In Malawi, maize (Zea mays L.) is the staple food crop that dominates the cropping systems. Average yields in the period, 2005 to 2012, ranged between 0.81 and 2.65 t ha^-1 (GoM, 2012). Poor soil  fertility  is  one  of  the  major reasons for low yields (Kumwenda et al., 1997; ICRISAT/ MAI, 2000; Blackie and Mann, 2005; MoAFS, 2012). Other constraints include recurrent droughts, poor management, foliar diseases, stalk  borers,  termites  and  the parasitic weeds species Striga (MoAFS, 2012; Kabambe et al., 2008).

For some parts of the country, especially in high rainfall upland areas, aluminium toxicity due to low pH is a major constraint.  Chilimba and Saka (1998) reported that over 40% of the  soils in Malawi belong to soil orders Oxisols and Ultisols (US Taxonomy), which are highly weathered and strongly acid with a pH (in water) of < 5.5. Many reports confirm occurrence pH levels below this critical minimum level (Maida, 1985; Chilimba and Nkosi, 2014; Snapp, 1998; Kabambe et al., 2013). There is a current recommendation to apply 2 ha^-1 agricultural limes as pre-plant broadcast incorporation for soils with pH_w (H₂0) < 5.5 (Chilimba, 2005; Chilimba and Nkosi, 2014; Chilimba et al., 2013).

Aluminium toxicity is considered as the most important plant-growth limiting factors in many acid soils, particularly those with pH below 5.0 to 5.5. Tully et al. (2015) reported that up to 24% of Sub-Saharan Africa is under Aluminium toxicity while 8.49 are subject to high P fixation. When acid soil are limed, exchangeable Al³⁺ and hydrohyaluminium cations such as Al(OH)₂⁺ are converted to insoluble Al(OH)₃; resulting   in removal of the Al₃⁺ from cation exchange competition (Marschner, 2011; McCauley et al., 2009). Availability of P and K are improved with liming. Excess aluminium interferes with cell division in plant roots, fixes phosphorus into less available forms, decreases root respiration, and interferes with uptake, transport, and use of Ca, Mg and P (Fairhurst, 2012; Goulding, 2016). In Cameroon, Yamoah et al. (1996) reported that application of lime and green manure improved stand count, root and stem weights and yields of maize, bean and potato in an acid soil and that high P application was unnecessary when lime was applied. Onwonga et al. (2008) reported that lime application increased soil pH within two months of application and that organic amendments had similar effects in Kenya. The et al. (2012) reported that continuous cultivation of maize resulted in a soil acidity in Cameroon, falling from 4.89 to 4.38 and drops in soil availability of Ca and K. There is potential to reduce the lime requirements and increase yield through microdosing of lime application (The et al., 2012; Kisinyo et al., 2015; One Acre Fund, 2016), including application by point placement or dollop method (One Acre Fund, 2016).  Malawi has large deposits of limestone and dolomite. An explorative study was therefore conducted to compare the effectiveness of applying reduced rates of agricultural lime on maize growth and yield at Bunda Campus of Lilongwe University of Agriculture and Natural Resources, central Malawi in 2015-2016 season.

Experimental design, treatment descriptions and general crop and management

A field trial was conducted utilizing a randomized complete block  to evaluate seven treatments and four replicates (Table 1); which were designed to compare the recommended application of lime at 2.0 t ha^-1 to reduce rates applied as point placements (dollop method), hereafter referred to as microdosing.

The maize variety DK8053, with a maturity period of 140 days, was used in the study. The maize was planted on ridges spaced at 75 cm apart with stations at 60 cm, 2 seeds per station, giving a target plant population of 44,444 plants/ha. Gross plot sizes were 5 ridges x 0.75 cm x 5.3 m long (19.875 m²), while the net plot comprised three middle ridges excluding end of ridge plants (10.575). The compound fertilizer 23:21:0+4S was applied at 100 kg/ha for basal dressing, one week after planting. The remaining 46 kg/ha N was applied as top dressing at 4 weeks after planting in the form of urea. Ridge preparation, lime application and planting were done in mid-December, 2015. The fields were kept free of weeds by weeding with a hand-hoe. The field preparations and fertilizer, lime application and planting were done in mid-December 2015. Figure 1 shows daily cumulative rainfall (mm) recorded at the site. In general, the season was considered good for maize production.

Study site

A field experiment was conducted at Lilongwe University of Agriculture and Natural Resources – Bunda campus, at the student Crop and Soil Sciences research farm in the 2015/2016 cropping season (November 2015 to April 2016). The farm is 1159 masl at 14°35' S and 33°50' E with an annual rainfall of 1031 mm. The soil types vary from clay loam to sandy loam (Jones and Kanyama, 1977).  The specific site for the study had mean pHw of 5.1 (SD 0.22, n=10).

Data recording and analysis

The data recorded in this report were maize plant height, grain yield and stover weight. Grain yield and stover were determined by harvesting all plants from net plot area. For stover, 6 plants were randomly sampled from the net plot area and oven dried to determine percent moisture content of stover, which was used to adjust field weight to dry weight.  Grain moisture content was determined using grain moisture meter and yield adjusted to 12.5% storage moisture. Plant height (from base to center of leaf whorl or flag leaf) was determined from an average of 6 plants randomly selected from the net plot. Data were subjected to analysis of variance procedure using Genstat 18th Edition (VSN International, Memel, Hemstead, UK) and mean separation by LSD at ≤ 0.05 probability.

Determination of gross margins (GM), value cost ratio (VCR) and benefit-cost-ratio (BCR) 

Gross margins (GM) were determined as gross income from sales minus production costs that vary. The costs that vary mainly consist of labor (land preparation, ridging, planting, fertilizer application, harvesting shelling, cleaning and packaging).  These are basic components described by several authors (Dzanja, 2008; Ngulube et al., 2001; Takane, 2008). Dzanja (2008) estimated the total labor requirement to be 139 mandays for maize and these were used in the calculations. However, Takane (2008) estimated labor requirement for maize to be 176 mandays. Another factor considered as a variable cost was packaging, which comprised a new sack for each 50 kg of harvest. Tables 2 and 3 show the total costs for each of the five rates of lime application. The costs of inputs and labor use were those of the October 2019 in Malawi. At this time the exchange rate of the Malawi Kwacha to US $ was 1: 745. The labor was valued  based  on  the  minimum  wage  rate  of US$ 1.29 for time (October 2019). Calculations were made for two output price scenarios of US$0.201 kg^-1 and US$ 0.268 kg^-1 as realistic average prices in the August-October 2019 period. The evaluation also involved calculation of Value Cost Ratio (VCR) and Benefit Cost Ratio (BCR).  The VCR was calculated as incremental value of maize divided by treatment/coast associated with implementing the treatment  (Kihara  et  al.,  2015).  Specifically  the costs were those associated with lime purchase and application (Table 2). Benefit–cost ratio was determined by dividing gross income of a treatment by total variable cost. There was no discounting of costs, as the comparison is for one-season data rather than a long- duration project (Ajayi, 2009). The economic calculations are presented in US $ to facilitate ‘universal” communication.

Agronomic results

There were no significant treatment differences in plant heights at 2, 4 and 6 weeks after planting. Significant differences were observed at 8 and 10 weeks after planting (Table 3). In both cases 500 kg ha^-1 microdose treatment gave tallest plants, which were significantly higher than no-lime control and 100 kg ha^-1 at 8 weeks only but similar to the rest of the treatments including commended 2 t ha^-1 treatment. For grain yield (Table 4), the 500 kg ha^-1 lime microdose gave the highest yield which was also significantly higher than all other treatments treatment. There were no significant treatment differences on stover yield.  

The increased growth and yield due to lime application are in agreement with local literature (Kabambe et al., 2013; Mtonga, 2013) and regional literature (Kisiyo, 2015; The, 2012) and is agreement with recommendation to apply lime for soils with pHw <5.5 (Chilimba and Nkosi, 2014). It of interest to note the significant yield increase to the lower lime rates as applied by point placement. Rates of 100-250 kg ha^-1 are substantially lower and would be relatively feasible for smallholder farmers compared to 2 t ha^-1 currently recommended. The et al. (2012) reported that application of lime at the rate of 250 kg ha^-1 increased maize grain yield in some varieties of maize, but not others. However, agricultural  lime  applied at high rates to a large volume of soil has advantage of residual effects which may last for 2-3 seasons. In Malawi, Kabambe et al. (2012) confirmed residual effects of 2 t ha^-1 after one season with maize and common beans. There are different ways which can be employed in the integrated management of acid soil in order to reduce amount and cost of lime while improving crop productivity. These include use of tolerant varieties (The et al., 2012), amendments with organic inputs including wood ash, compost, and legume rotations (Kabambe et al., 2012; Onyonga et al., 2008; Yamoah, 1996; Goulding, 2016) and agroforestry (De Pauw, 1994). Farmers may also opt for crop species with greater acid resistance (De Pauw, 1994; Goulding, 1989). The yields obtained in this study are higher than average smallholder farmer yields (MoAFS, 2012).  On farm yields from research plots with fertilizer application normally fall in the range 2.5-3.5 t ha^-1 (Ngwira et al., 2012; Kabambe et al., 2013, 2018). The results obtained are explorative in nature and mainly serve to provide a basis for wider on farm verification on the both the rates and method of application before recommendations to farmers

Gross margins, value cost ration and benefit-cost ratio

The results on Gross Margins (GM) (Table 5) show that all treatments  had  positive  GM,  the  highest  coming  at 500 kg ha^-1 lime and lowest with the 2 t ha^-1 lime treatment using both maize product prices. The results are in line with yield results. In a similar GM analysis Kabambe et al. (2018) also reported that GM of maize was positive for yields and prices of 4,000 kg ha^-1 and US$0.27 kg^-1 respectively. Recently, Ngwira et al. (2020) reported GMs of US$1000-1800 from on-farm maize trails in Malawi.

The results consistently demonstrate that 500 kg ha^-1 lime as micro dose resulted in better maize growth and yield compared to no lime treatment and 100 kg ha^-1 rate. However, both growth and grain yields were similar with the other treatments. The results suggest potential to reduce application rates and increase yield of maize through microdosing of agricultural lime and provide a solid basis for wider evaluation of the concepts for subsequent rolling out to famers. Profitability was clearly related to yield and the associated costs of production. These results are from an explorative study. As the concept of microdosing, as defined here, is the reduction of application rates and the precise point placement. It is recommended that it should be evaluated along with other integrated soil management options that are proven to ameliorate soil acidity and improve yields.

The authors have not declared any conflict of interests.