African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 10, Num. 1, 2002, pp. 39-49

African Crop Science Journal, Vol. 10. No. 1, 2002, pp. 39-49

YIELD RESPONSE OF BAMBARA GROUNDNUT  TO PLANT POPULATION AND SEEDBED TYPE

F.L. Mkandawire and K.P. Sibuga¹

^{University of East Africa, P.O. Box 2500, Eldoret, Kenya}
1Sokoine University of Agriculture, Department of Crop Science and Production, P.O. Box 3005, Morogoro, Tanzania

(22 April, 2001; accepted 16 October, 2001)

Code Number: cs02005

ABSTRACT

A field experiment to study the effect of plant population and seedbed type on yields of bambara groundnuts was conducted at Morogoro in Tanzania.  The experiment was designed with three seedbed types split-plot; flat, ridge and furrow in the main plots and bambara groundnut population densities (9, 13, 22 and 66 plants m^-2) in the  sub plots.  At 63 days after sowing (DAS), leaf area index (LAI) was highest for 66 plants m^-2 and/or when planted on the ridge.  In the short rain season, bambara groundnuts planted in the furrow at 13 plants m^-2 had the highest (P< 0.05) grain yields (519.2 kg ha^-1) and harvest index (62.5).  Shelling percentage values were highest across all seedbed types (average 74.2%) at 13 plants m^-2.  In the long rains, LAI was also the highest (P< 0.05) for 66 plants m^-2 at 63 DAS, but differences in LAI between seedbed types were not significant.  Crop performance at 22 plants m^-2 on a flat seedbed, earthed at weeding, was generally superior as it gave the  highest (P< 0.05) pod yield (798 kg ha^-1), grain yield (585.1 kg ha^-1), and harvest index  (75.0).  In both the short and long rainy seasons, yields were significantly reduced to the lowest level at 66 plants m^-2 regardless of the seedbed type.  

Key Words: Harvest index, leaf area index, population density, seedbed, Tanzania, Vigna subterranea

RÉSUMÉ

Une expérience de terrain pour étudier les effets de la population des plantes et le type de semis sur le rendement de l'arachide bambara était conduite à Morogoro en Tanzanie. Les expériences étaient préparées sur trois types de semis: split-plot; plat, billons et sillons dans la parcelle principale et l'arachide bambara de densité (9, 13, 22 et 66 plantes m^-2) dans les parcelles secondaires. Soixante et trois jours après la semence, l'indice de surface foliaire (LAI) était le plus élévé pour 66 plantes m^-2 et / ou pour les plantes sur des stries. Durant la courte saison pluvieuse, l'arachide de bambara plantée sur les sillons à densité de 13 plantes m^-2 avait le rendement le plus élévé de 519.2 kg ha^-1 (P<0.05) et l'indice de production 62.5. Le pourcentage d'ecales était le plus élévé à travers les semis (en moyenne 74.2%) pour la densité de 13 plantes m^-2. Pendant les longues saisons pluvieuses, LAI était aussi le plus élévé à la densité de 66 plantes m^-2.  La différence due aux types de semis ne s'est pas revélée significative. La performance des plantes à la densité de 22 plantes m^-2 sur des semis plats, était généralement supérieure concernant la production en gousses (798 kg ha ^-1) et l'indice de récolte (75.0). Pour les deux saisons, les rendements étaient les plus bas à la densité de 66 plantes m-2 quelque soit le type de semis.

Mots Clés: Indice de récolte, indice de surface foliaire, densité de la population, semis, Tanzanie, Vigna subterranea

INTRODUCTION

Bambara groundnut [Vigna subterranea (L.) Verde], also known as bambaranut, is a legume indigenous to tropical Africa and is primarily grown by subsistence farmers, mostly women (Rachie, 1979).  Studies from different parts in Africa report large variation in seeding rates (Linnemann, 1992).  Earlier studies in the highly leached soils of Bukoba in North-western Tanzania (Dunbar, 1969) indicated that farmers sow bambaranuts at an average spacing of 30 cm x 30 cm.  In West Africa, Ameyaw and Doku (1983) recommended a spacing of 60 cm x 30 cm.  Duke et al. (1977) similarly reported seed rate variations from 25 - 75 kg ha^-1, with inter-row and intra-row spacing varying from 30 cm -75 cm and 10-50 cm, respectively.  Bambara groundnuts' reaction to population density also varies with location and cropping systems.  At Chitala in Malawi, a population density of 167, 4000 plants ha^-1 gave the highest yields while in Thuchila, also in Malawi, high yields were obtained at a lower population density of 83,720 plants ha^-1 (Malawi, Agriculture Research Council, 1972; 1975, cited by Linnemann, 1992).  Matelerkamp (1988) reported that under conditions of moisture stress, high plant population density can depress yields.  

Comparative studies on seedbed types are few.  In Zambia, it was shown that planting on  flat (30 cm x 30 cm) with or without earthing-up resulted in no significant differences in yield (Kannaiyan, 1988; cited by Linnemann, 1992).  In another study at Ukiriguru in Tanzania, planting on the ridge or flat land resulted in no significant difference in yields (Tanzania, Ministry of Agriculture, 1970).

Regardless of the population density and seedbed type at the time of sowing, the performance of bambara groundnut is also determined by the day length.  For this crop, photo-regulation is an important trait with specific day-length requirements for successive stages of development.  Nikishitani et al. (1980) tested the response of 21 varieties of bambara groundnut from Indonesia and Africa to photoperiods of 8 to 24 hours.  The results indicated that even though 19 varieties flowered under both photoperiods, there were fewer matured pods per plant under short day conditions.  Other studies  by Linnemann (1993) have also indicated  that if exposed to continuous long days of 14 and 16 hours, some bambara groundnut accessions fail to produce pods.  These results apparently indicate the stronger effect of photoperiod on the beginning of fruit-set than on the beginning of flowering.

Based on studies conducted in Botswana, Harris and Azam-Ali (1993) indicated need to investigate the relationship between population density, planting date and water supply in V. subterranea. The wide variation in yield response to planting densities and seedbed types suggests need to establish optimum plant density and seedbed combinations under different agro-climatic conditions in order to enhance bambara groundnut production.  The purpose of this investigation was to determine the growth and yield response of a local bambara groundnut landrace to plant population density and seedbed types in an area receiving bi-modal rainfall.

MATERIALS AND METHODS

Field experiments were conducted at Sokoine University of Agriculture Farm situated at latitude 6^o45' South and longitude 37^o40' East near Morogoro (525 m.a.s.l.) in Tanzania during the short rain (RS) season (September-December, 1994) and long rain (RL) season (February-June, 1995).  The experimental design was a split-plot in a randomised complete block with three replicates.   The main plot treatments were three types of seedbed; flat  (S1), ridge (S2) and furrow (S3).  In the furrow, bambara groundnut seed was sown three quarters down the height of a 35 cm ridge.  On the ridge, seed was sown on the highest point of the ridge.  The sub-plot treatments were plant population densities designated  as  T1 = 9  plants m^-2  (94,444  plants  ha^-1), T2 = 13 plants m^-2 (133,333 plants ha ^-1, T3 = 22 plants m^-2 (222,222 plants ha^-1) and T4 = 66 plants m^-2 (666,666 plants ha^-1) achieved by maintaining a constant inter-row spacing of 30 cm and varying intra-row spacing of 35, 25, 15 and 5 cm, respectively.

Currently, there are no improved varieties of bambara groundnuts in Tanzania and farmers plant different landraces in different locations.  The seeds for planting in this experiment were a cream  landrace (non-eyed), bought from an open market in Dodoma (5^o45'S) in central Tanzania.  Two seeds per hill were panted at a depth of 4 - 5 cm on November 8, 1994 and March 3, 1995.  Germination and emergence occured 21 days after sowing (DAS).  Consequently, thinning to one seed per hill was done 27 days after sowing in both seasons. Prior to sowing, triple super-phosphate at a rate of 40 kg P2O5 ha^-1 was broadcast over each plot and incorporated into the soil.  Plants were earthed-up ten days after the first flowers were noticed.

Plots were 3 m x 3 m and consisted of 10 rows.  Four rows, two on each side of the plot next to the outermost boarder row, were used for destructive sampling in the process of assessing growth measurements.  One inner border row on each side enclosed the two center rows used for final harvest.  Sequential growth measurements were made on three plants uprooted randomly from the sampling area.  Sampling was done at 14 days intervals beginning 21 up to 105 DAS.  At sampling, three plants per plot were used to determine leaf area index (LAI).  Total above ground dry matter was determined at harvest by removing the shoot from all plants in the harvest area and drying in an oven at 60^oC for 72 hours.  

RESULTS

Weather conditions. During the short rain  season, the total amount of rainfall received was 333 mm  which fell intermittently; the crop suffered periodic droughts particularly in the reproductive phase.  Total rainfall received in the long rain season was 583 mm and was relatively better distributed throughout the season.  The average sunshine hours day^-1 was 7.1 and 5.5 in the short rain and the long rain seasons, respectively.  Between March and June, when the long rain season crop was in the field, the average ambient temperature of 23.6^oC was about 3^oC lower than the ambient temperature during the period for which the short rain  season crop was in the field.

Leaf area and dry matter production. Maximum leaf area index (LAI) for the different seedbed types was attained at 63 DAS (Figs. 1 and 2).  However, in the short rains, the  highest LAI (8.5) (P < 0.05) was recorded for bambara groundnut planted on the ridge (Fig. 1).  In the long rains, leaf expansion was considerably reduced reaching a maximum LAI of 2.9 for the crop planted in the furrow (Fig. 2).  All LAI values in response to changes in plant density were highest at 63 DAS in both seasons (Figs. 3 and 4).  The plant density of 66 plants m^-2 had significantly higher  LAI both in the short rain season (LAI = 9.5) and long rain (LAI = 5.6) season, while the lowest plant density of 9 plants m^-2 also maintained the lowest LAI in both seasons (Figs. 3 and 4).

Yield variables. During the short rains, pod yields were highest (P < 0.05) overall for bambara groundnuts planted on the ridge and lowest  for the nuts planted on the flat.  In response to plant density, pod yield was  highest  overall at the  population  density  of  9 m^-2 (538 kg ha^-1), but declined to the lowest level (380 kg ha^-1) with increase in plant density to 66 plants m^-2 (Table 1).  The interaction of plant density and seedbed type gave highest pod yields also at 9 plants m^-2 for the nuts planted on the ridge.  During the long rains, bambara groundnuts planted on the flat surface gave  the  highest (P < 0.05)  pod yields (597 kg  ha^-1) compared to other seedbed types.  During this season, the plant density of 22 plants m^-2 was generally favourable for high pod yields across  seedbed types, particularly on the flat seedbed (Table 1).

In both seasons, differences in grain yields were significant (P < 0.05) for seedbed type, plant population density and the interaction of the two factors (Table 2).  Planting in the furrow during the short rain season generally enhanced grain yield compared to planting on a flat or ridge seedbed.  At the highest plant density of 66 plants m^-2, planting on the flat seedbed gave the lowest (P < 0.05) grain yields.  Generally, planting on the flat or ridge was least favourable for grain yields during the short rains across all plant densities.

In the long rain season, yields were generally higher than those realised from the short rain  season crop regardless of the seedbed type or plant density.  Planting on the flat resulted into significantly (P < 0.05) high grain yields (449 kg ha^-1), while planting in furrows or on ridges significantly (P< 0.05) depressed yields to 360 and 324 kg ha^-1, respectively, (Table 2).  On the flat seedbed, grain yields increased with increase in plant density up to 22 plants m^-2 when the highest yields (585 kg ha^-1) were achieved and, thereafter, declined.  Averaged over seedbed types, 22 plants m^-2 was most productive and 66 plants m^-2 was the least productive (Table 2).

The highest  mean shelling percentage across all seedbed types was recorded at plant densities of 13 and 22 plants m^-2   for the short rain season crop and at densities of 9 to 22 plants m^-2 during the long rains (Table 3).  Shelling percentages were generally higher during the long rain   season for which planting on the flat or ridge was most favourable at 22 plants m^-2.

During the short rains, the biomass of above-ground vegetable growth was highest (P< 0.05) for 66 plants m^-2 across seedbed types (Table 4),  while the reverse was true  during the long rain  season.  The lowest  above-ground dry biomass, across plant population densities, was recorded for bambara groundnuts grown in  furrows (916 kg ha^-1) during the short rains or on the ridge (642 kg ha^-1) during long rains.  Harvest index values were significantly (P< 0.05) highest for bambara groundnuts grown on the flat or ridge during the long rains  (Table 5) or in furrows during the short rains.  Harvest index values were also highest  (P<0.05) for bambara groundnuts planted at 9 or 13 plants m^-2 in the short rain season or at 22 plants m^-2 during the long rains.

DISCUSSION

In both seasons, LAI increased with population density.  The highest LAI values for the long rains (5.8 cm²) and short rains (9.8 cm²) were  recorded at a population density of 66 plants m^-2.  The increase in LAI with increase in plant density, however, did not necessarily result in high yields.  This is in agreement with Weber et al. (1966) on soybean that conditions that maximise LAI are not conducive to high yields because of excessive competition for resources, especially light.  On the average, the short rain season crop recorded higher LAI values than in the long rains.  However, whereas in the short rain season LAI was significantly higher on ridges, there were no significant differences between seedbed types in the long rains during which values were consistently lower.

Overall, the rate of dry matter accumulation was higher during the short rains and this is ascribed to increased solar radiation during this season.  There is evidence that dry matter accumulation is directly related to the amount of solar radiation intercepted (Chavula, 1991).  This was probably the case during the short rain season for which LAI values were high.  The short rain season crop received a total of 333 mm  of rainfall  and  an average 7.1  hours of  sunshine day^-1  compared to 583 mm of rain and 5.5 sunshine hours  day^-1  for  long rain crop.  In addition, the long rain season experienced excessively wet conditions and a progressive reduction in the sunshine hours from 6.6 at the beginning of the season in February to 4.2 during the pod filling stage in May.  The reduced sunshine hours probably slowed down the rate of photosynthesis, reduced the LAI and consequently dry matter accumulation for the long rain season crop.  Hence, shoot and root biomass for the long rain season crop were relatively lower despite the higher precipitation received during this season.  

Increasing the plant population density reflected negatively on pod yields.  During the short rains, 9 plants m^-1 produced the highest pod yields, while  at 66 plants m^-2, pod production was generally lowest regardless of type of seedbed.  During the long rains, however, the optimum population  density  for pod  yield was 22 plants m^-2 on the flat seedbed with no apparent yield advantage due to increase in  plant density to 66 plants m^-2.  Grain yield followed a similar trend for the long rain season crop, but in the short rains, grain yields were significantly higher overall for 9 or 13 plants m^-2.  Similar pod yield decreases have been reported in other studies. For instance, cumberland (1978) reported high pod yields of bambara groundnut at densities of 7 and 14 plants m^-2. Similarly, Eliesen and Freira (1992) working with groundnuts and  Edje et al. (1971) with beans reported a decrease in  number of pods plant^-1 with increase in plant population.  Usually, with increase in plant density, the combined demand for nutrients and light results into competition for these resources unless the supply is either unlimited or has been somehow supplemented.  In this study, all treatments in a season were subjected to the same level of fertiliser and rainfall regime, hence, as plant population increased, intra-plant competition came into effect  and probably caused  the observed decline in pod yields.  

The higher grain yields for the long rain season crop compared to the short rain season crop is attributed to increased pod production and enhanced grain filling for the long rain season. It is anticipated that the photosensitivity of bambara groundnut also contributed to the differences in grain yields between the two seasons.  However,  this was probably not a significant factor given the minor difference in the latitude between the location from where the seed was obtained and the experimental site.  The cream landrace used in this study was obtained from Dodoma (5^o45'S) where the day length is 12.5 to 13.2 hours day^-1 (Brink et al., 2000) and then planted in Morogoro (6^o45'S) where the day length is 11.7 to 12.3 hours day^-1 (Collinson et al., 2000). The shorter days coincide with the long rain season. The results obtained indicate lower yields in the short rain  season, thus reflecting  the indirect effects of day length on the reproductive development of bambara groundnut.  Linnemann (1993), in a greenhouse experiment with accessions from Nigeria (7^o22'N), reported that lengthening the photoperiod between 10 and 16 hours delayed both flowering and first fruiting but the effects were stronger on fruit set than on  flowering.  Similarly in related trials with accessions from different locations in Africa, Linnemann and Azam-Ali (1993b) showed that long photoperiods of 14 hours or more delayed or even prevented fruit set in some accessions.   Generally, 14 to 16 hours day length  is reported to prevent pod set in some bambara groundnut accessions (Linnemann, 1993). However, Morogoro lies outside this bracket. By implications, therefore, the relatively lower pod and grain yields in the short rain season were to  a greater extent probably limited  by other factor(s) other than day length variation.  One such factor could be the low total precipitation and subsequent reduced moisture availability in this season.  

Shelling percentage is a reflection of the pod filling efficiency.  High shelling percentage values are an indication of effective pod filling.  Shelling percentage was generally lower in the short rains compared to the long rains.  The short rain season  also received relatively less rainfall (333 mm) compared to the 583 mm received by the long rain  crop.  The short rain season is when day length is relatively longer, with a difference of about 36 minutes between the shortest and the longest day.   Even though the difference in day length is rather small, the relatively extended day length (Linnemann, 1993), could have  delayed fruit set and thus pod filling occurred mostly at the "tail end" of the season when it was relatively drier.  Since podding was not completely prevented by the relatively long days in the short rain season, the overall implication is that in this season,  drought was probably much more limiting than day length.  The  results obtained in this study further indicate that if stress conditions prevail, such as the intermittent droughts experienced by the crop during the short rain season, a large leaf area alone cannot guarantee high grain yields.  Hence,  grain yields in the short rain season crop were low despite the higher LAI.  The lack of correlation between LAI and grain yield has also been reported in other crops such as groundnuts.  For instance,  Sibuga et al. (1990) reported no grain yield advantage with increase in leaf area.  Collinson et al. (2000), in bambara groundnut time of sowing experiments similarly associated low pod and grain yields to later planted crops which also suffered from drought that limited the convention of solar radiation into dry matter for grain production.  

Apart from the effects of drought, the short rain season crop also suffered  heavy infestation by powdery mildew (Sphaerotheca voandzeia) and Cercospora leaf spots (Cercospora canescens and C. voandzeia).  These diseases infested the crop at the critical pod filling stage (77 DAS) and caused premature senescence of leaves, which denied the crop its photosynthetic machinery for continued synthesis and transfer of photosynthates into the major sink (the seed).  The detrimental effects of Cercospora leaf spot on bambara groundnut yields have similarly been reported  in Ghana (Lamptey and Ofei, 1977).  However, despite the combined stress due to disease and intermittent drought, particularly during the short rain season, the range of shelling percentages reported here (52-86%) is much higher than the results reported in other situations.  Linnemann and Azam-Ali (1993a) recorded shelling percentages of bambara groundnut as low as 50% under conditions of moisture stress.  The ability of the crop to perform relatively better in this study confirms both the variable response of bambara groundnuts to different environments as well as the ability of bambra groundnuts to grow and produce grain even under stress conditions.

On the other hand, the results obtained in this study have also shown that planting in furrows during the short rains enhanced pod and grain production, while in the long rains, grain yields were enhanced if bambara groundnuts were planted on  flat beds.  It was observed that during the short rains, furrows acted as micro-watershed in which runoff water from the ridges and other parts of the field was concentrated, resulting into better performance of the crop during periods of intermittent droughts. On the contrary, during  the long rain season, furrows collected much water as to create occasions of water logged conditions for the furrow-planted crop.  In Zambia, Kannaiyan, (1988, cited by Linnemann, 1992) and Rweyemamu and Boma (1990) in Tanzania reported high yields of bambara groundnut and groundnut (Arachis hypogaea), respectively, obtained on a flat seedbed in the wet season and attributed this to good drainage.  Rweyemamu  and Boma (1990) further noted that planting in  furrows during the long rains was unfavourable for groundnut production.

On the basis of harvest index (HI) values, the long rain season was more productive than the short rain crop.  In the long rain season, the flat seedbed (HI = 57.4) and ridge seedbed (HI = 58.5) were, on the average, the most highly productive and the plant density of 22 plants m^-2 was most productive especially on the flat seedbed.  The HI values reported here collaborate with the HI values of 30 to 60 reported by Squire (1990) for indeterminate crops under wet conditions.  In the short rain season, on the other hand, the furrow seedbed with an average HI of 44.7 across all plant densities was the most productive.  This was, particularly the case at the lower plant densities of 9 or 13 plants m^-2, where the recorded HI values of  of 59.9 and 62.5, respectively, were significantly highest.  It has been postulated (Deloughery and Crookston, 1979) that a sparse stand will use the store of water in the soil more rapidly than a dense stand and will, therefore, have a greater partition factor for grain.  Thus, under dry conditions, the HI usually increases as the plant population density decreases.  In this study, plant densities of 9 and 13 plants m^-2 were significantly more productive than at higher densities during the short rains. On the other hand,  the intermediate density of 22 plants m^-2 was significantly more productive than plant densities below or above this level in the long rains.  High economic yields are pre-determined by dry matter production and partitioning into various sinks of which the grain is the most important.  Therefore, any attempts to manipulate plant spacing to maximise yields are considered successful if subsequent growth characteristics support dry matter partitioning into grain.

These results reveal that seedbed types and population densities influenced bambara groundnut grain yields during both seasons.  Planting bambara groundnuts in  furrows  enhanced grain yield during the short rains, while planting on the flat beds followed by earthing-up at first flower enhanced grain yields during the long rains.  The results also indicate that during the short rains when total precipitation is low, sowing at not more than 13 plants m^-2 is ideal, while a higher plant density of 22 plants m^-2 is recommended for the long rain season crop.

REFERENCES

• Ameyaw, M.V. and Doku, E. V. 1983.  Effects of soil moisture stress on the reproductive efficiency and yield of bambara groundnut (Vondzeia subterranea).  Tropical Grain Legume Bulletin 28:23-29.
• Brink, M., Sibuga, K.P., Tarimo, A.J.P. and    Ramolemana, G.M. 2000. Quantifying photothermal influences on reproductive development in bambara groundnut (Vigna Subterranea): Models and their validation.  Field Crops Research 66:1-14.
• Chavula, K. M. 1991.  Growth, light interception and yield of bambaranut [Vigna subterranea (L.) Verdc.] in response to soil moisture.  MSc. Thesis, University of Nottingham.
• Collinson, S.T., Sibuga, K.P., Tarimo, A.J.P.  and Azam-Ali, S.N.  2000. Influence of sowing date on the growth and yield of bambara groundnut landraces in Tanzania.  Experimental Agriculture 36:1-13.
• Cumberland, J. 1978.  Jugo beans:  Research Report No. 19.  Agriculture Research Division, Ministry of Agriculture; Malkerns Research Station, Swaziland.  11pp.
• Duke, J.A., Okigbo, B. N., Reed, C. F. and Weder, J. K. P. 1977.   Voandzeia subterranea (L.)  Thouars.  Tropical Grain Legume Bulletin 10:8-11.
• Dunbar, A. R. 1969.  Bambara groundnut (Voandzeia subterranea).  In:  The Annual Crops of Uganda.  East African Literature Bureau, Nairobi, Kenya.   85 pp.
• Edje O.T., Mughogho, L. K. and Ayonoadu, U.W.U. 1971.  Effects of row width and plant spacing on the yield of canning beans.  Lilongwe, University of Malawi.  Bunda College of Agriculture.  Research Bulletin 2:29-36.
• Eliesen, J. M. and Freira, M. J. 1992.  Effect of plant density and sowing pattern on the yield of a groundnut and maize intercrop.  In:  Proceedings of the Fifth Regional Groundnut Workshop for Southern Africa. Nageswara Rao, C. and Subrah, M.P (Eds.), pp. 121-124. 9-12 March 1992, Lilongwe, Malawi.
• Harris, D.  and Azam-Ali, S.N.   1993. Implications of day length sensitivity in bambara groundnut (Vigna subterannea) for production in Botswana.  Journal of Agricultural Science 120:75-78.
• Lamptey, P. and Ofei, I.O. 1977.  The effect of leaf spot caused by Cercospora canescens (Ellis and Martin) on the yield of bambara groundnut (Voandzeia subterranea Thours.)  Ghana Journal of Agricultural Science 10: 113-115.
• Linnemann, A. R. 1992.  Bambara groundnut (Vigna subterranea) literature review: A revised and updated bibliography.  Tropical Crops Communication No. 7.  Wageningen Agricultural University. 124pp.
• Linnemann, A. R. and Azam-Ali, S. 1993.   Bambara groundnut (Vigna subterranea) literature review:  A revised and updated bibliography.  Tropical Crops Communication No. 7.  Wageningen Agricultural University. 124pp.
• Linnemann, A. B. 1993. Phenological development in bambara groundnut (Vigna subterannea) at constant exposure to photoperiods of 10-16^th.  Annals of Botany  71:445-452.
• Metelerkamp, H. R. R. 1988.  Review of crop research relevant to semi-arid areas of Zimbabwe.  In:  Cropping in the semi-arid areas of Zimbabwe Proceedings of a Workshop held in Harare 24^th to 28^th August 1987.  1:285-286.
• Nikishitani , T., Muruki, K. and Inouye, J. 1988.  Effects of daylength on the flowering and fruiting in bambara groundnut (Vigna subterranea (L.) Verdc.). Japanese Journal of  Tropical Agriculture 32:80-84.
• Rachie, K.O. (Ed.). 1979.  Tropical legumes:  resources for the future.  National Academy of Sciences, Washington D.C. P331.
• Sibuga, K. P., Songambele, S. S. and Kakema, T.P. 1990.  Influence of sowing dates on growth and yield of groundnuts in Tanzania.  In:  Proceedings of the Fourth Regional Groundnuts Workshop for Southern Africa.  19-23 March, 1990, Arusha, Tanzania. pp. 183-188.
• Tanzania:  Ministry of  Agriculture.  1970.  Ukiriguru Research Station Annual Report. 1970.  115pp.
• Weber, C. R., Shibles, R. M. and Byth, D.E. 1966.  Effect of plant population and row width on soybean  development and production.  Agronomy Journal 58:99-102.

©2002, African Crop Science Society

The following images related to this document are available:

Photo images