i = G + L + T + GL + GT + TL + GLT + R:TL + RGTL, which corresponds to the statistical model

where T = FP treatment, t_ik, gt_ijk, tl_ikl and gtl_ijkl are, respectively, the effects of treatments, genotype x treatment, treatment x location, and genotype x treatment x location interactions; bl(tl)_in(kl) is the effect of blocks within treatments and locations. Other effects were as given for the statistical models described above. To allow a comparison of variance component estimations with pruning and without pruning, these traits were analyzed (excluding data for without pruning) using PLABSTAT with the model statement x_i = G + L + GL + R:L + RGL (statistical model as described above). The factors G and R were considered in all ANOVA calculations as random, whereas T and L were considered as fixed. The correlations of results without pruning and with pruning for SRFY and SRDY were calculated using the SAS procedure CORR (SAS Institute, 1997), using the Spearman correlation coefficient.

	RESULTS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
For most of the 33 agronomic traits evaluated, clear differences between species and significant differences between locations were observed (Table 3 ). Significant genetic variation between genotypes within species was also observed for most of the agronomic traits (Table 4 ). Without FP (Table 3), SRFYs at both locations were higher in the case of P. erosus (35.2 and 11.5 t ha^–1 at Songhai and Niaouli, respectively) than for P. tuberosus (21.1 and 6.7 t ha^–1 at Songhai and Niaouli, respectively) and P. ahipa (19.3 and 5.5 t ha^–1 at Songhai and Niaouli, respectively). Both plant height and yield traits (SRFY, SRDY, and BIOM [SRDY plus the dry weights of vines, leaves, shells, and seeds]) were clearly higher at the nonstress location (Songhai) than at the drought-stress location (Niaouli). For P. tuberosus and P. ahipa, there was no significant difference in SRDM between the two locations. In P. erosus, however, SRDM was significantly higher at the drought-stress location than at the irrigated location (25.5% and 16.1% at Niaouli and Songhai, respectively).

View this table: [in this window] [in a new window]   Table 3. Yam bean species means () across accessions for observed agronomic traits, without flower pruning, for each location.

View this table: [in this window] [in a new window]   Table 4. Variance components of observed traits without flower pruning, in three yam bean species.

The _G² component (Table 4) was significant for all traits observed under FP and often in all species. For only four traits was _G² significant for only one of the three species: storage root protein content, storage root starch content, and start of climbing in P. ahipa, and SRDM in P. tuberosus. Only in P. tuberosus were several traits observed that exhibited no significant _G²: the protein, starch, and sucrose contents of the storage roots, number of storage roots per plant, damage of storage roots by nematodes, seed yield, pod yield, seed number per pod, shell weight, harvest index for seeds, start of climbing, and damage to stem and leaves by fungi. Estimations of _G² of nutritional traits (protein, starch, sucrose, glucose, and fructose contents of storage roots) were, in most cases, nearly as large or larger than ² in all species. The only exception was the protein content of storage roots. It should be noted that estimations of ² for the starch, sucrose, glucose, and fructose content of storage roots in this study include G x L effects. The _GxL² was lower than the _G², particularly for SRFY, SRDY, SRDM (except in the case of P. erosus), BIOM, 1000-seed weight, beginning of flowering, time of maturity (except in the case of P. erosus and P. ahipa), and the protein content of storage roots.

Flower pruning (Table 5 ) had a significant and large effect on the root yield traits of all three species. Flower pruning increased SRFY by 48, 91, and 61%, and SRDY by 58.3, 100.5, and 65.8%, in P. tuberosus, P. erosus, and P. ahipa, respectively. Pachyrhizus erosus had the highest SRFY both without and with FP (23.4 t ha^–1 without FP and 44.6 t ha^–1 with FP), while SRFY was similar for P. tuberosus and P. ahipa (13 t ha^–1 without FP and 20 t ha^–1 with FP). Pachyrhizus erosus also produced a higher SRDY (8.5 t ha^–1 with FP) than P. tuberosus (6.8 t ha^–1 with FP) and P. ahipa (4.5 t ha^–1 with FP) in most cases. However, the SRDY of P. erosus was only slightly higher than that of P. tuberosus Chuin accessions, which exhibited an SRDY of about 8.0 t ha^–1 with FP and which had the highest SRDM (32.5%; results not presented). Total biomass decreased significantly as a result of FP in all species; however, SRDM and protein content, vine and leaves dry matter yield, damage to storage roots by nematodes and insects, and the number of storage roots per plant were not clearly affected by FP and usually differences were not significant.

View this table: [in this window] [in a new window]   Table 5. Yam bean species means () across accessions and locations for observed traits, without and with flower pruning (FP).

View this table: [in this window] [in a new window]   Table 7. Variance components of observed traits without flower pruning in three yam bean species.

View larger version (23K): [in this window] [in a new window]   Figure 1. Means of storage root yield of 34 yam bean accessions with and without flower pruning: (A) storage root fresh matter yield (r = 0.960, P < 0.0001) and (B) storage root dry matter yield (r = 0.946, P < 0.0001).

View larger version (75K): [in this window] [in a new window]   Figure 2. Processing of fresh storage roots to produce yam bean gari.

	DISCUSSION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

One limitation of the study is that the three species were grown in three separate, although adjacent, experiments. This was done to exclude border effects, since the species differ greatly with regard to growth type and biomass production, facts that were also clearly observed in this study (Table 3). Therefore, the effect of species in our study was confounded with the experimental effects. Accordingly, we have not presented any results of significant tests for between-species differences; however, to allow further descriptive test comparisons and calculations of the LSD, we do present the pooled error term (MSE) across species. However, we do not expect this to lead us to make untrue statements, for three reasons: (i) the experimental fields were reasonably homogeneous within both locations, (ii) block effects were not significant within the experiments, and (iii) the variance component for the blocks within experiments was always less than 10% of the variance components for the treatments and genotypes (Zanklan, 2003).

Storage root fresh matter yields were found to be different across contrasting locations (Tables 3 and 6), especially for P. erosus, which yielded (across 14 accessions) an average of 23 t ha^–1 without FP and 45 t ha^–1 with FP. In addition, P. erosus gave high seed yields without FP (Table 3), with an average yield of 5.2 t ha^–1 over all 14 accessions and both locations. This is comparable to the yields produced by high-yielding breeding lines of soybean in experimental trials. The storage root yield levels observed in this study were similar to, or slightly higher than, those obtained in comparable evaluations of yam bean. In Senegal, for example, Annerose and Diouf (1998) recorded an average SRFY of about 17 t ha^–1 without FP, while Belford et al. (2001) obtained an average SRFY of 14 t ha^–1 without FP in Sierra Leone. By the same token, Ratanadilok et al. (1998) recorded an average SRFY 28 t ha^–1 without FP in Thailand. The similarity of our yield results to those of studies in other regions indicates that our results may hold true for other countries in West Africa, as well as in other regions of the world. All three of the yam bean species studied here have been reported to be rather susceptible to bean common mosaic virus (BCMV) (Sørensen, 1996), although no BCMV symptoms were observed in our experiments in Benin. However, damage by fungal pathogens was noted. Pachyrhizus ahipa was the most affected of the species, possibly because P. ahipa genotypes display determinate growth and mostly produce leaves close to the humid soil, making them easier targets for fungal attack.

Yam beans are grown exclusively for storage root production, even though their seed yields and 1000-seed weights are high (Table 3) and the seeds that they produce have oil and protein contents similar to those of soybean seeds (Grüneberg et al., 1999). This is because the seeds have a high rotenone content (1% seed weight) and should never be consumed. Before use can be made of the seeds, studies are needed to determine technical or genetic methods of eliminating the rotenone. Seed production is still required, since seeds are used to propagate the crop, although only a few plants are needed to obtain seeds for the next growing season. We, for example, were able to ensure well-established experimental fields in this study by using only two seeds per planting position.

In current yam bean cultivation, seed production should be avoided by using FP since FP considerably increases storage root yields (Table 5). In the present study, for example, FP increased SRFY by 48% in P. tuberosus, 91% in P. erosus, and 61% in P. ahipa. Other authors have reported similar, or even more marked, increases in SRFY as a result of FP. In the case of P. erosus, FP has been observed to increase SRFY by 100, (Heredia-Zepada 1971), 250 (Belford et al., 2001) and up to 900% (Noda and Kerr 1983). Nielsen et al. (1999) noted that FP increased P. tuberosus SRFY by 52%. However, although FP is often practiced with P. ahipa and P. erosus, it is rarely applied to P. tuberosus (Sørensen, 1996). The results of this study, coupled with those of Nielsen et al. (1999), clearly show that FP should be used when cropping P. tuberosus. Flower and storage root formation during plant development are almost simultaneous events in yam beans, and the seed production is a metabolic sink that limits the amount of assimilate that can be allocated to storage roots (Heredia-Zepada, 1971; Nielsen et al., 2000). The high seed yields of yam beans (Table 3) explain why the FP treatment leads to such a large increase in storage root yields, by freeing up a large amount of resources.

Currently, yam bean is usually cultivated as a root vegetable on a small scale, although the crop is frequently found in vegetable markets in Central America and Southeast Asia. The storage roots of the Chuin yam bean type are processed by some families in the area in which it originated into a flour called farina de Chuin (in contrast to farina de yuca, the flour processed from cassava). This discovery has caused some researchers (Sørensen et al., 1997; Grüneberg et al., 2003) to conclude that the yam bean could be used to provide a protein-rich starch staple, and we feel that our study clearly supports this option. The protein content of P. tuberosus and P. erosus storage roots is higher than 10% DM and, in P. tuberosus, is combined with a high starch content (about 55% of DM). In addition, it is possible to prepare gari from the fresh storage roots of all yam bean species using traditional small-scale production methods. The yam bean gari produced contains considerably more protein (5.5% of DM) than cassava gari (1% of DM). Indeed, the quality of the yam bean gari produced, in terms of its protein and starch content, was an unexpected finding (Table 8). That said, the preparation of gari from yam bean storage roots does result in a decrease in the protein concentration, as well as an increase in the content of starch present. As expected, the contents of total dietary fiber and insoluble dietary fiber present were found to increase. These results indicate that yam bean protein contains a considerable quantity of water-soluble protein fractions. They also suggest that after gari production, a large amount of the yam bean starch remained trapped in the fibers and that starch granules present were rather small. This agrees with the findings of Bergthaller et al. (2001), who observed small-scale industrial starch extraction and found (i) that nearly 20% of the starch remained trapped in fibers and (ii) that the starch particles had a median size of about 13 µm. The dietary fiber content of yam bean is also of interest, as insufficient amounts of dietary fiber in people's diets have been closely linked with Type II diabetes, obesity, colon cancer, and cardiovascular disease. The dietary fiber contents of yam bean storage roots (15% of DM) and yam bean gari (24% of DM) are considerably higher than those of other starchy products such as wheat flour (4.7% of DM; Belitz et al., 2004) and rice (0.2–0.3% of DM; Ternes, 1995). The dietary fiber contents of yam bean storage roots are also higher than those of most other roots and legumes, including cassava (7.86% of DM), yam (11.4% of DM), and mung bean (Vigna vexillata; 11.2% of DM); however, this is not true for soybean (16.6% of DM) and sweetpotato (25.3% of DM) (Souci et al., 2000).

We also observed that yam bean gari prepared from different species differed in terms of its nutritional composition (Table 8). Thus, for example, considerably less protein was lost when P. tuberosus was used to make gari than when P. erosus was used. This study used the small-scale gari preparation procedures usually applied to cassava; research should be undertaken to determine if the technique can be modified to reduce protein losses. Nonetheless, the agronomic performances of the better-performing agronomic genotypes considered (Fig. 1), coupled with the fact that yam bean storage roots can be processed to produce gari (an easily stored and transported staple in West Africa) all make yam bean a promising crop that should be made available to end users in West Africa.

Yam bean also offers advantages in terms of breeding because it is propagated using seeds. As a result, it does not attract the costs, maintenance, and dissemination problems usually associated with the production of cloned root and tuber crops. In addition, yam bean is self-compatible and mainly self-fertilizing and can be crossed like Phaseolus vulgaris L. cultivars, which means that the breeding methods used with soybean and phaseolus beans can be applied. The initial low success rates of crossings (Grüneberg et al., 2003) have been improved, and about 1 out of 10 crosses (interspecific crossings) now result in F₁ seed production (Grüneberg and Zanklan, unpublished). Moreover, the total amount of genetic variation available within and between the three cultivated yam bean species can be used because they are not separated by crossing barriers. In this study, we observed significant genetic variation for all important agronomical traits (Tables 4 and 7). However, we also observed significant and unexpected large G x E interactions for most yield traits. To a certain extent, these G x E interactions can be explained by the different drought responses of different accessions (e.g., EC533 and ECKEW). As expected, the SRDM and protein content of yam bean exhibited low G x E interactions. This means that screening programs for these important traits could be conducted centrally at one or two locations. Such low G x E interactions of storage root quality traits have previously been reported for the DM, starch and ß-carotene contents of sweetpotato (Grüneberg et al., 2005). We therefore assume that it is also true for the other quality traits associated with yam bean (i.e., starch, sucrose, fructose, and glucose contents), which were evaluated in this study across locations but without plot replications. Genetic variation for these quality traits was evident in this study (Table 4), meaning that selection could be used to improve yam bean starch, sucrose, fructose, and glucose in the desired selection direction.

Although FP has a considerable effect on SRFY and SRDY (Table 5), the ranking of genotypes according to with and without FP was highly correlated for SRFY and SRDY (Fig. 1). Hence, yield selection in the early stages of a breeding program could be conducted without FP, which would ensure that sufficient amounts of seed from a selected entry is available for the next selection step. Nevertheless, genetic variances of root yield traits were more pronounced with FP than without FP (Tables 4 and 7), so it may be an advantage in later stages of a breeding program—when differences between entries are smaller and more difficult to detect—to conduct FP. In our study, no effect of FP on SRDM and protein content was observed (Table 5). We assume that this is also true for starch, sucrose, glucose, and fructose. However, we cannot totally rule out that this is possible because in this study these traits were only determined under the treatment without FP. In a breeding program, P. erosus accessions may become genetic sources of yield, wide adaptation, and drought tolerance, while P. tuberosus accessions—especially the Chuin cultivar group—are a source of high SRDM and starch content, and P. ahipa accessions are a source of early maturity and erect or semi-erect growth type.

The yam bean genotypes currently available, particularly the high-yielding P. erosus and P. tuberosus accessions, show considerable promise for use in low-input farming systems in Benin and West Africa. The crop is adapted to the ecogeographic region, fits into the region's food system, and should improve the supply of nutritious food available. Therefore, it merits further breeding research and development work. In addition, because yam bean contains a large amount of genetic variation and because interspecific hybridization is possible, rapid progress in cultivar development should be expected. Yam bean seeds also possess attractive quantities and qualities. However, until the rotenone problem is solved, seed production should only be undertaken to maintain and propagate the crop.

	ACKNOWLEDGMENTS

 
This research was supported by a scholarship from the German Academic Exchange Service (DAAD) to the first author. We would like to thank Marten Sørensen (KVL Copenhagen) for providing a major part of the planting material used. We also wish to express our gratitude to the Centre Songhai in Porto-Novo and the Niaouli experimental station belonging to the Institut National des Recherches Agricoles du Benin, both of which granted us access to their facilities during the fieldwork undertaken in Benin. Thanks are also due to Bo Ørting and Friedrich Kopisch-Obuch, who made valuable comments on a draft of the manuscript.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication March 10, 2006.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

• AAAC. 1984. Official methods of analysis of the American Association of Analytic Chemistry. 47.021-47.027. 14th ed. AAAC, St. Paul, MN.
• Adjahossou, D.F. 1998. Influence de l'ablation régulière des inflorescences sur les rendements en tubercule de cinq varietes du genre Pachyrhizus DC. (In French, with English abstract.) p. 97–114. In M. Sørensen, J.E. Estrella, O.J. Hamann, and S.A. Ríos Ruíz (ed.) Proc. of 2nd Int. Symp. on Tuberous Legumes, Celaya, Guanajuato, Mexico. 5–8 August 1996. MacKeenzie Press, Copenhagen, Denmark.
• Annerose, D.J.M., and O. Diouf. 1998. Recherches sur l'adaptation de la culture de Pachyrhizus DC. en zones semi-arides. (In French, with English abstract.) p. 377–387. In M. Sørensen, J.E. Estrella, O.J. Hamann, and S.A. Ríos Ruíz (ed.) Proc. of 2nd Int. Symp. on Tuberous Legumes, Celaya, Guanajuato, Mexico. 5–8 August 1996. MacKeenzie Press, Copenhagen, Denmark.
• Asp, N.G., C.G. Johansson, H. Hallmer, and M. Siljeström. 1983. Rapid enzymatic assay of insoluble and soluble dietary fiber. J. Agric. Food Chem. 31:476–482.[CrossRef][Web of Science][Medline]
• Belford, E.J.D., A.B. Karim, and P. Schröder. 2001. Exploration of the tuber production potential of yam bean (Pachyrhizus erosus (L.) Urban) under field conditions in Sierra Leone. J. Appl. Bot. 75:31–38.
• Belitz, H.D., W. Grosch, and P. Schieberle. 2004. Food Chemistry. 3rd rev. ed. Springer-Verlag, Berlin, Germany.
• Bergthaller, W., H.-J. Kersting, L. Velasco, and W.J. Grüneberg. 2001. Andean yam bean (Pachyrhizus ahipa) tubers, a new source of legume starch. p. 204–205. In Proc. of the 4th European Conf. on Grain Legumes, Cracow, Poland. 8–12 July 2001. European Assoc. for Grain Legume Research, Paris, France.
• Boehringer Mannheim. 1983. Methodenbuch der enzymatischen Lebensmittelanalytik. Boehringer Mannheim, Germany.
• Castellanos, R., J.Z. Zapata, F. Pena-Cabriales, J.J.E.S. Jensen, and G.E. Heredia. 1997. Symbiotic nitrogen fixation and yield of Pachyrhizus erosus (L.) Urban cultivars and Pachyrhizus ahipa (Wedd.) Parodi landraces as affected by flower pruning. Soil Biol. Biochem. 29:973–981.[CrossRef]
• FAO. 1979. Review on agriculture and development: Conventional crops imperil good protein. Ceres 12(68):4–5.
• Grum, M., and M. Sørensen. 1998. Pachyrhizus DC. (yam bean) symbiosis with Rhizobium bacteria: Genetic variation and performance. p. 419–429. In Proc. of 2nd Int. Symp. on Tuberous Legumes, Celaya, Guanajuato, Mexico. 5–8 August 1996. MacKeenzie Press, Copenhagen, Denmark.
• Grüneberg, W.J., P. Freynhagen-Leopold, and O. Delgado-Váquez. 2003. A new yam bean (Pachyrhizus spp.) interspecific hybrid. Genet. Resour. Crop Evol. 50:757–766.[CrossRef]
• Grüneberg, W.J., F.D. Goffman, and L. Velasco. 1999. Characterization of yam bean (Pachyrhizus spp.) seeds as potential sources of high palmitic acid oil. J. Am. Oil Chem. Soc. 76:1309–1312.[CrossRef][Web of Science]
• Grüneberg, W.J., K. Manrique, D. Zhang, and M. Hermann. 2005. Genotype x environment interactions for a diverse set of sweetpotato clones evaluated across varying ecogeographic conditions in Peru. Crop Sci. 45:2160–2171.[Abstract/Free Full Text]
• Heredia-Zepada, A. 1971. Effecto de desfloración de la jícama (Pachyrrhizus erosus) sobre el rendiemiento. p. 146–150. In Proc. of the Tropical Region, American Society for Horticultural Science, 19th Annu. Meeting, Managua, Nicaragua. 18–24 July 1971.
• ICC. 1999. Standard methods of the International Association for Cereal Science and Technology. ICC, Vienna, Austria.
• International Board for Plant Genetic Resources. 1985. Phaseolus acutifolius descriptors. IBPGR Secretariat, Rome, Italy.
• International Board for Plant Genetic Resources. 1987. Descriptors for bambara groundnut (Vigna subterranea). IBPGR Secretariat, Rome, Italy.
• International Board for Plant Genetic Resources. 1991. Descriptors for sweet potato (Ipomoea batatas). IBPGR Secretariat, Rome, Italy.
• Kale, P.R. 2006. Studies on nutritional and processing properties of storage roots of different yam bean (Pachyrhizus spp.) and wild mung bean (Vigna vexillata) species. Ph.D. diss. Universität Göttingen, Cuvillier Verlag, Göttingen, Germany.
• Lee, S.C., L. Prosky, and J.W. de Vries. 1992. Determination of total soluble and insoluble dietary fiber in foods: Enzymatic–gravimetric method, MES-TRIS buffer—Collaborative study. J. Am. Off. Anal. Chem. 75:395–416.
• Nielsen, P.E., M. Halafihi, and M. Sørensen. 1998a. Germplasm evaluation of Pachyrhizus tuberosus (Lam.) Spreng. p. 253–260. In M. Sørensen, J.E. Estrella E., O.J. Hamann, and S.A. Ríos Ruíz (ed.) 1998. Proc. of 2nd Int. Symp. on Tuberous Legumes, Celaya, Guanajuato, Mexico. 5–8 August 1996. MacKeenzie Press, Copenhagen, Denmark.
• Nielsen, P.E., M. Halafihi, and M. Sørensen. 1999. Pruning practices in Pachyrhizus tuberosus (Lam.). Spreng. Trop. Agric. (Trinidad) 76:222–227.
• Nielsen, P.E., V.T. Manu, M. Sørensen, and O. Stölen. 1998b. P-fertilizer effects on growth, nodulation, N and P-uptake and utilisation for fresh and dry matter yield in Pachyrhizus ahipa (Wedd.) Parodi. p. 469–488. In M. Sørensen, J.E. Estrella E., O.J. Hamann, and S.A. Ríos Ruíz (ed.) 1998. Proc. of 2nd Int. Symp. on Tuberous Legumes, Celaya, Gto., Mexico. 5–8 August 1996. MacKeenzie Press, Copenhagen, Denmark.
• Nielsen, P.E., M. Sørensen, and M. Halafihi. 2000. Yield potential of yam bean (Pachyrhizus erosus (L.) Urban) accessions in the kingdom of Tonga, South Pacific. Trop. Agric. (Trinidad) 77:174–179.
• Noda, H., and W.E. Kerr. 1983. The effects of staking and inflorescence pruning on the root production of yam bean (Pachyrhizus erosus Urban). Trop. Grain Legume Bull. 27:35–37.
• Ratanadilok, N., K. Suriyawan, and S. Thanisawanyangkura. 1998. Yam bean (Pachyrhizus erosus (L.) Urban) and its economic potential. p. 261–273. In Proc. of the 2nd Int. Symp. on Tuberous Legumes, Celaya, Guanajuato, Mexico. 5–8 August 1996. MacKeenzie Press, Copenhagen, Denmark.
• Santos, A.C.O., M.S.M. Cavalcanti, and L.C.B.B. Coello. 1996. Chemical composition and nutritional potential of yam bean seeds (Pachyrhizus erosus (L.) Urban). Plant Foods Hum. Nutr. 49:36–41.
• SAS Institute. 1997. SAS user's guide: Statistics, version 6.2, SAS Inst., Cary, NC.
• Sørensen, M. 1996. Yam bean (Pachyrhizus DC.). Promoting the conservation and use of underutilized and neglected crops. Institute of Plant Genetics and Crop Plant Research, Gatersleben and International Plant Genetic Resources Institute, Rome, Italy.
• Sørensen, M., S. Doeygaard, J.E. Estrella, L.P. Kvist, and P.E. Nielsen. 1997. Status of the South American tuberous legume Pachyrhizus tuberosus (Lam). Spreng. Biodiversity Conserv. 6:1581–1625.[CrossRef]
• Souci, S.W., W. Fachmann, and H. Kraut. 2000. Food composition and nutrition tables. 6th ed. Medpharm Scientific, Stuttgart, Germany.
• Sweeney, R.A., and P.R. Rexroad. 1987. Comparison of LECO FP-228 "Nitrogen Determinator" with AOAC copper catalyst Kjeldahl method for crude protein. J. Am. Off. Anal. Chem. 70:1028–1030.
• Tapia, C., and M. Sørensen. 2003. Morphological characterization of the genetic variation existing in a neotropical collection of yam bean, Pachyrhizus tuberosus (Lam.). Spreng. Genet. Resour. Crop Evol. 50:681–692.[CrossRef]
• Ternes, W. 1995. Naturwissenschaftliche Grundlagen der Lebensmittelzubereitung. 2nd edition, Behr's Verlag, Hamburg, Germany.
• Utz, H.F. 1997. PLABSTAT: Ein Computerprogramm zur statistischen Analyse von pflanzenzüchterischen Experimenten, version 2B. Institut für Pflanzenzüchtung, Saatgutforschung und Populationsgenetik, Universität Hohenheim, Germany.
• Velasco, L., and W.J. Grüneberg. 1999. Analysis of dry matter and protein contents in fresh yam bean tubers by near-infrared reflectance spectroscopy. Soil Sci. Plant. Anal. 30:1797–1805.
• Zanklan, A.S. 2003. Agronomic performance and genetic diversity of the root crop yam bean (Pachyrhizus spp.) under West African conditions. Ph.D. diss. Universität Göttingen. Cuvillier Verlag, Göttingen, Germany.

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Zanklan, A. S. Articles by Grüneberg, W. J. Search for Related Content PubMed Articles by Zanklan, A. S. Articles by Grüneberg, W. J. Agricola Articles by Zanklan, A. S. Articles by Grüneberg, W. J.