Banana {Musa spp. Colla) cropping Systems, production constraints and cultivar preferences in eastern Démocratie Republic of Congo

trees in the banana plantation or as hedgerows. Banana production can be improved in the study area using participatory research and development approach where smallholder farmers could evaluate appropriate technology introduced in the form of research. The key intervention measures are introduction of improved varieties with high yield and/or résistant to disease and sensitization on integrated pest management which could contribute into food security and poverty réduction.





Sectoral —Gross—Domestic—Product (GDP) contributions in the Démocratie Republic of Congo (DRC) include agriculture (55%), industry (11%) and services (34%) (CIA, 2009). Bananas (banana and plantain) constitute the second most important source of calories after cassava in eastern DRC. The consumption ranges from 136.9 to 173.9 kg/person/year (Ministry of Planning, 2005a). Most of the population in eastern DRC practices small scale agriculture with banana and plantains as the prédominant crops. The banana crop is mainly used for food and as a cash crop. Banana plantations also contribute to environment conservation as the crop grows throughout the year. Banana products (Juice, Wine, bunches and fibers) are consumed and marketed locally and, as a resuit, income generated from the— crop circulâtes at village level (Sebasigari, 1987; Jagwe et al, 2008).

The majority of culdvated bananas are triploid varieties belonging to the Eumusa section of the genus Musa, family Musaceae (Tenkuano et al., 2007). These varieties evolved from intra- and inter-specific crosses involving two diploid ancestor species, M. acuminata Colla (genome AA) and M. balbisiana Colla (genome BB), which originated from Malaysia and India, respectively (Stover & Simmonds, 1991; Lus’ty et al., 2006). In DRC, there has been only little research on banana since 1960s due to lack of manpower after the departure of Belgians after independence. However, the resumption of banana research programme around 1986 concluded that bananas were produced on small scale in farms characterized—by mixed banana varieties and intercropping, without any mechanization or renewal of plots for almost 30 years (Sebasigari, 1987). Unfortunately, there

was no continuation of banana research du. m insecurity in North Kivu and South Kivu régions after three years implémentation. However, Bakelana and Makangidila (1996) evaluated the banana and plantain production throughout the country and reported low vields ranging from 4.6 to 4.9 tons/ha; they concluded that banana production is constrained by the following problems: Traditional cultural practices, tield diseases and the lack of govemment funding for banana and plantain. The following diseases and pests, ranked according to the degree of severitv (l=rare, 2=mild, 3= severe and 4=v-ery severe) hâve been found on plantain and banana in Zaïre: P’usarium wiJt/Panama disease (3), Black and yellow Sigatoka (3), Virus disease hunci ■ top (2), Banana weevil (4) and Nématodes 3;. The Zaïre govemment has funded several agricultural projects in the past but almost none on banana and plantain. This information remaining general for the whole country elidn’t give any information on a particular province such as North Kivu or South Kivu.

There is need for on-farm banana production study to détermine the factors contributing to this low yield especially in eastern DRCongo. Therefore, the objectiva of this study was to détermine the banana production constraints that hinder food security, farmers’ cultivât preferences and the cropping Systems practiced.


A survey using an interview schedule was conducted in South-Kivu and North-Kivu provinces from 2iul to 29,h March 2008. The province of South Kivu with an area of 69 130 Km2 having 3.9 millions of population, is located at East of Démocratie Republic of Congo (formerly Zaire) especially between 10 36′ – 5° latitude South and 26″ 47′ – 29″ 20′ longitude East (l’NUD, 2009by. SK s population densities vary between 100 and 123 persons per km2, the mean of annual rainfall is distributed between 1437 mm and 1661mm with a long growing season (365 days) and altitude of 950 to 2019 m above sea level (masl) (Farrow et al., 2006). SK is bordered at East by the Republic of Rwanda through river Ruzizi and by Burundi and Tanzania through Lake Tanganyika, at South-East by Katanga province; at South, West and North­west by Maniema province and at North by NK province. SK is divided into 8 districts (territories) (Ministry of Planning, 2005b).

The province of North Kivu covering an area of 59 631 Km2 and having 4.9 millions of population, is also located at East of Démocratie Republic of Congo especially between 0°58′ latitude North- 02° 03′ latitude South and 27″ 14′-29″ 58′ longitude East (PNUD, 2009a). NK’s population densities vary between 59 and .22 persons per km-, the mean of annual rainfall is also distributed between 1268 mm and 1556mm with a long growing season (365 days) and the altitude of 909 – 1803 masl (Farrow et al., 2006) . NI< is bordered at East by the Republic of Uganda, at South East by the Republic of Rwanda, at North and West by the Oriental province, at South-West and South by Maniena and SK provinces respectively. It is divided in 6 districts (territories) (Ministry of Planning, 2009a).

In each province two districts/territories were purposively selected and within each district two divisions (sites) were randomly picked from a participatory rural assessment (PRA), according to agro-ecological conditions, market accessibiiity and population density (Table 1) (Farrow ét al, 2006).

At each site, fîve sub-sites (Table 1) were selected. In each sub-site six farmers according to wealth categories: two nch (with at least 2 ha of banana and cattle (at least 25 cows), two average (with at least 1 ha of banana and livestock (at least 1-3 cows or 10 goats) and two poor (having a ploi ol bananas with at least 50 mats (<0.5 ha) and wirhoui livestock) were selected for interview. The interviewées in different categories were identilied with the aid of village leaders, from winch sampling frame were préparée!.

A total ot 240 farmers were inieniewid enurc .study, interviews were c’uudiicicd selected farm where tarmers were to indicaie ihen preferences for the different banana accessions (:.< local cultivars) in term of fruit yield, faste, juice productivity and juice quality. Farmers’ preference on banana cultivars was recorded in term of characters which guide their planting material’s sélection. They were asked to list the ten most productive cultivars in terms of taste, juice quality and the fivc best for juice production. The respondents were also asked to idcniilÿ ban,un. production constraints that they were facing.

Direct observation (transect walk) in each farm (by the researcher) was conducted after each interview in order to (1) Describe the cropping System; (2) Assess agronomie pyâctices in place; (3) Verifv cultivât préférences by; derermining. iln cultivars présent w-‘ithin the tields. .1; \ en!’, nu listed banana production constraints.

Various methods described by Robinson (1996), Eckstein et al. (1995), Tenkuano et al. (201) »’) Blomme et al (2008) and Dens et at. (2008) wer< used to identify and characterize banana production factors. Diseased banana samples were collected and analysée! using methods described by GPU (2008) for infection and pest infestation.

Soil samples were collected at the depth of 20 cm in the banana field of two farmers pe: category per sub-site. These were analvzcd for their physica! and Chemical prophéties in the laborarorv ol the Department of Soil Science at Sokoine University of Agriculture, based on methods described by Page (1982).

The             product-moment corrélation

coefficient was used to assess similarité among banana-accessions. A corrélation math’. w , calculated and cluster analvsis carried oui :w m.- group average method using NTSYb paekagi. according to Karamura (1999). Descriptive siatistics and régression coefficients were computed using SPSS package, while ANOVA was donc using ihe GenStat package.


Table 1: Characteristics of Sites, population and Access to markets within time zones from PRA Sites in Sud- Kivu and North-Kivu Mountain


Sub sites P/km2 H1 mi2 H mi H mi H mi

123        7       26       14       37      10       58          9            51   365      1497   1555

100        2       7         9         9       5         29          4           45   365     1551   159‘

128        6       17       14       37      10       57          9            2     365     1661   2019

123        2       42       11       3        7         23          5            28 0 36.5    1554   1664

North-Kivu__________________________ Kasindi Butembo Kampala_____________ Béni____________ _________

Munoli Luseke,


Nguva, –                        59—T4—-0—— 8— 23—— 28—– 13—— 8—— 36 —365—1515—176( )



Maboya Kaliva,


Vughano,                        59   11        23         5     20         25       37         5        58      365       1470   1405


____________ V uvembe_________________                                                                                    -_______________

Mutwanga Kisanga,


Kyavitumbi,                 122    1     23        4     18        15      49         4        27     365      1268   1048


____________ Nzenga,________________________________

Mangodomu Mangodomu,


Kalibo,                            122     3      33         3     20         17       48         3        0        365       1482    1010


__________ Somicar                         ______________________________________________________

Source: Farrow et al. (2006) I= hours,^“ »minutes, 3= Growing season, – Annual rainfall, 5= Elévation



The results (Figure 1) reveal that there are more rich farmers in North Kivu (NI<)(25.84%) then in South Kivu(SK) (8.33%) even there are more average farmers in NI< (23.33%) then in SK (18.33%), but lesser poor farmers in NK (50%) then in SI< (73.34%). The self decision of farmers to maintain their roads by manual labour since 1990’s in NK could justify thé facility of movement of » banana from village (farms) to the nearest town (Butembo or Béni) and abroad (Kampala). Butembo is the farmers’ town, built by them, The name Butembo corne from the word “butembe” meaning “Ïnsete spp.” which is one of herbs at that site on Ndande ancestors’ arrivai in the area (Isale). The Ndande ancestor (Yira) consider that banana originate from

Ïnsete spp. Banana constitute their staple iood. Farmers in SK don’t hâve the seli wili oi maincaining their roads by manual labour and rheir roads hâve disappeared completely, hence high difficult condition of banana trade (Table I). The average banana field is small (0.7 ha per t’armer). A possible explanation for this (small banana field) is

 » that the study area-is mountainous with undulating land that is laborious for banana cultivation through building contour line which most farmers are not aware of. Although more suitable land exists where the terrain is fiat, the population is higher there and the land is used also for cattle grazing. The lack of sensitization on the appropriate technique of land use and family planning could be part of the raison.




Figure 1: Proportion of farmers (%) as categorized based on wealth status in North Kivu and South Kivu provinces in the Démocratie Republic of Congo.














The rich farmers “Oashamuka » who operate at commercial level own more than four hectares of banana (called Piukoma i.e., many hectares of banana in the Mashi dialect of South-Kivu) with less then 25 cows in SI< (due to war) but with more then 50 cows in NK, with their children studying up to University; also the cattle farming…supply manure used in the banana field. This resuit into production of big bunches of banana throughout the year which ensures self-sufficiency, The observed effect of manure use is similar to that documented « by « Michigan- (2008)-and- that-observed————————–

cultivars were widely grown by ail farmer categories. The lists of the ten most preferred cultivars per province are prescnted in the subséquent paragraphs (régression tables) combined with outcome of on fama vérification.

The study shows that most of farmers in the study area (85%) are between 20 and 65 years old. The âge of the respondents in South-Kivu province ranged from 24 to 90 years (Figure 1) with 44% of those interviewed being women. The proportion of women (44 %) in banana production in SK shows the level of their involvement in banana production Systems in term of food security. This women proportion support the report of PNUD (2009b) which State that the household’s proportion leaded by women in SK is less poor (66%) in term of food security compared to the proportion of poor house hold leaded by a men (87%). This explains the low implication of men in banana production which is mostly limited to brewery processing activity and management of it generated income without care of banana

cultivation activity in SK. Hence, men sensitization on the weakness of traditional culture which reserve agriculture labour to women according to DSRP (2008) for their involvement in agriculture production constitute the way of poverty réduction in SK. On the other hand, in North-Kivu province, farmers’ âge ranged from 20 to 81 years with 16% being women. The men proportion (84%) involved in banana production activity are in line with the increase proportion of medium to rich farmer in NK. The results on the distribution of âge show that banana farmers hâve long life expectance compare to other farmers having other crops (Cassava) as staple food. The oldest farmer in both locations was a man (aged 81 or 90 years) from the poor category; hence life expectation of banana farmers is above the total of life expectation such as presented by the National Statistic Service (DSRP, 2008). The development of three towns (Goma, Béni and Butembo) in NK province constitutes the indication of beginning of development due to men involvement in         agricultural activity.




B SKivu □ NKivu











Figure 2: Age structure of farmers in Souda and North Kivu provinces in DR Congo.




Soils analysis results (Table 2) show that apart from Lurhala and Mutwanga, which are characterized by silt-clay and sandy-loam soil types respectively, ail other sites hâve well drained clay soils. The pH in water (Table 2) in the study areas ranges from 5.1 to 6.5, which is medium acid and is within the optimum range for banana. The results show that except for Maboya site with the pHcaœ of bulk soil verv acid (3.98), ail other sites hâve acid (4.59) to medium (5.9) acid soils. There was no significant différence between pH in water or in bulk soil either between farmers of different categories or between sites at P<0.05 levels. The cation exchange capacity (CEC) of the soils is medium ranging from 10.2 to 23me/100g. This shows that the clay minerais in eastern DRC are kaolinites. The organic mat.ter (OM) concentration between 4.39 — 6.22 % is medium while the total nitrogen (N) of 0.41 to


0.58 % is high. The division ot Katana/Luhihi, which is the most important source of East African Highland Banana (EAPIB) in south Kivu, is characterized by soils with a pH of 6.2 in water and 5.8 in CaCb, which arc within the optimum range for banana and is composée! ot well drained clay with medium Chemical properties (Table 2). According to Pecrot (1958) this site is settled on recent volcanic soil. The site of Munoli, with soil properties similar to Luhihi (Table 1), is the main center of banana production specially the EAFIB f’cooking and beer banana) in North Kivu. This area, which includcs Isale locality/sub division, has a ceniurv old banana plantation. The respond. u with the âge ol 81 vears mentioncd that his banana plantation was an inheritance from his grand fadier. The Ndande tribe’s ancestors called “Yira” settled at Isale with their first banana (Mukingiro, Nzirabahima). Mangodomu and Mutwanga sites, which constitute the main center of plantain production, are characterized by soil with medium to high nutrient content. This soil property is similar to those described by Delvaux (1995). Low vields (2.8 5kg- Lunch) obtained by farmers u Maboya can partlv be explained-bv the observed low soil fertility fiable 2).




T a b le 2: Physico-chcmical properties of soils in eastern DK Congo.

Site Soil type pPÏH2O pH CaC12 CEC OM2 N3           P1 K5 Ca6 Mg’
Kabamba clay 5.45 4.54 13.16 5.12 0.43 49.94b 0.52 3.92 ’ ts
i a.hiitt clav 6.24 5.37 12.30 1.06 o.l ■ ‘.48c 1.44 fi.9 >  
1 atrhala silt-clay 5.61 4.59 12.56 5. ! a 11.40 18.8’Tl 0.51 2.82 . 4/
Burhale clay 5.48 4.65 15.26 5.88 0.57 68.36a 0.77 3.09 1.62
Maboya clav 5.10 3.98 10.20 4.72 0.58 48.89b 0.21 0.96 0.19
Mutwanga loam 5.96 5.05 18.60 6.22 0.53 41.9bc 0.49 6.94 4.03
Munoli clay 6.52 5.80 14.20 4.39 0.41 22.14d 1.77 7.76 4.51
Mangodomu clay 6.51 5.90 23.00 6.18 0.52 73.89a 0.74 15.2 8.40
    Cv%:9 Cv%:     Cv%: 1 Cv%:7   Cv%: Cv%:
    .1 13.7 Cv%: 23 ( ,’v « •'<>: 5 .4 Cv%: 41.9 81.2
    Lsd:l. Lsd: Lsd: 1 1 Lsd:10 81 Lsd: lsd:
    7 2.18 10.92 Lsd: 2 24 7 Lsd: 2 14.82 8.55

Note: These values shown are mean of 6 samples per sub-site. 1 Cation exchange capacity (me/lOOgi, Organic matter (%), 3 Total Nitrogen (%), 4 Extractable phosphores (mg/kg), 5 exchangeable Potassium (cmol/kg), 6 exchangeable Calcium (cmol/kg), 7 exchangeable Magnésium (cmol/kg).




The results show that there are 29 and 32 local cultivais grown by tarmers in South Kivu (SK) and North Kivu (NK), respectively. It was established that the planting matcrials’ origins is inheritance (40%) from parent and gifts from neighbours (60%, N=240). The Bashi tribe ancestors arrived in South Kivu with most of their highland banana such as the “Nshikazi” meaning “belonging to Bashi” (Nsbi tribe Bashi and ba^i — belong to, dialect Mashi).

Nshikazi (Shika) is also called “magizi”, which means “bitter”. The Bashi tribe got the plantains Iront the Barega tribe through dowry. Thus plantains are called “musheba ” meaning “wedding” in Mashi dialect. In North Kivu, apart from the inherited cultivars such as Mukingiro, the cultivars Kitoke/Mathoke and Nyakitembe are from Uganda;

while Nguma, Tunm CLundu), Nvaghengc are –m. i to originale Iront South Ki\ u. The cultivai- Ki mangango which is close to Yangambi km ? in le- of taste and growih habit and many other plantains is from the Congo forest basin.

Ligure 3 shows that most farmers’ land is allocated to beer banana types with 63.5 and 33.~ ’o (N=120) in SK and NK, respectively. The genomic groups (Ligure 4 & 5) used in beer making arc widc and include AAA EA-beer (43-70%), AAA-ibota (7 12%), VA A kamaramasengi (5-7%), A.A \ cavendish (12 15%) and AB-Ney poovan (2-14′ ■> and ABB-group 4 1 1%). The major beer banana cultivât in South Kivu is Nshikazi. On tts own this cultivai’ covers an average ot 42.9 % (Table 3) ot ali banana lields and contributcs up to 52% (L’igure L


0.58 % is high. The division of Katana/Luhihi, which is the most important source of East African I Iighland Banana (EAHB) in south Kivu, is characterized By soils with a pH of 6.2 in water and 5.8 in CaCb, which arc within the optimum range for banana and is composée! of well drained clay with medium Chemical properties (Table 2). According to Pecrot (1958) this site is settled on recent volcanic soil. The site of Munoli, with soil properties similar to Luhihi (« fable 1), is the main center of banana production specially the EAHB (cooking and beer banana) in North Kivu. This area, which includes Isale locality/sub division, has a centurv old banana plantation. The responde u with the âge of 81 vears mentioncd that lus banana plantation was an inheritance from his grand father. The Ndande tribe’s ancestors called “Yira ” settled at Isale with their first banana (Mukingiro, Nzirabahima). Mangodomu and Mutwanga sites, which constitute the main center of plantain production, are characterized by soil with medium to high nutrient content. This soil property is similar to those described by Delvaux (1995). Low yields (2.8-3 kg/ bu ne h) obtained by farmers ai Maboya can partlv be explained by the observed low soil fertility (Table 2).




Table 2: Physico-chemical properties of soils in eastern DR Congo,

Site Soil type pHn2O P H CaC12 CEC’ OM2 N’ P4 I<5 Ca6 Mg’


clay 5.45 4.54 13.16 5.12 0.43 -19.94b 0.52 3.92 2. N
1 aihihi clay 6.24 5.37 12.30 1.99 O.-I5 ,“,1S( 1.44 6.9 3 3.-
1 atrhala silt-clay 5.61 4.59 12.56 5.13 0.44 1S.8TI 0.51 2.82 1 1
Burhale clay 5.48 . 4.65 15.26 5.88 0.57 68.36a 0.77 3.09 l.<>2


clay 5.10 3.98 10.20 4.72 0.58 48.89b 0.21 0.96 0.39
Mutwanga sandy-


5.96 5.05 18.60 6.22 0.53 41.9bc 0.49 6.94 4.03
Munoli clay 6.52 5.80 14.20 4.39 0.41 22.14d 1.77 7.76 4.51
Mangodom u clay 6.51 5.90 23.00 6.18 0.52 78.89a 0.74 15.2 8.40







Cv%: 23 Lsd: 10.92 Cv%:

1 1

Lsd: 2









Lsd: 2









Note: These values shown are mean ot 6 samples per sub-site. 1 dation exchange capacity (me/lOOg), Organic matter (%), •’Total-Nitrogen (%), 4 Extractable phosphores (mg/kg), 5 exchangeable Potassium (cmol/kg), 6 exchangeable Calcium (cmol/kg), 7 exchangeable Magnésium (cmol/kg).




The results show that there are 29 and 32 local cultivars grown by farmers in South Kivu (SK) and North Kivu (NK), respectively. It was established thaï the planting matcrials’ origins is inheritance (40%) from parent and gifts from neighbours (60%, N“240). The Bashi tribe ancestors arrived in South Kivu with most of their highland banana such as the “Nshikazi” meaning “belonging to Bashi” (Nsbi = tribe Bashi and A/ty’ = belong to, dialect Mashi). Nshikazi (Shika) is also called » “magizi”, which means “bitter”. The Bashi tribe got the plantains from the Barega tribe through dowry. Thus plantains are called “musheba » meaning “wedding” in Mashi dialect. In North Kivu, apart from the inherited cultivars such as Mukingiro, the cultivars Kitoke/Mathoke and Nyakitembe are from Uganda;

while Nguma., Tunru (Tundu), Nvaghengc an s, ai to originate from bouth Kivu. The cultivât K:su mangango which is close to Yangambi km 5 in icrm of taste and growih habit and many other plantains is from the Congo forest basin.

Ligure 3 shows that most farmers’ land is allocated to beer banana types with 63.5 and 33.7  » , (N-120) in SK and NK, respectively. The genomic groups (Ligure 4 & 5) used in beer making are widc and include AAA-EA-beer (43-70%), AAA-ibota (7-12%), AAA kamaramasengi (5-7%), AAA cavendish (12 13″,,) and AB-Ney poovan (2-14″, and ABB-group 4 11%). The major beer banana cultivât in South Kivu is Nshikazi. On its own tins cultivât covers an average of 42.9 % (Table 3) of ail banana lields and contributcs up to 52% (Figure 4

of raw materials for making beer. The increasing contribution (13%, Figure 5)’ of Cavendish group in bcer making in eastern DRC inclicates adoption and adaptation of this variety by small scale farmers. Farmers attributed the entrent situation in land allocation (Figure 3) to two reasons: the first is the fact that harvested bunches are processed into juice and becr which can be kept and sold over a longer time and generate more income comparée! to cooking and dessert banana which spoil after a short time because of market problems. Secondlv, becr banana constitutes the tradicional income reserve, which is released progressively. Apart from the similarité in land allocation to beer banana in eastern DRC, there is a different scénario for orher banana types.






ÉË 60


.E 50


|     40

I 30

° 20 o »



J 63.5
• % S K % NK
Figure 3: Types of banana grown in South Kivu (SK) and North Kivu (NK) provinces in eastern DR Congo.

120/ i AB-Ney poovan




AB-Ney poovan 14%
Figure 4 (left): Genomic groups of beer banana in South Kivu province; Figure 5 (right): Genomic group of beer banana in North Kivu province, eastern DR Congo.























in South Kivu, high market demand for dessert banana bas led to expansion ol its production. The tesults (Figure 3) show that dessert bananas (21%) are second to beer banana (63.5%) in terms of land allocation, tollowed by cooking banana (10.9%) and plantains (4.4%), winch are mostly grown for subsistence. In North Kivu, the high local market demand (in Butembo and Béni) for cooking types and the demand in Kampala, Uganda, for plantains



hâve raised the allocation of these types to second (.32.2%) and third (28.7%) positions, respectively.

The results show that the most productive cultivar is Nshikazi (AAA-EAHB-beer) in South Kivu and Nguma (AAB-Frcnch-plantain) in North

Kivu provinces (‘Fable 3). The tenth most productive cultivar is Cisukari red (green red) and Kalole II (AAB-Kamaramasengi) for south Kivu and North Kivu respectively.




Table 3: Régression between banana préférence in term of bunch \ ickl and lantl allocation X,     Y

N o Most




respond. (N =120)

% in farm (N = 120) P SE? Unstand.(


t-val uc Signif.T
S Kivu   R2 0.25 R2adju.0.2 F = 39.45 d.f(118) Signif.  
    4       p<0.001  
1 Nshikazi1 64.5 42.9 0.501“ 0.015 0.093 6.281 0.000
2 Barhabesha2 19 2 0.120 0.393 0.512 1.305 ‘ 0.1 95
3 Kamelamasengi 14.9 3.7 0.114 0.169 0.256 1.512 0.133
4 Kisamunyu2


14 4.1 -0.010 0.168 -0.017 -0.099 0.921
5 Gros Michel3 9.1 2 -0.105 0.214 -0.215 -1.005 0.317
6 Musheba4 8.3 2.2 0.242’ 0.239 0.515 2.157 0.034
7 Malaya (Kitika)3 8.3 2.6 -0.041 0.233 0.070 -0.299 0.766
8 Chindege3 3.3 5.3 0.038 0.246 0.056 0.229 0.820
9 Yangambi Km ‘ 51 3.3 5.7 0.079 0.742 0.265 0.357 0.725
10 ( ihisukari/green « 1.7 3.8 0.613 0.246 0 502 1.903 0.106
  red ’          
N Kivu   R2 0.34 R2adju.0.2 F = 8.063 d.f (16) Signif.  
    9       p <0.025  
1 Nguma4 25 9.4 0.080 0.370 0.310 0.838 0.404
2 Vulambya1 21.7 9.3 0.081 0.316 0.27! 0.856 0.394
3 Tuntu1 (Tundu) 17.5 8.3 -0.058 0.267 -0.152 -0.571 0.569
4 Kitika sukari   3.6 -o.l 3″ o. I2B i) p 5   1!
  17.5       ! 1 % .
  kiri 5              
5 Mukingiro1 12.5 6.5 -0.116 0.721 ■0.499 -0.692 U. 494
6 Kitoke(Mathok


11.7 3.4 0.579[*] 0.910 2.584 2.839 0.012
7 Kisubi 8.3 4.1 -0.581 1.200 1.914 -1.596 0.171
8 mangango1 Kisubi musa1 7.5 3.8 -0.585 0.505 -0.630 -1.248 0.301
y Mudjuva2 3.3 2.5 0.519 0.202 0.213 1.052 0.370
10 Kalole 1I3 (Kamela) 2.5 2 -0.617 1.030 -1.143 -1.109 0.383

Constant \                                                         10.228       0.000





There was strong positive corrélation (R = 0.501, N=120) lor South Kivu and weak corrélation R= 0.34, N=120) for North Kivu between the proportion of cultivars on farm’ and the most productive cultivar. Thus allocation of large portion

of farmers’ field (Table 3) to the most productive banana cultivars in SK is explained by the high population densitv which needs food security, and the subsistence characters of this production. Bui the marke: demand in NK has influence the lann


allocation such as farmers allocate their land to the most productive including the market needs. The market needs in NK has thus reduced the goodness ol the déterminant coefficient (R= 0.34).

The best cultivar in respect to taste/flavour was Kamelamasengi (AAB-Kamaramasengi) (Table 4) for South Kivu and Vulambya (AAA-EAHB cooking) lor North Kivu provinces. Of the ten ranked cultivars, the least preferred in tenus of taste was Sukumba/Mugombozi (ABB-Bluggoe) and Kitokc/Mathoke (AA A-EAHB-cooking) in South Kivu and North Kivu, respectively.




Table 4: Régression between the best tasting banana and land allocation.

Prov/rank —  ———-

Best taste Cultivar




  SE; Y

l nstandard(b)

t-value Signif. 1
S. Kivu R2 0.044 R2adi„. 0.036 F = 5.424 d.f(118) Signif. p <0.025
1 Kamelamasengi ’’ 24.8 0.21P- o.2g O4>21 2.520 o.ii52.
2 Barhabesa2 16.5 0.042 0.024 0.01 1 0.458 04)48
3 Nshikazi1 12.4 0.154 0.045 0.074 1.652 ‘0.101
4 Kisamunyu2 10.7 0.176 0.056 0.102 1.826 0.071
5 Gros Michel3 9.2 -0.097 0.053 -0.049 -0.927 0.356
6 Musheba’1 7 0.205 0.038 0.068 1.774 0.080
7 Malaya-3 6.6 0.102 0.145 0.106 0.729 0.469
8 Ndundu’ 6.6 0.132 0.077 0.060 0.774 0.444
n Cisukari3 red 5.8 0.412 0.088 0.164 1.864 0.080
11) Sukumba1 3.3 0.417 0.371 0.417 1.123 0.304
N.Kivu R2 0.112 RTdju. 0.088 F = 4.645 d.f (37) Signif. p <0.050
1 Vulambya2 40.8 -0.088 0.020 0.018 -0.921 0.359
2 Pakuma2 25 -0.089 0.035 0.0.31 -0.898 0.371
3 Nyaghenge2 11.7 0.103 0.048 0.041 0.845 0.401
-1 Kalole3 10 0.334* 0.441 0.951 2.155 0.038
5 Musilongo4 10 -0.024 0.469 -0.042 -0.090 0.930
6 Nguma4 5.8 -0.850’ ■ 0.645 2.179 -3.229 0.032
*7 Kitikasukari kiril3 4.2 -0.387 0.549 0.800 -1.456 0.171
8 Nzirabahima co2. 3.3 -0.014 0.291 o.ol 3 -0.045 (t.965
9 Mudjuva 12 2.5 0.9.39 ( 1.46″ 1 Al ) 3 ys > < l.l »(‘: ;
10 Kitoke2 2.5






0.5 1 3 1.102


h 46 »


Significant at 0.05 levels. 1 Beer type, 2 Cooking type, ^Dessert type, ‘Plantain




A weak corrélation (R = 0.210 for South Kivu and R = 0.334 for North Kivu) was observed between cultivar with best taste/flavour and its proportion (%) on farm. The litrle influence of land allocation to banana cultivar by the criteria of taste/flavour in both provinces is explain by the tact that some best tasted banana producing small bunch are not grown by poor farmers who constituée the majority of the population.

Regarding the production of juice and beer, farmers hâve classified the five best cultivars as shown in Table 5. The best cultivar for juice/beer production was Nshikazi (AAA-EAHB) with a score of 76% while Ndundu (AAA-EAHB) had the lowest score (4.1%) in south Kivu. In North Kivu the best cultivar with a score of 35% was Tundu (AAA-EAHB) followed in decreasing order to a score of 15% for Kitika sukari kiril (AA A Cavendish) (Table 5). A positive but weak corrélation (R = 0.207, South Kivu and R — 0. loi, North Kivu) was observed between the best cultivât for juice/beer production and its proportion on farm. The corrélation results are in line with the observations oi Richards (1985) cited by Engels ■/ al. (1995) indicating that local farming practices are not a matter of traditions, but of active innovation and invention; in the case that farmers by their own expérience hâve set criteria lor banana’s land allocation assurmg their food securtiv and increas: e their income. This shows the tact thaï adoption ■’ improved banana could be eas\ through involvement ol farmers in the on-farm sélection proccss. in South Kivu, the most important criterion (Figure 6) used by farmers to select banana cultivars is the combination of flavoür, taste, and juice production (FF, 86%); whcreas in North Kivu the outbreak ol Banana Xanthomonas wilt disease that is devasraling most banana types has influenced tarmers’ criteria of selecting cultivars.




Table 5:: Régression between the best juice producing cultivar and land allocation.

Prov./   % respond.       t-value  
rank Best juice cultivar (N=120) P siç Unstand.(b) Signif.T  


R2 0.043 RMtju. 0.035 F = 5.308   d.f(118) Signif. p <0.025
! Nshikazi1 76 0.207* 0.039 0,(186 2.304 0.0.7. 5
2 Gros Michel2 9.1 -0.004 0.027 0.001 ■0.038 0.969
3 Malaya2 11.6 -0.092 0.055 -0.042 -0.758 0.4.51
4 Kamela masengi1 9.9 0.232 0.062 0.098 1.583 0.121
5 Ndundu1 ’ 4.1 0.144 0.518 0.346 0.667 0.512


R2 0.010 RMiju. 0.002 F = 1.175   d.f(114) Not Signif. At 0.05
1 4‘un du1 35 0.101 0.661 0.717 1.084 0.281
2 Mukingiro1 26.7 -0.011 0.952 -0.100 -0.105 0.917
3 Kisubi mangango1 21.7 -0.094 1.158 -0.877 -0.757 0.452
4 Ratatina1 20 -0.170 1.162 1.316 – F133 0.264
5 Kitika sukari kiri12. 15 -0.030 3.836 0.310 -0.081 0.938
    Constant 7.326 2.691   2.723 0.007

‘ Significant at 0.05 levels.1 Beer type,2Dessert,




Hcnce, farmers hâve generally adopted résistance to pest (RP, 65%) as the major criterion for selecting planring material in NK. After these major criteria, the oiher criteria in decreasing order of importance are bunch size (BU), the availability of planting material (AV), adaptation to poor soil fertility (RS), market demand/prices (MD), résistance to drought (RD), short production cycle (SC) and sustainable production (SP). Farmers argued that such factors/constraints are temporal, but food need to be sufficient first and appetizing for producer and consumer.

The phenogram (Figure 7) shows two clusters in respect to how tarmers use banana. The first cluster of 23 cultivars (Kalole 2 (Dl) to Cingulube (D53)) consists of beer making types, while the second cluster is subdivided into three sub-clusters. The first sub-cluster of 13 cultivars Kingulungulu (P3) to Cibulanana 1 (P61)) consists of roasting banana types; the second sub-cluster of 8 cultivars (Kitika kiri (D6) ro Kitika kikuhi 1 (D21)) consists of dessert types while the third sub- cluster of 1” cultivars (Kiwarc (C8) to Isanzi (P54i) consists o! soft cooking banana tvpes. The corrélation coefficients ol similarité (Figure ’.i h ; ■ shown that rlx three main characters contributcd up to 87% in grouping banana cultivais in the two clusters were characters in respect to consumption, cooking ability and maturity. The genomic composition of the first cluster is wide as stated previously (Figure 4 & 5). Farmers’ criteria oi classification alone arc not able to make différentiation between banana genome groups. For example the I’rench plantain Musheba and Walungu 16 (horn) are fitted in the first sub-cluster (dessert banana), due io their fast maturing trait, bunch size and fruit sweetness when it is ripened. Farmers bave started consuming raw when ripen some plantain, due ro food insécurité created by population densité and food scarcite. This information is usetui lor breeders during sélection of cultivars for on-farm évaluation in eastern DR Congo.


The results showed that rhe sélection of banana planting materials is donc during the wet scason. The responsibility for sélection is equally shared (50%) between men and women in South Kivu; while 90 % of cultivars sélection is done by the head of houschold in North Kivu. In both provinces, about 84 % of planting materials are selccted on the planting day and 16% 2 to 4 days before the planting day. lt is noted that beer bananas is mostiy selected by men because brewery activities and sale are perlormed by males. Women are involved in selecting cooking and other type of banana as t’hey are the ones who cook and/or transport and sell them in the markets.

According to the symptoms described by Mak et al. (2007) and Molina (2007) Gros Michel, Katncla masengi and Kisubi hâve disappeared in sonie places from castern OR Congo due to Pusarium wilt. From sample of diseased plants collected from the study area, different species of Pusarium hâve been isolated, such as Pusarium c.j. oxysporum (on AAA-Cavendish), Gliocladium, Pusarium sacchari and Pusarium. solani (on AAA- FAIIB, beer (Ndundu) and cooking (Barabesha) types). This new banana pathogen isolation show that according to farmers cropping System charactcrizcd b\ mixing banana with sugar c..u, ISaccharum otjiciiiarum), Solanum mgrum. Soiauum tuherosum among other, Soil pathogen previouslv known having different crops as plants host hâve started infecting banana. Moreover, other discases like banana bunchy top virus (BBTV) locally named “Sindika” as described by Robinson (1996) and nématode damage according to symptoms described by Chabrier et al. (2008); Munroe (20()9) and Agri-Net (2009) were observed. From samples taken, varions species of nématodes were isolated such as Me/oidogyne, blelicotylenchus multichictus and Radophofus similis. Ail other cultivars such as Ngorva, Cibirangondo, Musilongo, many types of Musbeba, Pomc, Buhake, Bumpavu, Kisubi, Lushuli, Masisi, Kiware, Mukingiro, Ngenge, Barhabesha, Ndundu and Isanzi may hâve disappeared due to weevil and nématodes damage. The most important disc.t^i^ that havi c< mi rshuted to the disappearanet t cultivars an BB 1 \ . \unlhomona.’ wilt and Bi Sigatoka. This is confirnied by the lact that I )e Langhe (1961) cited by Stover and Simmonds (1991) lias collected about 56 cultivars in the région and plantcd in INFRA-Mulungu station but now a day there is only 32 cultivars remaining. This suggests a collecting mission in order to find out if they could be found in other area (somewhere else) of die provinces. Hencc, germplasm conservation becomes an alternative for escaping cultivars disappearance in the région.

Survey results hâve shown that in eastern DRC, farmers use only suekers as planting material. They don’t use rissue cultured plants or corm bits. As stated previously, the source of chose suekers is from own existing banana field or from neighbours. There is no research institute or private sector producing clean banana planting material.

The agro forestry Systems applied by farmers in banana Systems both in South Kivu and North Kivu consists of scattered trees in plantation where most of the trees are Persea americana, Carica papaya, Mangifera indica, Cilrus sinensis, Psidium goyava, Citrus delictosa and Citrus limon* These trees are plantée! as sources of fruits. Moreover, they hâve rnult ipurpose tree hedgerows like Grevillea robusta, Accassia sp., Cassia siamea, [eucalyptus sp., Musanga sp., C.upressus sp., and Ficus spp. These species are used for building and are also a source of income. The farmers also practice alley planting only when mixing banana with Coffea arabica or Cinchona sp.

Farmers also mix’banana with Phaseolus rn/garis, Arachis hypogea, ! ‘igna unguiculala, 7.ea mays, Orypa sativa, Saccharum ojficinarum, Solanum nigruw, and Sorghttm bicolor. An inappropriate practice was observée! whereby farmers till (lanel préparation) at around 25 cm deep in the banana field before sowing beans. Blomme et al. (2008) has determined that 45% of banana root System is in the zone around 25-30 cm deep, which mean that when farmers till banana fields, they destroy about 45% of die banana root System. Damage to roots reduces water and minerai uptake of the plant, and constitutes one of the reasons of low yield and srnall bunches (Robinson, 1996). The average weight of banana bunch from these ficlds was 3 kg/ mat.

Farmers (60 %) with srnall fields do not practice fallowing while farmers with large fields practice fallowing in order to manage diseases and to restore soil fertility. About 92% of the int rviewed farmers in eastern DRC practice desuckering. It was established that 55 % of farmers in South Kivu leave 3 to 6 suekers per mat, while in North Kivu 84 % of the farmers leave between 4 and 7 suekers per mat. Thus, a mean of 4 suekers is tisually left per mat in eastern DRC after desuckering. Farmers’ reasons for desuckering include that the practice minimises shade, allows intercropping and increases banana yields. The majority (73%) of farmers in South Kivu conduct desuckering ail times during farm visit, whiles in North Kivu about 83% conduct desuckering during weeding. According to Robinson et al. (1996), Tenkuano et ai. (2007) and Dens et al. (2008), the high number oi suekers left per mai could be mit ‘ the reasons for low banana yield.

The method of detrashing (dry leave rcmoval) seems to be known to 75% of farmers in eastern DRC, who apply it during weed control. The results show that most farmers (60%) are not aware of the importance of debudding (male bud rcmoval) which is a rccommendecl practice to improve uniformitv of fingers and reduce infection by Xanthowonas wilt and cigar end rot. A few of the farmers who are aware explained that debudding increases finger and bunch size. During farm visits, it was observed that even those -farmers who are aware ot the importance of debudding do not do it on their farms.

About 72% of farmers practice propping in order to prevent breakages of pseudostems due to wind. Propping is conducted during bunch development when the need anses. Cultive propped are ali that produce big bunches and hâve tall stems. Amongst cultivars propped aie Barhabesha, Gros Michel, Kamela masengi, Nshikazi, Malava, Nakasimbu or Buganda (Yangambi km 5), Cindege, Musheba, Kashulye, Nsha, Bulengere, Cingulube, Poyo and Chibulanana in SK; and Tuntu, Mukingiro, Vulambya, Kitika sukari kiri 1, Nguma I, Kiware 1, Kitika sukari kikuhi2, Kitoke, kithavwira, Kalole, Vuhindi and Mudjuva 1 in North Kivu.

Aller harvesting banana, farmers split and spread the pseudostems in the fields. For them, dus practice increases soil fertility and reduces soil érosion. In both SK and NK farmers weed using hand hocs onlv. About 80% of the farmers are weeding 2 to 3 times per year, while 13% weed 4 times per year and 7% weeding 5 to 12 times pet’ vear. Rich farmers are those who are weeduiu scierai urnes IP urnes ‘vcarl (or haï in;’ required bunch. lot lhe market. ( )n ihe mhci -… the srnall bunches (mean ol 2.8 kg/bunch In an mixed weed with banana) seen in poor farmers and some medium larmers’ field can partly be explained by the low weeding frequency (2 to 3 times per year) practiced, while the rain fall through the year


(365 days) in the study area allow high growth of weed which create high compétition with banana root System reducing hence banana root up take of soil nutrient. This resuit of srnall bunch due to low weeding frequency is in line with the finding of


ln eastern DR Congo banana farmers allocate large portions of their fields to beer.making types and hâve adoptecl new cultivars, e.g. Cavendish to be part of this group. The preference for beer banana is driven by a need to avoid marketing problems associated with other banana types, since after transformauon into juice and beer; the products can be conservée! for longer time ’ to generate more income.

The high market demand has increased the cultivation of dessert banana to the second most grown type followed by cooking and roasting banana (plantains) in South Kivu, while in North Kivu the second most grown type is cooking banana followed by roasting and dessert banana. Among lhe banana types, Nshikazi (AAA-EA-beer) and F’rench plantain Nguma (AAB) are the most preferred cultivars in South Kivu and North Kivu, rcspectively.

The major criterion of selecting banana cultivars is the flavour/taste, juice production capacitv and bunch size. The outbreak of banana

  • ACKNOWLEDGEMENTS: The authors would like to thank the Canadian International Development Agency (CIDA) (through Bioscience in east and central Africa Network/NEPAD) for

Agri Net G, 2009. Farmers Manual of plantains production. Ministry of Agriculture, Guyana. 21pp.

Bakelana K, and Makangidila K, 1996. Country Report: The State of Banana in Zaire. INERA, Kinshasa. 20pp.

Blomme G, Sebuwufu G, Addis T and Turyagyenda LF, 2008. Relative performance of root development in enset and East African highland bananas. African Crop Science Journal 16(1): 51-58.

Chabrier A, Carie C, Quénéhervé P, Cabidoche Y, 2008. Nématode dissémination by water leached in soil: Case study of Paulopkoliis similis (Cobb) Thorne on nitisol under

Robinson <?/ ai. (1996) on the effect of weed frequency on bunch weight in Nigeria who determined that plantain yield decreases in areas where it rains ail die vear from 16.6tons/ha using Tl weeding/vear to 2.5tons/ha using 4 weeding/vear.

Xantbomonas wilt has introduced the résistance o disease as part of major sélection criteria. Some banana types such as horn plantain that hâve onlv one hand per bunch and EAHB-cooking types that hâve very short fingers and srnall bunch are grown onlv by a certain category of farmers (fiashamukd). Other cultivars such as Yangambi km 5, Kamelamasengi (Kalole), Musheba (Nguma), Cavendish (Malava or Kitika sukari kiri), green red (Cisukari or kirisirve), Barhabesha (Mudjuva), Ndundu (Tuntu; are widely grown under names by ail farmers.

Farmers use hand hoes for weeding. The results suggest diat planting suekers from existing fields spreads pests and diseases. Low banana yield expericnccd in the area is due to degeneration of local cultivais, inappropriate banana manageu u and high pest intestarion levels. Thm then :                                                                                     •

to introduce improved banana cultivais <ud extension services on integrated pest management. The mixed cropping Systems need investigation in order to déterminé the optimal combinations.

funding this research. We would also like to thank Wimba Mumbalima for helping during the collection of data.

simulated rainfall. Applied soil ecologv 40: 299-308.

CIA, 2009. “(. »entrai Intelligence Agency World tact books’’            https://

www.cia.gv Zlibrarv/publications /the

world- fact book/.

Delvcaux B, 1995. Soils. In: Bananas and Plant uns Go weu, SR I al.j. (ihapman&l lall. ! >u • ai pp. 25< > 25~.

Dens K R, Romero RA, Swennen R and Turner DW, 2008. Rcmoval of bunch, leaves, or pseudostem alone, or in combination, influences growth and bunch weight ot ratoon crops in two banana cultivars. journal of Horticultural Science & Biotechnology 83 (1): 113-119.


DSRP, 2008. Strategie document for growth and poverty réduction. SK Province, Bukavu, DRC. 68pp.

Ecktein K, Robinson jC, Davic S],                             1995.

Pltysiological responses oi’ banana (Musa AAACavendish sub-group) in the subtropics. Gas exchange, growth analysis and source-sink interaction over a complété crop cycle. J. Plor~.S c’i (70): 169-180.

Engels JMM, Arora RK, Guarino L, 1995. An introduction to plan gcrmplasm exploration and collecting: planning, methods and procedures, lollow-up. ln: Collecting Plant Genctic Diversity, Technical Guidelines. Guarino L. et al. (Eds). CABI/IPGR1, Rome. Pp 31-63.

Earrow A, Busingyc L, Bagenze P, 2006. Characterisation of Mandate Areas for the Consortium for Improved Agricultural Kivelihoods in Central Africa (CIALCA). Bukavu. 132pp.

GP( ,      2008. Global ’ Plant Clinic. (site visited on October 5, 2008)

Jagwc J, Ouma E, Van Asten P, and Abele S, 2008. Banana Marketing in Rwanda, Burundi and South-Kivu. CIALCA Project Survey Report. Bujumbura. Pp. 61.

Karamura D, 1999. Numerical taxonomie studies of the East African .Highland Bananas {Musa AAA-East African) in Uganda. Thesis for Award PhD Degree at Univers! ty ol Reading, Rcading, UK. 192pp.

Lustv C, Akyeampong E, Davey MW, Newilah N and Markham R, 2006.. A staple food with nutritious appeal. InfoA-L/.w 15 (1-2): 39 – 41.

Mak C, Mohamed AA, l.iew KW, Ho YH, 2007. Early screening technique for Fusarium wilt résistance in banana micro propagated plants. FAO, Rome. 10pp.

Michigan DA, 2008. Generally Accepted Agricultural and Management Practices for Manure Management and Utilization. Michigan Commission of Agriculture, bansing. 45pp.

Ministry of Planning 2005a. Monograph of North- Kivu Province, Piloting Unit of DSRP Process. Kinshasa-Gombe, DRC. 155pp.

Ministry of Planning 2005b. Monograph of South- Kivu Province, Piloting Unit of DSRP Process. Kinshasa Gombe, DRC. 129pp.

Molina G, 2007. Mitigating the threat of banana Fusarium wilt. IN1BAP, Rome. 4p.

Munroe I.A, 2009. Insect pests of plantain

banana. NARI, Guvana. 15pp.

Page Al., 1082. Methods of Soi) Analvsis Pari

Mandison, Wisconsin, USA. American Society. Inc.l 159pp.

Pecrot A, 1958. Carre des sols et de la végétation du Congo Belge et du Rwanda Urundi. 16 Dorsale du Kivu. A. sols, carte de reconnaissance. (Soils and vegetation’s map ot Congo Bclgium and Rwanda Urundi. 16 Ridge ol Kivu. A. Soils, récognitions map).

1N ! iA(I, Bruxelles.

PNUD, 2009a. The unit üf poverty réduction ol North Kivu province: Summary profile of poverty and households’ livelihood. PNUD, Kinshasa-Gombe, DRC. 20pp.

PNUD, 2009b. The unit of poverty réduction of South Kivu province: Summarv profile ol pnveriv and households’ livelila •! P\ I D K insbasa ( >< nul a , DIB

Robinson JC 1996. Bananas and Plantains. ( \B International, Wallingford, OxlO 8DE, UK. 238pp.

Sebasigari K, 1987. Morphological taxonomv of Musa in Eastern Africa, ln: Banana and Plantains Breeding Strategies. Persley, G.J.N De 1 .anghe E.A.(Eds). Proccedings ol an International Workshop held at Cairns, Australia 13-17 October 1986. pp.

! ~2- I “6.

Stover RII, and Simmonds NW, 1991. Bananas, Tropical Agriculture Sériés, third édition. I.ongman Scicntifc & Technical. London, 468pp.

Tcnkouano A, Yuyisteke D, Swennen R, 2007. Sink Compeiiiion and Desuckering Effects on lield Performance ot Triploid .md Tciraploid Plantain Génotypes, pa/ru. <.■: crop science improvement 20 (1 -2): 31-51.

Yamaguchi J and Araki S, 2003. Biomass production of banana plants in the indigenous farming System of the East African Highland. A case study on the Kamachumu Plateau       in northwest

Tanzania. Agriculture,          Ecosyste/us and

E.nrironmenl. (Article in Press). Pp 1-19.


cooking types respectively. l’hese letters N, Fr, Fa, FI, I, C, K/S, and FA are Ney poovan, French, False horn, Horn.

Ibota, Cavendish, Kamaramasengi/Silk, and East Africa sub-group respectively, 01 Kalole 2 (N, AB), 82 Katarina (N, AB), P3 Kingulungulu (Fr, AAB), 84 Kisubi musa (ABB), 85 Kisubi mangango (I, AAA), 06 Kitika sukari kiri I (C, AAA), C7 Kiware I, C8 Kitoke/Mathoke (AAA), C9 Mudjuva I (EA, AAA), 810 Mukingiro 1 (EA, AAA), Pl! Musilongo (Fa, AAB), Pl2 Nguma I (Fr, AAB), Cl3 Nyaghenge (EA, AAA), 814 Nyakitembe (EA, AA), 1 !3 Nzirabahima I (EA, AA), 816 Tuntu I (EA, AAA), Pl7 Vuhindi (Fr, AAB), BIS Ovutsipa kakamatahira (EA, AAA). ‘ Vulambya I (EA, AAA), C2° Pakuma (EA, AAA). D21 Kitika sukari kikuhi I (C, AAA), ( 22 Kithavwira (F,A. \ \ \ >.

827 Kirisirye (C, AAA), P28 Vungelengele (Fa, AAB), c29 Ndaminya mughendi I (FA, AAA), Vuhethcra ti i. AAB), p?! Nzirabahima Pi(Fr, AAB), D2 Banana tembo (K/S, AAB), ls » Nshika (EA, AAA), b 1 Ndundu (FA. AAA), 835 Yangambi km 5 (I, AAA), 036 Chindege (C, AAA), t3, Cinyamunyu (EA, AAA), u‘’s Chisukari (C. AAA), 039 Kamela (N, AB) , C4° Kagenge (EA, AAA) , B4! Mugombozi (ABB), 042 Malaya (C, AAA), P43 Musheba loc (Fr, AAB)’, 844 Kashulye (EA, AAA) , C45 Barhabesha (EA, AAA), 046 Gros Michel (AAA), 047 Karoline (N, AB), P48 Chibulanana (H, AAB), C49 Mbwazirume (EA, AAA), 830 Poyo (C, AAA), 031 Cisukari red (C, AAA), 832 Nsha (EA, AAA), 053 Cinguiube (C, AAA), p54 Isangi (Fr, AAB) , P55 Walungu 16 (H, AAB), 856 Nyamijala (FA, AAA), 837 Munyamimba (EA, AAA), C58 Inzirabushera (EA, AAA), 839 Sukumba (ABB), 860 Bulungere (EA).

Figure 7: See link


Journal of Animal & Plant Sciences, 2009. Vol. 4, Issue 2: 341 –

Publication date: 15 Septcmber 2009, http:/ /ww. biosciences,; ISSN 2071 –

Banana {Musa spp. Colla) cropping Systems,
production constraints and cultivar preferences in
eastern Démocratie Republic of Congo

*Dowiya, N.B., Rweyemamu, C.L. and Maerere, A.P.

Department of Crop Science and Production, Daculty of Agriculture, Sok,oine University o) Agriculture, l‘.< >. lé.- 5005, Chuo Kifcuu, Morogoro, Pawgania,

‘Corresponding author email: doïviyafçfpyahoo. corn: Co-authors and maerere/çfyahoo. co. uk

Key words

Banana, plantain, production factor, cultivars preference


Banana and plantain {Musa spp.) are important crops in the Démocratie Republic of Congo (DRC) for food security and household income. However, yield has been decreasing to an average of 4.6 tons/ha currently, and the factors leading to the décliné hâve not been fully determined. In the présent study a diagnostic survey was conducted in the South Kivu (SK) and North Kivu (NK) régions of the country to détermine the existing banana production constraints. Soil was sampled for analysis from 240 fields (30 samples/sitc) from 8 sites. The results indicated that apart from one site (Maboya, NK), soil fertility is not a limiting facmr for banana production in the area. Planting materials used by farmers are exclusiwh suckers from existing crops and there is no institution producing clean planting materials in the area. Tillage using hand hoes for the purpose of growing common bean causes destruction of banana roots leading to poor nutrient uptake which contribute to low yield (3kg/bunch). Weed control is done by hand using hoes. Inadéquate weeding and inappropriate desuckering practices increases compétition between plants, leading to production of small bunches. In order of importance, the most important criteria for cultivar selection were flavour/taste, juice quality, résistance to disease and bunch size. Diseases including black Sigatoka,-Eusarium wilt,-Xanthomonas wilt, bunchy top virus, nématodes and weevils were identifîed as among the main causes of cultivar disappearance in eastern DRC. Although farmers hâve generally little knowledge on banana plantation management, they hâve good understanding on value addition through beer brewing. Hence farmers allocate an average of 50% of banana fields to wine producing cultivars. Cultivars front subgroup like AAA-Cavendish, AAA-Gros Michel, AAA-Ibota, AB-Ney poovan, AAB- Kamaramasengi and ABB are well adopted for wine making. A total of twenty nine and thirty two banana accessions were found to be grown by farmers in SK and in NK, respectively. Based on similarity among these accessions using farmers’ criteria two clusmrs were identifîed consisting of wine/beer banana, and a combination of cooking, roasting ami dessert types. The good market demand for plantains hâve motivated farmers in North-Kivu to allocate the largest part of their farms (25%) to French plantain Nguma (AAB) while in South Kivu the largest area (64.5%) is allocated to Nshikazi (AAA-EAHB) for beer production. Farmers’ mixed cropping System includes legumes, cereals, and multipurpose

[*] Significant at 0.05 levels. ** Significant at 0.01 levels. 1 Beer type,2 Cooking type,3 Dessert,4 Plantain

A propos de inera21 24 Articles

Soyez le premier à commenter

Poster un Commentaire

Votre adresse de messagerie ne sera pas publiée.