African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 4, Num. 4, 1996, pp. 441-451

Reduction of P fertilizer requirement using lime and Mucuna on high P-sorption soils of NW Cameroon. *

C. YAMOAH**, M. NGUEGUIM^1 C. NGONG^1 and D.K.W DIAS^2

International institute of Tropical Agriculture, PMB 1520, Oyo Road, Ibadan, Nigeria
^1 Institute of Agronomic Research, P.O. Box 80 Barnends, North west Province, Cameroon
^2 Department of Agricultural Economics, University of Nebraska-Lincoln, USA

(Received 12 October, 1996; accepted 7 December, 1996)

*This article is contribution from the IITA-NCRE USAID-supported Project in Cameroon. "Center for Sustainable Agric. Systems, Dept. of Agronomy, 225 Keim Hall, Univ. of Nebraska-Lincoln, Lincoln NE 68583-0949, USA.

Code Number: CS96085
Sizes of Files:
    Text: 40K
    Graphics: No associated graphics files    

ABSTRACT

Soil acidity and high phosphorus (P) fixation are real problems in the Highlands of Central and Eastern Africa. Phosphatic fertilizers are imported and costly for the average farmer. We used lime and Mucuna green manure to reduce fertilizer P requirement of traditional food crops in NW Camcroon. Lime and phosphorus significantly improved stand count, root and stem weights as well as yields of maize (Zea mays L.), bean (Phaseolus spp.), and Irish potato (Solanum tuberosum L.) in three consecutive cropping seasons. Liming was most effective at low P rates and its effect on yield diminished with increasing P fertilization. Likewise, high P was unnecessary when lime was applied. Mucuna green manuring behaved similarly to liming and reduced maize P requirement by between 45 and 83%. Additionally, liming raised soil calcium (Ca), pH, effective cation exchange capacity (ECEC), and lowered exchangeable acidity (Al+H) as well as Al saturation. Consequently, maize and bean yields correlated positively (P<0.01) with pH, ECEC and exchangeable Ca but related inversely (P<0.01) with total acidity and Al saturation. Liming at 2 t ha^-1 was observed to be uneconomical at the current prices of lime and P fertilizer. Thus, our results suggest that farmers could use either a much lower rate of lime or Mucuna green manuring with less than 85 kg ha^-1 P fertilizer to sustain production on acid infertile soils in NW Cameroon.

Key Words: Acidity, Al saturation, liming, green manuring, Solanum tuberosum, Zea mays

RESUME

L'acidite du sol et la haute fixation du Phosphote (P) sont des veritables problemes des regions montagneuses de l'Afrique Centrale et de l'Est. Les engrais phosphoriques sont importes et sont tres couteux aux fermiers de la classe mayenne. Nous avous utilise des chaux et fumiers verts "Mucuna" pour reduire les exigences des engrais P des cultures traditionnelles au NW Cameroun. Les chaux et les engrais du phosphore ont cousiderablement ameliores la qualite des racines, les poids des tiges ainsi que les rendements du mais (Zea mays L.), l'haricot (Phaseolus spp.), et la pomme de terre (Solanum tuberosum) dans trois saisons des cultures consecutives. L'utilisation du chaux etait la plus effective a un faible niveau de taux et son effet sur le rendement a diminue avec l'augmentation d'engrais. De meme, la haute densite du phospore (P) n'etait pas necessaire apres l'utilisation des chaux. Les fumiers verts "Mucuna" ont remplace les chaux et reduit les besoins d'utilisation des phosphates pour le mays de 45 a 83%. En plus, l'utilisation des chaux a augmente la capacite effective d'echange sol calcium (Ca), pH, (ECEC) et a diminue la possibilite d'echange de l'acidite (Al + H) aussi bien que la saturation Al. Par cousequent, les rendements du mais et de l'haricot ont des correlations positives (P<0.01) avec pH, ECEC et echangeable Ca mais a l'inverse (P<0.01) avec l'acidite totale et la saturation Al. Les chauf a 2t ha^-1 etait constate etant moins economique par rapport aux prix actuels des chaux et des engrais P. Ainsi, d'apres nos resultats nous aimerions suggerer aux fermiers d'utiliser soit un taux moins eleve des chaux ou des engrais verts "Mucuna" avec moins de 85 kgs ha^-1 d'engrais P en vue de maintenir la production sur les sols infertiles d'acides au NW Cameroun.

Mots Cles: Acidite, saturation Al, l'utilisation des chaux, engrais verts, Solanum tuberosum, Zea mays

INTRODUCTION

Acid infertility and high P fixation are serious soil constraints in the Highlands of Central, East and West Africa (Vander Zaag et al., 1984; Nizeyimana and Bicki, 1988; Yamoah et al., 1990; Van Ranst et al., 1990). The highlands encompass the mountain agroecosystems of Northwest (NW) Cameroon, including the Bamenda, Bui, Bambui and Oku highlands. With the exception of potato (Solanum tuberosum L.), crop yields on farmers' fields are generally unsatisfactory in this environment, where temperatures are low and soil organic matter is relatively high (Van Ranst et al., 1989, 1990; Yamoah et al., 1995). Farmers in the region maintain yields through bush fallowing and use of household manure which is often inadequate. Earlier, on-farm studies and extension recommendations relied on P recommendation developed elsewhere in the country and did not consider the high P-fixation nature of the predominantly volcanic soils in the cool highlands. However, a recent fertility characterisation of soils in the highlands revealed that the soils have pH of less than 5.5 and standard P requirement greater than 1000 mg kg^-1 (Yamoah et al., 1995).

Several methods have been suggested to deal with acid infertility and P unavailability. These include liming, high P fertilizer application, use of organic manures and growing acid tolerant varieties (Hoyt and Turner, 1975; Kamprath and Foy, 1985; Hue et al., 1986; Hue, 1989; Hue and Amien, 1989). The rationale for using high P dose is to saturate P sorption sites and make subsequent P fertilizer available to crops, but this approach is not practical in Cameroon because P fertilizer is expensive compared to food prices. For instance, the price of triple superphosphate fertilizer in Cameroon in 1992 to 94 was $526.00 ton^-1 and it is not reasonable to expect subsistence farmers to apply P quantities beyond the minimum they can afford. Selection and breeding of acid-tolerant maize varieties have been initiated but it takes a while to develop truly promising lines. Liming is feasible because lime is available locally and is less costly than P fertilizer. In our view, the approach of using organic matter stands the chance of ready adoption because it is relatively easy to implement and less costly.

Studies elsewhere have shown that organic materials with low carbon to nitrogen ratios, such as alfalfa meal, Leucaena and cowpea, controlled soil acidity better than peat moss and guinea grass with high carbon to nitrogen ratio (Hoyt and Turner, 1975; Hue and Amien, 1989). The mechanism is that organic molecules and short-chain carboxylic acids such as oxalic and citric acids complex exchangeable and solution Al and and detoxify them. In addition, organic molecules enhance P availability to crops by binding exchangeable and hydroxy-Al, the key fixers of P in acid soils. We report results of field studies on the use of lime and Mucuna green manuring as amendments to reduce P fertilizer requirement for maize, bean and potato. Also, response functions and returns to P fertilisation under lime and Mucuna green manuring are compared.

MATERIALS AND METHODS

Agronomic field studies.

The first of the two field studies was conducted at the Bui Highlands of NW Cameroon, to examine the effects of liming on reduction of the crops' fertilizer P requirements. The area has an altitude of 2100 m and is also referred to as the High Lava Plateau. Mean annual temperature range is 12 to 24 C. Annual rainfall is 2000 mm and there is generally one long growing season from March to October. The soils are classified as Palehumult (Van Ranst et al., 1990), and are derived from trachytic parent material. Baseline soil properties are given in Table 1. The site was under Hyparrhenia, Sporobolus grasses, and fern bush fallow for about 10 years prior to initiation of the study.

TABLE 1. Baseline surface soil (0-20 cm) characteristics of the study area

---------------------------------------------------------------
pH (H2O)            5.34 (0.08)    pH (KCl)       4.65
Al (Cmol Kg^-1)     1.33 (0.52)    Al Sat. (%)   30.76 (18.5)*
ECEC (Cmol kg^-1)   5.13 (1.57)    Clay (%)      25.31 (5.89)
Extr P (ug g^-1)    3.95 (2.16)    Sand (%)       6.93 (2.28)

S.E. in Parenthesis
----------------------------------------------------------------

Lime and P treatments were combined in a factorial arrangement. There were five levels of P (0, 84, 209, 308, and 338 kg P ha^-l) as triple superphosphate. These were equivalent to the amounts needed to obtain no P in solution, 0.01, 0.03, 0.04, and 0.05 ppm of solution P based ,on sorption curves earlier developed for the site (Yamoah et al. 1995). Lime treatment was applied at two levels (0 or 2.0 tons CaCO3 ha^-l). The lime rate used was based on the study of (Kamprath, 1970) which showed that 1.65 tons ha^-1 of lime is needed to neutralise 1.0 cmol kg^-1 of Al in the exchange complex. Phosphorus and lime were manually broadcast once and hoed to a depth of about 20 cm at the onset of the study in 1992. The design was a randomised complete block (RCB) with four replications. Plot size was 5 x 6m. Nitrogen (120 kg ha^-1 as urea) was applied in two equal splits at planting and six weeks later. Potassium was applied as muriate of potash at the rates of 80 kg K2O ha^-1 in 1992, and at 50 kg K2O ha^-1 in 1993.

The study was conducted for four cropping seasons in 1992 and 1993, but yield data for the last season was dropped due to extensive damage by animals. Local varieties of Irish potato and bean were cropped twice and maize once a year in the following manner: the crops were planted on ridges with two bean rows spaced 15 cm within rows at the edges; one central maize row spaced 33 cm and two rows of potato at 50 cm next to the central maize row. After harvest of first season potato and bean, plots were prepared and replanted to the same crops while maize still remained in the field. Soils were sampled (0-15cm) toward the end of the 1993 growing season and analysed for exchangeable cations, extractable P, organic carbon, pH, and exchangeable acidity according to methods developed by Juo (1977) for tropical soils. Effects of lime and P treatments on crop growth were assessed by biomass sampling at 8 weeks after planting (WAP). Stand count at harvest was done on the two middle ridges covering an area of 15 m^2.

The second study was as a follow-up to the first and the objective was to determine the effect of Mucuna green manuring on reducing maize and bean fertilizer P requirement. Treatments were five P fertilizer rates (0, 50, 100, 150 and 200 kg P ha^-1) combined with and without Mucuna. All plots received 60 kg N as urea and 30 kg K2O ha^-1 as muriate of potash during each cropping season. Experimental design was RCB with four replications. This experiment was run for three cropping seasons including one season for establishing the green manure crop. Data were analysed statistically with MSTAT-C software.

Economic evaluation.

Yield response to phosporus fertilizer under lime and Mucuna green manuring were determined using simple log derivation of the Cobb-Douglas function; coefficients were estimated using the Ordinary Least Squares (SHAZAM) procedure. Models were also tested for auto-correlation. For each crop, two response functions were estimated; one for phosphorus without lime and the other for phosphorus with lime at a constant rate of 2 tons per hectare. The same analysis was done to estimate the yield response functions with and without Mucuna green manuring as a complement for P fertilizer. The only difference made here was the introduction of a dummy variable to represent the presence or absence of Mucuna prior to cropping the land. Prevailing market prices in 1992 to 1993 were used to calculate gross returns. The prices were: potato, $0.16 kg^-1; beans, $0.56 kg^-1 and maize, $0.24 kg^-1. Phosphorus fertilizer price was $526 ton^-1 and lime, $347 ton^-1.

RESULTS AND DISCUSSION

Crop yields.

Crop production on acid soils generally suffers from aluminum toxicity and Ca, and P deficiencies. In the Bui Highlands of NW Cameroon, liming significantly (P<0.05) improved the yields of maize, beans and potato (Table 2). As expected, the nature of response varied with crops and the amount of P fertilizer. For example, lime improved bean yield more than Irish potato and maize (Table 2). Averaged over P rates, the respective yield increases in the first season were 35.2% for bean and 26% for potato. But, during the second season, lime increased maize yield by 31% and bean yield by more than 100% compared to 3% for potato (Table 2). Bean, a legume, benefitted more from liming than Irish potato and maize. Yield increases due to liming were highest when no P was applied and diminished with increasing rates of P. Yield responses to lime at zero P were 65% for potato, 33% for maize and 72% for bean in season one of 1992. The respective yield increases for the highest P rate were about 6%, 30% and 30% (Table 2). The basic ingredient in most phosphatic fertilizers is monocalcium phosphate which contains Ca. Ordinary superphosphate has 18-21% Ca and triple superphosphate, 12-14%, implying that large dose of P fertilisation is synonymous to liming (Tisdale and Nelson, 1966). This may explain the lack of yield response to lime in the high P fertilised plots.

TABLE 2. Crop yield (kg ha^-1) as influenced by levels of P and Lime (CaCO3), 1992

---------------------------------------------------------------------------
P levels    Potato Lime levels    Maize Lime levels     Bean Lime levels
(kg ha^-1) --------------------  --------------------  --------------------
            0 tha^-1  2 tha^-1    0 tha^-1   2 tha^-1   0 tha^-1   2 tha^-1
---------------------------------------------------------------------------
                                  First Season

  0          987.5     1629.2      756.3      1006.8     612.5     1053.5 
 84         1637.4     2571.0      867.6      1167.7     625.3      868.7 
209         2316.5     3166.7      973.3      1197.5     711.0     1035.4 
308         2920.8     3208.4     1012.3      1348.4     820.0      828.0 
338         3312.5     3533.3     1072.1      1400.3     856.0     1116.5

S.E.         183.4      114.4      132.7

                                  Second Season

  0          907.0     1013.6        -           -       203.0      355.0 
 84          891.0     1035.9        -           -       182.7      479.1 
209          980.8      850.3        -           -       191.3      519.0 
308         1019.2     1030.3        -           -       260.1      461.7 
338         1021.8     1021.7        -           -       322.2      522.8

S.E.          91.6                   -                    52.7
--------------------------------------------------------------------------  

Similarly, crop yield response to P decreased in the presence of lime. To illustrate, the highest P rate increased bean yield by 40% without lime as opposed to a 6% increase when lime was present. This suggests that large quantities of P are unnecessary when the soil is limed. Furthermore, high P rates may cause yield reduction by limiting uptake of nutrients such as potassium and zinc (Fageria and Baligar, 1989). Potato responded significantly (P<0.01) to P fertilisation in the first season of 1992. In the second season of 1992, the highest P rate increased potato and bean yields by 11 and 58%, respectively, without lime, and <1 and 47% with lime. Crop yields for the first season of 1993 are reported in Table 3. Lime and phosphorus significantly (P<0.01) increased yield of potato in the first season. Bean yield increased with application of lime and phosphorus, but differences were not significant. Maize yield increased significantly with lime application (P<0.05), but not with P (Table 3). Lime, therefore, tended to have a longer residual effect than P. Furthermore, lime is available locally and costs about two thirds the price of fertilizer P. Thus, its use can be encouraged among farmers to reduce the dependence on expensive P fertilizers.

TABLE 3. Crop yields (kg ha^-1) as influenced by levels of P and Lime (CaC03), 1993

---------------------------------------------------------------------------
P levels     Potato Lime levels    Maize Lime levels    Bean Lime levels
(kg ha^-1) --------------------  --------------------  --------------------
            0 tha^-1  2 tha^-1    0 tha^-1   2 tha^-1   0 tha^-1   2 tha^-1
---------------------------------------------------------------------------
  0         1216.0    1742.0      2180.0     2331.2      479.6     418.6 
 84          187.6    2321.6      1521.6     2594.5      401.0     559.3 
209         2110.5    2914.5      2209.3     2799.4      423.3     553.4 
308         2597.9    2981.5      2184.9     3082.2      530.0     533.5 
338         2968.1    2941.3      2297.0     2505.9      643.7     565.1

S.E.              271.7                 273.7                  75.7
--------------------------------------------------------------------------

Crop growth.

Deleterious effects of soil acidity on crops include impairment of root development which is later manifested in poor growth and delayed maturity. Maize stem and root weights correlated significantly (P<0.01) with P only in the absence of lime (Table 4). The same trend appeared to hold true for bean stem and root weights (Table 4). Potato tuber weight did not correlate with P, either in the presence or absence of lime, indicating some degree of tolerance of potato to acid infertile soils (Table 4). Liming influenced the number of plants (stand count) at harvest. Differences in stand count were significant for first season bean in 1992, and maize in 1993 (Table 5). Low plant population at harvest was associated with yield reduction for almost all the crops. Poor plant growth in acid soils usually render them more suceptible to disease and pest infestation as noted by Burleigh et al. (1992).

TABLE 4. Correlation between plant growth parameters and applied P, 1993

--------------------------------------------
Growth Parameters    Correlation coefficient
(weight)            ------------------------    
                    +Lime        -Lime 
--------------------------------------------
Maize: Stems        0.50ns       0.95** 
       Roots        0.44ns       0.93** 
Potato:Stems        0.83*        0.66ns 
       Roots        0.66ns       0.76ns 
Bean:  Stems        0.77ns       0.83* 
       Roots        0.86*        0.87*

Crop growth and yield increases observed in this study is a reflection of improved soil conditions brought about by liming and P application (Table 6). Liming resulted in significant (P<0.05) increases in pH, Ca and ECEC, and reduction of aluminum saturation and exchangeable acidity (Table 6). Lime and P correct acid infertility by reducing aluminum saturation in the exchange complex and making P available (Sanchez, 1976). Liming gave more than a three fold increase in exchangeable Ca and caused a 64% reduction in aluminum saturation.

TABLE 5. Plants stand count at harvest as influenced by levels of P and lime for first (-1) and second (-2) season crops

---------------------------------------------------------------------------
1992     Maize       Potato- 1     Potato-2       Bean- 1         Bean-2 
---------------------------------------------------------------------------
  0     60    62     76    75      55    70      205    221      349    376
190     62    64     86    75      61    68      139    246      346    379
480     58    58     78    77      61    60      161    270      378    379
700     55    60     74    76      67    70      194    260      364    383
780     58    60     77    75      64    74      182    259      367    353

S.E.    1.3          1.7           9.9           9.9             9.9 
CV (%)  9.4         10.1          12.9          20.7             5.4

stand count in 1993: 
  0     60    73     20    22       -     -       61     66       -      -
190     65    70     21    23       -     -       65     73       -      -  
480     60    70     22    22       -     -       57     78       -      -  
700     66    70     20    20       -     -       69     76       -      -  
770     62    66     22    22       -     -       68     70       -      -  
S.E.    2.0          0.5                          2.6
CV (%) 13.5         11.0                         17.1
---------------------------------------------------------------------------

TABLE 6. Soil fertility changes as effected by lime and phosphorus, in a maize based system, 1993

----------------------------------------------------------------------
Soil Parameters           No Lime               2 t ha^-1 Lime
                  --------------------   -----------------------------
                    0     308     338      0      308     338     S.E. 
----------------------------------------------------------------------
pH (H2O)          5.13    5.02    5.02    5.41    5.44    5.26    0.16 
pH (KCl)          4.69    4.75    4.69    5.02    5.18    4.97    0.11 
Ca (Cmol/kg)      1.59    1.76    2.20    4.23    7.53    4.10    1.04 
Al + H (Cmol/kg)  0.81    0.63    0.98    0.32    0.32    0.32    0.16 
P (mg/kg)         0.41    0.85    1.29    0.38    1.06    0.85    0.14 
N (%)             0.27    0.26    0.26    0.27    0.21    0.27    0.02 
Org. C (%)       10.28   10.89   10.31   10.77   10.61   10.23    0.31 
ECEC (Cmol/kg)    2.72    2.89    3.40    4.84    8.20    4.67    0.95 
Al Sat (%)         33            34.3     9.03    6.48    8.06    6.84 

1 P rates, kg ha^-1
-----------------------------------------------------------------------

Correlation coefficients between soil fertility parameters and crop yields are presented in Table 7. Again, Ca, and pH exhibited good agreement (P(0.001) with maize yield. Calcium was responsible for 47% of the observed increase in maize yield. Compared to maize, correlations between these soil parameters and bean yields were weak, except P. About 44% reduction of maize yield and 15% of bean yield was attributed to aluminum saturation (Table 7). Correlations of these elements with potato performance were not significant. Thus, the soil conditions affected potato to a much lesser extent than maize and bean.

TABLE 7. Correlation coefficients (r) between soil fertility parameters and crop yields, 1993

-------------------------------------------------------
Soil parameters      Maize       Bean         Potato
-------------------------------------------------------
pH (H2O)           0.550**      0.420**       0.181ns 
pH (KCl)           0.674**      0.353*        0.305* 
Ca                 0.688**      0.406**       0.293ns 
Al + H            -0.498**     -0.252ns       -0.224ns
P                  0.155ns      0.430**       0.3418* 
N                 -0.369*       0.158ns       -0.339* 
ECEC               0.697**      0.415**       0.299ns 
Al Sat (%)        -0.661 **    -0.381*         -0.380*
-------------------------------------------------------

Mucuna and phosphorus study.

The results depicted in Table 8 validated the hypothesis that low C to N ratios green manures such as Mucuna reduce crops' P needs in acid soils. Mucuna established rapidly and produced 6.8 t ha^-1 dry matter (DM) in four months better than Canavalia (2.3 t ha^-1 DM) and Mimosa (0.8 t ha^-1 DM) previously tested at the same site. Accordingly, Mucuna green manuring consistently increased maize and bean yields. As shown in Table 8, maize yield with Mucuna green manuring and 50 kg P ha^-1 was as good as applying 200 kg P ha^-1. In the 0 P plots Mucuna green manuring increased maize yield by 45%, signifying supplemental N contribution from Mucuna to maize. Also, soil mycorrhizae associate readily with Mucuna roots and enhance their capability to extract soil P for use by subsequent crops. Mucuna resprouted in the course of the growing season and became problematic by harbouring birds and rodents that destroyed maize. Reseeding of Mucuna, however, ensured a continuous supply of N through fixation. Given that Mucuna is available locally, and free of cost, its widespread acceptance by farmers is expected to be faster than use of lime.

TABLE 8. Yields of maize and bean as affected by P and Mucuna

-----------------------------------------------------
P (kg ha^-1)          Maize               Bean
               -----------------    -----------------
               Control    Mucuna    Control    Mucuna 
-----------------------------------------------------
  0             1340      1947       198        283 
 50             1203      2083       131        361
100             1649      2870       220        604
150             1724      3159       127        491
200             2161      3316       269        384
S.E.                     421.1                 72.5
-----------------------------------------------------

Our analysis indicated that all three crops responded significantly to phosphorus applications in acid soils (Tables 9a, b and c). With lime, responsiveness increased considerably except in the case of bean where R^2 dropped 3% with lime. In general, lime appears to improve yields at each level of phosphorus. In the case of potato, P fertilizer accounted for 54% of variation in yield in the absence of lime and 58% with lime. Corresponding values for maize were 70% without lime and 79% with lime; bean, 48% without lime and 45% with lime. Yield response function for maize had a negative sign for phosphorus, indicating that a higher rate of phosphorus decreased yields regardless of lime. This may be the case where yields plateaued out at the lower level of phosporus when supplemented with lime. This is evident in Table 9b where maize crop showed moderate increase in yield with the first two levels of P fertilizer. The highest yield response to P was obtained between 0 and 84 kg ha^-1 with lime.

TABLE 9a. Yield response models for potato with and without lime plus P

---------------------------------------------------------------------------
Model Ip          Coefficient     P      Model IIp     Coefficient      P 
                  (Std.error)   Value                  (Std.error)    Value
---------------------------------------------------------------------------
Constant (alpha11)    7.0433    0.00   Constant (alpha12)  7.3496     0.00 
                     (0.2782)                             (0.3353)

LPF11 (beta11)        0.08121          LPF12 (beta12)      0.061525 
                     (0.0349)   0.03                      (0.0343)    0.09

R^2 (n=15)            0.54                                 0.58

TABLE 9b. Yield response models for maize with and without lime plus P

---------------------------------------------------------------------------
Model Im          Coefficient     P      Model IIm     Coefficient      P 
                  (Std.error)   Value                  (Std.error)    Value
--------------------------------------------------------------------------- 
Constant (alpha21)    7.3687    0.00   Constant (alpha22)   7.4416    0.00 
                     (1.7857)                              (1.4141)

LPF21 (beta21)       -0.06132   0.07    LPF22 (beta22)     -0.02190   0.38 
    (0.0297)            (0.0236)

R^2 (n=10)             0.70                                  0.79

TABLE 9C. Yield response models for bean with and without lime plus P

---------------------------------------------------------------------------
Model Ib           Coefficient     P      Model IIb     Coefficient      P  
                   (Std.error)   Value                  (Std.error)  
Value

--------------------------------------------------------------------------- 
Constant (alpha31)     6.0091    0.00   Constant (alpha32)  6.2654     0.00 
                     (0.3495)                             (0.3913)
LPF31 (beta31)         0.0385    0.34   LPF32 (beta 2)      0.07174    0.02 
                       0.0389)                             (0.0258)
R^2                    0.48                                  0.45

Yields responded significantly to phosphorus under Mucuna green manuring (Tables 10a and b). Gross returns increased approximately two fold for maize and more than doubled for bean (Tables 11 and 12). Twice as much P fertilizer or more is required to achieve the same level of returns without Mucuna. Model Imgm and Model IImgm clearly show that Mucuna contributed significantly to yields as expressed in the functions (P<.10, R^2 = 0.85 and 0.74), respectively.

TABLE 10a. Yield response models for maize with Mucuna green manure plus P

---------------------------------------------
Model Imgm          Coefficient         P
                   (Std. error)         Value
---------------------------------------------
Constant (alpha41)    7.1686            0.00 
                     (0.1474)    
LPFgm (beta41)        0.0523            0.17
                     (0.0339)    
D1 (Y41)              0.3290            0.09 
                     (0.1674)    
DPFgm (sigma41)       0.0017            0.23 
                     (0.0013)    
R^2                   0.85

TABLE 10b. Yield response models for bean with Mucuna green manure plus P

--------------------------------------------
Model Ibg^m           Coefficient       P
                     (std. error)     Value
--------------------------------------------
Constant (alpha51)     5.3333         0.00  
                      (0.1914)
LPFbgm (beta51)        0.0038         0.92
                      (0.0430)
D2 (Y51)               0.7062         0.00 
                      (0.2240)
DPFbgm (sigma51)       0.0019         0.28
                      (0.0017)
R^2                    0.74

TABLE 11. Gross returns ($ ha^-1) from intercropped maize, bean and potato as affected by lime and P

-----------------------------------------------------------------
P (kg ha^-1)    First season, 1992         First season, 1993
             ------------------------    ------------------------
             No lime   2 t ha^-1 lime    No lime   2 t ha^-1 lime 
-----------------------------------------------------------------
  0          227.50       -329:91        328.78      -336.46
 84          229.27       -345.44        215.74      -302.40
209          224.20       -345.97        258.39      -321.24
308          227.82       -422.46        250.27      -347.14
338          244.43       -362.91        284.43      -408.96
-----------------------------------------------------------------

TABLE 12. Gross returns ($ ha^-1) from maize and bean as affected by P and Mucuna green manuring, 1993

>
--------------------------------------------------
P (kg ha^-1)      Maize                Bean
            -----------------    -----------------
            Control    Mucuna    Control    Mucuna
--------------------------------------------------
  0         321.60     467.28    110.88     158.48 
 50         262.42     473.62     47.06     175.86
100         343.16     636.20     70.60     285.64
150         334.86     679.26     -7.78     196.06
200         413.44     690.64     45.44     109.84
--------------------------------------------------

Our analysis further indicated that liming is not economical at the prices of $526.00 ton^-1 for P and $347.00 ton^-1 for lime, even though liming resulted in significant yield improvements (Table 11). Nevertheless, supplementing lower rates of lime to improve crop productivity would be beneficial if lime prices were much lower than existing prices.

Assuming labour is free in subsistence agriculture in NW Cameroon, and a zero cost for Mucuna seeds, Mucuna as a green manure can cut down crop P fertilizer requirement and raise productivity more than lime. Thus, we conclude that it is possible to overcome acid infertility and maintain productivity of soils of NW Highlands of Cameroon with either lime or Mucuna green manuring and a moderate amount of P fertilizer. Choice of acid-tolerant crop varieties and use of compost and farm manure may further reduce the amounts of P and lime required and make farming a more attractive.

REFERENCES

Burleigh, J.R., Yamoah, C.F., Regas, J.L. and Mukaruziga, C. 1992. An analysis of factors associated with yield of climbing bean and Irish Potatoes in the Northern Highlands of Rwanda. Agriculture, Ecosystems and Environment 41:337-351.

Fageria, N.K. and Baligar, V.C. 1989. Response of legumes and cereals to phosphorus in solution culture. Journal of Plant Nutrition 12:1005-1019.

Foy, C.D., Lafever, H.N., Schwartz, J.W. and Fleming, A.L. 1974. Aluminum tolerance of wheat cultivars as related to regions of origin. Agronomy Journal 66:751-758.

Hoyt, P.B. and Turner, R.C. 1975. Effects of organic materials added to very acid soils on pH, aluminium, exchangeable ammonium, and crop yields. Soil Science 119:227-237. '

Hue, N.V. 1989. Effect of organic anions on P sorption and phytoavailability in soils with different mineralogies. In: Agronomy Abstracts pp. 242.

Hue, N.V. and Amien, I. 1989. Aluminum detoxification with green manures. Communications in Soil Science and Plant Analysis 20:1499-1511.

Hue, N.V., Craddock, G.R. and Adams, F. 1986. Effect of organic acids on aluminum toxicity in subsoils. Soil Science Society of American Journal 50:28-34.

Kamprath, E.J. and Foy, C.D. 1985. Limefertilizer plant interactions in acid soils. In: Fertilizer Technology and Use. 3rd Edition. Soil Science Society of America, Madison, WI., pp. 91-151.

Kamprath, E.J. 1967. Soil acidity and response to liming. International Soil Testing Program Technical Bulletin 4. North Carolina State University, Raleigh.

Kamprath, E.J. 1970. Exchangeable aluminum as a criterion for liming leached mineral soils. Soil Science Society of American Proceedings 34:252-254.

Nizeyimana, E. and Bicki, T.J. 1988. Soils and soil-landscape relationships in the mountainous region of Rwanda, East Central Africa. Agronomy Abstracts, p. 263.

Sanchez, P.A. 1976. Soil acidity and liming. In: Properties and Management of Soils in the Tropics. John Wiley and Sons. pp. 224-250.

Sanchez, P.A. 1977. Advances in the management of Oxisols and Ultisols in Tropical South America. In: Proceedings of the International Seminar on Soil Environment and Fertility Management in Intensive Agriculture. TokyoJapan, 1977. pp. 535-566.

Tisdale, S .L. and Nelson, W.L. 1966. Soil Fertility and Fertilizers. Macmillan Publishing Co., Inc. New York. pp. 214-226.

Vander Zaag, P., Yorst, R.S.,Trangmar, B.B., Hayashi, K. and Fox, R.L. 1984. An assessment of chemical properties for soils of Rwanda with the use of geostatistical techniques. Geoderma 34:293-314.

Van Ranst, E., Pauwels, J.M., Debaveye, J. and Zweier, K. 1989. Soil characterization and Maize Fertilization of PAFSAT Experimental farms in the Northern Part of the North-West Province. Technical Report No 1.

Van Ranst, E., Pauwels, J.M., Debaveye, J. and Zweier, K. 1990. Soil Characterization and Maize Fertilization of PAFSAT Experimental Farms in the Northern Part of the North-West Province. Technical Report No 2.

Yamoah, C.F.,Ngueguim,M. and Ngong, C. 1995. Fertility characterization of soils at six research sites in NW Cameroon. Fertilizer Research 41:49-57.

Yamoah, C .F., Burleigh, J.R. and Malcolm, M .R. 1990. Application of expert systems to the study of acid soils in Rwanda. Agriculture, Ecosystems and Environment 30:203-218.

Yamoah, C.F., Burleigh, J.R. and Eylands, V.J. 1992. Correction of Acid infertility of the Rwandan Oxisols for sustainable cropping with lime from an indigenous source. Experimental Agriculture 28:417-424.

Copyright 1996 The African Crop Science Society