Crop Science 40:1618-1623 (2000)
© 2000 Crop Science Society of America
CROP PHYSIOLOGY & METABOLISM
Semidwarf and Standard-Height Cowpea Responses to Row Spacing in Different Environments
Abdelbagi M. Ismail and
Anthony E. Hall
Dep. of Botany and Plant Sciences, Univ. of California, Riverside, CA 92521-0124 USA
anthony.hall{at}ucr.edu
 |
ABSTRACT
|
|---|
Yield potential of several grain crops has been increased by breeding semidwarf cultivars with greater harvest index and growing them at high plant densities. We evaluated effects of the semidwarf trait on yield potential of cowpea [Vigna unguiculata (L.) Walp.] in contrasting, but productive, environments. Three semidwarf and three standard-height cowpea lines with similar indeterminate habits and phenology were evaluated at row spacings of 51, 76, and 102 cm in four field environments where the extent of early vegetative vigor varied. Main stems of semidwarf lines were 33% shorter, and they produced 13% less vegetative shoot biomass than standard lines. Average grain yield of semidwarf lines was 10% greater than that of standard lines and was associated with 11% higher harvest index and 19% greater pod set. Semidwarf lines produced relatively greater yield than standard lines at narrower row spacings (15, 11, and 4% greater than standard lines at 51, 76, and 102 cm row spacing, respectively). Average yield of standard lines did not respond to row spacing. In a stressful soil environment that caused extreme dwarfing of both plant types, there was no difference in grain yield between semidwarf and standard lines. Semidwarf lines produced greater yield than standard lines at narrow row spacing in soil conditions that promoted moderate to vigorous early plant growth due to impaired reproduction of standard, but not semidwarf, lines when competition for light was strong.
|
INTRODUCTION
|
|---|
PRODUCTIVITY OF SEVERAL grain crops has been enhanced in productive environments by breeding semidwarf cultivars with greater harvest index and growing them at high plant densities. For cowpea, Kwapata and Hall (1990) suggested that selecting bush-type cowpea for greater harvest index could produce cultivars with the potential for greater productivity under high plant densities. Six pairs of breeding lines of cowpea have been developed that have similar genetic backgrounds and either have, or do not have, heat tolerance during flowering. The heat-tolerance trait was shown to be associated with dwarfing and greater harvest index (Ismail and Hall, 1998, 1999). One of the semidwarf lines with heat tolerance has been released as a new cultivar in California (Ehlers et al., 2000), where cowpea is mainly grown under wide row spacing (76–102 cm between rows), with 7 to 10 cm between plants within the row. In most other parts of the world where cowpea is grown, plant densities are much lower than in California and in some cases the row spacing is even wider. Consequently, it is of interest to determine whether grain yield of cowpea can be enhanced by growing semidwarf cultivars under narrow row spacing.
Studies with soybean [Glycine max (L.) Merr.] provide guidelines concerning the responses of semidwarf grain legume cultivars to narrow row spacing and high plant densities. Semidwarf soybean cultivars have been developed that produced greater grain yield than standard-height cultivars when grown with narrow row spacing in productive environments (Cooper, 1985). The semidwarf cultivars produced greater yield at a very narrow row spacing of 17 cm with little advantage over standard cultivars at 75 cm row spacing and low plant densities. One reason for the greater yield was a greater resistance to lodging than the standard cultivars. However, in more recent years, standard-height cultivars of soybean have been developed with substantial resistance to lodging.
The first semidwarf soybean cultivars that were developed were determinate (Cooper, 1985). A determinate soybean cultivar was shown to have greater yield response to an equidistant planting pattern because it produced more reproductive nodes than indeterminate cultivars (Egli, 1994); however, semidwarf determinate soybean cultivars were less capable of compensating for poor stands and stressful environmental conditions during early vegetative growth than taller, indeterminate cultivars (Cooper, 1985). A semidwarf determinate cultivar produced greater yield in narrow rows than taller, indeterminate cultivars in wider rows. Effects were greater when drought occurred early, and there was no advantage when drought occurred late (Cooper, 1989). Planting date may influence the comparison in that indeterminate cultivars exhited stronger reductions in yield with changes in environment due to later planting than did semidwarf determinate cultivars (Beaver and Johnson, 1981). Some contrasting results were obtained in a comprehensive study of 38 pairs of near-isolines of semidwarf determinate and standard-height indeterminate soybean lines grown at narrow (20 cm) and wider (60 cm) row spacing with both early and late planting (Robinson and Wilcox, 1998). They observed that on average, semidwarf determinate lines had less yield than standard-height indeterminate lines under virtually all conditions that were tested, even though they had less lodging; however, the lines used in these studies had not been selected for yield. A more definitive test would include breeding to select a genetic background that complements the semidwarf trait. The semidwarf determinate soybean cultivars are not widely grown by farmers in the USA. They are grown primarily on highly productive soils in the Midwest and under irrigation conditions where lodging can limit the yield of taller indeterminate cultivars.
Dwarfing in soybean also can be accomplished by using the brachytic stem trait. Four pairs of near-isolines, with and without the brachytic trait, were compared at two row spacings (76 and 28 cm), in four environments (Adams and Weaver, 1998). The brachytic lines were shorter but yielded less than the standard-height lines at both row widths in all environments. Unfortunately, overall yield was low and there was little lodging in these four environments, so the studies did not provide an answer to the critical question of the influence of the brachytic trait in environments with high yield potential. A more definitive test would also include breeding to select a genetic background that complements the brachytic trait.
For cowpea—a compact, erect determinate cultivar—Prima was shown to have a narrower optimum row spacing (34–40 cm between rows) than the vegetatively vigorous, semi-erect indeterminate cultivar, Pale Green, which showed optimum row spacing between 50 to 65 cm (Nangju et al., 1975). Cultivar differences in grain yield, which were attributed to interactions between drought and determinacy, occurred in contrasting environments (Nangju et al., 1975). Consequently, the influence of compactness on potential grain yield was not determined. Variation in within-row spacing has been shown to have less influence on grain yield of cowpea than differences in between-row spacing. For example, Grantz (1979) evaluated the semi-erect indeterminate CB5 at within-row spacings of 10, 20, 30, 40, and 61 cm, with a between-row spacing of 76 cm. Large grain yields of 3000 to 3400 kg ha-1 were obtained, but with no significant effects from within-row spacing with all treatments producing 
250 pods m-2.
We evaluated the influence of the semidwarf trait on yield potential of cowpea in contrasting environments. We compared semidwarf and standard-height, semi-erect cowpeas that had been selected for yield as well as for similar indeterminacy and dates of first flowering. We evaluated their response to row spacing in well-watered environments while grown in soils differing in the ability to influence the vigor of early vegetative growth.
|
Materials and methods
|
|---|
The six cowpea lines that were used in this study were semi-erect, indeterminate, had similar dates of first flowering, and were selected for yield and high adaptibility to the field environments used in this study. Two pairs of lines had been shown to differ in height and harvest index by Ismail and Hall (1998, 1999). Cultivar CB5 had standard height, was sensitive to heat at flowering, and was the major cultivar grown in California from the 1940s to 1990 (Hall and Frate, 1996). CB5 is still grown in a small area in California. Line H36 is a semidwarf, and was developed by crossing CB5 with a heat-tolerant parent followed by two cycles of backcrossing with CB5 while selecting for heat tolerance during flowering and yield. The second pair were selected from an F6 family that segregated for heat tolerance during flowering: the heat-tolerant semidwarf line, H8-8-27, which was released for use in California as CB27 (Ehlers et al., 2000); and the heat-susceptible near-isoline, H8-8-1N, which has standard height. We also included a current, major California cultivar, CB46 (Helms et al., 1991), which is more compact than CB5 (Hall and Frate, 1996) and sensitive to heat at flowering. Finally, we included 8517, which has standard height and is also sensitive to heat at flowering (Ismail and Hall, 1999). In one of the four experiments (Shafter in 1998), we had insufficient 8517 seed available and used another line of standard height (9802), that was sensitive to heat at flowering.
Three row spacings were used in this study that are currently used by farmers in California (Hall and Frate, 1996): a single row on beds 102 cm apart, a single row on beds 76 cm apart, and a narrow-row system consisting of two rows spaced 35 cm on beds 102 cm apart providing an average row-spacing of 51 cm. These three row spacing and bed configurations were consistent with furrow irrigation, which was used in this study, and is the major irrigation system used to produce cowpea in California. In all cases, an average of one seed was sown every 7 cm within the row using a cone planter. This resulted in seeding rates of 140 000, 188 000, and 280 000 seed/ha for the 102, 76, and 51 cm row spacings, respectively. Seed was inoculated with Rhizobia (EL type, Nitragin Co., Milwaukee, WI). Individual plots consisted of four beds that were trimmed in the middle of the season to 7.5 m long. The experimental design was a randomized, complete-block design with four replications.
Trials were sown on 22 May and 11 June in 1998, and on 11 May and 20 May in 1999, at Shafter and Riverside, respectively. Irrigation was provided to maintain well-watered conditions. Weeds were controlled by herbicides and occasional hand weeding, and insecticides were used to control lygus (Lygus hesperus Knight) infestations.
The two central beds were harvested when the first flush of pods were mature and dry. At Shafter, harvesting took place 94 d after sowing in 1998, and 108 d after sowing in 1999. At Riverside, harvesting took place 113 d after sowing in 1998, and 110 d after sowing in 1999. A random sample of seven plants was used to determine pod set, main stem length, average length of main stem internodes, and harvest index. Pod set was estimated based on the average number of pods per peduncle on the first five reproductive nodes on the main stem. Harvest index was determined as the ratio of grain weight to total shoot biomass after drying the samples at 65°C for 7 d. The remainder of the two central beds was allowed to dry in the field and then threshed. Subsamples of seed were dried at 105°C for 48 h to determine moisture content on a fresh weight basis. Grain yield was based on all plants that were harvested from the two middle beds of each plot and was reported on a 100 g kg -1 seed moisture content basis.
Statistical Analysis
Statistical analysis was conducted for each character studied based on a randomized, complete-block design with four replications for each environment, and in combined analysis across the four environments. Means and LSD values (after a significant F test) were calculated. Associations among characters were examined by simple correlation analysis.
|
Results
|
|---|
No lodging was observed for any line in any of the environments. Main stems of semidwarf lines were shorter in all environments, averaging only 64% as long as the main stems of the standard lines (Table 1)
. Main stems were longest at Shafter in 1999, and shortest in both years at Riverside, indicating that the Riverside environments were the most dwarfing (Table 1). Short main stem lengths of the standard lines at Riverside in 1999 indicated this was the most dwarfing environment. Overall, main stems were shortest at the narrowest row spacing, and similar at the intermediate and widest row spacings. Genotypic differences in plant main stem length were present among the semidwarf lines. CB27 had the shortest main stem, followed by H36 and CB46. There were no differences among the standard lines.
View this table:
[in this window]
[in a new window]
|
Table 1 Main stem length of semidwarf and standard-height cowpea lines grown at Riverside and Shafter during 1998 and 1999
|
|
Similar to the variation in main stem length, average internode length of the semidwarf lines was only 69% that of the standard lines (Table 2)
. Correlations between main stem length and internode number and internode length were 0.89 and 0.98, respectively, for genotypic means over the four different environments
. Variation in both main stem internode number and average internode length contributed to variation in main stem length, but with most of the effect due to variation in internode length.
View this table:
[in this window]
[in a new window]
|
Table 2 Average main stem internode length of semidwarf and standard-height cowpea lines grown at Riverside and Shafter during 1998 and 1999
|
|
Overall mean grain yield of the semidwarf lines was 10% greater than that of the standard lines (Table 3)
. Semidwarf lines produced relatively greater yield than standard lines at narrower row spacings (15, 11, and 4% greater at 51, 76, and 102 cm row spacing, respectively), indicating a relationship between plant habit and row spacing. Semidwarf lines produced their greatest yield at the narrowest row spacing and smallest yield at the widest row spacing, whereas the standard lines had similar yield at all row spacings. For genotypic effects averaged over the 51 and 76 cm row spacings, CB27 (3597 kg ha-1) and H36 (3564 kg ha-1) had the greatest average grain yield, followed by CB46 (3380 kg ha-1) and H8-8-1N (3268 kg ha-1), with 8517 (3078 kg ha-1) and CB5 (2916 kg ha-1) having the smallest yield.
View this table:
[in this window]
[in a new window]
|
Table 3 Grain yield of semidwarf and standard-height cowpea lines grown at three different row spacings at Riverside and Shafter in 1998 and 1999
|
|
Treatment effects on grain yield at Riverside in 1998 were similar to effects on overall mean grain yield with semidwarf lines producing greater grain yield than standard lines at the narrowest row spacing. In the most dwarfing environment (Riverside in 1999), plant habit had no effect on yield. The greatest grain yield was obtained at the narrowest row spacing for both semidwarf and standard lines with no relationship between plant habit or genotype and row spacing. In the more favorable soil environments at Shafter in 1998 and 1999, semidwarf lines produced the greatest grain yield with no row spacing effects (Table 3). There was an interaction between plant habit and row spacing due to the semidwarf lines producing relatively greater yields at the narrower row spacings.
Average vegetative shoot biomass of the semidwarf lines was 87% that of the standard lines, indicating that the semidwarf lines were more compact (Table 4)
. Vegetative shoot biomass was 13 and 10% greater at 51 and 76 cm row spacing, respectively, compared to shoot biomass produced at 102 cm row spacing, with no relationship between row spacing and either plant type or genotype.
View this table:
[in this window]
[in a new window]
|
Table 4 Vegetative shoot biomass, harvest index, number of pods per peduncle, number of seeds per pod and position of the first fruiting node of semidwarf and standard-height cowpea lines grown at Riverside and Shafter in 1998 and 1999. Values are means of the four different experiments
|
|
Average harvest index of the semidwarf lines was 11% greater than that of the standard lines (Table 4). Harvest index was also greater at the widest row spacing than at the narrowest row spacing. The semidwarf lines CB27 and H36 had the greatest harvest indices while the standard lines, 8517 and CB5, had the smallest harvest indices, and CB46 and H8-8-1N were intermediate. The semidwarf lines produced 19% more pods per peduncle than the standard lines. Pod set was 9% higher at both 76 and 102 cm row spacings compared with 51 cm row spacing (data not shown). H36 had the highest pod set followed by CB27 and CB46, whereas 8517 and CB5 had the lowest pod set. The interaction between lines and row spacing was mainly due to some lines (H36, CB46, and 8517) experiencing greater reductions in pod set at 51 cm row spacing compared with 76 and 102 cm than the other lines.
There was fewer seed per pod at the narrowest row spacing. Lines H8-8-1N, CB27, and CB46 had the most seed per pod followed by 8517, whereas CB5 and H36 had the fewest seed per pod. Semidwarf lines set their first fruits about one node lower on the main stem compared with standard lines. Genotypic differences in first fruiting node were greater among the semidwarf lines, with H36 having the lowest first fruiting node, followed by CB27, and then CB46. Standard lines produced their first fruits at similar nodal positions on the main stem.
|
Discussion
|
|---|
The much shorter main stems of the semidwarf cowpea lines (0.64 as long as the standard lines) was partially due to standard lines having a longer terminal internode that twined around so that differences in plant height were much smaller. Semidwarf lines were, however, more compact than the standard lines and produced less vegetative biomass (87% of the standard lines).
The greater grain yield of semidwarf lines in all but the most dwarfing environment suggests that the semidwarf lines are better adapted than standard lines to soil conditions that promote moderate to vigorous early plant growth. Correlation coefficients for the association of grain yield with vegetative shoot biomass for means of individual genotypes in the four different environments were 0.82
for the semidwarf lines, and 0.49
for the standard lines.
The greater grain yield of the semidwarf lines was associated with increased harvest index and pod set. The smaller vegetative shoot biomass and higher harvest index of the semidwarf lines suggested that these plant types were more efficient in partitioning assimilate supply into grain production than the standard lines. The semidwarf cultivars were bred by incorporating resistance to various stresses with final selection based strongly on grain yield. CB27 is the most recent cultivar developed in California and was released in 1999 (Ehlers et al., 2000), whereas, CB46 was released in 1987 (Helms et al., 1991), and CB5 has been in use since the 1940s (Hall and Frate, 1996).
The greater grain yield of the semidwarf lines with narrower row spacing (yield of the standard lines were similar at all row spacings) indicate that the semidwarf lines are more responsive to higher planting densities than the standard lines. These data support the suggestion of Kwapata and Hall (1990) that breeding bush-type cowpea with higher harvest index could produce cultivars with the potential for higher productivity under high plant density.
Grain yield responses to narrow row spacing compared with wide row spacing may be attributed to narrow rows, resulting in greater light interception, greater production of vegetative biomass and peduncles per land area, and a proportionate increase in pod production and grain yield. The responses of the semidwarf lines are consistent with this model with the greatest increases in vegetative biomass and grain yield occurring in the Riverside environments where early vegetative growth was slow (Fig. 1)
. The standard lines exhibited a substantial vegetative response to narrow rows in three of the four environments, but only exhibited a grain yield response in the most dwarfing environment of Riverside in 1999 (Fig. 1). These data suggest that reproductive development of the standard lines was impaired by the strong competition for light that occurred with the narrowest row spacing and low-stress soils. Presumably, branch nodes of the standard lines, which were not studied, produced few pods with the narrowest row spacing at Shafter in 1999, and Riverside in 1998. Branch nodes can make a major contribution to grain production. For CB5 growing on a double-row bed system with a 10 cm within-row spacing, branches produced 45% of the pods and 42% of the grain yield (Stoffella and Fousek, 1989).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 1 Grain yield responses (yield at 51 cm–yield at 102 cm row spacing) of semidwarf (open symbols) and standard-height (closed symbols) cowpea lines as affected by vegetative responses (vegetative biomass at 51 cm–vegetative biomass at 102 cm row spacing). Data are from four environments. R = Riverside, S = Shafter; 8 = 1998, 9 = 1999. The solid linear regression line has an r2 of 0.99
|
|
Care is needed when comparing the results of this cowpea research with research conducted on soybean. The narrow-row systems used in research on semidwarf soybeans had much closer rows (17–20 cm apart) than the narrowest row spacing used in this research. However, for both cowpea and soybean, the semidwarf trait confers a yield advantage on narrow-row systems with good soil conditions that results in adequate vegetative growth of the semidwarf cultivars and excessive vegetative growth of standard-height cultivars.
Semidwarf compact cultivars of cowpea conferred a yield advantage over standard-height cultivars in productive environments where early vegetative growth was not too slow and row spacing was less than 102 cm (Fig. 2)
. The optimal row spacing for semidwarf cowpea cultivars was not established in this study. Since furrow irrigation was used, the narrowest row spacing tested involved double rows on beds 102 cm apart to give an average row spacing of 51 cm. The greatest grain yield was achieved with semidwarf lines on this double-row bed system. Further tests are needed to determine whether the semidwarf lines will provide even greater yield with rows that are closer and more uniformly spaced than in this double-row bed system.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 2 Grain yield (±SE) of semidwarf (open symbols) and standard-height (closed symbols) cowpea lines averaged over three environments (Riverside 1998, Shafter 1998, and 1999) where early vegetative growth was moderate to vigorous
|
|
|
NOTES
|
|---|
Research partially supported by USDA-NRICGP, Award no. 98-35100-6129, and USAID Grant no. DAN-G-SS-86-00008-00 to AEH. The opinions and recommendations are those of the authors and not necessarily those of USAID.
Received for publication February 28, 2000.
|
REFERENCES
|
|---|
- Adams P.D., Weaver D.B. Brachytic stem trait, row spacing, and plant population effects on soybean yield. Crop Sci. 1998;38:750-755.[Abstract/Free Full Text]
- Beaver J.S., Johnson R.R. Response of determinate and indeterminate soybeans to varying cultural practices in the northern USA. Agron. J. 1981;73:833-838.[Abstract/Free Full Text]
- Cooper R.L. Breeding semidwarf soybeans. Plant Breeding Rev. 1985;3:289-311.
- Cooper R.L. High-yield-system-in-place (HYSIP) concept for soybean production. J. Prod. Agric. 1989;2:321-324.
- Egli D.B. Mechanisms responsible for soybean yield response to equidistant planting patterns. Agron. J. 1994;86:1046-1049.[Abstract/Free Full Text]
- Ehlers J.D., Hall A.E., Patel P.N., Roberts P.A., Matthews W.C. Registration of `California blackeye 27' Cowpea. Crop Sci. 2000;40:854-855.
- Grantz D.A. Earliness and drought adaptation of cowpea (Vigna unguiculata (L.) Walp). Riverside, CA: M.S. Thesis, University of California, 1979.
- Hall, A.E., and C.A. Frate. 1996. Blackeye bean production in California. Univ. of Calif., Div. Agric. Nat. Res. Publ. 21518.
- Helms D., Panella L., Buddenhagen I.W., Tucker C.L., Gepts P.L. Registration of `California Blackeye 46' cowpea. Crop Sci. 1991;31:1703.[Free Full Text]
- Ismail A.M., Hall A.E. Positive and potential negative effects of heat-tolerance genes in cowpea. Crop Sci. 1998;38:381-390.[Abstract/Free Full Text]
- Ismail A.M., Hall A.E. Reproductive-stage heat tolerance, leaf membrane thermostability and plant morphology in cowpea. Crop Sci. 1999;39:1762-1768.[Abstract/Free Full Text]
- Kwapata M.B., Hall A.E. Response of contrasting vegetable-cowpea cultivars to plant density and harvesting of young green pods. I. Pod production. Field Crops Res. 1990;24:1-10.
- Nangju D., Little T.M., Anjorin-Ohu A. Effect of plant density and spatial arrangement on seed yield of cowpea (Vigna unguiculata (L.) Walp.). Am. Soc. Hort. Sci. 1975;100:467-470.
- Robinson S.L., Wilcox J.R. Comparison of determinate and indeterminate soybean near-isolines and their response to row spacing and planting date. Crop Sci. 1998;38:1554-1557.[Abstract/Free Full Text]
- Stoffella P.J., Fousek D.J. Influence of within-row spacings on distribution patterns of yield components in cowpea. Sci. Hort. 1989;41:1-8.
TOP
ABSTRACT
INTRODUCTION
