Effects of varying seed proportions and harvesting stages on biological compatibility and forage yield of oats (Avena sativa L.) and vetch (Vicia villosa R.) mixtures

Berhanu Alemu, Solomon Melaku^* and N K Prasad^*

Bureau of Agriculture and Rural Development, Amhara Regional State, Bahir Dar, Ethiopia
 *Haramaya University, Department of Animal Sciences, P.O. Box 138, Dire-Dawa, Ethiopia
solmelay@yahoo.com

Abstract

The study on the effect of seed rates and harvesting stages on forage yield and quality of oats (Avena sativa L., accession CI-8237) and vetch (Vicia villosa R.) mixtures were carried out during 2003 cropping season at Adet Agricultural Research Center, Northwestern Ethiopia. The objectives of the study were to assess the optimum seeding rates and harvesting stages for maximum biomass yield and the biological potentials of the oats/vetch components in mixtures and pure stands. The experimental design employed was a split plot design with three stages of harvesting, namely, flag leaf stage (HS1), milk stage (HS2), and dough stage (HS3) as main plot treatments, and five seed proportions of oats/vetch, namely, 100:0 (SP1), 75:25 (SP2), 50:50 (SP3), 27:75 (SP4), 0:100 (SP5) as sub plot treatments in three replications.

The highest herbage dry matter yield (DMY) of 13.22 t/ha was recorded at HS3 and SP2. Significantly higher leaves/ plant (5.15) for oats was obtained at HS2, whereas highest branches/ plant (15.27) for vetches was obtained at HS3 and SP5. Seed proportion four for oats was found to have significantly higher (P < 0.01) leaf to stem ratio than the other SPs. Relative yield total was found to be more than one at HS2 and HS3, with the highest value of 1.32 at HS2 and SP4, indicating a yield advantage of 32% for oats / vetch intercropping compared to sole cropping of either of the two forage species. Therefore, it is concluded that cultivation of oats /vetch mixture with seed proportions of 25% oats and 75% vetch and harvesting at milk stage offers better relative yield advantage of the mixture and higher leaves per plant and leaf to stem ratio for oats.

Key words: biological compatibility, Ethiopia, harvesting stage, oats, seed proportion, vetch

Introduction

In most tropical countries, inadequate supply of feeds is the bottleneck to livestock production. Basically, this is due to the dependence of livestock on naturally available feed resources and little development of forage crops for feeding to animals. Like in other tropical countries, indeed in Ethiopia, most of the areas in the highlands of the country are nowadays put under cultivation of cash and food crops. This resulted in keeping large number of livestock on limited grazing area leading to overgrazing and poor productivity of livestock. Though, expansion in the cultivation of cereal crops increased the supply of crop residues for animal feeding, however, crop residues have low nutritive value and could not support reasonable animal productivity. Hence, shortage of nutrients for livestock is increasingly becoming serious. One of the alternatives to improve livestock feeding, and thereby their productivity could be the cultivation of grass-legume mixtures and offer them to animals during critical periods in their production cycle and when other sources of feeds are in short supply.

The nutritive value of pasture is mainly a function of species composition and its growth stage, which are controlled by climatic factors that affect mineral status, re-growth potential, sward structure, and botanical composition (Whiteman 1980). Stage of maturity at harvesting or grazing can be considered as crucial management practice that determines nutritional quality of either cultivated or natural pasture. In most of the cases, common pasture grasses in the tropics cannot satisfy even the minimum requirement of nutrients for animals due to harvesting at advanced stage of maturity (Beyene et al 1977). Forage species like vetches (Vicia dasycarpa), which is a herbaceous legume and oats (Avena sativa), which is a tall growing cereal has been promoted for mixed cereal/legume forage production by different livestock development programs (MOA 1989) in Ethiopia. Moreover, oats is cultivated for forage and grain production in the highlands of Ethiopia. It is adapted to a wide range of soil types, altitudes and rainfall conditions, and tolerates water logging than most of the other cereals (Alemayehu 1997). However, adequate studies were not conducted with regard to biological compatibility and biomass yield of oats / vetch (Vicia villosa) mixtures established using different seed proportions and at different stages of harvesting. Therefore, this study was conducted to assess the seed proportions and harvesting stages for maximum biomass yield, and compute the competition functions as related to forage yield advantage.

Materials and methods

Description of the area

The research was conducted at Adet Agricultural Research Center, Northwestern Ethiopia. The center is located at 11^o17' N and 37^o 43' E with an elevation of 2240 meters above see level. The soils at the station are alluvial, red, and black. The annual rainfall of the area is 1285 mm with a range of 860 to 1771 mm. Data collected for over 14 years indicated that the main rainy season extended from June to September with a mean of 108 rainy days. The mean annual minimum and maximum temperatures were 8.8 ^oC and 25.4 ^oC, respectively (Adet Agricultural Research Meteorological Station).

Land preparation, experimental design, sowing and germination test

The experimental plots were prepared on red soil using standard tillage procedures. The experiment was conducted using split plot design with three harvesting stages (HS) as main plots and five seed proportions (SP) as sub-plots consisting of fifteen treatment combinations in three replications. The three main plot treatments included flag leaf stage (HS1), milk stage (HS2) and dough stage (HS3). The sub plot treatments were 100 % oats + 0 % vetch (SP1), 75 % oats + 25% vetch (SP2), 50% oats + 50% vetch (SP3), 25% oats + 75% vetch (SP4), 0 % oats + 100% vetch (SP5).

Variety of oats used was (CI-8237) (O) and vetch (Vicia villosa R.) (V). The seed proportions were calculated on the basis of the recommended sole seed rates of 80 and 25 kg per hectare for oats and vetch, respectively. The plot size for harvesting stage was 12 m x 5 m and sub plot (seed proportion) was 2 m x 5 m. The spacing between replications, main plots, and sub plots were 1.5, 1.0, and 0.5m, respectively. Germination test was done on seeds of both crops before sowing in order to adjust the seeding rates in case of lower seed germination conditions. The seeds were broadcasted on a well-prepared seedbed and covered with soil, and diammonium phosphate (DAP) (18/46, N/P₂O₅) at the rate of 100 kg/ha was applied to the soil before sowing. The crops were weeded three times.

Data collection

Stand count was taken one week after emergence from three quadrates (0.5m x 0.5 m) in each plot. Stand count at tillering for oats was counted at 45 days of plant growth. Heading date for oats and flowering date for vetches were recorded. Ten plants of each species were harvested at ground level from each plot at 20 days interval starting from day 20 of plant growth to measure plant height and DM accumulated over the growing period (Getinet 1999). The samples were weighed and dried in a forced draft oven at 65 ºC to constant weight to determine the DM percentage at each sampling stage.

Total harvesting was done at three physiological growth stages of oats, namely, flag leaf stage (HS1), milk stage (HS2) and dough stage (HS3). Cutting was done immediately above ground level from a net plot size of 1.5m x 4.5m area. Plant height, number of leaves per plant for oats and number of branches per plant for vetch was counted at each harvesting stage. During total harvesting, total biomass weight was measured and then separated into oats, vetch and weeds to calculate proportions of each component. Leaf to stem ratios for oats was measured and calculated for each plot during each harvesting stage. Dry matter yield per hectare and DM percentage in all the samples were determined by drying fresh sample to constant weight in a forced draft oven at a temperature of 65 °C. The three physiological harvesting stages were made at days 58, 99 and 118 of plant growth.

Biological compatibility

Relative yield

Relative DM yield for grass and legume was calculated using the equation of De Wit (1960):

RY_G =DMY_GL / DMY_GG
RY_L =DMY_LG / DMY_LL,

where:

DMY_GG is the DMY of grass as a monoculture,
DMY_LL is the DMY of annual legume as a monoculture,
DMY_GL is the DMY of annual grass component grown in mixture with annual legume,
DMY_LG is the DMY of annual legume component grown in mixture with annual grass.

Relative yield total or land equivalent ratio

These parameters were calculated using the formula of De Wit (1960),

RYT or LER = (DMY_GL / DMY_GG) + (DMY_LG / DMY_LL)

where:

RYT= relative yield total
LER= land equivalent ratio

Relative crowding coefficient

This parameter was calculated to determine the competitive ability of the annual grass and legume in the mixture by measuring the component that has produced more or less DM than expected in a 50:50 grass legume mixture (De Wit 1960): The formula for the 50:50 grass - legume mixture is:

RCC_GL=DMY_GL / (DMY_GG - DMY_GL)
RCC_LG =DMY_LG / (DMY_LL - DMY_LG)

The formula for mixtures differing from 50:50 proportions was:

RCC _GL = DMY_GL X Z_LG / (DMY_GG - DMY_GL) X Z_GL

where:

RCC = relative crowding coefficient,
Z_{GL =} the sown proportion of grasses in combination with legumes,
Z_{LG =} the sown proportion of legumes in combination with grasses.

Aggressivity index

The aggressivity index (AI) of annual grass against the annual legume in a 50:50 mixture was calculated as described by McGilchrist (1965) and Trenbath (1986):

AI_GL = (DMY_GL /DMY_GG) - (DMY_LG /DMY_LL)
AI_LG = (DMY_LG/DMY_LL)- (DMY_GL/DMY_GL)

Data analysis

Data was analyzed using MSTAT-C statistical software for analysis of variance. Treatment means, where significant were separated by Duncan's multiple range test, and the statistical model used for this design was,

Y_ijk = µ + r_i + m_j + m_ij + s_k+ (ms) _jk + e_ijk,

where:

µ = over all mean effect

Y_ijk = the observation of i^th replication, j^th main plot and k^th sub plot.

r_i = i^th replication effect.

m_j = j^th main plot treatment effect (harvesting stage)

m_ij = main plot error or error (a)

s_k = _k^th sub plot treatment effect (seed proportion)

(ms)_jk = interaction effect

e_ijk = error component for sub plot and interaction or error (b).

Results and discussion

Seedling and tiller count

Different SPs resulted in significant differences (P<0.01) in stand count at emergence and number of tillers for oats (Table 1). The highest seedling counts of oats was obtained at SP1 (pure oats) and the lowest at SP4 with a range of values from 46 to 153 seedlings/m² and a difference of 107 seedlings between the highest (SP1) and the lowest (SP4) seedling counts.

The seedling count for vetch (Table 1) was also found to be significantly different (P<0.01) among the different SP treatments. The values ranged from 12 - 42 seedlings/m². The lowest and the highest counts were obtained at SP2 and SP5, respectively. The difference in seedling count for vetch between the highest and lowest was 30 seedlings /m². The highest seedling counts at emergence were observed for the sole plots of both species, and declined with a reduction of the respective species in the seed proportions of the mixtures, leading to the lowest seedling counts where either of the forage crops was low in the seed proportion of the mixture. Stand count after tillering for oats showed significant difference (P< 0.01) among the different seed proportions (Table 1), with the rank in the order of SP1 = SP2 > SP3 > SP4. The range was from 147 to 249 tillers /m² between the lowest at SP4 and the highest at SP1, respectively. When the numbers of tillers for the different SPs were standardized to 100, the highest value was obtained at SP4 and the lowest at SP1. These values showed that tillering capacity of oats increased with its relative decrease in SPs.

Days to heading for oats was significantly longer (P<0.01) in SP4 compared to the other SPs (Table 1). The delay in heading of oats in SP4 might be attributed to the transfer of a higher proportion of N fixed by vetch, which in turn delayed the maturity of oats by keeping the leaves green for a longer time than those plots with lower vetch proportions. Days to flowering for vetch (Table 1) were similar among the different SPs. The current study with a mean value of 99 days to flowering for vetch was not in agreement with the study of Hailu and Lulseged (1983) who reported first flowering in a similar variety of vetch to occur in a range of 131 to 148 days. The possible explanation for the differences between the two studies could be climatic difference between the two experimental sites.

Dry matter accumulation and plant height of oats and vetch through the growing period

Significantly lower (P<0.01) plant height was recorded at SP4 for oats on the twentieth day of growth compared to SP1 and SP2 (Table 2).

The range of heights between the twentieth and the eightieth days of plant growth were 19 to 122.8 cm. The DM accumulated at the twentieth day of growth of oats was significantly higher (P>0.01) for SP1 and SP4 as compared to SP2 and SP3. Accumulation of DM at fortieth day was significantly higher (P<0.01) for SP4 compared to SP1 and SP3, but differences were not observed among SP1, SP2 and SP3 (Table 2). Significantly higher (P<0.01) DM accumulation was observed on the sixtieth date of growth for SP2 compared to the other SPs, however, DM accumulation was similar for all SPs at eightieth day of growth. A relatively lower biomass accumulation in the later stages of growth for SP4 than the rest of the SPs could be attributed to the succulent nature of oats in this treatment due to delayed physiological maturity caused by N transfer from N fixation by the higher proportion of vetches in SP4. On relative term basis, SP2 resulted in longer plant height and higher DM accumulation when compared with the other SPs, and the lowest plant height and DM accumulation was recorded at SP4. This may be attributed to the fact that tropical grasses are highly efficient in their conversion of CO₂ and light energy to carbohydrate and in their use of water and nutrients for this purpose. The rapid growth of tropical grasses, associated with the C4 dicarboxylic acid pathway leads to the rapid accumulation of carbonaceous materials (Whiteman 1980). Thus, the relatively higher DM accumulation in SP2 could be attributed to the inherent characteristics of oats and the supply of N from the vetch component.

The sole planted vetch was significantly shorter (P< 0.01) at the twentieth and sixtieth days of growth than the other SPs which were mixed with oats (Table 3).

Moreover, the height of vetches tended to increase with increasing proportion of oats with the exception of that cut at the eightieth day of growth. The observation indicated that vetch in mixture with oats was highly affected by competition for light, and as a result the vetch in mixture with higher proportions of oats increased in height rather than expanding side ways by bearing branches. This argument is supported by the lower number of branches per plant for vetches in SPs with higher oats proportion and the vice versa. Thus, this result suggests the need to reduce the proportion of oats in oats /vetch mixed forage if increased proportion of vetch in the mixed forage biomass is required. No significant differences were observed (P>0.05) among the SPs in DM accumulation of vetch at the twentieth day of growth, whereas at the fortieth day of growth, significantly lower (P<0.01) DM accumulation were recorded for SP2 (Table 3). At the sixtieth and eightieth day of growth, significantly higher (P<0.01) DM accumulation was recorded for SP5 (pure vetch) compared to the other SPs. The results of this study showed that vetches in pure stands and with lower proportions of oats accumulated higher DM than those found in association with higher proportions of oats.

Dry matter yield and dry matter percentage of oats and vetch as considered separately

Dry matter yield of oats was affected (P<0.01) both by main effects and their interactions (Table 4).

Dry matter yield of oats at HS1 was significantly lower (P< 0.01) than HS2 and HS3, whereas HS2 and HS3 were not significantly different (P> 0.05) from each other. The lower DMY at HS1 for oats could be due to lower DM accumulation per plant which can be supported by the DM percentage value of 17.1% recorded for oats at HS1 (Table 5), whereas for the latter stages of harvests, the DM percentage increased with increased plant growth and development. The tendency of the reduction in DMY for HS3 than that of HS2 could be attributed to the shading of oats leaves with advance in plant maturity. Similarly, Smith (1960) reported a slight difference in DM weight between nearly ripe (9.38 t/ha) and ripe (8.74 t/ha) stages of Clintland oats variety. The difference in days among the three HSs, HS1 to HS2 (flag leaf to milk stage), HS2 to HS3 (milk to dough stage) and HS1 to HS3 (flag leaf to dough stage) were 41, 19, and 60 days, respectively.

The DMY of oats was significantly lower (P<0.01) in SP4 compared with the other SPs (Table 4). The lower DMY for SP4 could be due to the lower oats and higher vetch seed proportion as compared to the higher oats and lower vetch seed proportions in SP1, SP2 and SP3. The DMY for vetch at HS1 was significantly lower (P< 0.01) than HS2 and HS3, whereas the later HSs were not significantly different (P> 0.05) from each other (Table 4). The mean DMY of vetch was 2.02 t/ha with a range from 0.72 to 3.03 for HS1 and HS3, respectively. The second harvest coincided with 50% flowering of vetch and thereafter it increased aggressively in biomass resulting in high DMY at HS3. Significantly higher (P<0.01) DMY was obtained for vetch at SP5 (pure vetch).

The DM percentage of oats as well as vetches at HS1 (Table 5) was significantly lower (P< 0.01) than at HS2 and HS3, and the latter were similar in DM percentage for both species. The lower DM percentage at HS1 for both oats and vetches was due to low DM concentration and high contents of water in the plant tissues at early stages of physiological development.

The DM percentage for oats was significantly higher (P<0.01) for SP1 (pure oats) and lower (P<0.01) for SP4 as compared to the other SPs (Table 5). The observed lower DM percentage of oats for SP4 might be due to the delay in physiological development caused by the transfer of N from the high proportion of vetch in the treatment, which is similar to the results of other studies (Rai 2002; Pal and Shehu 2001) that reported the contribution of legumes to the total N uptake of maize in mixture. The DM percentage of vetch at SP5 was significantly lower (P<0.01) than SP2 and SP3 (Table 5), but there were no statistically significant difference (P>0.05) among SP2, SP3 and SP4 and also between SP4 and SP5.

Leaf to stem ratio of oats

Leaf to stem ratio (LSR) was highest (P<0.01) at HS1 compared to the later HSs, which were similar to each other (Table 5). In the case of SPs, SP4 resulted in higher LSR (P<0.01) compared to SP1, SP2 and SP3, which were similar to each other. There was an increasing trend in LSR from SP1 to SP4. The highest LSR (2.4) was obtained at HS1 and SP4 which might be due to the lower seed rate of oats that reduced competition between the two components. On the other hand, increased proportion of vetches might have contributed more to the leafiness of the oats component through more N transfer. This was clearly observed in the field on the different plots with higher vetch proportions. The oats on SP3 and SP4 were vigorous, deep green with wider leaf area. Leaf to stem ratio is an important determinant of feed intake by grazing livestock due to modulating ingestion, yield, distribution and density of biomass (Whiteman 1980). Therefore, treatments with higher leaf to stem ratio may improve feed intake under practical grazing conditions.

Combined dry matter yield of oats and vetch

Combined DMY was affected both by the main effects and their interactions (Table 6). The interaction between HSs and SPs showed that significantly higher (P< 0.01) DMY was obtained at HS2 and HS3 and SP1, SP2 and SP3. These results are similar to that of Abate (1998) who reported an increased DMY with increase in the stage of growth of grass-legume mixtures.

A significant increase in DMY of pasture with advancing stage of maturity was reported by Teshome (1987). Generally, an increasing trend in DMY from HS1 to HS3, and a decreasing trend from SP1 to SP5 were observed. Stage of maturity of forages at harvesting time is the most important factor that influences the DMY and nutritional quality (McDonald et al 2002). Forage at the time of harvest is the cumulative result of plant growth and the environmental factors influencing the distribution of energy and nutrients derived from photosynthesis (Van Soest 1994). The lower seed rates of oats did not seriously affect its DMY in the current study, which is in agreement with other reports (Getnet 1999; Droushiotis 1989) indicating lack of increased forage yield between high and low seed rates in pure stand of oats due to increased tiller development in oats at lower seeding rates.

Competition and yield advantages of oats vetch mixtures

Relative yield total or land equivalent ratio

The land equivalent ratio (LER) or the relative yield total (RYT) computed for the yield of different mixtures (Table 7) indicated that harvesting of the mixtures at flag leaf stage was biologically not desirable since it resulted in relative yield total less than one. However, the mixtures produced yield advantages when harvested either at the milk or dough stages. The highest RYT (1.32) with an advantage of 32% more yield was obtained at HS2 and SP4.

Intercropping of oats with Chinese cabbage also gave the highest RY under tropical conditions (Prasad and Singh 1991). Minale et al (2002) and Tilahun (2002) also reported that intercropping maize with faba bean gave higher RYT than unity. This RYT does not only give a better indication of the relative competitive ability of the component species, but also it showed the actual advantage due to intercropping. The intercropping system resulted in higher cumulative total biomass yield than either of the sole crops, which resulted in higher RYT values than the sole cropping. The higher cumulative total biomass yield was probably due to increased light use efficiency of the intercrops, which has resulted in higher cumulative leaf area of the intercrops. This finding was in agreement with that of Ibrahim et al (1993) who reported a similar result in sorghum/lablab intercropping. Jaballa (1995) also reported that intercropped treatments had higher combined leaf area than monocultures and the intercrops gave higher biomass yield per unit area than sole crops. Reddy (2000) reported that increase in leaf area would lead to an increase in light interception and photosynthesis, which results in increased productivity.

Relative crowding coefficient

Competition function of the two component species in relation to RCC (Table 8) indicated that when the crop mixtures were cut at flag leaf stage, under different SPs, oats was found as a dominant species over vetch and product of crowding coefficient of SP4 appeared to be within acceptable range resulting in crowding coefficient greater than one (k=1.01).

Harvesting at milk stage also resulted in dominance of oats over vetch. The product of crowding coefficient at this harvesting stage indicated an advantage of mixing oats and vetch either in 50:50 or in 25:75 since these two patterns resulted in products of crowding coefficient greater than one, namely (1.52) and (2.70), respectively. The trend in the product of RCC of the mixtures cut at dough stage was almost similar to harvesting at milk stage. It appeared that among all the stages of harvesting, harvesting at dough stage of the oats/vetch mixture (25:75) accounted for the maximum product of RCC (k=26.77) as well as RY of both the components were greater than one.

The dominance or aggressivity index

The agressivity index of the different plant associations and their stage of harvesting (Table 9) indicated that oats was always aggressive over vetch. The aggressiveness of the oats increased with advance in the stage of harvesting. This might be due to faster growth of oats through better utilization of space, nutrients and light energy in comparison to vetch.

Conclusions

• The result of the current study showed that reduced seed proportion of oats was associated with increased tillering, which contributed to comparable DMY with seed proportions containing higher proportions of oats.

• he higher proportion of vetch in the forage mixture delayed the physiological maturity and increased the LSR of oats, thus resulting in better nutritive value for animals. The RYT indicated higher yield advantage of 32% when oats and vetch were intercropped at SP4 and harvested at HS2. On the other hand, higher RCC value at HS3 in SP4 indicated that SP4 had higher yield advantage.

• The higher agressivity index (AI) value (2.91) at HS3 in SP4 indicated that oats was benefited from vetch, so that its vigor and aggressivity increased at later stages of growth. Therefore, it is recommended that cultivation of oats /vetch mixture with seed proportions of 25% oats and 75% vetch and harvesting at milk stage offers better relative yield advantage of the forage mixture as well as higher LSR for oats.

References

Abate Tedla, Crosse S, Umunna N N, Osuji P O, Azage Tegegne and Khalili H 1998 Comparative yield and nutritive value of forages from two-cereal legume based cropping systems. Tropical Agriculture (Trinidad) 75: 415.

Alemayehu Mengistu 1997 Conservation based forage development for Ethiopia. Institute For Sustainable Development. Addis Ababa, Ethiopia. pp. 1-2, 57-60, 86-88.

Beyene Chichaiblu, Capper C E and McDowell R E 1977 Laboratory evaluation and nutritive value of some Ethiopian feed stuffs and formula feeds. Association for Advancement of Agricultural Sciences in Africa. IIV (2): 9-23.

De Wit C T 1960 On competition. Verslag Landbouwkundig Onderzoek, 66 (8), 1- 82.

Droushiotis D 1989 Forage production from barley, oats and triticale, under semi-arid conditions. XIV. International grassland congress, Niece, France, pp. 1519 - 1520.

Getnet Assefa 1999 Feed resource assessment and evaluation of forage yield, quality and intake of oats and vetches grown in pure stand and mixtures in the highlands of Ethiopia. M.Sc. Thesis. Uppsala. pp.1-26.

Hailu Keno and Lulseged Gebrehiwot 1983 Prospects of vetches (Vicia spp.) as forage legumes for the highlands of Ethiopia. Ethiopian Journal of Agricultural Science 5(2): 131-138.

Ibrahim Y M, Gaffar M O and Wahab D A A 1993 Intercropping of pioneer sorghum with Lablab purpureus L. under irrigation at Shambat. Annals of Arid Zone. 32(3): 157-159.

Jaballa O R 1995 Response of upland rice varieties to nitrogen fertilization and intercropping. Ph.D. Thesis. Philippines University, Los Banos, Philippines.

Mc Donald P, Edwards R A, Greenhalgh J F D and Morgan C A 2002 Animal Nutrition. (5^th edition), Longman scientific and technical, John Wiley and Sons. Inc., New York, pp. 418-433.

McGilchrist C A 1965 Analysis of competition experiments. Biometrics, 21, 975-985.

Minale Liben, Tilahun Tadesse and Alemayehu Asefa 2002 Determination of nitrogen and phosphorus fertilizer levels in different maize-faba bean intercropping patterns. Paper presented at the 7^th Eastern and Southern Africa Regional Maize Conference and Symposium on low-nitrogen and drought tolerance. Nairobi, Kenya.

MOA 1989 Forage Extension Manual, Fourth Livestock Development Project. Animal and Fisheries Resources Development Main Department, MOA (Ministry of Agriculture), Addis Ababa, Ethiopia. pp. 89.

Pal U R and Shehu Y 2001 Direct and residual contributions of symbiotic nitrogen fixation by legumes to the yield and nitrogen uptake of maize (Zea mays L.) in the Nigerian Savannah. Journal of Agronomy and Crop Science187: 53-58.

Prasad N K and Singh A P 1991 Biological potentials and economic feasibility of Wheat: Lucerne intercropping system. Indian Journal of Agriculture Science 61: 838-840.

Rai M M 2002 Principles of soil science. 2^nd edition Macmillan, Delhi, pp 84, 146,195.

Reddy R S 2000 Principles of crop production. Kalyani Publishers, New Delhi-110 002, India. Pp 45-47.

Smith D 1960 Yield and chemical composition of oats forage with advanced in maturity. Agronomy Journal 52:637-638.

Teshome Shenkoru 1987 Nutritive, dry matter yields and in vitro dry matter digestibility as affected by cutting interval and fertilizer application on in pace soil types at Alemaya University of Agriculture. M. Sc. Thesis, Alemaya University of Agriculture. pp. 160.

Tilahun Tadesse 2002 Effect of planting density and arrangement of component crops on productivity of maize/ faba bean intercropping systems. M.Sc. Thesis. Alemaya University of Agriculture. pp. 113.

Trenbath B R 1986 Resource Use by Intercrops. In: Francis C A (editor) 1986. Multiple Cropping. Macmillan Publishing Company, New York. pp. 57-81.

Van Soest P J 1994 Nutritional ecology of the ruminant. O and B books, Inc Cornell University U.S.A. pp. 373.

Whiteman P C 1980 Tropical Pasture Science. Oxford University Press. New York. pp. 242.

Received 29 September 2006; Accepted 10 November 2006; Published 1 January 2007

Go to top