Development and Dissemination of a Cowpea Cultivar for Cover Crops

2004 Annual Report for SW02-034

Project Type: Research and Education
Funds awarded in 2002: $43,686.00
Projected End Date: 12/31/2007
Matching Federal Funds: $18,974.00
Region: Western
State: California
Principal Investigator:
Dr. Milt McGiffen, Jr.
University of California

Development and Dissemination of a Cowpea Cultivar for Cover Crops


Cowpea cover crops are cost effective because they enrich the soil with carbon and over 100 lbs per acre of nitrogen, and can reduce pest populations. Adoption has been slowed by a lack of pest-resistant varieties adapted for the Western United States. We have identified new sources of resistance to nematodes, weeds, insects, and diseases. In the last year we were able to develop and identify promising candidates for varietal release. The candidate cultivars produce high biomass and seed yields and have resistance to cowpea aphid, root knot nematode, and seed shattering. The ability to withstand drought was similar for all genotypes. An erect growth habit was best able to out-compete weeds. The candidate cultivars will be tested on farms in specific cropping situations to evaluate their suitability for varietal release. Information is being disseminated through talks, publications, and organic production training sessions and a manual.

Objectives/Performance Targets

1) Identify cowpea cover crop cultivars that resist nematodes, cowpea aphid, Fusarium wilt, and shattering when grown in the Western USA.

2) Disseminate seed of improved varieties and related information through the California Foundation Seed Service and commercial seed companies.

3) Demonstrate and optimize the merits of cover crops in specific cropping systems.

4) Disseminate information about cover crops and their advantages, and about seed production of cowpea as a new crop for limited resource and other growers.


Objective 1: Identify new cover crop cultivars. The breeding program has incorporated pest resistance and many other desirable traits into breeding lines, and field testing has verified nematode and aphid resistance and other advanced traits. Many of those results were presented in last year’s report. From those results, we have identified two breeding lines that appear to be significant improvements over Iron Clay, the current standard cultivar for cowpea cover crops.

We are proposing to release CC-85 and CC-110 as new cover crop varieties. On-farm testing began in the summer of 2004 at coastal and inland valley sites. These experiments generally verified the nematode and aphid resistance found in previous field station experiments. We found that climate and time of planting limits the locations where cowpea is an effective cover crop, but that growth is generally excellent when daytime temperatures exceed 90F, such as the Central Valley and desert valleys of California.

Objective 2: Disseminate seed of improved varieties and related information: We have reaffirmed earlier commitments for final on-farm tests of varieties being considered for release with the grant application’s original cooperators. Lockwood Seed and Grain has enthusiastically reaffirmed their interest in testing and disseminating our new cover crop cultivars. We have contacted the California Foundation Seed Service to find ways to make these varieties widely available at an affordable cost to growers. Growers are being notified of the advantages of the new varieties through meetings, commodity boards, and presentations. Seed of the new cowpea genotypes will be increased this summer in the Coachella Valley to provide sufficient quantity for distribution in 2006.

Objective 3: Demonstrate and optimize the merits of cover crops: On-farm experiments and grower contacts have identified biomass production and pest resistance as highly desirable traits for new cover crop varieties. A major focus of the breeding program has been selecting for genotypes with high biomass production and resistance to nematodes, aphids, and diseases. The two novel genotypes that we have recently developed, CC-85 and CC-110, were selected based on their biomass production and enhanced nematode and aphid resistance.

Weed control is the greatest expense in producing many crops. Minimizing weed control costs is even more essential for cover crops than cash crops as there is not a direct financial reward for growing cover crops. If the cost of controlling the weeds exceeds the potential nitrogen and other benefits, then the cover crop is usually plowed under prematurely. However, breeding for resistance to weeds is a concept that has often been suggested, but seldom attempted. A major reason why no food crop or green manure had been specifically bred with traits that resist weeds is a lack of knowledge. Breeders do not know which genetic traits would confer weed resistance, or how to detect traits that could potentially help crops out-compete weeds.

In the many field trials of genotypes conducted prior to the start of Western SARE funding, we observed how cowpea growth habits affect the ability of cowpea to shade the soil. Cowpea growth ranges from erect to prostrate. It appeared that the rapidly spreading vines of the prostrate genotypes would have the greatest ability to shade out weeds. However, a preliminary experiment found that an erect genotype, Iron Clay, had the greater ability to compete with sunflower than semi-erect 288 or prostrate growing 779 (Wang et al. 2004). IC biomass was less affected by sunflower, received more light when growing with sunflower, and caused a greater decline in sunflower biomass and leaf area than 288 or 779.

We used replacement series to compute the aggressivity index (AI) of six cowpea genotypes to confirm the most competitive growth habit. We also investigated the relationship between AI and growth parameters to identify the growth parameters associated with competitive ability (Table 1).

Twelve replacement series experiments were conducted with sunflower or purslane competing with one of six cowpea genotypes of similar vegetative vigor and maturity but different growth habits (Table 1). These were selected because earlier evaluations (Ehlers, unpublished data) showed them to be promising cover-crop cowpea genotypes with similar maturity and vegetative vigor. Common purslane, a short statured weed, and common sunflower, a tall species, were selected to represent cowpea competitors with different growth types. Five proportions of two species (cowpea with sunflower or cowpea with purslane) were used: 100:0, 75:25, 50:50, 25:75, 0:100

Table 1. Six cowpea genotypes used in the study
Genotype Abbreviation Full genotype name Growth habit Use Origin
CB5 California Blackeye 5 Erect Dry grain California
IC Iron-clay Erect Forage Southeastern US
SPH Speckle Purple Hull Semi-erect Dry Grain and forage Southeastern US
288 IT89KD-288 Semi-erect Dry grain and forage IITA, Nigeria
779 UCR 779 Prostrate Forage Botswana
730 UCR 730 Prostrate Forage Botswana

The experiment was a randomized complete block design with four replications. Cowpea and weed seeds were planted on July 10, 2003, and Aug 16, 2004, then thinned to the desired density and proportion five days after germination. All plants were harvested when cowpea had reached their maximal pre-flowering size at 690 degree-days after planting (single sine method (Zalom et al. 1983) using an 8.5 oC base temperature (Hall, 2001)), i.e. on Aug 18, 2003, or October 1, 2004.

Shoot biomass was removed at the soil level and plants were separated by species. Dry weights of each species were obtained by leaving plants at 7 oC with ventilation until a constant weight was reached. To compare the competitive ability of cowpea genotypes against sunflower or purslane, relative yield total (RYT) and aggressivity index (AI) were calculated by:

RYT = yield of species A in mixture + yield of species B in mixture
yield of species A in pure stand yield of species B in pure stand

AI= yield of species A in mixture – yield of species B in mixture
yield of species A in pure stand yield of species B in pure stand

Data on plant dry weight RYT and AI were subjected to analysis of variance (ANOVA) and treatment means separated using Duncan’s test at the 0.05 probability level. An ANOVA of treatment and year showed no significant treatment and year interaction, so plant growth data for the two years were combined. Because the ANOVA showed that weed species affected cowpea growth and cowpea genotypes affected weed growth, separate comparisons were derived for each weed species and cowpea genotype. To test if the same order exists in this glasshouse experiments with more cowpea genotypes, we used the isotonic regression method presented by Robertson (1988).

When grown with sunflower, neither cowpea genotype nor growth type affected the RYTs (Table 2). When grown with purslane, cowpea genotype did affect the RYTs, but the differences disappeared when averaged over cowpea growth types. The overall averages for six cowpea genotypes with sunflower or purslane were very close to 1 (0.98 for cowpea and sunflower; 0.97 for cowpea and purslane), indicating that cowpea used the same resources as sunflower or purslane.

Table 2. Relative yield total of cowpea and sunflower or purslane across different proportions
Growth type Cowpea genotypes RYT with sunflower RYT with purslane
Erect CB5 0.95 a* 0.96 a 0.87 b 0.94 a
IC 0.97 a 1.00 ab
Semi-erect SPH 1.01 a 0.98 a 0.94 b 0.95 a
288 0.94 a 0.96 ab
Prostrate 779 0.96 a 0.99 a 0.93 b 1.01 a
730 1.03 a 1.10 a
Overall average 0.98* 0.97

* Means are separated by Duncan’s test. Means within a column followed by the same letter are equivalent.

When grown with either sunflower or purslane, there were significant growth type, genotype, and proportion effects (all p<0.001) on AI, but the interaction of growth types and proportion or genotypes and proportion was not significant (all p>0.19). When grown with sunflower, erect genotypes and semi-erect genotypes had higher AI than prostrate genotypes, (Table 3), indicating that erect and semi-erect cowpea genotypes were more competitive with sunflower than prostrate genotypes. When grown with purslane, erect and prostrate genotypes had higher AI than semi-erect genotypes, indicating that erect and prostrate cowpea genotypes were more competitive with purslane than semi-erect genotypes.

Table 3. The effect of cowpea varieties on cowpea aggressivity indices
Growth type Cowpea genotypes Aggressivity index with sunflower Aggressivity index with purslane
Erect CB5 -0.35 a* -0.40 a 0.52 a 0.51 a
IC -0.45 ab 0.50 a
Semi-erect SPH -0.41 ab -0.38 a 0.39 ab 0.27 b
288 -0.36 a 0.16 b
Prostrate 779 -0.52 ab -0.56 b 0.42 a 0.35 b
730 -0.59 b 0.29 ab

* Means are separated by Duncan’s test. Means within a column followed by the same letter are equivalent.

The relative yields of cowpea were averaged by growth type, then graphed with those of competing sunflower or purslane (Figure 1). When competing with sunflower, the relative yield of prostrate cowpea genotypes decreased faster than that of erect or semi-erect genotypes, and the relative yield of sunflower increased faster when competing with prostrate cowpea genotypes. This indicates that the erect and semi-erect cowpea genotypes were more competitive to sunflower than prostrate cowpea genotypes. When competing with purslane, the relative yield of semi-erect cowpea genotypes decreased faster than erect and prostrate genotypes. The relative yield of purslane increased faster when competing with semi-erect cowpea genotypes. Erect and prostrate cowpea genotypes were more competitive to purslane than semi-erect genotypes. The relative yield results are consistent with the statistical ranking of AI cowpea growth types when competing with sunflower or purslane (Table 3).

Relative yield to monoculture

Proportion of cowpea to sunflower or purslane

Fig 1. Relative yield of cowpea and sunflower or purslane to their monoculture yield in different proportion
CB5 and IC are erect type; 288 and SPH are semi-erect type; 730 and 779 are prostrate type
( : Cowpea relative yield to monoculture; : Sunflower or purslane relative yield to monoculture)

It appears that the competitive advantage purslane gained by emerging one day earlier than either cowpea or sunflower (five days after planting) was insufficient to overcome purslane’s relatively short stature. Cowpea’s ability to compete with sunflower may have been related to an ability to tolerate shade.
The isotonic regression confirmed the results of Wang et al. (2004) that the order of competitive ability for cowpea genotypes erect> semi-erect > prostrate when grown with sunflower, and erect > prostrate > semi-erect when grown with purslane (full discussion of statistical analyses available upon request).
Results of the replacement series were related with data from growth analysis by multiple correlation and stepwise regression. Plant growth was characterized by directly measuring parameters (seed weight, plant weight, height, and leaf area) or deriving parameters (relative growth rate (RGR), unit leaf rate (ULR), leaf area ratio (LAR), specific leaf area (SLA), leaf weight ratio (LWR), and plant height growth rate (HGR)) using functional methods (Chiariello et al, 1991). Degree-days were calculated using the single sine method (Zalom et al. 1983), with base temperatures of 8.5 oC for cowpea (Hall 2001), 7 oC for sunflower (Robinson 1971), and 10 oC for purslane (Zimmerman 1977). The overall AI was the dependent variables, and the independent variables included mean values of RGR, ULR, LAR, SLA, LWR, HGR, plant dry weight, plant height, and initial seed weight. Correlations matrices of all parameters were calculated (Tables 4 and 5), and stepwise regressions were performed.

Table 4. Correlation matrix among AI and growth parameters when cowpea grown with sunflower
AI RGR ULR LAR SLA LWR Height HGR seed weight dry weight
AI 1.000
RGR 0.238 1.000
ULR 0.911 0.346 1.000
LAR -0.802 0.179 -0.858 1.000
SLA -0.825 0.189 -0.794 0.934 1.000
LWR -0.514 0.323 -0.619 0.826 0.589 1.000
Height 0.970 0.400 0.902 -0.706 -0.687 -0.481 1.000
HGR 0.950 0.469 0.882 -0.650 -0.644 -0.410 0.995 1.000
seed weight -0.906 -0.371 -0.826 0.629 0.701 0.282 -0.857 -0.839 1.000
dry weight 0.951 0.486 0.847 -0.593 -0.640 -0.288 0.968 0.971 -0.931 1.000

Correlation and regression were performed to determine which growth parameters best predicted the aggressiveness of cowpea genotypes and weeds. When grown with sunflower, the parameter least correlated with AI was RGR (Table 4). This is consistent with the study by Roush and Radosevich (1985). Plant height had the strongest relationship with AI. Stepwise regression showed that SLA, plant height, and seed weight best explained the variation of AI, indicating larger initial size, higher position, and larger leaf area per unit leaf weight to capture more light were the more important determinants of competitive outcome. The equation was AI = 0.50 – 62.48 * SLA + 0.01 * Height – 0.22 * Seed_weight. The R2 for the equation was 0.996.

When grown with purslane, the least correlated parameter was plant height, and the most correlated was ULR (Table 5). The equation selected by the stepwise procedure included ULR, SLA, and biomass, indicating the efficiency to produce new growth, larger leaf area per unit leaf weight, and plant size were the most important plant growth parameters in determining competitive ability. The equation was 1.37 – 1.30 * ULR –78.49 * SLA + 0.063 * Dry weight. These three variables explained 97.0% variation of AI.

Table 5. Correlation matrix among AI and growth parameters when cowpea grown with purslane
AI RGR ULR LAR SLA LWR Height HGR seed weight dry weight
AI 1.000
RGR -0.911 1.000
ULR -0.929 0.986 1.000
LAR 0.916 -0.895 -0.954 1.000
SLA 0.770 -0.747 -0.838 0.943 1.000
LWR 0.875 -0.770 -0.847 0.939 0.843 1.000
Height -0.611 0.687 0.719 -0.744 -0.605 -0.811 1.000
HGR -0.803 0.928 0.931 -0.880 -0.733 -0.820 0.900 1.000
seed weight -0.917 0.988 0.999 -0.943 -0.824 -0.832 0.719 0.934 1.000
dry weight -0.873 0.988 0.986 -0.900 -0.771 -0.773 0.719 0.942 0.992 1.000

Impacts and Contributions/Outcomes

Our previous work to introduce broad-based nematode and other pest resistance has led to new genotypes that will reduce pest populations during the cover crop season and in subsequent cash crops. We have identified two genotypes that appear to be pest resistant and yield well. We are proposing CC-85 and CC-110 as new cover crop varieties. We will need another year of on-farm and other tests to verify their performance in different cropping systems, but CC-85 and CC-110 are a substantial improvement over previous releases in terms of pest resistance and yield

Our research on the competitiveness of cowpea with weeds indicated that erect types were better competitors with weeds than cowpeas with semi-erect or prostrate growth habit. As a result of these findings we have selected aggressive erect growing genotypes that should be even better competitors with weeds. We plan to test these genotypes in on-station tests in 2005.

Acreage devoted to cowpea cover crops continues to grow. The increased interest in reducing the environmental impact of agriculture (TMDLs, dust abatement) is spurring increased use of cover crops for both crop and environmental management tools.


Cowpea cover crop use is growing and currently exceeds several thousand acres. The improved genotypes should increase the acreage of use and encourage new practices that include using cover crops instead of pesticides.

The increased demand for cover crops must be met with increased seed production. Cowpea cover crop seed is now being produced in the Southeast. Shipping charges to Western areas add significant cost to cowpea cover crop seed. The new genotypes created by this project will allow the production of cowpea cover crop seed in the low-elevation desert. This could create new economic opportunities for growers and seeds purveyors of these largely depressed farming communities.

We have completed cost studies that compare the economic return of vegetable production systems with or without summer cover crops (Ogbuchiekwe et al. 2004). Yield and net return were greatest when cantaloupe and lettuce were planted after the incorporation of a cowpea cover crop. Profits depended upon whether lettuce and cantaloupe were grown organically and the price paid growers for their crops. The new pest-resistant cover crop varieties developed by our current research should increase profitability by decreasing reliance on synthetic pesticides and fertilizers.


Jeff Ehlers
Associate Research Specialist
University of California
Department of Botany and Plant Sciences
Riverside, CA 92521-0124
Office Phone: 9097874332
Philip Roberts
Professor and Nematologist
University of California
Department of Nematology
Riverside, CA 92521
Office Phone: 9097877291