Journal of Heredity Advance Access originally published online on September 18, 2006
Journal of Heredity 2006 97(5):456-465; doi:10.1093/jhered/esl025
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Generation Means Analysis of Climbing Ability in Common Bean (Phaseolus vulgaris L.)
From the Faculty of Agricultural Sciences, Universidad de Nariño, Ciudad Universitaria Torobajo, Pasto, Colombia (Checa); Centro Internacional de Agricultura Tropical (CIAT), A.A. 6713, Cali, Colombia (Ceballos and Blair); and Universidad Nacional de Colombia, Palmira Campus, Carrera 32 Chapinero, Via Candlemas, Palmira, the Valley of the Cauca, Colombia (Ceballos)
Address correspondence to Dr. Matthew W. Blair at the address above, or e-mail: m.blair{at}cgiar.org.
Climbing common bean (Phaseolus vulgaris L.) genotypes have among the highest yield potential of all accessions found in the species. Genetic improvement of climbing beans would benefit from an understanding of the inheritance of climbing capacity (made up of plant height [PH] and internode length [IL] traits). The objective of this study was to determine the inheritance of climbing capacity traits in 3 crosses made within and between gene pools (Andean x Andean [BRB32 x MAC47], Mesoamerican x Mesoamerican [Tío Canela x G2333], and Mesoamerican x Andean [G2333 x G19839 [GenBank] ]) using generation means analysis. For each population, we used 6 generations (P1, P2, F1, F2, BC1P1, and BC1P2) that were evaluated at 2 growth stages (40 and 70 days after planting). Results showed the importance of additive compared with the dominant–additive portion of the genetic model. Broad-sense heritabilities for the traits varied from 62.3% to 85.6% for PH and from 66.5% to 83.7% for IL. The generation means analysis and estimates of heritability suggested that the inheritance of PH and IL in climbing beans is relatively simple.
Climbing bean varieties are morphologically distinct from bush bean varieties of common beans (Phaseolus vulgaris L.) characterized by tall growth, long internodes, and climbing ability. They are an important component of traditional agriculture in several parts of Latin America, especially Mexico, Guatemala, Colombia, Ecuador, and Peru (Voysest 2000), and have spread to the Great Lakes region of Africa (Sperling and Munyaneza 1995). Climbing bean cropping systems are classified as monoculture or intercropping, whereby farmers produce 2 or more species in the same area (Woolley and Davis 1991). Climbing beans are often grown in association with maize, either in relay or in simultaneous plantings, and maize provides the support required for the climbing beans to grow upward. This agronomic system is among the most representative of Latin America and has also been adopted in Eastern Africa as well (Francis and Sanders 1978; Woolley et al. 1991). In monoculture, climbing beans are planted with the support of wood or bamboo stakes or trellis systems. In this system, staking can provide a use for culled trees in agroforestry projects and has been cited as providing a stimulus for tree planting in Africa (Sperling and Munyaneza 1995). Trellising, a widespread system in the Andean region, is an alternative that reduces the need for stakes but requires an investment in wires and string for tying up bean vines (Obando 1992; Sañudo et al. 1999). Trellising of climbing beans is economically justified because yield may surpass 4500 kg ha–1. Although monoculture production is more laborious due to the need for hand labor, it is an important source of agricultural employment for on-farm and seasonal labor. Therefore, climbing beans are particularly useful for small landholdings in situations where labor is not limiting and where demand for beans is high.
One trait inherent to climbing beans is climbing ability. Climbing capacity in turn is closely related to a series of component traits including total plant height (PH) and internode length (IL), which together with the determinacy character help to make up what is known as growth habit in common beans (Debouck 1991). Climbing capacity in beans also depends on the variation in outgrowth of lateral branches and the degree of vine circumnutation (winding movements), which determines the ability or lack thereof of the plant to climb on staking material. Interaction of vegetative growth pattern with distribution of flowers and pods during reproductive growth can determine whether most of the seed production occurs along the entire length of the climbing bean or only in the lower or upper parts of the plant (Singh 1982). These traits permit the classification of common beans into 4 growth habits: type I growth habit characterized by plants that present terminal reproductive buds and inflorescences (determinate), erect branches, and no climbing ability; type II growth habit with erect growth but terminal vegetative growing buds that are indeterminate; type III growth habit that is also indeterminate but more prostrate, with terminal growth buds; whereas type IV growth habit includes tall indeterminate plants with long vines, terminal vegetative buds, and strong climbing ability (Singh 1982). A fifth growth habit (sometimes called type V) is postulated for determinate climbing beans, but genotypes with this morphology are rare (Evans 1973). Environmental conditions such as light quality and daylength greatly influence growth habit expression (Kretchmer et al. 1977, 1979), and the differences between growth habits II and III or between III and IV can be difficult to distinguish in less favorable environments (Singh 1982).
Two gene pools are recognized in common bean based on their centers of domestication in Central and South America, namely, the Mesoamerican (also known as Middle American) and Andean gene pools. The existence of the 2 gene pools is supported by differences in morphology (Singh et al. 1991; Debouck 1999), phaseolin seed protein (Gepts et al. 1986), allozymes (Koenig and Gepts 1989; Singh et al. 1991), and DNA markers (Becerra Velasquez and Gepts 1994; Sonnante et al. 1994; Blair et al. 2006). Common bean domestication is thought to have started with indeterminate wild climbing beans (type IV), whereas other growth habits (type I, II, or III) were thought to be selected by farmers in parallel in both gene pools (Gentry 1969; Evans 1973, 1976; Smartt 1985; Singh 1989; Gepts and Debouck 1991). Meanwhile, cultivated climbing beans have similar agroecological adaptation to the wild climbing beans from which they were derived; they are found mostly in medium to high altitude (2000 to 2800 m a.s.l.) regions of the Andes and Central America (Singh 1989; Voysest 2000). Along with agroecological adaptation, climbing beans like bush beans were selected for seed size and pod fiber characteristics (Evans 1980; Smartt 1988; Gepts and Debouck 1991), but distinct from bush beans, climbing beans are closer to the wild phenotype in terms of growth habit and photoperiod sensitivity (Singh 2001).
The inheritance of growth habit in common bean is thought to be controlled by a mixture of qualitative and quantitative genes and has long been a subject of research. Research on climbing ability began when Emerson (1904) crossed indeterminate and determinate growth habit plants and identified a single-recessive gene for determinacy. In another early study, Norton (1915) found dominant gene action for axillary inflorescence over terminal inflorescence, long main stem over short main stem, and climbing habit over nonclimbing habit. Subsequently, Kooiman (1931) established that the trait for determinate or indeterminate growth was simply inherited and attributable to a Mendelian gene. Lamprecht (1947), in a reanalysis of Emerson's studies, proposed the symbol Fin and fin (from finitus or finite in Latin) for the alleles that control growth type (indeterminate and determinate) and the symbol tor (from torquere or to turn or twist in Latin) for the allele that controls the twining habit of the plants as they climb. Coyne (1967) postulated that the recessive fin allele reduces PH and increases its harvest index, whereas Wallace and Yan (1998) suggested pleiotropic effects of fin on seed size and maturity. Wallace and Yan (1998) also identified a second gene controlling photoperiod response, with alleles ppd and Ppd. The recessive ppd allele was found to control daylength insensitivity and modify partitioning of available photosynthates in growth organs, whereas the dominant Ppd allele inhibited partitioning toward reproductive organs, thus increasing biomass and days to harvest. The same genes (fin and ppd) are important components of the domestication syndrome, which were genetically mapped by Koinange et al. (1996). Most of the previous research on the inheritance of growth habit, PH, and IL have been based on crosses between genotypes contrasting in determinacy (Frazier et al. 1958; Coyne 1967; Ortega Ybarra 1968; Kornegay et al. 1992), whereas little is known about the inheritance of these traits when genotypes of indeterminate growth habits but contrasting heights are crossed.
Climbing ability is a key to breeding commercial varieties with good yield for tropical environments; therefore, the study of this trait is crucial for crop improvement programs interested in climbing beans. The objectives of this study, therefore, were 1) to evaluate the inheritance of climbing ability traits in crosses between indeterminate common bean genotypes contrasting for nonclimbing (type II or III) versus climbing bean (type IV) growth habit and 2) to estimate, through the use of generation means analysis, the magnitude of the additive and dominance genetic effects. An innovative aspect of this study was the methodical design of 3 crosses, 2 within the Andean and Mesoamerican gene pools and 1 across the gene pools.
 |
Materials and Methods |
|---|
 Top Materials and Methods ResultsDiscussionReferences |
|---|
Genetic Materials
Parents included 3 genotypes (BRB32 [type II], G19839 [GenBank] [type III], and MAC47 [type IV]) from the Andean gene pool, and 2 (Tío Canela [type IIa] and G2333 [type IV]) from the Mesoamerican gene pool (http://isa.ciat.cgiar.org/urg/main.do; Blair et al. 2006). BRB32 is an advanced breeding line from Centro Internacional de Agricultura Tropical (CIAT) that has medium-sized, red-mottled grain and carries the bc3 gene, which confers resistance to the bean common mosaic virus. G19839 [GenBank] is a germplasm accession from Peru that tolerates low-phosphorus soils and has large, yellow and black–speckled grain color. MAC47 is an advanced climbing bean breeding line from CIAT that is adapted to medium altitudes and has dark red–mottled seed. Tío Canela is an improved variety from the Escuela Agricola Panamericana (Zamorano) in Honduras that is resistant to the bean golden yellow mosaic virus and has small red grain. G2333 is a climbing bean landrace from Mexico that also has small red grain. Crosses were made between these parents across growth habits and within or between gene pools by emasculation and pollen transfer. Thus, the following crosses were made: G2333 x G19839 [GenBank] (Mesoamerican x Andean), BRB32 x MAC47 (Andean x Andean), and Tío Canela x G2333 (Mesoamerican x Mesoamerican). A total of 50 pollinations were made per cross, and F1 plants were confirmed to be hybrids based on flower and seed color. F1 plants were used to obtain seed for the F2 generation and to carry out backcrosses of the F1 hybrids with their respective parents to create the BC1P1 and BC1P2 generations using a total of 40 pollinations per backcross. Thus, a total of 6 treatments were obtained, corresponding to the 2 parents and the F1, F2, BC1P1, and BC1P2 generations.
Experimental Design
The 6 treatments (P1, P2, F1, F2, BC1P1, and BC1P2) for each population were planted in March 2002 at a field site in Darién, Valle del Cauca, in the south of Colombia (latitude/longitude 03°55.839'N, 076°28.488W, elevation 1450 m a.s.l., yearly rainfall 1288 mm, average yearly temperature 20 °C). This site has Andisol soils of loam texture and pH of 5.6. The 3 populations were planted in separate experiments each in a randomized complete block design with 3 replications. The experiments were planted as bean monocultures using trellis systems in which each individual plant was tied with a string made of polypropylene to a heavy weight wire that was suspended horizontally and in parallel above the row at a height of 2 m above soil level on sturdy wooden bamboo posts that were placed every 5 m and at every sixth row. Seed was planted two to a planting hole at distances of 0.2 m between planting holes within rows and 1.0 m between rows. Plantlets were thinned to one plant per planting hole when the seedlings reached a height of 10 cm.
The number of plants evaluated varied depending on the treatment and was larger for the treatments consisting of generations with greater segregation such as the F2 (with 80 plants per repetition) and the BC1P1 and BC1P2 (with 30 plants per repetition each) than for treatments that do not segregate such as the P1 and P2 parents and the F1 generation (with 20 plants per repetition each). Agronomic management followed technical recommendations for CIAT bean crops with preventative treatments for diseases and insect pests, as well as manual hilling and harvesting.
Traits Evaluated and Data Analysis
Each plant was evaluated for PH, IL, and number of guide shoots (GS) per plant. PH was measured in meters from the insertion of the plant stem at the ground level to the last trifoliate leaf axis along the main stem. IL was evaluated by selecting an internode from the midpoint along the main stem of the plant and recording its length in centimeters. Similarly, the number of GS was evaluated by counting the number of stems that climbed around the guide string. PH, IL, and GS were evaluated twice during the growing season on an individual plant basis, first at 40 and then at 70 days after planting (DAP). In the evaluations, those plants at the end of each row were not considered to avoid border effects.
Analysis of variance (ANOVA) for each of the 3 populations was conducted separately. The number of GS was transformed for normalization through the formula square root of x (Steel and Torrie 1980). In those variables for which the ANOVA showed significant differences between generations, separation of means was carried out with Tukey's w procedure for multiple comparisons (P 
0.05). Those variables that showed significant differences by orthogonal contrasts between parents P1 and P2 were submitted to generation means analysis using the methodology proposed by Mather and Jinks (1971):
 |
k, 
k = coefficients determined by the degree of relationship of generation k; a = additive genetic effects; d = dominant genetic effects; aa = epistatic effects of additive x additive type; ad = epistatic effects of additive x dominant type; dd = epistatic effects of dominant x dominant type. To estimate the additive and dominant parameters, a stepwise linear regression analysis was carried out using the statistics package SAS version 8.2 (SAS Institute 1999). Regression analysis was weighted based on the inverse of the variance of means and the matrix of parameters or coefficient of genetic effects (Mather and Jinks 1971). R2 and the "goodness-of-fit" (F-test) were used to determine which parameters were acceptable within the model (Ceballos et al. 1998). The formula used for the F-test with the following sums of squares (SSq) and degrees of freedom (df) was as follows:
 |
In addition to generation means analysis, we estimated the following genetic parameters for each population x trait combination using formulas from Mather and Jinks (1971):
Environmental variance or error:
 |
Genotypic variance in F2:
 |
Additive variance in F2:
 |
Variance of dominance in F2:
 |
Broad-sense heritability:
 |
Narrow-sense heritability:
 |
|
Results |
|---|
|
TopMaterials and MethodsResultsDiscussionReferences |
|---|
Treatment Mean Comparisons
The ANOVAs for the 3 populations (BRB32 x MAC47, Tío Canela x G2333, and G19839 [GenBank] x G2333) showed highly significant differences among treatments for PH and IL, both at 40 and 70 DAP (Table 1). On the other hand, no significant differences were observed among treatments for GS except in the Tío Canela x G2333 population at 70 DAP; and therefore, this trait was excluded from further analysis. Table 2 shows Tukey's multiple means comparison tests for the 6 treatments across the 3 populations.
|
|
For the G2333 x G19839 [GenBank] population, the means of P1 (G2333), F1, and BC1P1 generations were not significantly different from each other for both traits at both evaluation dates. Meanwhile, the means of the F2, BC1P2, and P2 (G19839 [GenBank] ) generations were significantly different from each other and from the P1 (G2333), F1, and BC1P1 generations for PH at both evaluation dates. For IL at 40 DAP, the differences between the BC1P2 and P2 generations were less statistically significant as were the differences for the F1 and F2 generations for IL at 70 DAP (Table 2). These results, as well as observations from Figure 1, suggest that for this cross, complete dominance is important in controlling both PH and IL and that this trend is more evident early in the season at 40 DAP rather than late in the season at 70 DAP. Means of the backcross generations were observed to be similar to the means of their respective recurrent parents which themselves were contrasting for both PH and IL and were located at the opposite extremes of the continuum observed in the population distribution histograms.
|
For the Tío Canela x G2333 population, P2 (G2333) showed higher values for PH and IL at 40 and 70 DAP, followed in descending order by BC1P2, F1, F2, BC1P1, and finally P1 (Tío Canela), which presented the lowest means for all traits (Table 2). Significant differences among treatments were obtained for all comparisons for PH in both evaluations, except for the comparison between P2 (G2333) and BC1P2 at 70 DAP. For IL, significant differences in the mean comparisons were more frequent at 40 DAP than at 70 DAP. As for the cross between G2333 and G19839 [GenBank] , the parents of the Tío Canela x G2333 population were extreme relative to the other generations, while the F1 and F2 generations occupied intermediate positions (Figure 1). However, for the evaluation of PH at 70 DAP and for IL at both evaluations, the F1 of Tío Canela x G2333 tended toward the mean of G2333, the taller parent, indicating mid-parent heterosis for this variable.
For the BRB32 x MAC47 population, significant differences were found for the comparison of generation means between the 2 parents for PH and IL in both evaluations (Table 2). Parent P2 (MAC47), and in some cases the BC1P2 backcross, presented higher means for PH and IL than did the F1 and F2 generations, which in turn were significantly taller than the P1 parent (BRB32). The F1 and F2 generations had similar mean values for both traits. In the frequency distribution (Figure 1), intermediate values were found for the F1 and F2 generations suggesting incomplete or partial dominance. In comparison, extreme values were found for the parents, confirming their contrasting character in the trait x evaluation combinations. Moreover, the histograms for PH, especially at 70 DAP, showed a slight tendency for F1 plants to have values similar to the taller parent although this trend was not clearly observed in the evaluations for IL at either evaluation date.
Generation Means Analysis
For the cross G2333 x G19839
[GenBank]
population, the trait PH at 40 DAP showed a better fit to an additive inheritance model [m + a] (R2 = 95.8%) than to an additive–dominance model (Table 3). Meanwhile, for PH at 70 DAP the additive–dominance model [m + a + d] had a significantly higher fit (R2 = 99.46%) than the additive model [m + a] (R2 = 88.71%), showing that there was a highly significant contribution of the effects of dominance to the model based on the goodness-of-fit test. Results for IL were similar to those observed for PH in that an additive model was more appropriate at 40 DAP (R2 = 96.0%), whereas an additive–dominance model was more appropriate at 70 DAP (R2 = 98.8%). Goodness-of-fit tests showed that the contribution of the effects of dominance to the SSq of the model for IL was not significant at 40 DAP but was at 70 DAP.
|
In the Tío Canela x G2333 population, the additive model also explained the observed variation for PH at 40 DAP, with a coefficient of determination of 97.75%. Goodness-of-fit indicated that using the additive–dominance model would not be justified because the effects of dominance were not significant. PH at 70 DAP fits an additive–dominance model explaining 99.35% of the observed variation (5.68% higher than the R2 of the additive model). These results indicated that the effects of dominance contribute significantly to the SSq of the model, as was observed in the results of the goodness-of-fit test. For the Tío Canela x G2333 population, the IL trait showed a behavior very similar to that observed for PH, showing better fit to the additive model at 40 DAP (R2 = 91.86%) and to the additive–dominance model at 70 DAP (R2 = 99.49%).
For the BRB32 x MAC47 population, the additive model explained 97.70% and 94.72% of variation, respectively, for PH at 40 and 70 DAP; and the goodness-of-fit test for the effect of dominance was significant for both evaluations. Hence, the additive–dominance model was the most appropriate for explaining the variation for PH at both evaluation dates. For IL, the BRB32 x MAC47 population fits an additive model for both 40 and 70 DAP evaluations, explaining 99.29% and 98.50% of the observed variation, respectively. The contributions of the effects of dominance to the SSq of the model in the 2 evaluations were negligible. In addition to the multiple regression models for the additive [m + a] and additive–dominance [m + a + d] gene action, models that included epistatic interactions, such as aa, ad, and dd, were also considered (data not shown). However, the additive and dominant–additive models were sufficient to explain more than 95% of the observed variation as described above.
Trait Heritabilities
Broad-sense heritabilities calculated from genotypic and environmental variances for PH at 40 DAP were highest in the G2333 x G19839
[GenBank]
population and lowest in the Tío Canela x G2333 population and ranged from 62.3% to 81.5% (Table 4). Narrow-sense heritabilities calculated from additive and environmental variances were similar to broad-sense heritabilities, with the highest values for G2333 x G19839
[GenBank]
and BRB32 x MAC47 (66.9% and 65.4%, respectively) and the lowest for Tío Canela x G2333 (52.5%). Broad-sense heritabilities for PH at 70 DAP ranged between 80.6% and 85.6% for the 3 populations, being highest for G2333 x G19839.
[GenBank]
Narrow-sense heritabilities for PH at 70 DAP ranged between 56.0% and 79.6% but were highest in Tío Canela x G2333. For the evaluation of IL at 40 DAP, broad-sense heritabilities ranged between 83.6% and 68.7%, whereas narrow-sense heritabilities fluctuated between 82.6% and 59.2% and were highest in the BRB32 x MAC47 population. For IL at 70 DAP, broad- and narrow-sense heritabilities fluctuated between 66.54% and 79.80% and 52.65% and 75.26%, respectively, in both cases being highest for the BRB32 x MAC47 population.
|
|
Discussion |
|---|
|
TopMaterials and MethodsResultsDiscussionReferences |
|---|
In all 3 populations, the means of the parents (P1 and P2) showed a tendency to be more extreme and contrasting than the means of the F1 and F2 generations for both PH and IL. Tukey's multiple comparisons test demonstrated that differences between the parents of each population were indeed real and significant. As expected, the backcrosses BC1P1 and BC1P2 showed means that tended to be located close to those of their respective recurrent parents. These results confirmed the choice of parents for the present study as contrasting which is a prerequisite for generation means analysis as proposed by Mather and Jinks (1971). The same methodology (generation means analysis) has been used in common bean to study the inheritance of other complex traits such as leafhopper insect resistance (Kornegay and Temple 1986), rate of ethylene production (Sauter et al. 1990), bean pod morphology (Chung et al. 1991), aschochyta leaf blight tolerance (Hanson et al. 1993), leaf trichome density (Park et al. 1994), and most recently heat tolerance (Rainey and Griffiths 2005).
Complete dominance was suggested for both PH and IL in the cross G2333 x G19839 [GenBank] because there were no significant differences between P1, BC1P1, and F1 generations despite significant differences between the 2 parents. Meanwhile, in the Tío Canela x G2333 and BRB32 x MAC47 populations, the mean of the F1 generation for PH suggested only partial dominance. It is interesting that dominance was more important in the population developed from a cross between gene pools than in the populations derived from crosses within the same gene pool, where dominance was found to be partial. Another interesting observation was that dominance effects were less evident at flowering and became more notable as the plants developed into latter growth stages. This was shown by the analyses of generation means for both PH and IL in the G2333 x G19839 [GenBank] and Tío Canela x G2333 populations when comparing the evaluation at 40 DAP versus the evaluation at 70 DAP. These 2 crosses have in common the Mesoamerican parent G2333, which suggests that the later manifestation of the effects of the dominance observed may be a characteristic of this material or the gene pool it belongs to. It is important to note that for the population BRB32 x MAC47, in contrast to the other 2 populations, no dominance was observed in IL, whereas partial dominance was evident for PH at both evaluation dates, although more significantly at 70 DAP as in the other populations. Our results agree with those of Detongnon's (1985) for determinate genotypes of common bean, which indicated that both additive and dominant effects are involved in the expression of PH and IL. Additive or additive–dominance effects have been more prevalent than other epistatic effects in generation means analysis conducted for common beans (Kornegay and Temple 1986; Sauter et al. 1990; Chung et al. 1991; Hanson et al. 1993; Park et al. 1994; Rainey and Griffiths 2005).
The comparison of means for the different generations in each of the 3 crosses, together with the analysis of frequency distributions, showed that mid-parent heterosis was highest in the cross G2333 x G19839 [GenBank] , where the progenitors came from different gene pools (Andean and Mesoamerican). This was likely due to the greater genetic distance presumed to exist between G2333 and G19839 [GenBank] compared with the crosses carried out within the Andean (BRB32 x MAC47) and Mesoamerican (Tío Canela x G2333) gene pools. Transgressive segregation was not highly evident for either of the traits evaluated in the F2 or BC1 generations of any of the populations. However, more transgressive segregation was evident in the Andean intra–gene pool and the Mesoamerican x Andean inter–gene pool populations than in the Mesoamerican intra–gene pool population. The relationship between overdominance, complete, or partial dominance and patterns of heterosis as well as the existence of transgressive segregation for the 2 traits of PH and IL in climbing beans should be further studied.
For the 3 populations, additive effects were established as being of greater importance compared with dominance, especially in the first evaluation for both PH and IL. Similarly, in the second evaluation the additive model was more acceptable than the additive–dominance model for 2 of the 3 populations. In addition, in the evaluations where the additive–dominance model did give a better fit, the contribution of the additive effects to the SSq of the model was greater than the contribution of the effects of dominance. These results agree with Hamad (1975) who found that for climbing snap beans most of the variation in PH was due to additive gene action and that the additive gene action was more important than the dominant gene action for this variable. Confirming this for all 3 populations and for all the traits measured, the additive variance had a value that was very close to that of genotypic variance (Table 4), highlighting the importance of additive effects in the expression of these traits results that agree with those of Ortíz de La Cruz (1989) who studied the inheritance of growth habit using diallelic crosses between genotypes with type I, II, III, and IV growth habits and showed that additive factors predominated in the inheritance of most of the traits.
In our results, broad-sense and narrow-sense heritability values were high and suggested a large participation of the genetic effects on the phenotypic expression of the traits and that selection for the traits would be expected to be highly efficient. These results are similar to those reported by Ortega Ybarra (1968), who found that in 3 crosses between genotypes with bush (Goiano), prostrate (Costa Rica), and indeterminate (Mexico 50) growth habits, narrow-sense heritabilities for the length of the main stem ranged from 50% to 68%. Our results show similar values of heritability for PH and IL in the 2 evaluations, in contrast to Detongnon (1985), who suggested that heritability for PH was higher than for IL. This difference may have stemmed from the fact that all the parents in our study were indeterminate, whereas this previous author worked with beans of determinate growth habit. According to Bliss (1971), the genetic factors that control PH, duration of growth, and flowering time are complex and often affected by the environment. Although we preferred to carry out this trial in a single site where climbing beans are well adapted, it would be interesting to determine if generation means analysis results would be similar in additional environments because genotype x environment interactions have been observed for both PH and IL (Scully et al. 1991). To counteract the limitation of estimating heritabilities from generation means analysis, we have also studied the heritability of PH and IL for a recombinant inbred line population developed from the cross G2333 x G19839 [GenBank] grown over 4 contrasting environments (Checa et al. 2004).
The general results of this study are promising for climbing bean improvement because the additive fraction of genetic variance is useful in breeding self-pollinating crops such as common bean. The results suggest that given the similar inheritance of the PH and IL traits, these can be combined into a climbing ability score which will be useful for a rapid assessment of adaptation for climbing beans. Further crosses should be used to confirm the inheritance of climbing ability in different genetic backgrounds within each gene pool.
|
Acknowledgments |
|---|
We are grateful to Yercil Viera for field trial management, to Steve Beebe, Daniel Debouck, and Andy Jarvis for helpful discussions and reviews of the manuscript and to Patricia Zamorano for formatting. This work was supported by Fontagro grant ATN/SF-7382-RG.
|
Footnotes |
|---|
Corresponding Editor: Reid Palmer
Received February 1, 2006
Accepted July 6, 2006
|
References |
|---|
|
TopMaterials and MethodsResultsDiscussionReferences |
|---|
-
Becerra Velasquez VL and Gepts P. (1994) RFLP diversity of common bean (Phaseolus vulgaris L.) in its centers of origin. Genome 37:256–263.
Blair MW, Giraldo MC, Buendía HF, Tovar E, Duque MC, Beebe SE. (2006) Microsatellite marker diversity in common bean (Phaseolus vulgaris L.). Theor Appl Genet 113:100–109.[CrossRef][Web of Science][Medline]
Bliss FA. (1971) Inheritance of growth habit and time of flowering in beans, Phaseolus vulgaris L. J Am Soc Hortic Sci 93:715–717.
Ceballos H, Pandey S, Narro L, Perez-Velázquez JC. (1998) Additive, dominant and epistatic effects for maize grain yield in acid and non-acid soils. Theor Appl Genet 96:662–668.[CrossRef]
Checa OE, Blair MW, Beebe SE. (2004) Genotypic variation in climbing ability traits in a common bean RIL population. Annu Rep Bean Improv Coop 47:301–302.
Chung WJ, Baggett JR, Rowe KE. (1991) Inheritance of pod cross-section in beans (Phaseolus vulgaris L.). Euphytica 53:159–164.[CrossRef]
Coyne DP. (1967) Photoperiodism: inheritance and linkage studies in Phaseolus vulgaris. J Hered 58:313–314.
Debouck D. (1991) Systematics and morphology. In van Schoonhoven A and Voysest O (Eds.). Common beans: research for crop improvement(CAB Intl, Wallingford, UK) pp. 55–118.
Debouck D. (1999) Diversity in Phaseolus species in relation to the common bean. In Singh S (Ed.). Common bean improvement in the twenty-first century(Kluwer Academic Publication, Dordrecht, The Netherlands) pp. 25–51.
Detongnon J. (1985) Inheritance of stem elongation tendency in determinate forms of Phaseolus vulgaris L. [PhD dissertation]. (Oregon State University, Corvallis, OR).
Emerson RA. (1904) Heredity in bean hybrids. Ann Rep Nebr Agric Exp Stn 17:33–78.
Evans A. (1973) Exploitation of the variability in plant architecture in Phaseolus vulgaris. In Wall D (Ed.). Potentials of field bean and other food legumes in Latin America(CIAT, Cali, Colombia) pp. 279–286.
Evans A. (1976) Beans. In Simmonds NW (Ed.). Evolution of crop plants(Longman, London) pp. 168–172.
Evans AM. (1980) Structure, variation, evolution, and classification in Phaseolus. In Summerfield RJ and Bunting AH (Eds.). Advances in legume science(Royal Botanic Gardens, Kew, UK) pp. 337–347.
Francis CA and Sanders JH. (1978) Economics analysis of bean and maize system: monoculture versus associated cropping. Field Crops Res 1:319–335.[CrossRef]
Frazier WA, Baggett JR, Sistrunk WA. (1958) Transfer of certain Blue Lake bean pod characters to bush beans. Proc Am Soc Hortic Sci 71:416–421.
Gentry HS. (1969) Origin of the common bean, Phaseolus vulgaris L. Econ Bot 23:55–69.
Gepts P and Debouck D. (1991) Origin, domestication, and evolution of the common bean (Phaseolus vulgaris L.). In van Schoonhoven A and Voysest O (Eds.). Common beans: research for crop improvement(CAB Intl, Wallingford, UK) pp. 7–53.
Gepts P, Osborn TC, Rashka K, Bliss FA. (1986) Electrophoretic analysis of phaseolin protein variability in wild forms and landraces of the common bean, (Phaseolus vulgaris L.): evidence for multiple centers of domestication. Econ Bot 40:451–468.[Web of Science]
Hamad IA. (1975) Inheritance of yield. Yield components, numbers of days to flowering, plant height an incidence of interlocutor caviation of pods in snap beans Phaseolus vulgaris L. [PhD dissertation]. (University of Wisconsin-Madison, Madison, WI).
Hanson PM, Pastor-Corrales MA, Kornegay JL. (1993) Heritability and sources of Ascochyta blight resistance in common bean. Plant Dis 77:711–714.
Koenig R and Gepts P. (1989) Allozyme diversity in wild Phaseolus vulgaris: further evidence for two major centers of genetic diversity. Theor Appl Genet 78:809–817.[CrossRef][Web of Science]
Koinange EMK, Singh SP, Gepts P. (1996) Genetic control of the domestication syndrome in common-bean. Crop Sci 36:1037–1145.
Kooiman HN. (1931) Monograph on the genetics of Phaseolus. Bibliogr Genet 8:195–413.
Kornegay J, White JW, Ortiz de La Cruz O. (1992) Growth habit and gene pool effects on inheritance of yield in common bean. Euphytica 62:171–180.
Kornegay JL and Temple SR. (1986) Inheritance and combining ability of leafhopper defense mechanisms in common bean. Crop Sci 26:1153–1158.
Kretchmer PJ, Laing DR, Wallace DH. (1979) Inheritance and morphological traits of a phytochrome-controlled single gene in bean. Crop Sci 19:605–607.
Kretchmer PJ, Ozbun JL, Kaplan SL, Laing DR, Wallace DH. (1977) Red and far-red light effects on climbing in Phaseolus vulgaris L. Crop Sci 17:797–799.
Lamprecht H. (1947) The inheritance of the slender-type of Phaseolus vulgaris and some other results. Agri Hort Genet 5:72–84.
Mather K and Jinks L. (1971) Biometrical genetics(Cornell University Press, Ithaca, NY).
Norton JB. (1915) Inheritance of habit in the common bean. Am Nat 49:242–261.
Obando L. (1992) El fríjol (Phaseolus vulgaris L.) en los agroecosistemas de ladera de Nariño, Colombia. Curso internacional sobre cultivo de fríjol (Phaseolus vulgaris L.) en zonas de ladera de la región andina(ICA, CORPOICA, PROFRIZA, Cali, Colombia) pp. 51–64.
Ortega Ybarra S. (1968) Contribución al estudio de la herencia del hábito de crecimiento en Phaseolus vulgaris L. Agron Trop (Venezuela) 28:87–115.
Ortíz de La Cruz O. (1989) Estudio de heredabilidad del hábito de crecimiento y otros caracteres morfológicos en fríjol [dissertation]. (Universidad Nacional, Palmira, Colombia).
Park SJ, Timmins PR, Quiring DT, Jui PY. (1994) Inheritance of leaf area and hooked trichome density of the first trifoliolate leaf in common bean (Phaseolus vulgaris L.). Can J Plant Sci 74:235–240.
Rainey KM and Griffiths PD. (2005) Inheritance of heat tolerance during reproductive development in snap bean (Phaseolus vulgaris L.). J Am Soc Hortic Sci 130:700–706.
Sañudo B, Checa OE, Arteaga G. (1999) Manejo agronómico de leguminosas en zonas cerealistas. (PROFRIZA, UDENAR, FENALCE, Pasto, Nariño).
SAS Institute. (1999) SAS/STAT user's guide, release 8.1 ed(SAS Institute, Cary, NC).
Sauter KJ, Davis DW, Li PH, Wallerstein IS. (1990) Leaf ethylene evolution level following high-temperature stress in common bean. Hortscience 25:1282–1284.
Scully BT, Wallace DH, Viands DR. (1991) Heritability and correlation of biomass, growth rates, harvest index, and phenology to the yield of common beans. J Am Soc Hortic Sci 116:127–130.
Singh SP. (1982) A key for identification of different growth habits of Phaseolus vulgaris L. Annu Rep Bean Improv Coop 25:92–95.
Singh SP. (1989) Patterns of variation in cultivated common bean (Phaseolus vulgaris Fabaceae). Econ Bot 43:39–57.
Singh SP. (2001) Broadening the genetic base of common bean cultivars. A review. Crop Sci 41:1659–1675.
Singh SP, Gutierrez JA, Molina A, Urrea C, Gepts P. (1991) Genetic diversity in cultivated common bean. II. Marker based analysis of morphological and agronomic traits. Crop Sci 31:23–39.
Smartt J. (1985) Evolution of grain legumes. IV Pulses in the genus Phaseolus. Exp Agric 21:193–207.
Smartt J. (1988) Morphological, physiological and biochemical changes in Phaseolus beans under domestication. In Gepts P (Ed.). Genetic resources of Phaseolus beans(Kluwer, Dordrecht, The Netherlands) pp. 143–161.
Sonnante G, Stockton T, Nodari RO, Becerra Velasquez VL, Gepts P. (1994) Evolution of genetic diversity during the domestication of common-bean (Phaseolus vulgaris L.). Theor Appl Genet 89:629–635.[CrossRef][Web of Science]
Sperling L and Munyaneza S. (1995) Intensifying production among smallholder farmers: the impact of improved climbing beans in Rwanda. Afr Crop Sci J 3:117–125.
Steel RGD and Torrie JH. (1980) Principles and procedures of statistics. 2nd ed. (McGraw-Hill, New York).
Voysest O. (2000) Mejoramiento genético del fríjol Phaseolus vulgaris L: legado de variedades de América Latina. (CIAT, Cali, Colombia)1930–1999.
Wallace DH and Yan W. (1998) Plant breeding and whole system crop physiology, improving adaptation, maturity and yield(CAB Intl, Wallingford, UK).
Woolley J and Davis JHC. (1991) The agronomy of intercropping with beans. In van Schoonhoven A and Voysest O (Eds.). Common beans: research for crop improvement(CAB Intl, Wallingford, UK) pp. 707–735.
Woolley J, Lépiz R, Portes e Castro TA, Voss J. (1991) Bean cropping systems in the tropics and subtropics and their determinants. In van Schoonhoven A and Voysest O (Eds.). Common beans: research for crop improvement(CAB Intl, Wallingford, UK) pp. 679–706.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. M. Warner and A. E. Walworth Quantitative Inheritance of Crop Timing Traits in Interspecific Hybrid Petunia Populations and Interactions with Crop Quality Parameters J. Hered., February 8, 2010; (2010) esp131v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||