INTERCROPPING Proceedings of the Second Symposium on Intercropping in Semi-Arid Areas, Held at Morogoro, Tanzania, 4-7 August 1980 INTERCROPPING

IDRC-186e

INTERCROPPING Proceedings of the Second Symposium on Intercropping in Semi-Arid Areas, held at Morogoro, Tanzania, 4-7 August 1980 INTERCROPPING

Proceedings of the Second Symposium on Intercropping in Semi-Arid Areas, held at Morogoro, Tanzania, 4-7 August 1980

Editors: C.L. Keswani and B.J. Ndunguru

University of Dar es Salaam Tanzania National Scientific Research Council International Development Research Centre The International Development Research Centre is a public corporation created by the Parliament of Canada in 1970 to support research designed to adapt science and technology to the needs of developing countries. The Centre's activity is concentrated in five sectors: agriculture, food and nutrition sciences; health sciences; information sciences; social sciences; and com- munications. IDRC is financed solely by the Parliament of Canada; its policies, however, are set by an international Board of Governors. The Centre's headquarters are in Ottawa, Canada. Regional offices are located in Africa, Asia, Latin America, and the Middle East.

Keswani, C.L. Ndunguru, B.J.

University of Dar es Salaam, Dar es Salaam TZ Tanzania National Scientific Research Council, Dar es Salaam TZ International Development Research Centre, Ottawa CA IDRC-186e

Intercropping : proceedings of the Second Symposium on Intercropping in Semi-Arid Areas, held at Morogoro, Tanzania, 4-7 August, 1980. Ottawa,

Ont., IDRC, 1982. 168 p. : ill.

/Intercropping/, /semi-arid zone/-/agricultural research/, /Africa/, /culti- vation practices/, /plant breeding/, /plant protection/, /crop yield/, /research results/, /research methods/.

UDC: 631.584(213) ISBN: 0-88936-318-8

Microfiche edition available Contents

Foreword R. Bruce Scott 7 Addresses to the Participants Welcoming address N.A. Kuhanga 10 Opening address Hon John S. Malecela 12 Agronomy Summary and conclusions B.J. Ndunguru 16 Comparative development and yield and other agronomic characteristics of maize and groundnut in monoculture and in association O.T. Edje 17 Evaluation of soil-testing methods for available potassium in some soils of Morogoro B.R. Singh, A.P. Uriyo, M. Kilonde, and John J. Msaky 27 Intercropping maize or millet with soybean, with particular reference to planting schedule E.N. Nnko and A.L. Doto 33 Some observations on the effects of plant arrangements for intercropping K. W. May and R. Misangu 37 Agroforestry: preliminary results of intercroppin acia, Eucalyptus, and Leucaena with maize and beans J.A. Maghe e and J.F. Redhead 43 Intercropping under marginal rainfall conditions in Kenya Hassan M. Nadar 50 Influence of plant combinations and planting configurations on three cereals (maize, sorghum, millet) intercropped with two legumes (soybean, green- gram) D.B. Nyambo, T. Matimati, A.L. Komba, and R.K. Jana 56 Density of dry beans (Phaseolus vulgaris) interplanted with maize (Zea mays) - summary W. de Groot 63 Evaluation of phosphorus placement methods and nitrogen carriers under conditions of maize-bean intercropping - summary Andrew P. Uriyo, Budh R. Singh, and John J. Msaky 65 Effect of planting schedule and intercropping systems on the production of green-gram (Phaseolus aureus Roxb.) and bulrush millet (Pennisetum americanum (L.) Leeke) - summary K. W. May 66 Influence of intercropping methods on foliar NPK contents and yields of maize and cowpeas - summary H.O. Mongi, M.S. Chowdhury, and C.S. Nyeupe 67

3 Modifying the competitive relationship in maize-bean mixtures in Kenya - summary O.E. Hasselbach and A.M.M. Ndegwa 68 Physiological aspects of maize and beans in monoculture and in association - summary O.T. Edje and D.R. Laing 69 The relative importance of above- and below-ground resource use in deter- mining yield advantages in pearl millet/groundnut intercropping - summary M.S. Reddy and R. W. Willey 70 Effects of moisture availability on intercropping and yield advantages - summary M. Natarajan and R. W. Willey 71 Performance of a maize-legume intercrop system in Sri Lanka - summary H.P.M. Gunasena 72 Effect of minimum tillage, mulches, and fertilizers on intercropped cowpeas with maize - summary A.A. Mashina and R.K. Jana 73 Increased resource exploitation through intercropping with cassava - summary G.F. Wilson and T.L. Lawson 74 Groundnut-maize interplanting in southern Mozambique - summary A.D. Malithano and J. van Leeuwen 75

Plant Breeding Summary and conclusions A.L. Doto 78 Genotype evaluations and implications for adapting plant material for intercropping K. W. May and R. Misangu 79 Soybean-cereal intercropping and its implications in soybean breeding M.M. Makena and A.L. Doto 84 Genotype identification for intercropping systems - summary D.S.O. Osiru 91

Plant Protection Summary and conclusions C.L. Keswani 94 A study of crop/weed competition in intercropping N.R. Mugabe, M.E. Sinje, and K.P. Sibuga 96 Intercropping of maize and cowpea: effect of plant populations on insect pests and seed yield A.K. Karel, D.A. Lakhani, and B.J. Ndunguru 102 Effect of intercropping on the severity of powdery mildew on green- gram C.L. Keswani and R.A.D. Mreta 110 Bean production in monoculture and in association with maize: the effect of diseases and pest incidence - summary H.A. Van Rheenen, O.E. Hassel- bach, and S.G. Muigai 115

4 Effect of intercropping on some diseases of beans and groundnuts - summary J.K. Mukiibi 116 Effect of insecticide spray on insect pests and yield of sorghum and simsim in pure stand and in intercropping - summary D. Kato, A.K. Karel, and B.J. Ndunguru 117

Farming Systems Summary and conclusions B.J. Ndunguru 120 The use of farming systems research for understanding small farmers and improving relevancy in adaptive experimentation M.P. Collinson 121 Asian experience in cropping systems research Gordon R. Banta 126 An experimental approach for improving present cropping systems in tropical Africa Peter Vander Zaag and Pierre Tegera 131 Farming systems economics: fitting research to farmers' conditions J.W. Gathee 136 On-farm experiments: some experiences C.N. Muriithi 141 Interaction between agronomic research and agricultural economic analysis to develop successful dryland cropping systems in Kenya N.M. Nadar and Gordon E. Rodewald 146 Farming systems and farming systems research in Morogoro - summary P. Anandajayasekeram 155 Farming systems research in Uganda: past performance and future prospects - summary I. Fendru 157 Mixed cropping in Tabora region - summary J.E. Mansfield 158 Farming systems research questions - summary C.D.S. Bartlett and E.A.M. Okarie 160

Concluding Remarks and Participants Concluding Remarks R. Bruce Scott 162 Participants 164

5 A Study of Crop/Weed Competition in Intercropping

N. R. Mugabe, M. E. Sinje, and K. P. Sibuga Department of Crop Science, Faculty of Agriculture, Forestry and Veterinary Science, University of Dar es Salaam, Morogoro, Tanzania

Adverse effects of weeds on crops were appa- control methods, which mainly involve hoe weed- rent even to pioneer plant growers (Donald 1963). ing, are still the common practice in Tanzania. The Consequently, weed control is one of the oldest latter is a painstakingly slow and laborious exer- basic agricultural practices in crop production. cise, which is usually accomplished (if at all) after Subsistence farmers of the tropics spend more weeds have long since exerted their depressive time and energy on weed control than on any effects on crops. other aspect of crop production (Terra 1959; Traditional cropping systems in Tanzania in- Kasasian 1971). clude various methods of mixed cropping. Most of Crop yields are depressed when weeds com- these mixtures involve cereals and legumes. If pete with crops for available nutrients, water, light, cereal/legume intercrops could effectively carbon dioxide, and agents of pollination (Donald smother weeds, a great deal of expense and en- 1963). Hence, most weed control methods are ergy would be saved by the small farmers of designed to minimize weed/crop competition. Tanzania. With this in mind, an experiment was Crops and weeds have different competitive carried out to find the best cereal/legume com- abilities, with some weeds being more competitive binations that would effectively smother weeds. than others. Nieto et al. (1968) reported that maize infested by weeds for the first 30-50 days suffered greatly Materials and Methods from weed competition. Season-long infestations The experiment was conducted at the university reduced maize yield from a potential of 4770 kg/ farm of the Faculty of Agriculture, Forestry and ha to 382 kg/ha and bean yield by 98%. Some Veterinary Science, Morogoro, Tanzania, during crops, however, are capable of outcompeting the March-July growing season of 1976. It was weeds. Bulrush millet was found to suffer very little carried out near the Ngerengere River. This area from weed competition (Kasasian 1971). In the of the university farm is considered to be low lying. same study, sorghum did not compete well with Despite the erratic rainfall of that year, the area weeds. was not seriously affected by moisture stress. Various techniques, including crop associa- The area was plowed and harrowed using a tions, herbicides, heavy seeding rates, and fertiliz- tractor. Plots of 5 m x 3 m were prepared using ers, have been employed to improve crop com- hoes. Maize cultivar MAS, a tall, high-yielding petition over weeds (Fine and Fletchall 1964; variety; millet cultivar Serere 17, a medium-height Esipov and Scherbakova 1970; Hammerton variety; and sorghum cultivar 9D x 99, a short 1962). Evans and Sreedharan (1962) showed variety, were the cereals grown. The legumes that a denser population, which usually results grown were soybean cultivar 1H/192, of determin- from intercropping systems, helps to control ate growth habit, which is recommended for weeds. They concluded that intercropping Morogoro, and green-gram, an erect, drought- systems have a much more competitive ad- resistant, local variety. Each cereal was grown in vantage over weeds than solid seedings. monoculture and intercropped with each legume, In East Africa, not much research has been and the legumes, in turn, were grown in mono- conducted on weed control. Traditional weed culture. Some of the materials and methods em-

96 ployed in the experiment are shown in Table 1. ling weed growth, followed by sorghum-green- The experiment was a split plot design in a rando- gram and maize-green-gram mixtures. Compared mized complete block layout with four replica- with other cereal - legume mixtures, or even with tions. The main plots were cereals and the sub- millet and maize monocultures, sorghum and plots were legumes. maize intercropped with soybean were not effec- Fertilizer was applied at planting at rates of 50 kg tive in controlling weed growth. Under mono- P/ha in the form of triple superphosphate and 50 culture, both legumes were heavily outcompeted kg N/ha in the form of ammonium sulphate. The by weeds. latter was applied again at the same rate 5 weeks These results agreed with the observations of after cereal planting. Sorghum, millet, and green- Rao and Shetty (1977) that crops differ in their gram were sown with 6 seeds/hole and maize and growth rates, spreading habits, height, canopy soybean were sown with 2 seeds/hole. Thinning to structures, and inherent competitive characters, 1 plant was carried out 2 weeks after sowing each and, therefore, their weed-suppressing ability. crop. Legumes were sown between rows of cere- Rao and Shetty (1977) also found that pearl mil- al. Crops were left unweeded in order to compete let, through its tillering habit, and maize, as a result with the weeds throughout the growing season. of its high leaf canopy, as well as by virtue of their Weeds were counted and harvested at crop fast growth, were highly competitive over weeds. harvest from 1 m2 sample areas in two opposite In the present study, bulrush millet cultivar Serere corners of the rectangular plots to give the average 17 grows quickly, producing many tillers. This dry weight of weeds per plot. Grain yield of both growth habit gives it an early start toward forming cereals and legumes was determined at final a closed canopy, which, in turn, suppresses weed harvest from an 8-m2 area. An analysis of variance growth. Maize variety MAS, having the genotypic and an F-test were conducted for each of the qualities mentioned above, is a tall variety. The tested parameters, followed by Duncan's multi- combination of these qualities gave maize an early ple-range test to determine the significant factors. start toward producing a bigger and higher canopy, which intercepted much of the incident light and enabled it to compete better for other inputs and, consequently, created an environ- Results and Discussion ment unfavourable for weed growth. On the other Intercropping resulted in less harvestable weed hand, sorghum cultivar 9D x 99 is a slow- dry matter than monocropping. These results de- growing, short variety. This habit made sorghum, monstrate that intercropping compared with comparatively, much less competitive with weeds, monocropping is more effective in controlling some of which were fast growing, tall, and tiller weeds. This agrees with Evans and Sreedharan forming, e.g., Rottboellia exaltata. The legumes, (1962) and Rao and Shetty (1977) who observed being low-growing crops, were completely out- that intercropping has a greater competitive ad- competed by weeds, most of which were species vantage over weeds than monocropping. They reputed for their competitive ability over low- attributed this enhanced competitive ability of growing crops (Ivens 1967). The importance of intercropping mainly to the high population pres- choosing crops with superior competitive qualities sure provided by the component intercrop species over weeds for intercropping programs designed together. The crop densities that resulted from the to control weeds was demonstrated by these re- two cropping systems under study (Table 2) sults. appear to confirm these findings. Millet was found to combine well with both Bulrush millet in monoculture or under inter- legume types. Maize and sorghum combined well cropping was the most effective crop for control- with green-gram but not with soybean. These re-

Table 1. Some of the materials and methods employed in the experiment.

Plant spacing Plant population (cm) Planting date per hectare Maize 75 x 30 18 March 1976 42000 Sorghum 60 x 15 18 March 1976 107000 Millet 60 x 15 18 March 1976 107000 Green-gram 60 x 15 30 March 1976 107000 Monoculture 60000 Intercropped Soybean 60 x 15 30 March 1976 120000 Monoculture 94000 Intercropped

97 Table 2. Mean weed species count per square metre Crop combinations Millet Millet Maize Maize Sorghum Sorghumo 0 0 E Green- + + + + + + Dominant weed species MaizeCN Millet0 Sorghum Soybean gram green-gram soybean green-gram soybean soybean green-grama CU 71 E cINI11cc H MI rI°III ,TZIr-U Y 0Eozs-5 .c U o ov0EvE bl i 0 O Heckelochioc granulans 27.0 7.0 3.8 I 7.3 5.3 3.8 3.3 6.3 CO C'?

0 Q - -

° 6 - Rottboellia exaltata 5.3 3.3 1 3 2.0 11.3 - - N -' - - - - Commelina0 0 c benghalensisb 6 0 4.0 2.7 3.0 3.3 2.3 2.0 2.0 - 2.3 3.3 1.8 E=Qn, 0 o o 2.8 Launea comuta - 4.7 3.3 2.0 - - - - - m - TricodesmaU p c) E zeylanicuma g S 0 E 6.0 - 2.0 -I - - 3.4

o - - - - Aeua7 O lanatero 4.5 I

S ------o)

C C o p o 0 E - - - - I Pennisetum polystachum ------1.8 1.5 - 98 This 0.05) total balance have used ulation, pointed suits capacity suppress also ductive the more ity. study. various in crop tribution whole in ping from growth then correlated true, if the significant the intercropping ling systems (Table ceeded maize loa these study, two with smothering dered growing habit, Commelina tive a influenced existed rapid what Broad-leaved The With suppressing the The two growth granularis weed, the relative concluded could weed indicates mixtures associate a are agree legumes poor i.e., competitive crops. the in was a than must Ivens quality 6), experiment strong growing rate, most and cropping regard total and, out work in crops in Rao in luxuriant had the crops, some for of tillering showing that especially with counts) suppressing considered difference favour competitor the the have negative of these in crops. that with seed crowding Although have (1967) two and higher in and number an of benghalensis, and is abundant was bearing nature and existing all that were weed crops weed to and two population aspects intercropping, whereas that shown Haizel and been inherent crop and systems cropping relative other species helped growing Shetty plant Rao production. of against did of The more crop are high quantity cropping neither described plant a the during found spreading crops. plus correlation of growth. growth in of this one grass species, not vigorous R. capable crop weeds. on coefficient, (1973) in given of and the association treatments, competition population, inherent mixtures weed weeds Launea replacement sorghum weed effective exaltata shift Table in (1977) (Tables mechanism the crop R. differ shifting of weed weed densities systems increase growth in early the of varieties, weeds of Shetty with the exaltata. in is systems weeds the R. the proved variety, higher of numbers and weed the management that 2. various significantly maize Table species highly that operative competitive stages exaltata cornuta smothering by and between observed 3, is nature balance The by the relative its of of variety two (Tables very this likely (r 4). grew (and, in there that (1977) is in reducing suppressing such species maize numbers Heckeloch- such such for rate. = crop-weed crop in the and differences 5. plant scrambling capable negatively However, of crops intercrop- crop over cropping competi- between of terms The - is as resulted but control- growth. to was present against hence, factors Haizel repro- weeds weeds 0.98), as consi- in 3, crops could under weed holds being (P pop- to have who abil- low- that suc- that that and dis- the the 4). the no be its = of of in Table 3. Mean count of thin-leaved bulrush millet suffered less from weed competition weeds per square metre. where rainfall was not a limiting factor. lntercropped with Total grain yield positively responded to reduced weed growth as evidenced by the harvest- Monoculture Soybean Green-gram able weed dry matter and grain yield of crops Maize 31.41 33.45 23.64 (Table 6). Millet and maize in monoculture and Millet 26.55 36.30 23.30 intercropped with both legumes resulted in Sorghum 29.95 37.00 81.15 significantly higher grain yields than the rest of the Soybean 32.92 treatments (P = 0.05). The crop associations Green-gram 28.22 most successful in smothering weeds were None: Standard error = ± 4.28. millet-soybean and millet-green-gram respective- ly. Similarly, they gave the best combined yield Table 4. Mean count of broad-leaved under nonweeded conditions (Table 6). From the weeds per square metre. intercropping system, legume yield contributions lntercropped with were almost negligible, i.e., the effective compo- nents of the combined yields were cereals. Monoculture Soybean Green-gram Although the maize-soybean mixture did not ef- Maize 35.10 39.65 29.65 fectively reduce weed growth (weed dry weight), Millet 32.42 35.00 41.65 its grain yield was fairly stable. This would seem to Sorghum 34.06 38.84 36.84 suggest a critical time element to have been in- Soybean 37.78 volved in the maize-weed competition. Maize, Green-gram 37.39 being a tall and fast-growing cereal, had an early NOTE: Standard error = ±6.31. competitive advantage over weeds enabling it to pass the critical period of 20-50 days after planting The environmental factors over which the crops during which maize would suffer from weed com- and weeds could have effectively competed were petition (Nieto et al. 1968). The highest dry weight sunlight and nutrients. Despite the erratic rainfall of weeds was obtained from green-gram mono- experienced during the season, the area where the culture. The effect of such weed-crop competition experiment was conducted was not low enough in resulted in almost total destruction of the crop, soil moisture to affect the growth of the crop types resulting in a grain yield reduction of 97.3%. understudy (although no soil measurements were The results of the percentage reduction in grain taken). The prevalence in this area of such weed yield due to weed competition (Table 6) demon- species as R. exaltata, Commelina spp., and strated that leaving sorghum under monoculture Launea spp., which are associated with low-lying or intercropped, or leaving the tested legume wet areas (Ivens 1967), is proof of the area's varieties under either of the two cropping systems moisture-retention potential. Soil moisture was, studied, through season-long weed infestation is therefore, not the main limiting environmental tantamount to destroying the crop. Furthermore, factor for crop-weed competition during the sea- the results demonstrated that even the best weed son. Evidence that light was the main limiting suppressors suffered appreciably from weed com- factor in the crop-weed competition is indicated by petition, resulting in 46-57% grain yield reduction. the success against weeds that fast-growing crops This is in concordance with Rao and Shetty (1977) such as maize and bulrush millet had as a result of who indicated that intercropping could be a quickly forming closed canopies. This is in agree- potential biological tool to manage weeds but, ment with Kasasian (1971) who observed that alone, it could not completely control weeds. It

Table 5. Dominant weed species composition (by numbers). Mean count Frequency of weed species per square metre in all treatments Heckelochloa granularis 7.46 8 Rottboellia exaltata 4.62 5 Pennisetum polystachum 4.50 2 Tricodesma zeylanicum 3.78 3 Launea comuta 3.17 4 Commelina benghalensis 2.65 10 Aeuna lanater 1.63 1

NOTE: The frequency was obtained by taking into account the occurrence of the weed species in the treatments for both systems.

99 Table 6. Comparative total grain yield of crops (kg/ha) under nonweeded and weed-free conditions.' Yield under Yield under % reduction weed-free nonweeded in yield Crop combination conditionsb conditions due to weeds Bulrush millet + soybean 2550 1375 46.1 Bulrush millet + green-gram 2756 1355 50.8 Bulrush millet monoculture 2610 1122 57.0 Maize + green-gram 3310 1326 59.9 Maize + soybean 3330 1311 60.6 Maize monoculture 3510 1354 61.4 Sorghum + green-gram 1184 194 83.6 Sorghum + soybean 869 94 89.2 Monoculture sorghum 3629 138 96.2 Monoculture soybean 652 19 97.1 Monoculture green-gram 293 8 97.3 a Total per unit area for one crop under monoculture or two crops under intercropping. b These data were obtained from a similar experiment carried out near the present experiment (conducted during the same period and season) to study the effect of time of weeding on the growth and yield of the same cereals and legumes under intercropping. follows, therefore, that intercropping to control tive proportions of different weed species and their weeds may be a useful method for small farmers in distribution in different crops and crop mixtures, the tropics where, at present, minimum weed con- type of cropping system, planting configuration, trol is practiced. and overall crop management.

Donald, C. M. 1963. Competition among crops and Summary and Conclusions pasture plants. Advances in Agronomy, 15, 1-114. Esipov, A. and Scherbakova. 1970. The thicker the crop the fewer the weeds. Weed Abstracts, 19(3), The results from this investigation showed that 955-1382. intercropping was superior to monocropping in Evans, A. C. and Sreedharan, A. 1962. Studies of inter- terms of smothering weeds. This implies that, cropping. II. Castor-beans with groundnuts or soya- when good crop combinations are used, inter- beans. East African Agricultural and Forestry Journal, cropping may be an effective method of control- 28, 7-8. ling weeds. The results also highlighted the many Fine, R. and Fletchall, O. H. 1964. Some studies on the nature of corn-weed competition. Weed Abstracts, facets im- of crop-weed competition, showing the 14, 66-67. portance of an integrated approach to weed man- Haizel, A. K. 1973. Competitive relationship between agement taking into account all the ecological and maize (Zea mays L.) and the weed Rottboellia ex- biological interactions between crops and weeds. altata. Journal of Agricultural Science, 6, 25-31. Bulrush millet under monoculture or when in- Hammerton, I. L. 1962. A preliminary study of the com- tercropped with legumes was the best crop for petition between Kale and three weed species of suppressing weed growth and minimizing grain genus Polygonum. Weed Research, 2, 274-282. yield reduction due to weed competition. On the Kasasian, L. 1971. Weed control in the tropics. London, other hand, the two legume varieties under England, Leonard Hill, 124-215. Ivens, G. W. 1967. East African weeds and their control. very low grain yield accom- monoculture gave a London, England, Oxford University Press, 250 p. panied by increased weed dry matter production, Nieto, J. H., Brando, M. A., and Gonzalez, J. T. 1968. suggesting a shift in the crop-weed balance in Critical periods for crop growth cycle for competition favour of the weeds. from weeds. Weed Abstracts, 17, 374. The main environmental factor affecting com- Rao, M. R. and Shetty, S. V. R. 1974. Some biological petition in the crop-weed populations studied aspects of intercropping systems on crop-weed ba- appeared to be sunlight. This factor, but also to lance. Presented at the Weed Science Conference/ some extent plant nutrients, set the stage for Workshop, Andhra Pradesh Agricultural University, crop-weed competition. The shift in balance in this Rajendranagar, Hyderabad, India, 17-20 January 1977. competition for sunlight, either toward the crops Terra, G. J. A. 1959. Agriculture in economically under- or the weeds, was determined by many other developed countries especially in equatorial and sub- factors, including: types of crops and weeds and tropical regions. Netherlands Journal of Agricultural their respective growth habits, plant density, rela- Science. 216-231.

100 Discussion outcompete weeds for nutrient uptake, light inter- ception, etc. Nadar (question): You stated that intercropping Bujulu (question): Why did you count and was superior to monocropping in terms of harvest weeds at the time of crop harvest when smothering weeds. Is the efficiency of intercrop- there was a possibility of missing short-lived weeds ping due to higher populations or to intercropping and counting weeds that might have germinated per se? Have you compared monocropping and at a time when the main crop was mature enough intercropping at the same population level? not to suffer any economic stress from weed com- Mugabe (answer): The population in intercrop- petition? ping was additive. We are in agreement with other Mugabe (answer): Harvesting weeds at crop workers that the greater population under inter- harvest was the most appropriate time because we cropping could have suppressed the weeds. wanted the crop to compete with the weeds throughout. VanderZaag (question): Why did you not use a control treatment (no crop planted) to determine Collinson (question): How would the cowpea dry matter of weeds produced without crop com- and Phaseolus compare to soybean and green- petition and use this as an index for comparison gram as weed suppressants? with the other treatments? Mugabe (answer): This type of comparative Mugabe (answer): This is a good suggestion to work has not been done. It is suspected that dif- see the population of weeds without crops. ferent cultivars of cowpeas will compare differ- ently depending upon their growth habit (cowpea Fendru (question): What are the possibilities of exhibits variable growth habit). increasing the ability of various crops to compete more effectively with weeds both below and Banta (question): Why were the yields of the above ground? legume and the cereal not presented separately? Mugabe (answer): Such possibilities rest with Mugabe (answer): Your point is well taken. In plant breeders to produce crops with the ability to future this will be done.

101