EFFECT OF PHOSPHORUS (P) RATES AND WEEDING FREQUENCY ON THE GROWTH AND GRAIN YIELD OF EXTRA EARLY COWPEA ( VIGNA UNGUICULATA IN THE FOREST-SAVANNA AGRO-ECOLOGICAL ZONE OF SOUTHWEST NIGERIA

: Field experiments were conducted at the Teaching and Research Farm of the Federal University of Agriculture, Abeokuta (7 o 20’N, 3 0 23’E) during the 2014 early and late cropping seasons to evaluate the effect of weeding frequency and phosphorus fertilizer application on the growth and grain yield of the early maturing cowpea variety ( Vigna unguiculata L. Walp). The experiment was laid out in a split-plot arrangement fitted into a Randomized Complete Block Design with three replications. The main plot consisted of three phosphorus fertilizer rates (0, 15 and 30 P 2 O 5 kg ha -1 ) while the subplot comprised five weeding frequencies (no weeding, weed-free, hoe weeding at 3 weeks after sowing (WAS), hoe weeding at 3 and 6 WAS and weeding at 3, 6 and 9 WAS). The results showed that plots treated with phosphorus fertilizer at 15 kg ha -1 produced the highest number of leaves and the tallest plant in the late season while the highest grain yield was recorded in the early trials. Weeding at 3, 6 and 9 WAS during the early season trial gave the highest grain yield compared with other weeding treatments. Unchecked weed infestation reduced yield by 53.10 % and 49.9 % in the early and late seasons respectively compared to the maximum obtained from weed-free plots. This study concluded that application of 15 kg P 2 O 5 ha -1 and weed removal at 3, 6 and 9 WAS were effective for effective weed control and optimum grain yield in cowpea production.


Introduction
Cowpea (Vigna unguiculata L. Walp) is one of the most important grain legumes in less developed countries of the tropics, particularly Asia and Africa (Mortimore et al., 1997;Van Ek et al., 1997). Cowpea has been reported to be grown on an estimated area of 12.3 million ha in Africa in 2014 with the bulk of production occurring on 10.6 million ha in West Africa, most especially Niger, Nigeria, Burkina Faso, Mali and Senegal (FAOSTAT, 2016). The world's estimated annual cowpea production is put at 5.4 million tonnes with Africa producing 5.2 million (Agro Nigeria, 2015). In 2010, Nigeria produced 2.2 million metric tonnes of cowpea production making it the largest producer and consumer of the crop with estimated yield put at 687 kg ha -1 (FMARD, 2011) compared to the mean of 450 kilograms per hectare obtained globally. Tarawali et al. (2002) have noted that cowpea has a significant role in contributing to food security, income generation and sustainable environment for millions of small-scale farmers who cultivate it in West Africa. The seed is the largest contributor to the overall protein intake of several rural and urban families (Agbogidi, 2010;Kyei-Boahen et al., 2017).
Cowpea fixes atmospheric nitrogen through symbiosis with nodule bacteria (Shiringani and Shimeles, 2011). The seed also contains bioactive antioxidants such as vitamin C, carotenoids and phenolic compounds which represent a crucial group of bioactive elements in foods that prevent the development of diseases such as atherosclerosis and cancer (Omae et al., 2011).
In spite of the great potential for cowpea in Southwest Nigeria, the yield of cowpea obtained by farmers is generally low due to high level of diseases, pest infestation (Adigun et al., 2014), lack of knowledge of good agricultural practices, use of low-yielding varieties coupled with low soil fertility and weed management problems (Adigun et al., 2014;Osipitan, 2017). Yield loss resulting from weed infestation in cowpea is often aggravated by the low level of soil nutrient particularly phosphorus. Weeds could cause yield losses ranging from 40% to 80% in cowpea (Tijani-Eniola, 2001;Li et al., 2004;Osipitan, 2017). Phosphorus plays a vital role in the growth and development of cowpea. Phosphorus deficiency in cowpea has also been reported to affect nodule formation, accumulation of N, seed formation as well as grain filling (Tang et al., 2001).
The timing of weed operation and availability of nutrient facilities in cowpea appear to be very vital in the determining the outcome of the competitive interaction between crop and weed as well as assisting the farmer to reduce the number of weeding on his/her farm, reduce the cost of production and maximize his/her profit. Das (2011) reported that the more abundant the soil nutrient, the less important the weed competition. However, in tropical soils, soils are low in nutrient and competition becomes critical. Moreover, applying fertilizer to improve crop yield will tend to provide maximum benefit unless weeds are properly managed.
Hand weeding is the most common agricultural practice among small-scale cowpea producers. The persistent use of manual weeding by cowpea farmers could be as a result of the high cost of herbicides which may not even be within their reach. Moreover, most farmers have not adequately timed their weed control with the use of fertilizer in cowpea. It should be noted that proper timing of weeding and fertilizer application can modify the weed flora and provide a platform for effective weed control.
Quite a number of research work done on the weed control in cowpea has not really addressed the interaction of P application and weeding frequency in cowpea. Hence, this research work is developed to investigate the growth and yield performance of the early maturing cowpea variety (IT97k-568-18) as influenced by the application of P fertilizer and weeding regime.

Materials and Methods
The experiment was conducted at the Teaching and Research Farm of the Federal University of Agriculture, Abeokuta (Latitude 7 0 15 I N, Longitude 3 0 25 I E) in Southwest Nigeria during the early (April-July) and late (September-November) cropping seasons of 2014. The plot was cropped with maize in the previous season. The soil was sandy loam in texture with (84.4 g kg-1, 89.4g kg-1) sand, (4.8 kg-1, 5.8g kg-1) clay and (6.8 kg-1, 4.8g kg-1) silt particles in the early and late seasons respectively. The soil pH was moderate ranging between 6.65 and 5.65. However, the soil was very low in organic carbon (1.00% and 0.82%) and thus low in organic matter content and total nitrogen (0.08% and 0.10). Exchangeable potassium (0.23 cmol kg-1 and 0.17cmol kg-1) and available phosphorus (Bray -1 P) of the soil were low (6.02ppm and 8.75ppm). Exchangeable magnesium was low (0.90 cmol kg -1 and 0.71cmolkg -1 ), however exchangeable calcium was high (8.31 cmol kg -1 and 5.12 c mol kg -1 ).The climate of the study area is of the sub-humid type with the total amount of rainfall of 466.9 and 246 mm in the early and late seasons of 2014 respectively. The mean temperature was 26.75 o C and 26.97 o C in the early and late seasons respectively.
The early maturing cowpea variety (IT97k-568-18) used was sourced from the International Institute of Tropical Agriculture Ibadan, Oyo State, Nigeria. The experiment was a split-plot fitted into a Randomized Complete Block Design (RCBD) with three replicates, given a total number of 45 experimental units. The treatment consisted of the main plot of phosphorus rates at three levels: 0, 15 and 30kg P 2 O 5 ha -1 and the subplot of five weeding frequencies (weedy check, weedfree, weeding once at 3 weeks after sowing (WAS), weeding twice at 3 and 6 WAS and weeding thrice at 3, 6 and 9 WAS). Weedy-check in this study implies leaving the plot weed-infested throughout the crop life cycle (that is no weeding was applied) while weed-free was achieved by weeding the plot every week.
The experimental site in each crop cycle was ploughed and harrowed at the two-week interval and the land was marked out into various plots each measuring 5 m x 4 m. Three seeds of the cowpea variety (Var. IT97K-568-18) treated with CIBAPLUS were dibbled into the soil at a depth of 3-5 cm on the 18 th of April 2014 for the early-season cowpea while the late-season crop was sown on the 5 th of September 2014. The seeds were sown at a spacing of 50 cm x 20 cm (Dugje et al., 2009) and were thinned to 2 plants per stand two weeks after sowing (WAS). The seeds were sown at inter-row spacing of 50 cm and intra-row spacing of 20 cm between and within ridges respectively, representing 200,000 plants ha -1 . Hoe weeding was carried out according to the treatment structure using a West African hand hoe (a local farm tool that is used for weeding, ridging and heaping). Application of Cypermethrin plus Dimethoate was done every ten days beginning from 14 days after planting to control insects using a CP3 knapsack sprayer with a green deflector nozzle at a pressure of 2.1kg/cm 2 . Also, Mancozeb 80WP (fungicide) was applied at 4 WAS to prevent fungal diseases. Both the insecticide and the fungicide were applied at the rate of 0.6 L/ha. Harvesting of the earlyseason cowpea started on the 21 st of June 2014 while harvesting of the late-season cowpea started on the 8 th of November 2014.
Data were collected on plant height (cm), number of leaves/plant, leaf area (cm²), number of days to 50% flowering, days to 50% podding, dry pod weight (kg/ha), number of seeds per pod, 100-grain weight of cowpea, grain yield of cowpea (kg ha -1 ) and weed dry matter production. The data collected were subjected to analysis of variance (ANOVA) and significant means were separated using Duncan's Multiple Range Test (DMRT).

Results and Discussion
Weed-crop competition can be influenced by the time of weed removal and soil fertility maintenance. Generally, weeds have higher nutrient and water use efficiency, therefore a realistic weed management procedure and enhancement of the soil nutrient status will go a long way in enhancing the completive ability of the crop against weeds. The most noticeable significance of weed competition in crops is a reduction in the economic yield of the affected crop (Alagbejo, 1987;Dadari, 2003). The results of this study showed that application of fertilizer levels and weeding frequencies on cowpea influenced growth, yield and yield components, pod weight and the number of seed per pod as well as weed biomass and flora composition.
Effect of P rates and weeding frequency on the growth and grain yield of extra early cowpea 51 Effect of phosphorus fertilizer application and weeding frequencies on the growth and development of cowpea Effects of phosphorus fertilizer rates and weeding frequencies on plant height and number of leaves of cowpea during the 2014 early and late cropping seasons are presented in Table 1. It was observed that application of phosphorus fertilizer had no significant effect on plant height at 3, 6 and 9 WAS during the early cropping season and at 3 WAS during the late cropping season (Table 1). However, plant height significantly differed from one another during the late season at 6 and 9 WAS, where the application of 15 and 30 kg P 2 O 5 ha -1 respectively gave similar and higher values compared with plots without phosphorus application. The effect of weeding frequencies was not significant on cowpea plant height in both cropping seasons. With respect to the cowpea number of leaves, a significant effect of phosphorus fertilizer application was observed only at 6 WAS during the late cropping season. Cowpea plants on plots treated with 15 and 30 kg P 2 O 5 ha -1 produced higher number of leaves than those plants on unfertilized plots. The effect of weeding frequencies was significant on the number of leaves at 6 and 9 WAS in the early cropping season and 9 WAS in the late cropping season. In both cases, all the weeding frequencies produced a significantly similar number of leaves compared with weedy check plots where reduced values were obtained. Interactions of phosphorus fertilizer rates and weeding frequencies were not significant on cowpea plant height and number of leaves at the various weeks of observation (Table 1). As presented in Table 2, cowpea responded to fertilizer application with respect to the leaf area at 6 WAS, but no significant effect was observed at 3 and 9 WAS. At 6 WAS cowpea plants on plots applied with 15 and 30 kg P 2 O 5 ha -1 had similar leaf area but significantly higher values than the control (0 kg P 2 O 5 ha -1 . Weeding frequencies did not show any significant effect on the leaf area except at 6 and 9 WAS in the late season. Maximum and comparable leaf areas were recorded on the weed-free plots and the plots weeded at 3, 6 and 9 WAS at 6 and 9 WAS in the late season ( Table 2). The increase in plant height and leaf production observed as a result of fertilizer application during the late season trial at 6 and 9 WAS could be attributed to the fact that the crop was able to utilize the residual nutrient in the soil. Furthermore, the higher number of leaves and leaf area could have allowed the greater reception of light which encouraged the photosynthetic process of the plants required for pod filling and improved yield. Similar findings have been reported by Muleba and Ezumah (1985) and Singh et al. (2011).  Fertilizer rates significantly affected the number of days to 50% flowering during the early and late cropping seasons. A delayed number of days to 50% flowering was observed in both seasons when phosphorus fertilizer was increased from 15 kg P 2 O 5 ha -1 to 30 kg P 2 O 5 ha -1 compared to no fertilizer treated plots (Table 3). However, weeding frequencies did not have a significant effect on days to 50% flowering and days to 50% podding in the early and late cropping seasons ( Table 3).

Effects of phosphorus fertilizer application and weeding frequencies on cowpea yield and yield components
Cowpea pod and grain yield were significantly affected by fertilizer rates in the early season but not in the late season (Table 4). Application of 15 kg P 2 O 5 ha -1 gave pod and grain yields in the early season which were very similar to the maximum values obtained from plots treated with 30 kg P 2 O 5 ha -1 (Table 4). However, there were no significant effects of phosphorus fertilizer on the number of seeds per pod and 100-grain weight in both early and late season trials. Weeding frequencies significantly affected cowpea pod yield in early and late seasons (Table  4). Weeding at 3, 6 and 9 WAS gave significantly higher pod yield but similar to the weed-free in the early season. In the late season, though weeding thrice ranked highest with respect to pod yield, it was not significantly different from the weed-free, weeding once and weeding twice (Table 4). In the same vein, weeding at 3, 6 and 9 WAS gave the maximum and comparable grain yield and comparable values with weed-free plots. There was no significant interaction between fertilizer rate and weeding frequency on pod yield, number of seeds/pods, 100-grain weight and grain yield. The highest grain yield and yield characters such as pod yield recorded by the application of 30 kg P ha -1 could be attributed to the fact that cowpea responded to the application of phosphorus fertilizer. Similar research findings have been obtained by Okeleye and Okelana (1997) and Majeed et al. (2001). Reduction in cowpea yield in the early season compared to that of the late season could be attributed to the rainfall pattern during the period, which was higher than the one in the late season. This is in agreement with the report of Coulibaly and Lowenberg-DeBoer (2002), who documented that cowpea performed well in agro-ecological zones where rainfall distribution ranges between 500 mm and 1,200 mm/year. Therefore, the higher values recorded for yield and yield components of cowpea in the late season than early season could be attributed to the fact that rainfall during the growing period (September to November) was adequate and optimum in the late season. However, in the early season, excessive rainfall aggravated the preponderance of pest and diseases. This study, hence, showed that timely weed removal is important in cowpea production. Adigun et al. (1991) observed that the period between 3 and 6 WAS is particularly critical for weed removal in the wet season due to vigorous weed growth and competition with the crops. It is also worth noting that weed-free plot recorded lower grain yield than the plants on the plots weeded at 3, 6 and 9 WAP in the early season. Flower abortion and destruction of the crop roots at the time of weeding could have accounted for this lower yield in the weed-free plot. In the late season, on the contrary, weeding regimes irrespective of the time of weed removal gave similar effects on the grain yield. Uncontrolled weed infestation throughout the crop life cycle in this study resulted in about 50% to 53% reduction in potential grain yield of cowpea. The significantly high percentage yield reduction could be attributed to the fact that the cowpea crop is sensitive to weed competition, especially in Nigerian savanna agroecological zones as indicated by Magani (1990). Moreover, the field was highly infested with broadleaf weeds such as Tridax procumbens, Phyllanthus amarus and these weeds are known to compete adversely with the cowpea crop. This finding was in conformity with the report of Li et al. (2004), who stated that yield losses ranging between 50% and 86% could occur due to unchecked weed growth throughout the life cycle in cowpea. It is, thus, become imperative that early weeding should be done to guarantee a higher yield in cowpea. Adigun et al. (2014) have observed that early weeding starting from 3 WAS is very crucial for cowpea production while the critical period of weed removal for optimum yield in cowpea is between 3 and 9 WAS in the forest-savannah transitional zone of Southwest Nigeria.
Effects of phosphorus fertilizer rates and weeding frequency on weed dry matter production In the early and late seasons, the fertilizer rates did not have a significant effect on weed biomass throughout the weeks of observation except at 6 WAS during the late-season trial (Table 5). In the late-season trial, fertilizer application significantly affected weed dry matter with the application of 15 kg P 2 O 5 ha -1 giving the highest weed dry matter which was similar to the value obtained with no fertilizer application. Application of 30 kg P 2 O 5 ha -1 gave the least dry matter production. Weed dry matter was significantly reduced by 77.6, 80.7 and 87.9% when weeding was carried out once at 3 WAS, twice at 3 and 6 WAS and thrice at 3, 6 and 9 WAS respectively. In the late season, on the other hand, weed dry matter was reduced by 81.4, 77.2 and 90.9 % in the plot weeded at 3 WAS, 3 and 6 WAS and 3, 6 and 9 WAS respectively (Table 5). Higher weed biomass was recorded during the late season trial at 6 WAS when the plots were treated with 15 kg P 2 O 5 ha -1 High weed relative density of Cyperus esculentus (12.29%), Dactylotenum aegyptium (13.65%), Digitaria horizontalis (21.84%) and Kyling abulbosa (8.19 %) observed on these plots at the early stage of crop growth at the second peak of rain in the ecology could have accounted for the increased weed biomass. These weed species were known to have higher competitive ability than field crops in the rainforest and savanna ecology (Adeyemi et al., 2008).

Conclusion
Application of phosphorus fertilizer at 15 kg P 2 O 5 ha -1 and weed removal at 3 and 6 weeks after planting in the late season are appropriate and adequate for cowpea production in the forest-savannah transitional zone of Southwest Nigeria. Hoe weeding at 3, 6, and 9 WAS significantly (p < 0.05) reduced weed biomass by 85% compared to control plots while application of 15 kg P 2 O 5 ha -1 increased grain yield by 34.8 and 30.7 % in early and late seasons, respectively. The use of phosphorus fertilizer and timely weed removal are, thus, a sine qua non in promoting weed control effectiveness in cowpea and improving grain yield.