The experiment was conducted at the INERA/Mvuazi Center, located in Kongo-Central province, 207 km south-west of Kinshasa (5˚27'08.77''S et 14˚53'20.34''E at 427 m altitude) during 2011-2012 growing season. The climatic and edaphic characteristics of INERA/Mvuazi correspond to those of the tropical zone with warm and humid climate of type AW₄ according to the classification of Köppen with 6 months of rainy season intercalated by a short drought (2 months), and followed of 4 months of dry season. Temperature variation and precipitation amount during the study period was presented in Figure 1. The soil of Mvuazi valley develops a humiferous horizon 15 - 25 cm thick, black (fresh) or brownish (dry) with a well-pronounced subangular or grumulous structure. This horizon is often overhauled by crops.

The experiment was in factorial design with 4 replications and treatments consisted to intercropping and sole cropping. Eight maize varieties, which five came from Zimbabwe (07 SADVE, 08 SADVE1, 09 SADVE-F2, ZM523, VP0538) and three local check of DR Congo (Mudishi1, Mudishi3 and Samaru) were sown at a spacing of 1 m × 0.25 m in sole crop and intercrop with a density of 80.000 plants∙ha⁻¹, according to Paliwal et al. [29] , in plots of 6 m × 4 m. Soybean (variety Vuangi) and cowpea (variety Diamand) were simultaneously planted with maize, at a spacing of 0.33 m × 0.055 m in pure culture and sole crop and intercrop with a density of 360.00 plants∙ha⁻¹. The pure culture for each variety was used as a control treatment.

The competition effects between 3 species used in this experiment were calculated by using the indices of competition. Seeds of each species were weighed and adjusted on the constant levels of moisture, and maize was the principal crop and the legumes like components of intercrops. The maize yield of each plot was converted, for intercropping system, in maize equivalent yield (MEY) which was computed by converting yield of intercrop into the yield of maize on the basis of prevailing market prices of individual crops as [30] :

MEY = Y maize + Y legume × P 1 P 2 (1)

where, Y_maize and Y_legume are the yield of maize and legume in an intercropping system and P₁ and P₂ are the selling price of maize and legume, respectively. The advantage of intercropping and the effect of competition between the two species used in a mixture were calculated using different competition indices. The land equivalent ratio (LER) was used as the criterion for mixed stand advantage as both vetch and cereal were desired species [31] . In particular, LER indicates the efficiency of intercropping for using the resources of the environment compared with monocropping [32] . The value of unity is the critical value. When the LER is greater than 1 the intercropping favors the growth and yield of the species. In contrast, when LER is lower than 1 the intercropping negatively affects the growth and yield of the plants grown in mixtures [33] [34] . The LER was calculated as:

LER = ( LER mais + LER legumineuse ) (2)

LER maize = ( Yml / Ym ) , et     LER legumineuse = ( Ylm / Yl ) (3)

where Yl and Ym are the yields of legumes and maize, respectively, as sole crops and Ylm and Yml are the yields of legumes and maize as intercrops, respectively. The second index is the area time equivalent ratio (ATER) provides more realistic comparison of the yield advantage of intercropping over monocropping in terms of time taken by component crops in the intercropping systems. ATER was calculated by formula:

ATER = 1 T ∑ 1 n d ∗ Yi Ym (Hauggaard-Nielsen et al., 2001) [10] (4)

where Yi and Ym are yields of component crops in intercrop and monocrop system, respectively, and n is the number of crops involved. d is the growth period of the crop in days and T is the time in days for which the field remained occupied i.e. the growth period of the longest duration crop. Numerical values of ATER approaches that of LER for a mixture consisting of crops of approximately identical growth periods i.e. when t = d in comparison, productivity can also be expressed in terms of the resources use efficiency of the most limiting resource e.g., water, nutrients, energy or till. The interpretation of ATER involves that ATER > 1 implies yield advantage; ATER = 1 implies of no effect of intercropping; ATER < 1 shows yield disadvantage [10] .

The relative crowding coefficient (RCC or K) is a measure of the relative dominance of one species over the other in a mixture [18] [35] . The K was calculated as:

K = ( K maize × K legume ) (5)

K maize = Yml × Zlm / ( ( Ym − Yml ) × Zml ) (6)

K legume = Ylm × Zml / ( ( Yl − Ylm ) × Zlm ) (7)

where Zml and Zlm were the proportions of maize and legume in the mixture, respectively. When the value of K is greater than 1, there is a yield advantage; when K is equal to 1, there is no yield advantage; and when it is less than 1.00, there is a disadvantage [15] .

The next index was agressivity (A) which is often used to determine the competitive relationship between 2 crops used in the mixed cropping [28] . The agressivity was formulated as fallows [15] :

A legume = ( Ylm / Yl × Zlm ) − ( Yml / Ym × Zml ) (8)

A maize = ( Yml / Ym × Zml ) − ( Ylm / Yl × Zlm ) (9)

For cereal example: if A_maize = 0, both crops are equally competitive, if A_maize is positive, then the maize species is dominant, if A_maize is negative, then the maize is the dominated species. Also, competitive ratio (CR) is another way to assess competition between different species. The CR gives a better measure of competitive ability of the crops and is also advantageous as an index over K and A [15] [36] . The CR represents simply the ratio of individual LERs of the two component crops and takes into account the proportion of the crops in which they are initially sown [37] . The CR is calculated according to the following formula:

CR maize = ( LER maize / LER legume ) ( Z legume-maize / Z maize-legume ) (10)

CR legume = ( LER legume / LER maize ) ( Z maize-legume / Z legume-maize ) (11)

The next index that was used was the actual yield loss (AYL) index, which gave more accurate information about the competition than the other indices between and within the component crops and the behavior of each species in the intercropping system, as it is based on yield per plant [12] . The AYL is the proportionate yield loss or gain of intercrops in comparison to the respective sole crop, i.e. it takes into account the actual sown proportion of the component crops with its sole stand. In addition, partial AYL_legumes or AYL_maize represent the proportionate yield loss or gain of each species when grown as intercrops, relative to their yield in sole crop [15] . The AYL [38] was calculated as:

AYL = AYL maize + AYL legume (12)

AYL maize = ( ( Yml / Zml ) / ( Ym / Zm ) ) − 1 (13)

AYL legume = ( ( Ylm / Zlm ) / ( Yl / Zl ) ) − 1 (14)

The AYL can have positive or negative values indicating an advantage or disadvantage remained in intercrops when the main aim is to compare yield on a per plant basis. Moreover, none of the above competition indices provides any information on the economic advantage of the intercropping system. For this reason, the monetary advantage index (MAI) was calculated as:

MAI = ( valueofcombinedintercrops ) × ( LER − 1 ) LER (15)

The higher the MAI value the more profitable is the cropping system [13] . Also, intercropping advantage (IA) was calculated using the following formula [12] :

IA maize = AYL maize × P maize (16)

IA legume = AYL legume × P legume (17)

where P_maize and P_legume are the commercial value of maize (the current price is 1000 FCD per kg) and legumes (the current price of cowpea and soybean are 1500 FCD per kg) respectively. Data were analyzed using R 3.1.3 (2015) following standard Analysis of Variance procedures, and least significant difference (LSD) test at 5% probability level was used to compare all varieties of maize and each treatment means. Principal component analysis (PCA) was used to represent the correlations between yield and competitiveness―monetary indices was carried out to show the profitability of DMC systems.

Regarding maize grain yield, a highly significant difference was observed (p < 0.001) between treatments (Table 1). Monocrop was higher (2350.2 kg∙ha⁻¹), compared to intercrops (1342.9 kg∙ha⁻¹ for cowpea and 2081.9 kg∙ha⁻¹ for soybean). Therefore, the best yield was recorded in monocrop of Samaru with 2849.44 kg∙ha⁻¹. This is similar to that obtained by KASETSART [39] in trials at the SUWAN farm in Thailand, where the yield of maize grown in monoculture was 31.9% higher than maize + soybean intercrop. A reasonable leaf area index (LAI) is essential to maintain high photosynthesis and yield [40] . Prasad and Brook [41] reported that the increase in maize density had a significant effect on LAI in maize-soya intercropping.

Means followed by the same letter in the columns do not differ by Tukey test (p ≤ 0.05); Statistical significance level: 0 “***” 0.001 “**” 0.01 “*” 0.05 “†” 0.1 “ns” 1.

Variances analyses of legumes yield showed a highly significant difference (p < 0.001) between the treatments and significant difference (p < 0.05) between the interactions and varieties (Table 1). The soybean monocrop yield was greater (1802.23 kg∙ha⁻¹) than yield in maize intercrop (809.31 kg∙ha⁻¹). And then, the sole yield of cowpea was higher (178.2 kg∙ha⁻¹) than yield in maize intercrop (48.39 kg∙ha⁻¹). Hardter et al. [42] has already reported these, although maize and cowpea intercrop was lower than in rotation, sole crop continuously have lower productivities. While the pure leguminous plant cultures maintained their superiority on the mixed farming system only for soybean, the same proportion (50% of soybean and 50% of cowpea) used in monoculture put in association, increased the outputs in grains of each variety of maize, but decreased those of soybean and increased that of cowpea.

The highest total mixed crop yield was obtained when Mudishi 3 was mixed with soybean (3108.89 kg∙ha⁻¹) (Table 1). MEY was to calculate and was higher in the system of soybean intercrop (Table 1) compared with monocultures.

Maize intercrops with soybean and cowpea had obtained higher maize equivalents yields than the total intercrops yields (Table 1). This can be allotted to the complementary effect of legumes in intercropping system by the nutritive transfer. Similar results were obtained by Ofori and Stern [33] .

Various competitive relationship of intercropping were computed and presented in Table 2. Results showed that maize partial LER were higher than 0.50 for all intercrops which indicates that there was an advantage for maize, except for Mudishi 1-cowpea (0.31) and VP0538-cowpea (0.46). The variance analysis showed that there was significant difference between Treatments (p < 0.05) for LER_maize. On the other hand, all partial LER of cowpea was lower than 0.5 except intercrops with 07SADVE, Samaru and VP0538 which indicates that there was a disadvantage for cowpea. The partial LER of soybean were higher than 0.5 for all intercrops which indicates that there was an advantage for soybean [19] , except intercrops with 07SADVE and Samaru.

LER showed the positive influences on growth and yield of maize and legumes intercrops (LER > 1) in maize-soybean intercrops with the greater for 08SADVE 1-soybean (1.66). This indicates that 66% (0.66 ha) more area would be required by a sole cropping system to equal the yield of intercropping system [25] . In these cases, total LER was significantly different (P < 0.01) and values were higher than one showing the advantage of intercropping over sole stands in regard to the use of environmental sources for plant growth [32] . Similar results were reported for mix-proportions of pea-barley [19] , bean-wheat [10] , and maize-faba bean [43] . Partial LER values also showed that, compared to soybean, cowpea appears to have more beneficial land use efficiency in all mixtures.

LER doesn’t consider the duration of the crops in the field and it is based on the harvested products, and not on desired yield proportion of the component

Means followed by the same letter in the columns do not differ by Tukey test (p ≤ 0.05); Statistical significance level: 0 “***” 0.001 “**” 0.01 “*” 0.05 “†” 0.1 “ns” 1.

crops. Moreover, the choice of sole cropped yield for standardizing mixture yield in the estimation of LER is not clear [28] , therefore, area time equivalent ratio (ATER) provides more realistic comparison of the yield advantage of intercropping over sole cropping in terms of variation in time taken by the component crops of different intercropping systems [44] .

The data presented in Table 2 showed that in all maize-legumes intercropping arrangements, the ATER values were lesser than LER values indicating the over estimation of resource utilization perhaps due to the wide variations in the maturity periods of the crops of which maize stayed longer on the land and had enough time to compensate for the legumes competition. ATER is free from problems of over estimation of resource utilization contrary to LER. ATER values showed an advantage of 57% in 08SADVE 1-soybean (Table 2). This could be due to the reason of a significant effect on LAI in maize-soybean intercropping [41] .

Although a statistical comparison among the partial K values of legumes and maize was not performed, it seems that the maize partial K was higher than legumes partial K in the case of the maize-cowpea mixtures and in maize-soybean mixture, indicating that maize is more competitive than its associated crop, except the cowpea intercropped with 09SADVE F2, Samaru and ZM523, and the soybean intercropped with 07SADVE and 08SADVE1 (Table 1). Similar results reported by Banik et al. [12] in chickpea-wheat intercropping and Dhima et al. [15] in cereal-vetch intercropping. The total K was above one in the case of all intercrops with cowpea and soybean, which indicates a definite yield advantage due to intercropping [12] . However, in the maize-cowpea mixture (09SADVE F2 and Mudishi 1) and in maize-soybean mixtures (07SADVE and Mudishi 1), the total K was below one, which indicates that there was a yield disadvantage [13] [36] . K values followed a similar trend with the LER values.

In all mixture, positive A_maize values showed that maize was the dominant species (Table 2). Intercropped maize had higher competitive ratios (CRs) in all mixtures; however, cowpea had higher CR values than those of soybean with 09SADVE F2 (98.63), Mudishi 3(57.96) and 08SADVE1 (56.22) varieties. The results of competitive ratio index were also in corroboration with those of the aggressivity index. The values of A and CR for cowpea were indicated that cowpea was more competitive than soybean in maize mixtures. According to Yilmaz et al. [22] , although increasing the cereal rate in mixtures elevated the crowding efficiency over legumes, doubling the rate per se may commence competition among maize plants, which probably resulted in weaker growth, thereby, lower CR and A_maize values. This was also confirmed with the negative AYL_maize values that were negative only when the maize ratio was the highest (Table 3). This was probably due to the fact that nitrogen fixing ability of the legumes did not compensate vigorous growth of cereals over a certain proportion.

In particular, AYL_maize had positive and the higher value only in Mudishi 3-soybean intercropping, it was increase yield in this association with 0.41% (0.0041) yield advantage compared to sole cropping. Which indicate a yield advantage for maize probably because of the positive effect of soybean on Mudishi 3 variety when grown in intercropping [12] [38] . According to Banik et al. [12] , the AYL index can give more precise information than the other indices on the inter- and intra-specific competition of the component crops and the behavior of each species involved in the intercropping systems. Quantification of yield loss or gain due to intercropping with other species or the variation of the plant population could not be obtained through partial LERs, whereas partial AYL shows the yield loss or gain by its sign and as well as its value.

Monetary advantage index followed the trend similar to LER (Table 3). The MAI values were positive under maize-soybean intercropping system (Table 3), which shows a definite yield advantage compared with the other intercropping

Means followed by the same letter in the columns do not differ by Tukey test (p ≤ 0.05); Statistical significance level: 0 “***” 0.001 “**” 0.01 “*” 0.05 “†” 0.1 “ns” 1.

systems tested in this study. In particular, the highest positive MAI values were for the Mudishi 3-soybean (+922.92), 08SADVE1-soybean (+917.03) and ZM523- soybean (+913.41). The lowest monetary benefit was recorded in SAMARU-soy- bean (+353.38). The other intercropping with cowpea showed negative MAI values, except 07SADVE-cowpea (+542.65) and SAMARU-cowpea (+404.19). Similarly, the IA, which is also an indicator of the economic feasibility of intercropping systems, indicated that most advantageous mixture was only for maize in Mudishi 3-soybean (+4.11). The fact that MAI and IA values were positive indicates that these intercropping systems had the highest economic advantage, whereas all the other mixtures, which had negative values, showed an economic disadvantage. These findings were also in agreement with the results of LER and the other competition indices (Table 1 and Table 2). Similarly, Ghosh [13] found that when the LER and K where higher there is also significant economic benefit expressed with higher MAI values.

The differences found between mixtures in this study can be attributed to aggressivity of maize and to other factors such as morphology, physiology and the different requirements for nutrients. In particular, the growing maize intercropped with cowpea can affect nitrogen fixation because of reduced light interception of cowpea due to shading by maize. This can result in poor nodulation, growth and competitive ability of cowpea compared to soybean in these mixtures. Also, the advantages of the intercropping systems found in this study can be attributed to the better utilization of growth resources [33] . For example, Osunde et al. [45] found that without the addition of fertilizer the proportion of N derived from N2-fixation was about 40 percent in the intercropped soybean and 30 percent in the sole crop, and with application of 40 kg∙N∙ha⁻¹ N2-fixed by the sole cropped soybean was significantly greater than that fixed by intercropped system. However, to find out the real causes of these differences among mixtures examined in this study further research is required to study those factors.