Yield performance of cowpea as influenced by insecticide types and their combinations in the dry savannas of Nigeria

Insecticide spray offers the most effective control of insect pests where resistant varieties are not available as is the case with the legume pod borer and pod sucking bug complex. Field experiments were conducted in the northern Guinea and Sudan savannas (Samaru and Minjibir, respectively) zones of Nigeria during 2005 to 2007 rainy seasons to evaluate the efficacy of common and new insecticides in the Nigeria markets for field control of insect pests of cowpea. Post spray insect counts showed that the new insecticide combinations gave control of thrips, Megalurothrips sjostedti (Order: Thysanoptera, Family: Thripidae) statistically similar to Monocrophos. Maruca vitrata (Order: Lepidoptera, Family: Crambidae) population densities were significantly reduced by the application of the insecticide regime of Imidacloprid (first spray), Cypermethrin (second and third) which however was not different from Cypermethrin-Dimethoate. The no spray plots gave significantly higher fodder yield in Samaru. The highest mean grain yields at Minjibir were obtained from plots sprayed with Imidacloprid + Cypermethrin + Cypermethrin (1391 kg/ha) while highest mean grain yield (924 kg/ha) in Samaru was from Monocrophos applied three times. Considering safety issues and environmental concerns, Imidacloprid (first spray) and Cypermetrin (second and third sprays) is recommended in a three insecticide spray recommendation for cowpea cultivation, especially in the Sudan Savanna of Nigeria and three sprays of Cypermethrin-Dimethoate in the northern Guinea Savanna.


INTRODUCTION
Cowpea [Vigna unguiculata (L.) Walp.(Family: Fabacea)] is an important grain legume in West Africa and in many parts of the tropics (Singh et al., 2002).It provides an inexpensive source of protein and minerals for the urban and rural people of the region and is an important crop in the cereal-legume cropping systems, especially of the drought prone dry savannas of West Africa.However, the on-farm yields of cowpeas are very low in this region.The *Corresponding author.E-mail: h.ajeigbe@cgiar.org.low on-farm grain yields averaging 25 to 300 kg/ha in West Africa (Rachie, 1985;Mortimore et al., 1997), and 150 to 400 kg/ha in Uganda (Sabiti et al., 1994) compared to 1500 to 3000 kg/ha in research Institute (Ajeigbe et al., 2005;Singh and Ajeigbe, 2002) are mainly due to a number of insect pests.Consequently, significant increases in yield are attained as a result of spraying.For example, Edema and Adipala (1996) attributed approximately 70% yield reduction to insect pest alone.
In Nigeria, Algahli (1992) realized yield increase of 50 to 200%, while Kyamanywa (1996) working in Kenya got a 15-fold increase in cowpea grain yield due to insecticide sprays.Insect pests cause maximum damage to cowpea from seedling stage to grain storage (Ajeigbe and Singh, 2006).The most damaging of all insect pests are those that occur during flowering and podding, and these include flower thrips, M. sjostedti (Order: Thysanoptera, Family: Thripidae), the legume pod borer, M. vitrata (Order: Lepidoptera, Family: Crambidae) and a complex of pod and seed suckers, of which Clavigralla tomentosicollis (Order: Hemiptera, Family: Coreidae) is the dominant species in Africa (Jackai and Adalla, 1997).According to Jackai et al. (1985), it is not possible to grow cowpea commercially without the use of insecticide sprays.This is due to the lack of acceptable level of resistance to pod borers and pod sucking bugs in the available germplasm lines including the crossable wide relatives (Fatokun et al., 1997).Therefore, the improved varieties still need 2 to 3 sprays of insecticides (Ajeigbe and Singh, 2006;Asante et al., 2001).Faced with the insect pest problems, some farmers have resorted to indiscriminate use of insecticides to reduce pest damages, sometimes applying as many as 5 to 7 times instead of the 2 to 3 times as recommended (Ajeigbe and Singh, 2006).Several insecticides are available in the market and some are extremely dangerous and may pose hazardous consequences to the user and the environment.Also, the use of insecticides must be minimized due to high cost and harmful effects on humans and the environment (Giliomee, 1997).
This study was undertaken to compare the effectiveness of common and new insecticides in the Nigerian market for cowpea insect pest management.The aim is to determine which insecticide type, combination and/or application sequence will provide best control of cowpea flower thrips, M. sjostedti and legume pod borer, M. vitrata and considering safety issues.Thus, a relatively safe insecticide could be recommended for high cowpea grain production.

Location and land preparation
Trials were conducted during 2005, 2006 and 2007 crop seasons at the International Institute of Tropical Agriculture (IITA) Research farm, Minjibir (Sudan savanna) near Kano (12° 08" N, 8° 40" E ) and the Institute of Agricultural Research (IAR), farm in Samaru (Northern Guinea savanna), near Zaria (11° 11" N, 7° 38" E) in northern Nigeria.The plots were disc harrowed twice, a basal dose of 100 kg NPK 15:15:15/ha was broadcasted and ridged prior to planting.Planting of the trial in each year was done in July, in Minjibir and August in Samaru following a good rain the previous day.A randomized complete block design with four replications was used.Each plot consisted of 4 rows, 5 m long, with an inter-row spacing of 0.75 m and hill-to-hill distance of 0.2 m within the rows.A blank row was left between each plot.To ensure proper stand establishment, four seeds were planted at each hill and thinned to two plants/hill a week after germination.The plots were kept weedfree using hand hoe at 3, 6 and 9 weeks after planting.During the spray Ajeigbe et al. 5931 operations, a polythene sheet was held between the plot boundaries to stop the drift of insecticide from plot to plot.

Varieties and insecticide treatments
The cowpea variety used was an improved medium maturing dual purpose variety (IT90K-277-2).This variety is very popular in the region and has also been released in several countries (Singh et al., 2006).Sprays were applied using a hand sprayer at the preflowering, 50% flowering (10 to 14 days after first spray) and at podding (10 to 14 days after second spray) stages of the crop.The treatments included 9 insecticide treatments and a control as follows: 1 Chlorpyrifos is an organophosphate insecticide.It is used to kill insect pests by disrupting their nervous system.Chlorpyrifos has an advantage over other products in that it is effective against a wide range of plant eating insect pests.It is classified by the World Health Organisation as a Class II, 'moderately hazardous' pesticide.It is relatively non-persistent in the environment.Oral LD50 for rats is 95 to 270 mg/kg.
Monocrotophos is an organophosphate non systemic insecticide with contact and stomach action.It is acutely toxic to birds and for that reason has been banned in the U.S. and many other countries.It is highly hazardous, and has been responsible for deaths resulting from accidental or intentional exposure.It is highly toxic orally, as well as by inhalation or absorption through the skin.Acute oral LD50 is 23 mg/kg.However, it does not have long persistence in the environment.
Endosulfan is a neurotoxic organochlorine insecticide.It is highly toxic and an endocrine disruptor and it is banned in the European Union, and several other countries; however it is still used extensively in many countries.It is acutely neurotoxin to both insects and mammals.LD50 is 30 to 80 mg/kg weight of rat.
Dimethoate is an organophosphate insecticide used to kill insects on contact.Dimethoate is moderately toxic and non-persistent with oral LD50 ranging from 150 to 400 mg/kg.It is a safer alternative to more persistent Monocrotophos.It was used in the trial as a commercial formulation involving Cypermetrin + dimethoate EC at 30 g + 250 g ai ha -1 .

Data collection
Data were collected during rainfall; this is to assist any relevant interpretation of results.Field data was collected on the two middle rows of each plot while the two outer rows served as border rows.At onset of the spray treatment, a day before each of the insecticide spray, five flowers were randomly picked per plot and placed in bottled vial filled with ethylene.They were later taken to the laboratory for insect count; the flowers were opened and observed under the microscope (Asante et al., 2001).The number of legume pod borer, thrips and other unidentified larvae or adult insect in the flower were counted and recorded as Maruca count and thrips count, respectively.At maturity, cowpea pods were harvested from the net plots and sun dried to a constant weight.The cowpea pods were threshed and the grain was weighed.The stover left after pod harvest was cut and sun dried to a constant weight and the weight recorded.Recorded weights were extrapolated to yield per ha.The data collected were analyzed using GENSTAT Discovery 3 program (2007).Analysis of variance (ANOVA) was carried out, and least significant difference (LSD) at 5% were calculated where the ANOVA showed significant differences.

Rainfall
The monthly total rainfalls (mm) for the three years in the locations are given in Figure 1.Rainfall data during the growing season showed that higher rainfalls were received at Samaru compared to Minjibir.The month of August generally had the highest rain in both Minjibir and Samaru.Rains generally start in April in the northern Guinea Savanna (Samaru), peaks in August and ends in September with a few showers in October while it starts in May in the Sudan Savanna (Minjibir), peaks in August and ends in September.

Insect infestation
The mean number of insect count on cowpea flower in the different environments is presented in Table 1.Significant differences were found among the different environments for number of insect found in cowpea flowers twelve days after insecticide spraying.M. vitrata infestation rates (8, 6, and 4, first, second and third count, respectively) were significantly higher in Minjibir in 2007 than Minjibir 2006 and Samaru 2007.However, thrips infestation rates were significantly higher in Samaru in the first count, compared to other locations and counts.The effect of the different insecticide treatments on the number of insect in the flowers of cowpea in Sudan and   Chlorpyrifos.The control (no spray) had significantly higher number of thrips in its flower compared to other treatments before the third spray.

Grain yields
The mean grain yield of cowpea as affected by year, location and insecticide are given in produced the highest grain yields in 2005, and these were significantly higher than yields produced by four of the remaining 8 treatments.In 2006, nospray (245) and Courage*3 (726) produced significantly lower yields than the other treatments, while there were no significant difference among them.In the three years mean, Mono*3 (924) and Mono + Cyper + Cyper (812) produced the highest grain yields which were significantly higher than yields obtained from 7 and 4 other treatments, respectively.

Fodder yields
The effect of different insecticide spray and their combinations on cowpea fodder yield is given in Table 4.In Minjibir, there was no significant difference among the treatments in 2005 (mean of 1382 kg/ha) and 2006 (mean of 1476).However, significant differences were observed in 2007.Fodder production among the different insecticide treatments in 2007 ranged from 1064 kg/ha in Courage + Cyper + Cyper to 3940 kg/ha in Courage*3.Courage*3 produced significantly higher mean fodder than other treatments across the years.The nospray treatment (1831 kg/ha) produced significantly higher fodder than three insecticide treatments while there was no significant difference among the other treatments.In Samaru, there was no significant difference in fodder production among the various insecticide treatments in 2005 (mean of 601 kg/ha), while significant differences were observed in 2006 (mean of 1022), in 2007 (mean of 514) as well as across the years mean (712 kg/ha).Mean fodder production across the three years among the different insecticide treatments ranged from 525 kg/ha in Cyper*3 to 1256 kg/ha in the nospray treatment.The nospray treatment produced significantly higher fodder than the insecticide treatments in 2006, 2007 and across year mean.

Relationships between insect infestation and yields
The correlation coefficients among the various variables measured are given in Table 5. Maruca infestation had a significant negative correlation with grain yield (-0.41), threshing (%) (-0.55) and harvest index (-0.72).The correlation between thrips infestation and grain yield (-0.34), threshing (%) (-0.39) and harvest index (-0.29)were also negative.However, the correlations among the different insect infestations were low: between Maruca and thrips infestation (0.09), Maruca and other flower insects (0.04), while the correlation between thrips and other flower insects was -0.17.A positive correlation (0.39) was found between Maruca infestation and fodder yield.

DISCUSSION
As expected, the average total rainfall and duration were higher in Samaru which is in the NGS than Minjibir in the Sudan savanna.Planting was done later in Samaru so that the cowpea matures at the end of the rains.Apart from the total rainfall received, its duration and prevailing temperature has significant effect on insect abundance.Muthoni et al. (2008) noted higher incidence of flower thrips on legumes during shorter rainy season than longer ones, especially when temperature was higher during the short rainy seasons.Higher and active population of M. sjotsdti can be found at temperature as low as 15°C and as high as 30°C.However, the highest thrips populations were found when the mean temperature ranged from 27 to 29°C (Alghali, 1992).The pre-spray flower insect infestations were fairly uniform as shown by the lack of significant differences in the insect count before the first insecticide sprays.Maruca infestation appeared to be higher in Minjibir than Samaru, though Samaru received higher rainfall than Minjibir.This was due to the difference in planting dates, while planting was done in July in Minjibir; it was done in August in Samaru and the cowpeas flower towards the end of the rains after the peak of Maruca population in the area.While the Maruca and thrips counts did not show significant difference among the insecticide treatments in Samaru, the cowpea grains produced showed significant differences, implying that the control of other insect pest (e.g, pod sucking bugs) are important in this location.This confirm the observation of Amatobi (1994) who reported higher incidence of pod sucking bugs in Samaru compared to Minjibir.It is generally known that the insect pest complex on cowpea is higher in the more humid areas compared to the drier areas as evident in this trial (Amatobi, 1995;Tarawali et al., 1996).
Significant differences that were observed for grain yield among the years and between the two locations did not follow the pattern of flower insect count revealing the importance of other factors in yield reduction.These other factor will include the foliage pests, pod sucking bugs and possibly rainfall distribution in these areas.Asante et al. (2001) noted that two sprays of Cypermetrin + Dimetheoate increased grain yield from 225 to 900 kg/ha for cowpea planted in July and August in Kano, they also found out that insecticide application was more profitable for cowpea planted in late July than those planted in June or early July.Integrated insect control requires that insect population build-up should be monitored, and insecticide should only be used when the population attained action threshold levels.The thrips populations in this study exceeded the economic threshold of seven thrips per inflorescence (Anonymous, 1996) only before the first spray, and recommended three thrips per flower as reported by Salifu (1992) and Schipps et al. (2000).This implies high level of controls of thrips by most of the insecticides used, though there are indications of less control by tricel.Significant differences were observed among the different insecticide chemicals for grain yield even when there was no significant difference among them for flower insect count, signifying the importance of the foliage flowers.Insecticide application is the most popular and widely practiced control measures against cowpea insect pests, regardless of the scale of farming operations (Amatobi, 1995).Many systemic and contact insecticides and their combinations have been tested for the control of major pests, such as aphids, Maruca, thrips and pod sucking bugs (Jackai et al., 1985;Amatobi, 1994;Jackai and Adalla, 1997).Because of environmental concern and safety issues in the use of insecticides by small scale farmers in tropical Africa, only the least toxic insecticides should be recommended and as such, integrated pest control should be used.Kurungi et al. (1999) noted that cowpea management practices that combine early planting close spacing and minimum insecticide application was most effective in reducing pest infestation.Several authors Ajeigbe and Singh (2006), Asante et al. (2001) and Kamara et al. (2010) noted that three strategic insecticides sprays are effective in the control of insect pest of cowpea.The present study compares cowpea productivities under different insecticide, using improved early maturing varieties under close spacing with three strategic insecticide applications.Integrated pest management, using a combination of resistant varieties planting dates and less toxic insecticides, should be advocated.The Maruca counts in flowers reveal that Monocrotophos may be less effective in controlling this insect species compared to Cypermetrin and Cypermethrin-Dimethoate; however, the efficiency of the different insecticides in the control of pod sucking bugs was not investigated, and damage caused by pod sucking bugs may account for the differences in yield between the different insecticides (Amatobi, 1994).
Three insecticides sprays in this study is comparable to the well timed minimum application of two sprays of Sherpa plus (Cypermethrin-Dimethoate) recommended by Amatobi (1994), Ajeigbe and Singh (2006) and Javaid et al. (2005) for economic production of grains on short season cowpea varieties (60 to 70 days), and three sprays for medium duration varieties (Ajeigbe and Singh, 2006;Asante et al., 2001).For dual purpose cowpea variety (IT90K-277-2), three sprays is worthwhile, this justifies the additional grains that could be obtained in addition to the fodder which is of immense value in the traditional crop-livestock systems of farmers in northern Nigeria.Cowpea farmers routinely apply insecticide as both preventive and curative measures against insect pests.Some of them also apply insecticide at higher rate and frequencies (Hirose, 1990).Thrips have been found to be key pest in areas where pesticide inputs are high.This suggest that misuse of insecticide have favored development of resistance to insecticides, the mechanisms of which is however unknown (Talekar, 1991).Similarly, Ekesi (1999) reported that M. vitrata field populations at Samaru were found to be resistant to Cypermetrin, Dimetheoate and Endosulfan.The resistance ratio ranged from 15 to 92 fold.
Less toxic chemicals like Cypermetrin are as effective as, or more effective than Monocrotophos in the control of some insect pests (pod borer) of cowpea.Since Imidacloprid provide effective control of aphids and thrips in addition to being less toxic to the user as well as environment, it can be used as the first insecticide spray on cowpea.Considering safety issues and environmental concerns, Imidacloprid (first spray), Cypermetrin (second and third sprays) is recommended in a three insecticide spray recommendation for cowpea cultivation, especially in the Sudan Savanna of Nigeria and three sprays of Cypermethrin-Dimethoate in the northern Guinea Savanna.Well timed and monitored insecticide application on cowpea reduce 50% of cost of insect pest control and have unquantifiable benefit to the environment when compared to calendar spraying (Afun et al., 1991).Some of the insecticides reduced cowpea fodder yields in both locations.However, the reduction in fodder yields were compensated for by the higher grain yields obtained in these treatments, implying that total biomass production may be similar.
A higher and significant correlation between Maruca infestation grain yield compared to thrips and other flower insect infestation implies that Maruca was a more important insect pest of cowpea in this region.Recent advances in transgenic cowpea have shown promising in their resistance to Maruca (Higgins et al., 2010), the use of which would further reduce the need for insecticide protection significantly.However, if the transgenic varieties have no resistance to thrips and pod sucking bugs, insecticide use may not be completely eliminated in cowpea production.New and less toxic insecticides, and their combinations, provide effective control of thrips (M.sjostedti) and pod borer (M.vitrata) on cowpea in the dry savannas of Nigeria.Considering safety issues and environmental concerns, Imidacloprid (first spray), Cypermetrin (second and third sprays) is recommended in a three insecticide spray recommendation for cowpea cultivation, especially in the Sudan Savanna of Nigeria, while three sprays of Cypermethrin-Dimethoate is recommended for the northern Guinea Savanna.Maruca pod borer (M.vitrata) was a more important insect pest of cowpea in this region than thrips (M.sjostedti).However, provision of resistance to Maruca through a welcome development, may not eliminate the needs for spray in cowpea except if linked or backed up with concurrent provision of appreciable level of resistance to thrips.

Table 1 .
Effect of environment on number of insect in cowpea flowers (insect/5 flowers).

Table 2 .
Effect of different insecticides on number of insect in cowpea flowers (insect/5 flowers).
various plots during the first insect count which was done before the first spray.Insect count in cowpea flower in Minjibir, before the second insecticide spray showed significantly (P < 0.05) lower Maruca under plot sprayed with Cypermethrin and Cypermethrin-Dimethoate (2 and 3, respectively) while flower thrips count were signifycantly lower under the plots sprayed with Monocrotophos, Cypermethrin (0) Imidacloprid and Cypermethrin-Dimethoate (1).There was no significant difference among the treatments in insect infestation in the third flower count before the third spray.Similar observations were made in Samaru in the northern Guinea savanna zone.However, there was no significant difference among the various plots during the first and second insect counts.Significantly lower numbers of thrips were found in flowers of plots sprayed with Cypermethrin and

Table 3 .
Effect of different insecticide and combination on grain yield of cowpea in savannas of Nigeria

Table 4 .
Effect of different insecticide and combination on fodder yield of cowpea in savannas of Nigeria.

Table 5 .
Correlation coefficient of selected variable.