73 i f5
Testing & Liaison Unit
Farming Systems Programme
PMB 25, Bues
National Cereals Research and Extension Programme
1988 Farming Systems Survey of hanyu Division, South West Province
Backgrounds In order to help IRA researchers on and off station direct their
work more towards the potentials and needs of food-crop farmers in the South
West and Littoral, the TLU at Ekona undertook a series of farming systems sur-
veys. Tqe series began in 1986 and ended in 1988.
Objectives: To provide useful information to IRA breeders and agronomists, to
MINAGRI extension workers and planners, and to the TLU for its on-farm trial
Methods: Preliminary analysis of Manyu Division soils, climate and demogra-
phic data was followed by a formal survey of 95 farmers in 19 villages of all
the most populated ecozones, excluding most of Akwaya; it was coupled with
qualitative description of one to two fields per village. Statistical and
qualitative results were combined to produce a descriptive report based on
differences by ecozone.
Results and Discussions Considering the four divisions of South West Pro-
vince, Manyu is the largest in land area, and the third most populated divi-
toen, after imes and Pako. The most important Manyu food crops in economic
terms are cocoyams and maize, followed by plantains, cassava and taro. Other
food crops grown are bananas, yams, groundnuts, egusi melon, dwarf beans,
irish potatoes and rice.
Access to land is mostly by inheritance of customary rights as village mem-
bers. The fields are distant, averaging an hour or more from the house.
Women do the majority of farm work on food fields after clearing. Manyu far-
mers spend most cash on land clearing and preparation, school fees and year-
end holidays. The main source of cash is food sales, not cocoa and coffee.
The extension service is inadequate in size and supervision (due to lack of
transport). Manyu markets are small and local. Other constraints to food
production are poor soil, crop destruction by animals, crop diseases (espe-
cially cocoyam and plantain) and insect pests (especially maize and egusi) and
the lack of chemicals (pesticide and insecticide).
Priority zones for agricultural research are the highlands of Fontem Sub-Divi-
sion and the central lowlands of Mamfe Central and Eyumojock. The former
shares many similarities and market access with Dschang and should be handled
from there; problems of soil management and input access are threatening a
very productive system. The latter is a moderately productive granitic-soil
region with a need for greater market access and for research focused on
fertility enhancement and varietal adaptation; it can be divided into a high-
er-density, sandier cassava/egusi-based section and a more forested, heavier-
soil taro/maize-based one. Akwaya's population needs urgent assistance in
mealy-bug control and better cassava and maize varieties to prevent severe
deprivation, but its marketing prospects are meager.
This report is written to be used by many types of peo-
ple. Those with interest in a particular crop or crop
should scan sections D.1 and D.2 and E.1 and E.2. (en-
vironment, cropping systems and field management) as
well as the crop itself'in section F. Administrators
with little time may want to start with the recommenda-
tions in section 6 and with sections A, B, and D.1 and
D.2. Social scientists will be more interested in Sec-
tions C and D.
TABLE OF CONTENTS
A. Background information . . .. . . . . *
B. Objectives . . . . . . . . . * *
C. Research Methodology ...... .......... . . . . .
D. Results: Environment and Resources ...... .. . . .
1. Manyu Division . . . . . . . . . . .*
2. Agro-ecological Zoning . . . . .. . . . . .
Mamfe Forest: (9); Mamfe West (10); Eyumojock (10); Highlands (10);
Akwaya Centre (11); Other Zones (11)
3. The Farming Household . . . . . . .
4. Land . . . . . . . . . . .
5. Labour . . . . . .
6. Cash Flow and Input Use . . . . . . .
7. Extension . . . . . . . . .
8. Livestock . . . . . . . . .
9. Food Markets . . . . . . .. .
10. Perceptions of Farming . . . . .
E. Results: Manyu Cropping Systems . . . . . . .
1. Food-Crop Field Selection and Management . . .
2. Food-Crop Field Design and Crop Associations . .
3. Tree Crop Fields . . . .. . ...
F. Crop Agronomy and Utilization . . ... . . .
1. Maize . . . . . . . . . .
Importance (35); Ecozones (35); Associations (35);
(35); Calendar (35); Varieties (36); Field
Processing and marketing (36); Storage (37)
2. Cassava . .. ... .
Importance (37); Ecozones (37); Associations (37);
(38); Calendar (38); Varieties (38); Field
Processing and marketing (38); Storage (39)
3. Cocoyams . . . . . . . . . . .
Importance (39); Ecozones (39); Associations (39);
(39); Calendar (39); Varieties (40); Field
Processing and marketing (40); Storage (40)
4. Taro . . . . . . . . . . . .
Importance (41); Ecozones (41); Associations (41);
(41); Calendar (41); Varieties (41); Field
Processing and marketing (42); Storage (42)
a a . . JA
. . . 17
. . . 20
. . . 23
. . . ... 26
. . . 27
. . . 29
* . a a
. a . a .
S . . 34
. . . . 35
. . . 37
. . . 39
. . . 41.
5. Plantains and Bananas . . . . . . . 4
Importance (42); Ecozones (42); Associations (43); Planting methods
(43); Calendar (43); Varieties (43); Field Problems (43);
Processing and marketing (43); Storage (44)
6. Yams . . . . . . . . . . . .
Importance (44); Ecozones (44); Associations (44);
(44); Calendar (44); Varieties (44); Field
SProcessing and marketing (45); Storage (45)
7. Groundnuts . . .. . .... . .
Importance (45); Ecozones (45); Associations (45);
(46); Calendar (46); Varieties (46); Field
Processing and marketing (46); Storage (46)
8. Egusi melon . . . . . . . . . .
Importance (47); Ecozones (47); Associations (47);
(47); Calendar (47); Varieties (47); Field
Processing and marketing (47); Storage (48)
9. Beans . . . . . . . . . . .
Importance (48); Ecozones (48); Associations (48);
(48); Calendar (48); Varieties (48); Field
Processing and marketing (49); Storage (49)
10. Potatoes . . . . . . . .
Importance (49); Ecozones (49); Associations (49);
(49); Calendar (50); Varieties (50); Field
Processing and marketing (50); Storage (50)
11. Rice . . . . . . . . .
Importance (50); Ecozones (51); Associations (51);
(51); Calendar (51); Varieties (51); Field
Processing and marketing (51); Storage (51)
. . . . 45
. . . 47
. . . . 48
. . . . 49
. . . 50
G. Conclusions and Recommendations . . . . . . . . . .. 51
1. Zones . . . . . . . .
2. Management and investment . . . . .
3. Crops . .. . . . . . . . . . ..
. . . 52
. . . 53
. .. . 54
1: Average Soil Characteristics on Farmers' Food-crop Fields at Survey Sites
in Manyu Division, 1988 . . . . . . . . . . .
2: Demographic Data Varying by Zone . . . . . . . .... . .14
3: General Characteristics of Manyu Fields by Zone . . . . . 16
4: Hours of Work Contributed by Each Demographic Grouping per Day (Busy
Season) by Zone . . . . . . . . ... ........ 18
5: Labour Associated with Husband's & Wife's Activities . . . .. 19
6: Average Household Crop Sales and Other Income Sources, by Zone . . 21
7: Extension Contacts, by Zone . . . . . . . . . . 24
8: Livestock Ownership, by Zone . . .. . . . . .. 25
9: Market Distances, Transport and Prices, by Zone . . .. . . . 27
10: Fallowing, Clearing and Burning Practices (Food Fields) . . . . 30
11: Crop Associations . . . . . . . .... . . . .. 32
1: Monthly Rainfall Variations in Mamfe Town . . . . . . . . 6
2: Intercropping Patterns . . . . . . . . . . . .32
Manyu Division . . . . . . . . ... . . . . . . 12
1988 FARMING SYSTEMS SURVEY OF IANYU DIVISION, SOUTH WEST PROVINCE
TESTING & LIAISON UNIT, IRA-EKONA
Principal Researchers: Susan W. Almy and Manfred T. Besong
A. Background information:
The Testing & Liaison Unit (TLU) at Ekona is charged with determining the
utility of the technology developed by the Institute of Agronomic Research
(IRA) for food-crop farmers in the South West and Littoral Provinces; and with
bridging the gaps between IRA researchers on the one hand, and farmers and
extension agents on the other. These two provinces are a root-crop and plan-
tain region, in which cereals play a relatively minor role. Therefore the
Ekona TLU has been charged with addressing the problems of the entire system,
although with an emphasis on maize, the predominant cereal.
Until the TLU began work in this area, cereals breeders and agronomists
operating from Nkolbisson had to extrapolate from conditions in the Centre-
South to decide what to try in the coastal lowlands. Information on other
food crops was somewhat more available, because Ekona plantains and root crops
researchers have access to local farmers, but their principal contacts natu-
rally tend to come from the Ekona subzone of Fako Division. Statistics on
provincial and divisional production and sales will become available with the
1988 publication of the Ministry of Agriculture's Agricultural Census of 1984,
but this provides no diagnostic information and also cannot address the vari-
ety of ecological and sucio-economic systems to be found inside the province.
A description of food-crop farmers and farms was imperative for the TLU's
work, and important for input to IRA work on-station.
Limited by resources, the Ekona TLU decided to start in Fako Division
(South West Province) in 1986, apd to take on one more division each year. The
Acknowledgements: The survey team consisted of Barnabas Akumbo and George
Donnah of MINAGRI (Manyu), Jato Johnson of MINAGRI (Ndian), and Augustine
Igwacho and Anna Ngundu of the TLU-Ekona. Station technical support was pro-
vided by IRA-Ekona personnel, including Tsegazeab Woldetatios, C. Fri Poubom
N. and Mboussi A. Messia of the TLU; Martin Tchuanyo, Jerome Ambe Tumanteh and
Joseph Wutoh of Root Crops; and Patrick Kofi of Plantains. Special thanks go
to Jean Zambo and Frederic Tchuenteu of Soils for soils identification and
analysis. George Gamze provided secretarial and drafting assistance.
' TLU IRA-Ekona p.l
rl; Division survey was carried out in October-November 1987, and data col-
Jcction for the Manyu and Ndian surveys was conducted jointly in April-May
1989. A full provincial report will follow the release of the divisional
reports by mid-1989.
1) to provide information on cropping systems and cultivation practices of
principal food crops to guide on-farm and on-station agronomy and variety
trials for the Division;
2) to provide information to the breeders (maize, cocoyam, cassava) on exis-
ting varieties, cropping systems, treatment in production and post-harvest,
field and storage problems and preferences;
3i i' establish contact with the Ministry of Agriculture extension service and
with interested farmers, in preparation for on-farm trials and training;
4) tc collect information of interest to the Divisional and Provincial Delega-
tions of the Ministry of Agriculture, to aid in their work..
C. Research Methodologys
This action is written for those with a particular interest in survey me-
thk-g. .. or in the validity of this survey. Others should skip to the next
-- t;i cn.
Secondary data were collected to map soils, altitudes, rainfall, roads,
villages and towns in Manyu Division, and supplemented by a road tour in the
company of staff from the Sub-Divisional Delegations of the Ministry of Agri-
ci;ture. These data were used to delineate nine potentially different agro-
erc.oqal zones, four of which (in south and east Akwaya Sub-Division and
o~Cthe,- E/umojock Sub-Division) are sparsely populated and were inaccessible
within the resources available.
This preliminary data collection induced some small modifications in the
'ri2inal questionnaire as applied ip Fako Division and revised for Meme, in
th, -dle of palms and field types. However, the final questionnaire is very
close 'n form and method of administration to the ones.used in Fako and Meme.
The dzta quality is not as good as for Meme. Firstly, the survey had to be
mountean more quickly and for a shorter time than anticipated, due to other TLU
comamtmen:1t-. Although the training was not scanted, the survey team was fre-
quentl. ower-tired and the evening review of completed questionnaires was not
as -oud as it might have been. Secondly, it unfortunately coincided with the
removal cf all village extension workers under technician status to Mamfe for
a course, which removed one important source of information. Finally, many
!'a.yi farl-ers are not fluent in Pidgen, and often questions had to be trans-
TLU IRA-Ekona p.2
lated or expanded by their children or neighbors. In order to compensate,
during analysis, answers to dubious variables were compared among team mem-
bers, and sometimes one interviewer's answers to a particular set of questions
was eliminated from analysis. Extreme responses from any individual respon-
dent were closely scrutinized and eliminated if not reasonable within their
The final instrument is eight pages long and takes an experienced enume-
rator an hour to an hour and a half to administer, depending on the farmer's
understanding and the complexity of his/her crops. It covers all food and
cash crops, their cropping system, land use and fallow, and for food crops,
agronomic methods, field losses and harvests, type and source of inputs, input
needs, storage, processing and marketing methods and difficulties, labour and
cash bottlenecks and solutions, extension contacts, and demographic data. The
enumerators were IRA and Ministry of Agriculture technicians from Manyu and
Ndian. The two surveys were combined to make the full provincial survey avai-
lable as early as possible for potential areal development plans.
Sampling was stratified by eco-zone and randomized within villages. In
each of the five accessible zones, three to five villages were selected accor-
ding to population size. Later reclassification of zones left one small zone
(Eyumojock) with only two villages, insufficient for reliability. Villages
were selected to cover the geographical range of the zone and the variety of
village sizes. Local extension agents did the sampling, following instruc-
tions to start at one end of the .village or town quarter and select every
tenth house, or its neighbor if the tenth had no food-crop farmer working at
least half-time. In most places, this procedure was followed. In Fontem Sub-
Division, where the people live scattered (but near the roads) the Sub-Dele-
gate tried to count the houses within sight of the road and pick the tenth.
Ninety-five farm households in nineteen villages were interviewed in Manyu.
Although a Population Census was carried out in 1987, the results will not be
released for several years. In order to estimate the total population of
farming households per zone, we worked from the Agricultural Census estimates
of number of farmers per Division, less 10% (since the Census includes very
small part-time farmers in urban or formal employment). We used the Nominal
Rolls to total the number of tax-payers in all the villages of each zone,
factored in the percentage of non-farmers encountered in each zone during the
sampling1, and estimated the percentage of all Manyu Division farmers in each
zone. All results reported for Manyu as a whole are weighted by the estimated
size of farming population in each zone relative to the zonal sample. That
is, the answers of a single Eyumojock farmer count less in the Manyu total
than those of a single Highlands farmer. Also, the Divisional totals reported
reflect only the areas surveyed, and exclude the 21% of population living in
most of Akwaya and in the Korup Forest.
Attempts were made to interview the farming couple together, and, failing
SAnd an estimate, derived from Meme towns, that approximately 67% of
households in the capital were engaged in farming, and 95% in the Sub-Divisio-
TLU IRA-Ekona p.3
this, the woman, since she is the primary food-crop farmer. This was suc-
cessful, although at times the men were not present when needed for cash-crop
information. Forty-seven percent of the respondents were women alone, and 4%
were men alone. Individual respondents were continually reminded to answer
for their spouses' concerns as well. In Manyu polygamy is widespread; the
husband's labour may be shared between 2-4 women's households, co-wives may
work fields jointly or separately, have common hearths or separate ones. We
defined the "household" which we interviewed as the core group that both wor-
ked fields together and shared a house, whether a couple or a man and two
wives and a mother or a widowed woman with grandchildren. The man was still
included if he did not contribute to the farm but lived in the house.
While the interviewers were administering the survey questionnaire in five
farmers' homes, the team leader visited one to two fields picked as typical by
the local extension agent, together with the latter and the farmer. Crop
association patterns were described, densities measured, soil taken for analy-
sis, and the cropping and fallow history of the field, clearing, land prepara-
tion, planting, weeding and harvesting sequence described. Pests and diseases
were described and often sampled; as part of the Ekona breeding effort for
cocoyams, petri dish samples of root rot on Xanthosoma and Colocasia specimens
were taken. Farmer and agent were also asked about local marketing, tran-
sport, access to inputs and whether particular cultural practices were common
to the area. Usually several other fields and fallows en route were also
described in brief. Rapid follow-up visits to some fields and to most zones
during July, August and October have also been incorporated into the data.
We needed to approximate production and land area for purposes of weighing
the relative importance of different crops and cropping patterns. South West
farmers have no conception of land area, and it was impossible to follow the
Census procedures (visiting every farm to measure it). In the Meme Survey, a
proxy was chosen for land area based on the number of man-days required to
weed a field in the first weeding after planting. The team leader measured
sample fields and compared with weeding time, controlling for field type, and
the measure correlated well enough for use. Unfortunately, in both Manyu and
Ndian the weeding time correlated better with the state of health of the far-
mer than to any measure verifiable in the questionnaire. Very approximate
measures have been created using weeding time, but they are not reliable.
Area per crop under the normal intercropping situation was estimated by a
simplification of the 1984 Agricultural Census formula2, in which major3 crops
I Direction Nationale du Recensement Agricole, Ministere de 1'Agricul-
ture. 1983. 1984 Agricultural Census (methodology). Yaounde. Offset.
3 "Major" crops are those which a farmer cites as occupying the field
without prompting. "Minor" crops are those which she remembers when asked if
there are any small other crops inside, or which she specifies as "just a few
stems". Two interviewers distorted these definitions, one insisting to the
farmer that only one crop be identified as the major crop, and the other trea-
ting all crops as major. All their interviews were eliminated for the major/-
minor crop calculations and anything based on them.
TLU IRA-Ekona p.4
were assigned weights of 1 and minor crops of .1, all crop weights in a field
were added to create a field weight, and the crop area was taken as size es-
timate times crop weight divided by field weight.
Production was measured by self-report of the farmers of the previous
year's harvests. Units were the volumes carried out of the field or filled at
home: baskets, standard market basins and tins, hand-trucks, mokutas (jute
bags used for cocoa bean sales) and 20-10-10 50kg fertilizer bags. Baskets
and non-standard basins were equated to one of the other measures by the far-
mer. Experimentation with dried grain and truck measurements led us to make
an equation of one fertilizer bag to 2 tins, 1.43 basins, .39 of a mokuta
(worn, stretched type) and .17 of a hand-truck. Finally, volumetric units
appropriate to each crop were weighed in the market to arrive at approximate
production figures. (The Agricultural Census of 1984 followed a similar pro-
cedure but weighed the units in the village and took reports of harvested
quantities three times in the year.)
Although data on production and proxy data on land area has been col-
lected and is being reported, they should be regarded as estimates only, of
less accuracy than those provided by the Ministry of Agriculture censuses
begun in 1984. The Census has concentrated large staff and technical resour-
ces on these two questions, whereas the survey reported here is focused pri-
marily on opportunities and constraints, both agronomic and socio-economic,
rather than quantification of present production. The production data is...
considered to be fairly reliable.
All statistics provided in the text refer to the questionnaire data, unless
otherwise indicated. The descriptive material from the field visit sample was
used to interpret the numbers rather than to provide them, since by its very
nature this sample was less representative of the population. "Survey" infor-
mation means questionnaire data, throughout the text.
Zoning was refined after the survey. Villages originally assigned to a
zone were compared with each other and with neighboring villages and their
zonal totals on a series of indicators: main crops, number of crops, timing
and nature of field operations, fallowing, densities, crop losses, ethnicity
and household size. Those differing from each other more than from their
neighbors were recombined, and neighboring zones were compared to determine if
they should be merged. The distinction between Eyumojock zone and Mamfe For-
est and Mamfe West was the most difficult to judge; the zone shows traits of
both and the two villages sampled are almost as different from each other as
they are from either of the latter two zones.
D. Results: Environment and Resources:
1. Many Division:
Manyu Division is by far the largest in South West Province, with 10.180
square kilometers, but its population of 153.000 people (1976 Census) makes it
only the third most populous. Most of the population at the time of the sur-
TLU IRA-Ekona p.5
vey was concentrated in the Highlands of Fontem Sub-Division and in the large
villages south and west of Mamfe Town. Fontem has 317 of the Division's 1987
tax-payers, although it covers only 6% of the land area. The Mamfe town area
(referred to herein as the Mamfe West ecozone) covers about 7% of the land
area and contains 21% of the tax-payers. A smaller population concentration
exists around the crossroads of the Mamfe-Kumba and Fontem roads, from Ntale
to Tinto and Defang. Otherwise, the Division is sparsely populated.
Rainfall in Manyu follows the unimodal South-West pattern, but starts,
peaks and ends later than in most of the province. Average annual rainfall is
3.290mm at Mamfe Town, probably4 declining slightly to the south and west.
Fontem's rainfall distribution is similar to Dschang's. In the year of the
survey, 1988, the rains came over a month late, steadying in late May and
continuing strong through the end of October. August is generally the only
cool month in the Idwlands, and September is considered particularly hot and
Chart 1: Monthly Rainfall Variations in Mamfe Town
(Source: National Meteorological Service)
Large and small rivers run throughout the division, providing space for
dry-season farming (where the hippos are not excessive) and hindrances to
travel and communication.
Soils are poorer than to the south, except in the Highlands and parts of
Akwaya. There has not been an adequate soil survey of the Division, but frag-
mentary reports and analysis of samples brought back from farmers' fields
visited during the survey indicate that the lowlands, including Akwaya, are
derived primarily from granitic substrata, with admixtures of sedimentary sand
and occasional volcanic outcroppings. The highlands of Fontem Sub-Division
seem to be volcanic, laid over a granitic base that becomes more dominant as
one descends to the lowland plain.
4 Rainfall data is only available for Mamfe Town. These estimates are
partly based on observation and discussion of local conditions relative to the
nearest sites with data outside the division.
TLU IRA-Ekona p.6
The four Sub-Divisions of Manyu have distinct identities. Fontem includes
the entire range of mountains between the lowlands of Mamfe Central and West
Province. It is made up of one ethnic group, the Bangwa, related to the Wes-
terners, who farm in distinctive ways, even when planting side by side with
lowlanders on the borders. The mountains rise abruptly from the plain, their
steep slopes resulting in precarious roads and scattered households clinging
to the few flat sites, from the lowest settlements at 300m altitude to the
highest at 2500m.
Mamfe Central and Eyumojock Sub-Divisions form a continuous lowland plain,
cut by plentiful small rivers and inselbergs. The distinction between them is
primarily ethnic; Mamfe Central contains the Upper and Lower Bayang, and Eyu-
mojock the Ejagham and Obang. However, the settlement and farming patterns of
the Lower Bayang (around Mamfe Town) and the eastern Ejagham (to their west)
are practically the same.
Mamfe Central enjoys road access to almost all settlements, whereas most of
Eyumojock villages outside the Mamfe West area are far removed from motorable
roads. Most such villages are also small and restricted in agricultural ac-
tivities by prohibitions on reserve forest as well as difficulty of marketing.
Akwaya Sub-Division, the largest, is cut off from road access both inter-
nally and externally. Three large and many smaller rivers cut across it,
making foot travel nearly impossible several months a year. Much of the Sub-
Division is also covered with steep hills, slowing travel. Trekking to the
administrative centre from Mamfe Town takes 4-6 days; from the nearest roads
in Wum, North West Province, 1-2 days; and from the nearest road in Nigeria,
2.5-4 hours. Because of travel difficulties, the Sub-Division is divided
economically into six sub-regions, each looking to a different market or none
The road network in Mamfe Central and especially in Fontem is dependent on
the rains. Landslides close the main Fontem road, which passes from Mamfe
Central through Menji (the administrative centre) to the city of Dschang, for
weeks, and only the best drivers would brave it during May to October. Driv-
ing the Mamfe Central roads in rainy season requires a good four-wheel vehicle
or a lot of pushing; in August-September the trip from Kumba can take 3-4
days. The worst section of the road was improved in 1985-86, but by 1988 it
was back to normal. Food markets are thus limited to internal supply, and the
towns to be supplied are not large.
Even cocoa is not a major product, although it has been established in the
area almost as long as in Meme to the south. Fontem Sub-Division has more
substantial coffee production, but until 1988 this was being sold across to
Dschang, since the South West Farmers Cooperative did not handle Arabica cof-
Still, despite the lack of markets and the depredations by forest animals,
there is substantial food production in Manyu, and much potential for expan-
sion. Although the soils are poor, rainfall is abundant, and there is still
enough land for long fallows. Manyu also contains a wider diversity of crops
TLU IRA-Ekona p.7
- -<< TLU IRA-Ekona p.7
than found in Meme or Fako.
e. Agro-ecological Zoning:
Agro-ecozones are sub-regions that are sufficiently distinct from one ano-
ther in their agricultural goals and methods that they should be considered
separately in deciding on appropriate technology and extension. In the Manyu
survey, five zones were delineated, with an estimated 79% of the farming
population: Mamfe Forest, a lowland zone still extensively forested with mar-
ket access and ample land, and an estimated 3.250 households; Mamfe West, a
relatively densely populated semi-urban area around Mamfe Town with concentra-
tions of cassava and cocoa, and 4.100 households; Eyumojock, a sparsely popu-
lated, forested zone restricted in farmland use by the Reserve Forest, with
1.100 households; the Highlands, a densely populated, mountainous area with
5.600 households; and Akwaya Centre ("Akwaya"), an isolated grasslands area
oriented towards Nigeria, with 800 households. The areas omitted from the
survey, briefly described under "Other Zones" below, are the Manyu section of
the Korup zone (500 households) and five other zones of Akwaya Sub-Division
(Messaga, with 1.000 households; the Mbulu Highlands, with 400; the Assumbo
Highlands, with 400; the Akwaya Interior, with 1.250; and the Cross River,
Table 1: Average Soil Characteristics on Farmers'
Sites in Manyu Division, 1988=
Food-crop Fields at Survey
s r=range, a=sean, P test is Bray-2 tppa).
TLU IRA-Ekona p.8
No. pH (HO) X Organic C X Total N Avail. P K* meq/100g X Al Sat
Zone Sites r B r a r a r a r a r a
naAfe Forest 6 4.2-5.3 4.8 1.0-2.6 1.6 .05-.15 .10 4-20 9 .10-.20 .16 2-44 21
HaAfe West 4 4.6-4.8 4.7 1.5-3.2 2.4 .07-.17 .13 6-14 9 .11-.21 .15 18-30 25
Eyueojock 2 4.2-5.1 4.7 1.1-2.4 1.8 .07-.13 .10 2-3 3 .14-.32 .23 10-55 33
Akwaya 3 5.0-5.8 5.4 2.4-5.4 3.6 .12-.14 .13 4-8 6 .18-.21 .19 6-9 7
low & aid 4 5.2-5.8 5.5 2.3-6.9 3.9 .14-.39 .23 3-13 7 .22-1.33 .60 1-9 5
high 2 4.6-5.2 4.9 9.3-9.6 9.5 .63-.92 .78 34-48 41 .93-1.32 1.13 1-3 2
The three central zones, Mamfe Forest, Mamfe West, and Eyumojock, have
similar soils and farming systems. The differences between them are probably
due more to density of population and market access than to different systems
of indigenous agricultural technology or basic ecology, although the amount
of, and access to, forest does have an effect. The description of the Eyumo-
jock zone is least reliable, because one of the three villages sampled proved
to fit better into Mamfe West, leaving only two sites, both of which are on
motorable roads, whereas about half the population within the presumed zonal
boundaries are not within reach of roads. It is possible that Eyumojock zone
should be considered as continuous with Korup.
Great reliance is placed on the Sub-Delegates of Agriculture for establis-
hing zonal boundaries in those parts of the Division to which it was impossi-
ble to penetrate.
Manfe Forest: Mamfe Forest zone covers the larger part of Mamfe Central Sub-
Division, and should possibly be considered as continuous with Nguti zone in
Meme. It stretches from Eyang at the Meme border, to Ebensuk at the Fontem
border, to the border with North West Province in the north, and covers about
21% of the surface area of Manyu.
Soils in general are poor, although in some areas (especially the eastern
side of the river Mbu around Tinto) old volcanic upswellings have left slight-
ly better conditions. The principal basement material seems to be granitic,
with an admixing of old sedimentary material. Soil sampled by the Kemden
river in an area used for alluvial planting differs little in composition from
that in a nearby interior farm, and that sampled from recent fallows is no
different from that of older fields. The soil of the southern part of the
zone is extremely hard to till, resulting in a practice of planting crops in
newly cleared, fallowed land on stick mounds (see section E.1); this practice
is applied elsewhere in the zone for planting virgin forest clearings.
Despite being located along motorable roads, Mamfe Forest village markets
are extremely localized and small. Farmers buy and sell crops frequently, but
primarily to and from neighbors. Oriented towards local subsistence, they
grow a wide range of food crops, with emphasis on maize and cocoyam or taro,
all intercropped on the same mounds at high densities. In order to avoid
heavy forest cutting, they return to fallowed land within 3-4 years, and often
plant several years in sequence on the same plot. One newly cleared field
coming out of a four-year bush fallow inspected in Eyang had no topsoil at
all. Wide distances between villages and small village size indicate that the
restricted fallows are not due to population pressure.
There is little occupational diversification within the villages; most
secondary jobs are as internal vendors of fish, palm oil or other local com-
modities. Many people, however, receive remittances in cash or kind from
employed relatives on the coast6.
A. The situation seems to be little changed from the early 1950's when
Ardener et al. described Tinto.
TLU IRA-Ekona p.9
1amfel West: Mamfe West stretches from Bachuo Akagbe through Mamfe Town and
Kembong to where the Reserve Forest approaches the Eyumojock road at Ayukaba.
Despite its high population, it covers only about 7% of the surface area of
the Division. The villages are large, well built, connected by numerous fee-
der roads, and often supplied with electricity and piped water by community
groups. Many people have relatives in employment outside Manyu, but farming
itself is considered a remunerative activity. Farmers have a larger area in
tree crops (cocoa, coffee and cultivated oil palm) than any other zone, and
sell a larger proportion of their food production, especially cassava, plan-
tains and egusi. Considerable monocropping is done, particularly of tree
crops and cassava.
The soils are similar to Mamfe Forest's, some extremely sandy and others
clayey, apparently granitic in origin. Forest is available only on the south-
western perimeter, and most of the land is in bush fallow, permanent or annual
crops. Fallows are relatively long, and follow one to two years of cropping
and one to two years more of cassava harvest, during which harvest period
little or no weeding is done. Mounds are less densely planted than in Mamfe
Forest. An independent palm oil cooperative maintains a large factory, and
there are also numerous schools, village industrial and commercial opportuni-
ties for secondary employment.
Eyumojock: Spanning the Reserve Forest from Mamfe West to the feeder streams
of the Cross River, the Nigerian border, and the Korup region to the south,
Eyumojock covers 8% of the surface area, largely with small isolated villages
nestled in ungazetted sections of the forest. Although the principal Came-
roon-Nigerian trade road passes through its northern sector, settlements along
it are severely restricted by the gazetted forest. Until Eyumojock Sub-Divi-
sion was created, there was little to cause a vehicle to stop between the
border and Mamfe Town, and even now, there is very little commerce or urbani-
zation in the zone.
At least in the northern part of the zone, farming strategies are influen-
ced by the Nigerian economy. When Nigerian food is cheap, farmers clear lit-
tle land for food crops, and concentrate on their cocoa fields. When Nigerian
labour is cheap, they pay laborers to come and clear food land for them, and
expand. Hunting and smuggling are doubtless important, especially in the more
remote southern section. Soil type is in the same range as Mamfe West's sandy
SHighlands: The Highlands zone corresponds closely to the borders of Fontem
Sub-Division7, and completely to the extent of the Bangwa people. Even when
farming down in the same altitudes as Bayang neighbors, they maintain their
own systems of land preparation and crop emphasis. Despite a high population,
the zone covers only 6% of the surface area of the Division. The soil type
7. A good description of the Sub-Division is found in Tanyi Clive Ndansi,
1987. "Alelleh-Tindoh", a small Encyclopedia of Nweh-Mundani. Mimeo, Fontem:
G.S.S. (author c/o Box 315, Tiko).
TLU IRA-Ekona p.10
changes gradually from a rich, somewhat acidic volcanic at the highest farming
altitudes of 2500m to the poorer, sandier, granitic soils of the lowlands at
400m. Rains fall earlier than in the rest of Manyu, and planting is in Febru-
ary and March.
Slopes are extreme, and roads, fields and settlement patterns are all ad-
justed to them. Villages are spread out through several valleys each, with a
house wherever there is sufficient flat space. Roads curve in and around the
hills, often being threatened by landslides. Beds are planted straight up the
slopes in many cases. Although fallows are relatively long and wild oil palm
and cultivated eucalyptus left in many fields, the forest does not regenerate
easily and clearing of long fallowed land involves little tree felling. Le-
gumes are emphasized, along with the staple aroids and maize, and coffee is a
major crop, planted separately from the food. The difficulty of the terrain
makes food crop sales more risky than coffee, despite the proximity of the
urban market in Dschang, and most marketing is internal.
The highest subzone, in the north of the zone, produces large quantities of
temperate-zone vegetables and irish potatoes, which attract Dschang trucker-
traders. The area visited consists of profitable farms and ranches of several
hectares each requiring large amounts of capital to maintain (for fences and
spraying). Younger sons, however, are forced out of the area or relegated to
Akwaya Centre: Akwaya zone encircles the administrative centre of Akwaya Sub-
Division, a triangle of very hilly lowland facing an underpopulated area of
Nigeria's Cross River State. It covers only 4% of the Division's surface.
Villages (including Akwaya Town) are small and distant from each other. The
vegetation is savanna (shrub trees and tall grass), and long fallows are used
to obtain a reasonable crop. Upland rice and groundnuts are grown as major
crops, primarily for sale to the nearest market across the border; the only
other sources of cash are the small public sector and religious missions. The
soil has high sand content and is probably granitic, but less acid than in the
Mamfe area. Large fields are planted to compensate for low yields, and weeded
earlier and more often, but there is only one planting season. The staple
food, a cocoyam-cassava fufu, has already been threatened by the cocoyam root
rot and is now menaced by a plague of cassava mealy bug. A local Catholic
mission health-and-agricultural project claims that malnutrition is common.
Other Zones: The Korup zone, in southwest Manyu, covers 17% of the Division's
surface, as well as most of Mundemba Sub-Division in Ndian and part of north-
western Meme. Huge, sparsely populated, inaccessible except by foot, heavily
forested and with mostly poor, acidic, granite-based soils, it supports small
villages specializing in hunting and shifting cultivation. The focus of the
Korup Project for the National Forest Park established in 1987, it has been
0. Cf. J. Mullan et al., 1988. Report of Rapid Rural Appraisal Survey,
Akwaya, S.W. Cameroon. Mimeo, Akwaya: Catholic Mission.
TLU IRA-Ekona p.11
IJC I ^ ^ T J
Map of Manyu Division
by George Gamze
TLU IRA-Ekona p.12
x x Survey sites i c
AP Agric Post c
S -H- Adllministrative boundary c
,--- Zonal boundary c
S --- Road '
S 0 Sub-Div. headquarter c
3 Divisional headquarter, c
?baoooooooo 0on nonomooo oooo 60ofc
the subject of a number of socio-ecological surveys', and it was not conside-
red necessary to include it in the IRA surveys. Among the three Divisions,
there are probably about 1.400 farming households.
The Messaqa zone stretches along the northeastern border of Akwaya with the
North West Province, with about 3% of the surface area. According to the
Akwaya Sub-Delegate, it is a hilly lowland region, from which coffee is car-
ried out to Wum on foot. The Mbulu Highlands to the south of it is mostly
over 1000m altitude, and covers 4% of the surface; its coffee is carried out
to Momo in North West. The Assumbo Highlands are an interior chain extending
from Mbulu north and west of Akwaya Centre, to the Nigerian border, sparsely
populated and with exceedingly difficult terrain, and covering about 3% of the
surface. The Akwava Interior, with 19% of the surface area, is a large,
sparsely populated lowland area to the southwest, too distant from any popula-
tion centre to engage in crop sales. The Cross River zone is a lowland area
with 8% of surface area surrounding the Cross River as it turns away from
Nigeria towards Mamfe; cut by numberless feeder streams, it is said to contain
cocoa farmers who ship their produce to Mamfe town by canoe.
3. The Farming Household:
Manyu families are distinguished by the practice of polygamy, found among
38% of households surveyed. The survey defined the "household" as the group
sharing fields working on the same farms. In most places surveyed, the
husband's other wives lived in the same house or same grouping of houses as
the couple interviewed but worked different farms. 6% of polygamists had a
wife outside the village, and 36% had an average 1.5 other wives inside it.
In some cases (6%) co-wives worked the fields together, and were treated as
the same household. In line with the practice of polygamy, junior wives may
be married very young, and several women household heads interviewed were
under 20. Men were always treated as part of the household unit if they help-
ed with the wife's farming, whether or not usually resident in her house.
These definitions understood, 88% of households surveyed had the husband
present (and 92%, any adult man), while 99% had the wife present (and 95%, any
adult woman10). The border areas with Nigeria had the lowest rates of male
presence, and Mamfe West, the highest. Within the core households, 52% were
female, but including the husband's other households, the figure rises to
55%, or 62% of all adults. Men were least outnumbered in Mamfe Forest zone.
Core household size averaged 6.6 persons, while the husband's extended house-
hold averaged 9.4. Within the core, somewhat over a third each were adults
V. Among them (a) Ruth Malleson 1987. Food Survey of Mundemba Town and
Ndian Estate. Mimeo, Mundemba: Korup National Park Socio-Economic Survey
Paper No. 1; (b) World Wildlife Fund / Land Resources Development Centre,
1987. The Korup Project: Soil Survey and Land Evaluation. Mimeo, W.W.F. Re-
port No. 3206/8.
1o Because 4% of the wives were underage.
TLU IRA-Ekona p.13
(including the shared husbands) and children under 10, and a quarter were
adolescents (10-19 years). Since adolescents who were away at school but
returned to help in the holidays were counted, this indicates very early
departures of youths from their families. 76% of the adolescents and 56% of
children were in school. Although education levels for the children were
lower than those of the Fako and Meme surveys (7.2 years of schooling for the
most educated child still at home), those for the parents were higher (3.4
years for the husband and 2.0 for the wife), indicating a more widespread
primary education system in the previous generation than enjoyed by the south-
Table 2: Demographic Data Varying by Zone (% of households or mean household)
Mamfe Mamfe Eyumo- Akwaya High- Manyu
Forest West jock land
Core family type:
couple and own
children: 52% 60% 10% 27% 60% 52%
c. & other depen-
dent children: 80% 80% 50% 60% 84% 77%
no adult man 4% 0% 40% 20% 12% 10%
Husband polygamous: 32% 40% 20% 33% 44% 38%
core: 6.6 6.4 6.3 5.7 7.0 6.6
extended: 8.7 11.0 7.5 6.8 9.6 9.4
No. adults(core): 2.9 2.5 1.7 2.2 2.2 2.4
% women(core): 48% 59% 57% 47% 49% 52%
% women(extended): 54% 62% 68% 65% 64% 62%
adolescents: 1.0 1.2 .8 1.6 2.1 1.5
none: 56% 60% 70% 40% 72% 63%
mid-level: 20% 20% 0% 14% 20% 18%
husband's: 2.7 3.0 4.2 2.5 4.0 3.4
wife's: 1.1 1.3 2.7 0.5 3.2 2.0
best child: 8.2 7.4 7.6 4.6 6.9 7.2
in school: 84% 73% 48%* 69% 84% 76%
artificially decreased by two households run by adolescent wives.
Almost all households were native to Manyu, 93X to their present villages,
and 6% to elsewhere in the Division. Only 1 person came from outside the
province. Husbands had been in their present villages 38 years and wives 27,
TLU IRA-Ekona p.14
while both had been farming about 20 years (23 and 19), The primary ethnic
groups were Bangwa (36% in the survey, in the Highlands), Bayang (32%, in
Mamfe Forest and West), and Ejagham (22%, in Mamfe West and Eyumojgck), The
Roman Catholics accounted for 49% and the Presbyterians 28% of the population,
13% declared themselves non-religious and 6% were Apostolics. Only a third
had off-farm trades or income. 16% in the low-paid, informal sector (seller
of fish, beer or palm oil, wine tapper, herbalist), 3% in the low-paid, formal
one (third-class chief, laborer in government employ), 8% in the middle-paid,
informal sector (tailor, off-licence owner, native doctor) and 10% in the
middle, formal sector (teacher, extension agent, government clerk, secoQd-
Half the households consisted of a simple nuclear family (a couple Ind
their children). Another third consisted of couples with additional family
members siblings and siblings-in-law (15% of households), nieces and malhews
(9%), grandchildren (6%), co-wives (6%) and the couple's mothers and aunts
(5%). 13% were headed by separated or widowed women (most 10% of all hauge-
holds without an adult man to help), often with a sibling, mother or grind-
mother, or grandchildren added. One household was headed by a widower, with
an adult daughter to help.
In summary, Manyu families are usually polygamous, with several wives gha;*-
ing one husband and maintaining their own households and food crop fields,
Adolescents leave home earlier than elsewhere, hence are less available for
fieldwork. Farmers are indigenous to their villages, and relatively few have
Almost all access to land in Manyu is by inheritance of customary rights as
village member. The few exceptions to this are in the Highlands and CAkaya.
The Bangwa of the Highlands follow the Western Province stem family pattern;
only one son is picked to inherit land and wives, and the rest have either to
leave farming or to beg land from kin (3% of fields in the survey), rent (9%)
or sharecrop (4%). Eight percent in the survey owned no land and 20% supple-
mented owned land with temporary usage. The figure would be higher, btt it
seems that the southern and western Bangwa adhere less to this rule thaw the
ones closer to Dschang. In Akwaya, the attempt to develop Akwaya Town 4nd a
road through to Nigeria has attracted immigrants from nearby areas, who have
been denied permanent land rights; 16% of farms surveyed were lent without
remuneration (but often on the worst soil) and 5% were rented. In all zones,
however, most people claimed that land used to be easier to find whew they
started farming than it was now; 20% in the Highlands and 12% in Mamfe F~rest
claimed the reverse.
Fields are distant, averaging an hour or more from the house. Most vil-
lages line the roadsides, with the fields stretching out on all sides in be-
tween fallow areas. Most Highland households reside apart on the part of
their farmland nearest to a road, but the terrain is so steep that the trek to
the farther fields takes long.
(LU IRA-Ekona p.15
Households surveyed planted or maintained an average 3.4 fields, each re-
quiring a mean 4 person-weeks to weed11. Of these, 70% were pure food fields,
11% mixed, 7% trees with plantains or bananas only, and 11% pure trees.
Fifty-nine percent of households had cocoa, 52% coffee and 6% (mostly in Mamfe
Forest) palm plantations. Fifty-eight percent18 harvested wild palms, about
half each for oil and for wine.
Table 3: General Characteristics
of Manyu Fields by Zone
Mean no of fields:
Mean size (in weeks
(minutes from home):
% with tree fields:
% growing food inside
trees > 5 years:
% full owners:
% own & other
% no land owned:
I. % food only
% trees & food
% trees & plantains
% trees only
II. % steep
11 Unlike Meme, it proved impossible to relate weeding time to
field; there was too much variation within field types, mostly caused
ing of first weeding and health of farmer.
Le 85% in Mamfe Forest and Mamfe West, 60% in the lower regions and 25%
in the Highland zone.
TLU IRA-Ekona p.16
Fallowing systems are still intact. Farmers claimed they fallowed food
fields an average 3.6 years for every 1.9 years cultivating, for a fallow-to-
cultivation ratio of 2.4. Only two farmers claimed not to fallow food fields
at all. Actual food field fallows were close to claimed ones; an estimated 8%
of food fields weren't fallowed, and an overall ratio of 2.3 is probable from
lengths of actual fallow. Fallows and fallow ratios surveyed were lowest in
the Mamfe Forest and Highland zones, but in the latter, 30% of food fields
were new (1-2 years old), signifying that the farmers were abandoning land
completely to go elsewhere. Forty percent of Akwaya's (but only 10-15% of
other zones') food fields were new. An average 60% of those with food+tree
fields fallowed inside them, averaging ratios of .8 uniformly across zones.
77% felt that their land used to be more fertile, varying little across zones
or by claimed fallow rate.
To summarize, land access is usually by customary right. Fallows and set-
tlement patterns result in a pattern of fields distant from the house. Most
fields are for food crops, but most farmers also have tree crops.
All adult women and most adult men (exclusive of a few civil servants and
aged men) contribute to farmwork, as do almost all adolescents and over a
third of the children. When working separately, men tend to work with a njan-
ggi13 group, while women work with their children. Under half the men sur-
veyed assisted with processing of crops (for sale or consumption), while all
women worked in processing; two-thirds of male and almost all female adole-
scents and a third of the children helped process.
Labour bottlenecks were reported principally in the months of March and
April, when the major activity was land preparation (mounding and late burn-
ing) a-nd early planting. Two-thirds cited these months, two-fifths cited the
late-planting month of May, and one-third the clearing and early-burning
months of January and February.
In the busiest season of the year the average woman reported working 10
hours a day, while the average man reported 6.8 hours. Male adolescents pro-
vided 4.8 hours and female ones 6.4, while children of either sex contributed
1.2. Reported totals varied very little by zone. Overall (counting adole-
scent work as 80% of an adult's and children's as 50%), the average household
put in 29.5 person-hours a day during busy season, 75% in farming.
Labour inputs varied by type of field and operation. The husband and wife
both participated in clearing 70% of fields, but husbands cleared almost all
tree fields and wives almost none, while husbands cleared three-fifths of pure
food fields and wives almost all. The husband might cut the forest or bush
t3 Njanggi is the traditional form of labour exchange, in which a group
of male or female friends work successively on one another' farms, usually
being provided with food and drink afterwards by the host.
FLU IRA-Ekona p.17
and the wife follow to burn and prepare the mounds or beds, or they might do
both together. Husband and wife worked together in clearing only 40% of
fields, mostly food ones. Clearing was aided by njanggi groups (63%, usually
male to cut and female to hill), adolescents (41%, more in food fields), and
paid labour (32%, more in tree fields).
Table 4: Hours of Work Contributed by Each Demographic Grouping
Season) by Zone14
per Day (Busy
Total Group Contribution:
Men 20+yrs 8.2
Boys 10-19 2.8
Boys <10 0.1
Girls <10 1.0
Men 20+yrs 5.4
Boys 10-19 2.9
Boys <10 0.2
Girls <10 1.2
They are not
TLU IRA-Ekona p.18
'1 Hours are those reported by the respondents) for all
during a typical day or week in the busiest period of the year.
based on actual timing or daily recall and are indicative only.
Planting was by contrast a simple operation, often done alone. Husbands
planted 30% of fields (four times as many tree as food fields) and wives 86%
(all food fields and a quarter of tree ones), cooperating on a sixth. Often
it was specified that the husband planted the trees and plantains in a mixed
field and the wife planted the rest. The main helpers were adolescents, mst-
ly in food fields.
Weeding was done primarily by the person responsible for the field and
his/her helpers. Husbands weeded four-fifths of pure tree fields and a tenth
of pure food fields; wives weeded all food fields and three-fifths of tree
fields. Adolescents helped in half the fields, usually assisting their moth-
er, while the husband resorted to njanggi when working along. Njanggi was
used in under a third of fields, of all types.
Harvesting was done mostly by women helped by adolescent children. Hus-
bands harvested almost all pure tree fields and only a few food fields; wives
all food fields and two-thirds of tree fields, and adolescents slightly over
half of all fields.
Children under ten helped mostly in harvest (a quarter of fields), usually
helping to carry produce to the road or house, and in weeding (a fifth), in
both cases more in food fields. A few of the older ones helped in clearing
Table 5: Labour Associated with Husband's & Wife's Activities
(% of fields)
Clearing Planting Weeding Harvesting
% of fields in which:
wife works separately 30 69 74 66
couple works together 40 17 17 24
husband works separately 30 13 7 9
Who helps when:
wife works separately:
dependent family members 67 49 64 68
njanggi 54 6 26 3
paid labour 28 0 4 2
couple work together:
dependent family members 64 65 68 76
njanggi 71 6 25 23
paid labour 27 4 16 7
husband works separately:
dependent family members 37 5 28 70
njanggi 61 0 65 11
paid labour 42 0 34 13
TLU IRA-Ekona p.19
Paid labour was used by 69% and njanggi labour by 94%. Faced with over-
work, 91% said they resorted to njanggi and 48% to paid labour. Rates of pay
varied according to task a man cutting forest could earn 5.000fr/day but
clustered around a median of 1.00fr and a mean of 1.500fr, from a low of
800fr in the border areas (where many recruit Nigerians), 1.500 in the High-
lands, to 1.700fr and 1.BOOfr in Mamfe Forest and Mamfe West. Njanggi groups
ranged in size from 2 to 56, with a median of 14.5. The host provided palm
wine and food at a cost averaging 5.500fr, or 380fr per participant-day, low-
est in Mamfe Forest (285fr) and highest in the Highland zone (450fr). But
several people commented that njanggi groups work shorter days than paid la-
bour. Both are used most often in clearing, and secondly in weeding.
To summarize, food-crop farming is carried out almost entirely by the
women, with help from their husbands in some of the clearing. Women help weed
and harvest tree fields as well. Njanggi is widely used in clearing and land
preparation (the latter being the main labour constraint) and by men for weed-
ing and harvesting. Paid labour is used mostly by men, and for clearing, land
preparation and weeding.
6. Cash Flow and Input Use:
Manyu farmers said they needed cash most for land clearing and preparation,
school fees, and the Christmas-New Year holidays, when clothing and household
goods are replenished. The month of December (for New Year's) was cited by
64% of farmers, followed by September (for school fees) by 37% and the land-
preparation months of March (32%), April (33%) and May (20%), and finally the
clearing month of January (20%). Two-thirds cited agricultural labour (whe-
ther paid or njanggi costs) as a reason for needing cash, almost as many men-
tioned Christmas or New Year, a third school fees and supplies, and a few food
or seed. An average 58.600 francs were spent on paid labour, from a low of
9.600fr in Akwaya to a high of 89.200fr in Mamfe West. Most families bought
rice and/or garri occasionally all year round, and one to three supplementary
foods (plantains, maize, yams, egusi or groundnuts) in months of scarcity,
especially March to May.
The main sources of cash were food sales (mentioned by 47% of households),
cocoa and coffee sales (32%), off-farm work (9%) and animal sales (4%). Only
one mentioned palm oil sales as important, although there are many local small
producers of the commodity. A few said they had no source of cash so did
Fifty-six percent saved or borrowed sums to meet cash crises in agricul-
ture, most in a "meeting"-'. The average amount was 61.500fr, with Akwaya and
the Highlands least often saving/borrowing. Akwaya savers amassed the least
I' An institution introduced after World War II, in which friends or
relatives meet regularly to save money and loan it out to each other at inte-
rest. Sometimes the savings or profits are divided or used for a group cele-
bration at Christmas.
TLU IRA-Ekona p.20
capital, and Mamfe West the most. Three-quarters used meetings, 12% saved at
home, 8% in a bank (Highlands and Mamfe West), 3% had loans from kin and 2%
from Young Farmers' Clubs.
Table 6: Average Household Crop Sales and Other Income Sources, by Zone
Mamfe Mamfe Eyumo- Akwaya High- Manyu
,Forest West jock land
food crop value
food crop sales (FCFA):
irish potatoes 0
% w/ animal sales 36%
% w/ cocoa/coffee 84%'
% w/ other job 44%
t1 Production totals were costed at the mean prices reported for sales
in the survey, treating Akwaya separately because its price structure is so
different: see Table 9 (cassava in root form), also: bananas 400fr/bunch, yams
3250fr/basin (Akwaya 2000fr), groundnuts 2650fr/tin (unshelled)(Akwaya 480fr),
beans 2420fr/tin, irish potatoes 7200fr/jute bag, and rice 2550fr/jute bag.
TLU IRA-Ekona p.21
Only the Highland farmers used purchased inputs (other than labour) on
their land. Outside the Highlands, one rice farmer in Akwaya reported putting
an unidentifiable Nigerian fertilizer on his rice, and none used fertilizers
on cocoa or coffee. A quarter of Highland farmers used fertilizers on coffee,
and a third used them on food; two-thirds of the latter were from the large-
scale food-crop-farming sub-zone of the upper Highlands. The food fertilizer
used was N-P-K 20-10-10, with isolated cases of animal manure and sulphate of
ammonium. Maize and irish potatoes, cocoyams, carrots, garlic and leeks tin
the highest sub-zone) got the majority of the fertilizer.
Fungicides and insecticides were used uniformly across zones by about a
quarter of surveyed farmers on their cocoa and coffee. However, many farmers
were limited,to the small demonstration quantities available from the Ministry
of Agriculture. One Eyumojock farmer put Gamaline 20 on plantains. A third
of Highland farmers again mostly from the highest subzone used chemicals
on their food, including Gamaline 20, Dacobre 500 Sandoz, Ridomil Plus, Round-
up, Timol, Difolatan and other unidentified insecticides, fungicides and
herbicides. Irish potatoes received most of the chemicals, followed by car-
rots, garlic, leeks and maize.
Input sources in the Highland zone are both UCCAO and an agricultural mis-
sion group in West Province. Availability was severely reduced in 1987 due to
the transfer of responsibility for Arabica coffee purchases (and input supply
and feeder road maintenance) from UCCAO to FOFCOOP in Fontem. Many farmers
have been accustomed to buying fertilizers, chemicals and seed in their local
cooperative stores from UCCAO, and are badly worried. The taxi service to
Dschang is costly and scarce and sometimes the roads are impassable.
Outside the Highland zone, the cooperatives had no chemicals and rarely any
fertilizer. The farmers and agents said some fertilizer was available in
Farmers were asked what kinds of inputs they would buy if they were commer-
cially available within their Sub-divisions. Tools were the most popular,
especially cutlasses (83%), hoes (76%), sprayers (28%), diggers (24%), files
(19%) and spades (12%). Smaller numbers wanted axes and hand trucks. Among
chemicals 53% wanted insecticides, 25% herbicides (mostly in Mamfe West and
the Highlands) and 18% fungicides. A few wanted nematocides or rat killer.
Few were knowledgeable about the names of chemical products; they would ask
for something to kill insects or grass or solve black pod. In the border
areas, anonymous powders were occasionally sold from Nigeria. Two-thirds
volunteered that they would buy fertilizers if available. Half asked to buy
maize seed, a third plantain suckers, a quarter each cocoyam and cassava plan-
ting material, and a seventh, yam setts.
Overall, cash needs are highest at the end of the year and in March-April,
and are satisfied by food- and tree-crop sales. A majority use local "mee-
tings" to save money, especially in the central lowlands. Only the Highland
farmers use chemicals on food, or fertilizer on any crop whatever. The High-
land source of fertilizers and chemicals is West Province, and has been threa-
tened by cooperative reorganization. Basic tools are in great demand, and
most are interested in fertilizer and chemical purchases.
TLU IRA-Ekona p.22
Manyu Division's extension service is staffed predominantly by untrained
field assistants, making it difficult to provide good advice to the farmers.
In 1988, new graduates had been posted to the Division for several years, but
they were understandably reluctant to be sent to areas distant from Mamfe
Town, and those who were, were often absent from post. The Sub-Delegates'
work vehicles could not be maintained because lack of money and facilities for
repairs, so they could seldom get to the outlying areas to check on presence
and activity. As a result, many junior staff were also absentee.
At the time of the survey, there were only five Agricultural Posts in Manyu
in addition to the four Sub-Delegations: Kembong, Tinto, Kemden, Ebeagwa, and
Nkongle (vacant). In addition there were three Extension Units in the more
populous areas of Mamfe Central, where a technician was based. Akwaya and
Fontem had placed a technician each outside the Posts. Otherwise, the farmers
were served by field assistants. Because of the lack of motorable roads, the
Sub-Delegates of Fontem, Eyumojock and Akwaya were forced to trek long dis-
tances in order to know their areas and follow up and encourage staff.
All this said, it is remarkable that as many as 28% of farmers surveyed had
been visited by an extension agent. The 19 survey sites included two with a
Sub-Delegation, two with a Post, four a technician, four a posted field assis-
tant, and seven without a local agent. Three-fifths of the farmers knew both
name and location of their agent (a quarter in Akwaya and three-quarters in
the Highlands), and only a quarter knew neither. Half the visits by agents
were for cocoa or coffee, a quarter for food crops (all in Fontem), and the
rest about the Best Farmer Competition and Agricultural Show in Maroua.
Female-headed households were slightly (p =.10, t-test) less likely to receive
extension visits than couples.
Two-fifths of farmers had asked an agent for help, usually for tree crops
(71%) and for supplies (67%) rather than advice. Food-crop help was sought
only in Akwaya and the Highlands. Half of those who asked said that they had
Almost two-fifths had been to a meeting convened to discuss agriculture. A
third of these were said to concern cocoa or coffee, 45% food crops (agronomy
and promises to distribute materials), and a fifth organizational matters
(cooperative elections, the Maroua Show, the Young Farmers' program, and an
effort in Akwaya to get to know the area's farmers and problems). Most (ex-
cept in Akwaya where the extension effort was just starting) felt that they
had learned something useful to their farms from the meetings.
The allocation of motorcycles to the technicians and most active demonstra-
tors, as has been done in Fako and Meme, would certainly be cost-effective in
terms of expanded coverage of farms and villages. In Fako and Meme taxis are
plentiful, allowing the agents to reach other villages and the Sub-Delegations
for instruction and materials. In Manyu and Ndian, taxis are rare and expen-
sive. Under the circumstances at the time of the survey, Manyu agents spent
more time walking to the farms assisted than in providing assistance, and were
limited to walking distance from their homes. A motorcycle would be more
'TLU IRA-Ekona p.23
useful than a car for agent supervision in most areas off the motorable roads,
Repair facilities for the delegates' cars are very poor, and budgets for the
repairs insufficient. For the village-bound agents, even bicycles would be a
help, if motorcycles are impossible during the crisis.
Table 7: Extension Contacts, by Zone
S Mamfe Mamfe Eyumo- Akwaya High- Manyu
Forest West jock land
% visited by agent:
never 76 75 90 87 60 72
yearly/less 16 20 10 7 12 15
more often 8 5 0 7 28 13
% who have asked
help of extension:
never, won't 56 55 40 60 40 49
", but could B 0 10 0 8 5
would ask other 4 10 20 7 0 6
have asked ext. 32 35 30 33 52 40
% who have been to an
agric meeting: 36 30 20 40 44 38
% who were helped:
when asking 25 71 67 17 54 52
by meetings"' 75 100 100 33 100 91
Poultry ownership is widespread in Manyu, but other animals are primarily
found in the Highlands, particularly iD the highest sub-zone, where ranching
was the original land use pattern and large herds of goats and sheep are com-
mon (averages of 82 and 99 respectively in the survey). Almost all Highland-
ers surveyed owned some animal, and even outside the highest sub-zone, 50%
owned goats, 85% poultry, and 55% swine. The highest sub-zone also raised,
sold and ate cattle and horses. Highland owners were much more likely than
others to sell their animals.
Outside the Highlands, goats are found mostly in the Mamfe Forest zone,
where a third of surveyed households had an average of five animals each.
Help with something applicable to improving the farm. Meetings also
dealt with topics not directly applicable but potentially also helpful to the
TLU IRA-Ekona p.24
Sheep were found in six households, and pigs only in three. About half of
owners had sold one or more of their animals of each type within the year.
Three-quarters or more owners had also eaten one or more of each type during
the year. Bush meat (wild game) and fish are frequent protein supplements in
Animals are mostly left to find their own food, but sometimes crops are
used for supplementary feed. Farmers fed maize (35%), purchased rice (19%)
and cocoyams (6%) to poultry; cocoyams (11%) and bananas (7%) to swine; and
cassava leaves and peelings to goats and sheep (10%). The highest sub-zone
maintains fenced pastures with cultivated grass and salt licks for their
Table 8: Livestock Ownership, by Zone
( ) only one case.
* among owners.
Because the animals are free-grazing, food-crop fields are difficult to
maintain near villages that do not impose restrictions on goats and sheep.
Some village chiefs ban these animals, some enforce rules about tethering
animals with heavy fines, but most do nothing. Goats especially are a sign of
men's wealth and a requisite item for ceremonial welcomes to dignitaries, so
the chiefs tend to favour them over the convenience of closer food fields.
Women in some places manage to leave dense bush and cane fence between their
field and the village, or plant across a stream, but the goats usually find a
TLU IRA-Ekona p.25
way across. However, surveyed farmers blamed wild animals more often than
domestic ones for crop damages, particularly citing the cutting-grass1e.
Manure is only applied in rare cases. Outside the Highlands, one peron
used goat manure, two poultry manure and one sheep manure. In the Highlands,
a fifth used manure: four poultry, three each sheep and goat, two pig and one
cattle and horse manure. Outside of the highest sub-zone, the Highlands usagQ
fell to 11%, almost that of tho lowlands (7% of those with animals).
9. Food Markets:
Markets all over Manyu are small and local. The three largest markets ore
Mamfe Town, an urban consumers' market supplied by traders, and Afap (in Mamfr
West) and Mfaka (border of the Highlands and Mamfe Forest), both farmeaPr
markets supplying the traders. All of them together would fit easily inte one
small-sized Meme or Fako market. The other lowland markets open for 2-3 hours
once a week, and attract only villagers and traders reselling food from Afif
and Mfaka. The Akwaya market is insufficient to feed the town, and government
and mission staff import food stuffs from Nigeria by headload. Highland
markets are larger and last longer, serving their local areas and some few
Dschang traders; the state of the roads makes the Dschang trade uneconomic in
Farmers surveyed in the Mamfe Forest, Mamfe West and Eyumojock zones wev*
to markets in their own or neighboring villages, travelling a half hoyr *r
less on average. Afap andr F.yumojock markets drew some farmers in their re-
spective zones. Over a fourth of Mamfe Forest farmers didn't sell in a9f
market at all; cash came f~-n food sales to neighbors, cocoa and coffee, ow
remittances from emigrant relatives. Market prices were higher than in *~)
Highlands, although in Eyumojocl light-weight produce that could be easily
transported to and from Nigeria was cheaper.
Akwaya is totally cut off from Cameroonian markets. Rice, groundnytso
egusi, garri and cocoyams are all carried down to Amana, the nearest road!iOp
market in Nigeria, for sale at the low Nigerian prices. The internal mark*
is very small but even so, inadequately supplied. Prices cited in the survey
were a third or less of Manyu average for crops sold mostly in Nigeria (e*e
when sold inside Akwaya), and a half to two-thirds of Manyu average for crop
mostly sold locally. Still, because of the lack of other income opportunities
practically all farmers marketed food crops.
A quarter of Highland farmers surveyed did not sell in markets. The rolae
terrain increase the effective distance for sellers to trek and trader-buyjr4
to drive. Arabica coffee brings in an alternative income, food prices quctte
were relatively low, and people carried home more unsold produce than else-
where. Surveyed farmers sold less of their food production than elsewhere in
18 Thryonomys swinderiani.s.
TLU IRA-Ekona p.26
Manyu, except for the massive sales of high-altitude vegetables.and potatoes
sold to Dschang traders from the highest sub-zone's Nkongle village market.
Table 9: Market Distances, Transport and
Prices, by Zone
Eyumo- Akwaya High-
time to market
farmers not going
to any market:
basin of roots
basin of garri
egusi (jute bag)
10. Perceptions of Farming:
At the end of the interview, farmers were asked for their general opinions
of farming prospects in the area. Only one farmer intended to leave farming,
and only a few felt that farming was in decline. Most said it was increasing
with the advent of young people, more Ministry of Agriculture help, or simply
because there was no better alternative these days.
Farmers saw the future in food crops. Asked what they would want to inr
crease if they had more labour, 62% named only food crops (from 54% in thl
Highlands to 93% in Akwaya) and the rest named both food and tree crops. Food
crops preferred corresponded closely to those already grown over half named
cocoyams, maize, cassava or plantains (in that order), and over a quarter
taro, yams or groundnuts. Relative to percentage already planting a crop, th*
most popular candidates for expansion were cassava and cocoyams, followed d by
19 Often sold as basins of dried cobs; this is the grain equivalent. .
TLU IRA-Ekona p.27
plantains and maize, yams and taro.
Farmers noted a decrease in the fertility of the land, which might be due
to pest and disease as well as soil problems. Three-quarters noted a decline
since they had started farming, and a fifth an improvement. Land was con-
sidered by 90% to be harder to obtain than at the beginning of their careers.
The primary complaint of farmers was about crop destruction by animals,
voiced by 53% (only a quarter in the Highlands, half in Akwaya and over two-
thirds elsewhere). The cutting grass was most often implicated, along with
bush pigs (especially in Mamfe Forest), monkeys", bush fowl, goats, rat
moles, porcupines and birds. Many farmers asked for rodent poisons, govern-
ment action or Forestry permission to kill animals.
Crop diseases and insect pests were the second and third most named pro-
blems, by 37% and 35% of respondents. These were often confused, with root
and tuber rots being attributed to ants or cassava mealybug to disease. In-
sect damage was perceived least often in the Highlands and Akwaya, and disease
in Mamfe Forest and the Highlands. Cocoyam and plantain pests or diseases and
maize and egusi pests were most often specified, except in Akwaya, where the
concern with "Apollo" (cassava mealy bug) eclipsed all others.
The lack of chemicals almost matched these concerns (34%, but only a quar-
ter outside the Highlands). The Highland farmers were vehement about the
disappearance of chemical, tools and other supplies with the changeover from
UCCAO to FOFCOOP, and some farmers elsewhere also thought that chemicals were
becoming even less available than before. Most farmers meant tree-crop chem-
icals, but many also wished for chemicals to fight food-crop insects and
A fairly uniformly distributed 20% of farmers complained about lack of
money, usually specifying capital to pay labour to open more lard.
Markets and roads to markets were a problem only to Akwayans (47%) and
Highlanders (44%), but serious there. The Highlands has an elaborate road
network, dug by the farmers themselves with help from UCCAO, but few of the
roads are passable during the rains from May through October.
Some Mamfe Forest and Highland farmers complained of soil infertility or
lack of fertilizers. Some in Mamfe West complained of the lateness of the
rains, which in 1988 only came in earnest in late May. A few in most zones
complained of lack of planting material and extension help but more stated
that the current increase in extension help was going to improve local farming
Le The most common monkey is Cercopithecus mona, the porcupine Atherurus
africanus, the rat mole may be Lophuromys sikapusi and the squirrels include
flying ones (probably Anomalurus durbianus) and a rodent resembling a hairy
anteater. Identifications from A. H. Booth, 1970, Small Mammals of West Afri-
ca. London: Longman.
TLU IRA-Ekona p.28
E. Results: Manyu Cropping Systems:
1. Food-Crop Field Selection and Management:
Most Manyu fields in cultivation are food-crop fields: 77% in the survey,
including 7% with secondary cocoa or coffee trees. This actually understates
the case, because it omits second- and third-year cassava fields (often with
plantains and resprouted taro or cocoyam) planted the year before and still in
harvest. However, food-crop fields were smaller, averaging 3 weeks to weed as
versus 6 weeks for tree-crop fields. Farmers planted two to three food fields
a year, with 15% planting only one and 8%, four or five.
Farmers surveyed believed that plantains and bananas, cocoyams, taro, and
yams would do better in new fields, but other crops in any field. Mamfe Fo-
rest farmers felt cocoyams and taro could do well in older fields but egusi
and beans only in new ones; Akwaya farmers preferred new fields for egusi
Shifting cultivation is still the rule throughout Manyu, and the need to
leave land in fallow and to escape damage from goats causes most fields to be
planted distant from home, an hour's trek on average. A fifth of fields were
planted within 30 minutes' walk, and a sixth over two hours'. Distances are
furthest in the Highlands (where steep slopes slow times) and Eyumojock (where
reserve forest intervenes). Quite a few villages have restricted goats, and
there women plant some fields close to their houses. These are usually smal-
ler copies of the more distant fields, often planted with the first rains to
reduce the hungry season. Those surveyed contained long-cycle crops as often
as the more distant fields did.
Most Manyu farmers surveyed and visited practiced bush fallow; that is,
they planted a particular field once every four to six years, harvested
through the second or third year as the field reverted to bush, and then clea-
red it again before it became forest. Most fields visited had scattered trees
and living stumps in the cleared fields; the farmers had burnt around them to
inhibit the foliage. These trees act as yam stakes, hold the soil, and hasten
the reversion to bush.
In Mamfe Forest, much of the soil is extremely hard to work, and farmers
incorporate the woody parts of the cleared bush ("sticks") as well as the
leaves ("grass") into their mounds for first planting. Burning is done pri-
marily to "discourage" tree foliage. The farmers in visited fields claimed
that only tuber crops could do well in the stick mounds; they planted the same
fields a second and third year (these times incorporating only the "grass") to
obtain maize, egusi, and groundnuts. Half of those surveyed who planted maize
did so twice a year, although the second crop was smaller. Visited farmers
said that this second crop was primarily to maintain seed stocks for the main
season, as their storage losses were high.
Highland farmers also usually utilize the grass in making their beds. In
the highest sub-zone, which has very rich soil, farmers plant the same fields
continuously for years, rotating between potatoes and carrots, cabbages, leeks
and garlic. Visited farmers claimed to supplement with fertilizer if yields
TLU IRA-Ekona p.29
decline, or to convert back to pasture when (rarely) fungal and cutworm pests
become unmanaqgeaihl. Most of the other Highlanders surveyed planted one year,
sometimes two seasons, and then fallowed three to five years, depending on how
good the response had been, or abandoned the field and opened another. In
visits it appeared that most newly opened fields had been abandoned by a
relative some years before. Forest cover is light.
Table 10: Fallowing"1, Clearing and Burning Practices (Food Fields)
Mamfe Mamfe Eyumo- Akwaya High- Manyu
Forest West jock land
Yrs. in contain. use 2.1 1.6 1.2 1.5 2.4 1.9
Interim fallow (yrs) 2.0 4.5 4.0 4.4 3.1 3.4
Fallow 1 cultivation 1.2 3.2 3.5 3.3 1.8 2.3
. fields <3 years old 12 10 15 41 30 21
V fields >4 yrs old
never fallowed 21 0 0 6 5 8
Clearing months NDJF JF NDJF DJF ONDJ NDJF
Burning months JFM JFM FM FM JF JFM
X burning 81 75 92 73 32 65
Some Mamfe West farmers visited incorporated grass in their mounds but
usually burned over the field before mounding, as did the Eyumojock farmers.
Both tended to return to the same fields for many cycles, to plant once, har-
vest several years and fallow for three to six years.
Akwaya has more land and a savanna environment; fields and fallows visited
reverted to meter-tall grass and scanty, stunted trees over a period of 5-6
years, and stayed that way. Farmers open new fields frequently, not keeping
track of who has farmed them before or when, but judging readiness by vegeta-
tion. The few trees are felled, leaving denuded slopes, the field is burned
over, and planting is done at random across the flat. Fields are often plan-
ted only one year before reversion to fallow, but can be planted up to 4
years, with declining yields.
Once planted, a Manyu field is weeded one to three times during the year,
averaging 1.7 times in the survey, and varying according to the presence of
long-cycle crops. The first weeding is usually at two months (or 1.5 in the
second season) for fields with short-cycle crops such as maize, groundnuts and
agusi, and 2.5-3 months for fields dominated by long-cycle crops. The second
Claimed years of continuous use between fallows, actual average fal-
lows for fields over 4 years of age, and adjustment of fallowicultivation
ratios by percentage of fields never followed.
TLU IRA-Ekona p.30
weeding is generally around September-October. The weeds in visited fields
were left where they fell or purposefully placed inside the beds and mounds,
usually to discourage animals from uprooting the tuber crops. There is no
idea of plant hygiene. Borer-infested maize in visited fields was sometimes
cut out, but left next to the surviving plants. Rosette-infested groundnuts
were left in place to infect the others. Root-rotten cocoyams and taro were
uprooted as they died only if they were large enough to eat, but not other-
wise. However, in some places most women had stopped growing cocoyams altoge-
ther for ten years in response to the rot, and then tried again.
The final (third) weeding is often done in preparation for harvest, to
locate and collect the earliest-maturing yams, taro and cocoyams. It is thus
of little use for stimulating growth. However, the density of planting on
mounds and beds often leaves little room for weeds to compete.
Women, assisted by their children, do almost all the planting, weeding and
harvesting, and much of the clearing, of the food crop fields.
2. Food-Crop Field Design and Crop Associations:
Intercropping is the rule in Manyu. Food fields surveyed averaged 2.9
major crops and 2.0 secondary crops. Three-quarters of the fields had at
least 3 crops, over a half had 4, and a third had 6 or more. 20% of all
fields were monocropped, but only 11% of the food fields most monocrops were
trees. The most common food monocrop was cassava. In fields visited, cassava
was planted sole where the production of the previous year's mixed planting
had failed to justify the labour of clearing it. A fifth of cassava fields
were monocropped. The only other foods monocropped were plantains (3% of
plantain fields) and irish potatoes (20%). Even effective monocrop (in which
small amounts of other crops are interplanted with a sole major crop) was
limited: 7% of cassava fields, 5% of maize and vegetables, 20% of irish pota-
toes, and 1-3% of plantains, bananas, yams and cocoyams.
Crop associations in surveyed fields followed almost no rules: everything
was intercropped with everything else (Chart 2, Table 11). Bananas tended to
be found primarily with plantains, and plantains with cocoyams and/or taro.
Cocoyams and taro were often together. Beans, groundnuts or equsi were usual-
ly with maize in the Highlands, Akwaya and Eyumojock. Otherwise, every crop
had a nearly equal chance of being paired, tripled, quadrupled or quintupled
with every other one.
Most planting was simultaneous in each section of a visited field. Usual-
ly, the plantains and bananas were either found and preserved in the field
when it was cleared, or planted there by the husband before or after his wife
planted the other food crops. Cassava was sometimes planted in August as the
taro and maize were removed, or one to two months after other plantings. Many
farmers visited said that cassava (or tree seedlings) that were crowded during
the first few months by other crops, could recover adequately after their har-
vest. Others delayed planting cassava, cocoyam and taro in mixed fields until
they were sure of steady rains, thus giving the short-cycle crops and yams a
TLU IRA-Ekona p.31
Chart 2: Intercropping Patterns
be eg n
Table 11: Crop Associations (number of fields with the crop(s))
95 56 114
45 27 71 64
20 16 37 32 57
74 51 61 69 37 19
59 27 36 38 20 11 32
92 57 59 57 26 11 51 37
84 58 61 64 30 13 52 27 59
155 102 143 138 115 65 95 68 98 total fields
mz cs cy tr pl ba ym be gn
(mz=maize, cs=cassava, cy=cocoyam, tr=taro, pl=plantain, ba=banana,
ym=yam, be=bean, gn=groundnut, and eg=egusi melon)
TLU IRA-Ekona p.32
head-start of about a month.
Mound-planting typically followed standard patterns, but these varied from
farmer to farmer visited. One common one, especially in Mamfe West and Eyumo-
jock, was the placing of a cocoyam or yam in the center of a mound, and then
surrounding it with 2-3 alternated pairs of maize and egusi or groundnut.
Another common in Mamfe Forest was alternation of taro with maize, groundnuts
and egusi, with the taro densities being extremely high (20-50.000pphme).
Cassava was planted into the sides of the mound, nearly parallel to the
ground, and covered over with grass to hide it from cutting-grass rodents.
Mounds varied in size (sometimes in the same field) from 30cm to 2m across,
20-60 cm tail. Small mounds would have one stick of cassava and/or one yam,
large ones two or rarely three. Distances between edges of mounds varied from
field to field, between 20 and 60cm. Estimated area in mounds in fields visi-
ted varied from 20-25% to 60-80% as mound and alley size varied. Densities
were high relative to agronomic recommendations, except for second-year cas-
sava. Each mound or small group of mounds was planted to all crops at once
(unless some crop was to be planted in another month), and different parts of
a field were planted sequentially over a month or more.
Bed-planting was found mostly in the Highlands fields visited. Usually
there was an alternation of crops in a rough line down each side of the length
of the bed, but it was not unusual to find thai in planting down a whole bed
the woman had forgotten where she had just put the maize or bean seeds, with
the result that the mature bed would have maize erupting from under a cocoyam
stem and vegetable struggling to find a way around a bean. Beds seen were
usually planted down-slope, which is easier on the legs and back and is felt
to limit bed erosion. But the extension service has had some success in con-
vincing people to plant across the contour to preserve the soil. Most beds
were about 90cm wide and 3m long, but some were continuous down long slopes or
up to 2m wide. Alleys also varied from 30 to 70cm wide. Highland beds measu-
red occupied 60-70% of the surface area, and densities on them varied greatly.
Akwaya fields observed to be dominated by a single crop in first season,
usually groundnuts or rice (or sometimes both). Smaller amounts of cassava,
maize, taro and cocoyam, egusi and water yams were planted inside, after the
rice had sprouted. The cocoyam and taro were concentrated in the wettest part
of the field. Most fields had a small stream-bed passing through where early
crops were planted in the first rains. The only exception to planting on the
flat was small, 10cm-high mounds built over the yams. First-season planting
was very dense, but after short-cycle crops had been harvested, the cassava
and cocoyam were left at low densities.
3. Iree Crop Fields:
One quarter of fields in cultivation by surveyed farmers were for tree
Spph = plants per hectare
TLU IRA-Ekona p.33
crops about 60% in cocoa, a third in coffee and the remainder in oil palms.
Almost half of these fields were monocropped, while a third were inter-planted
with plantains or bananas (usually secondary) and a fifth with these and small
amounts of other food crops. Tree crop fields were larger than food fields,
taking twice as long to weed on average. They also tended to be more distant
from home, averaging 35% farther; more food fields were under an hour away,
while over a quarter of tree fields were over two hours' trek.
The tree fields were generally long established, averaging 18 years of age.
More than half were over 15 years old, and only a tenth under five years. The
secondary food crops were rotated through short fallows in two-fifths of the
few mired fields, and otherwise continuously cropped. Fertilizer and manure
were not applied except by some Highland farmers. Chemicals were used by a
quarter of farmers, and many complained of unavailability. Densities were low
in the lowlands (700 to 800 pph) and high in the Highlands (1600pph for cocoa
and 2000 for coffee). However, Highland farmers did not follow the Dschang
pattern of intercropping already dense coffee with food crops.
Planting was done in June, July or August, often after secondary plantains
and other crops had been planted in the same field in April. Weeding was done
an average of 2.7 times a year (maximum four), generally in November-December,
May-June and September-October. Parts of the field were burned over in 37% of
cases, usually in January, in preparation for new secondary planting. Cocoa
harvesting occurred from August to January, peaking in October and November,
while coffee harvesting was from November to January, peaking in December.
Men did most of the planting, spraying, weeding and harvesting, but women
helped with the latter two operations.
F. Crop Agronomy and Utilization:
This section provides details on the incidence, cropping patterns and ca-
lendar, varieties, field problems, processing, storage and marketing of each
food crop of any importance in the Division. Readers may prefer to skip to
particular crops and then the section on recommendations. But first there are
some specific cautions to note.
Almost fourth-fifths of farmers said they planted at the usual time first
season, a quarter planting early. Almost all planted at the usual time second
season. Two-thirds said they planted more than usual first season, but 60%
planted less in the second season. These changes from usual habit must be
recalled in considering area, production and the calendar. Calendar dates are
early cr late in a particular month, not specific days.
As described in Section C., production and especially area figures are
tentative. Farmers' estimates of field problems and losses are relative to
their general expectations for a crop; they are not real "yield gaps", the
difference between what should be obtainable in an area and what is obtained
TLU IRA-Ekona p.34
currently. Crop sales were discussed in two separate parts of the interviews
at one point, the farmer was asked how much of the 1987 harvest was sold, and
in another, in what forms the crop is eaten and sold. As a result, some far-
mers mentioned how they usually sell the crop but did not report selling the
crop in 1987.
lpr rb6-ncei Approximately 5.050 tons of maize were harvested in the 1987 crop
year, equivalent to 690 million francs at local prices or 18.3 million kilo-
calories of food energy. This made it third in economic importance and second
in nutritional importance in Manyu Division.
Ecozonesi The most important zones were the Highlands, with 41% of total
production, and Mamfe Forest, with 32%. Mamfe Forest produced an average
415kg per household, the Highlands 305kg, Akwaya 190kg, Mamfe West 160kg and
AMoEieatiesl( Maize was planted in 48% of fields by 99% of far-mers, as a
major crop in 76% of the fields. Among fields with the crop, 5% was inter-
cropped as a sole major crop with secondary crops and 95% as major or secon-
dary crop with other major crops. In Mamfe Forest and Mamfe West, it was
grown in association with cocoya-rs, taro, groundnuts and aeusi in more than
half its fields, and with cassava, yams and beans in over a third. In the
Highlands, beans and groundnuts were the most common intercrops, in Eyumaooek
egusi, yams and groundnuts, and in Akwaya, groundnuts.
Planting methods Maize was planted on mounds (47%), beds (39%, mostly in ie
.Highlands), and sometimes on the flat (mostly in Akwaya). Bed-planting and
flat-planting were each used by a quartertof Mamfe Forest farmers. Mound-
planting was almost universal in Mamfe West and Eyumajock. Planting distances
varied from 25 to 160cm, with a median of 70cm and a mean of 77cm. An average
of 2.5 seeds were planted per hole, ranging from 1 to 5 but with 94% planting
2 to 3. Densities varied around a mean of 42.000pph, ranging from a high
average of 64.000pph in Mamfe Forest to under 30.000 in Akwaya and Eyumojock.
Converted to monocrop equivalent, the crop occupied an estimated 7% of the
total field area, or 15% of food-crop fields.
Calendar. Planting was usually done in first season (98%), generally in late
March and April at the first rains, in the lowlands (later in the border
zones), and in February or March in the Highlands. Second-season planting was
done by 36%, mostly in Mamfe Forest and the Highlands, in August. Third-sea-
son planting in November in swampy areas was carried out by 6%, also in Mamfe
Forest and the Highlands. The first season's first weeding was done by 86% of
-farmers growing the crop, and a second by 18%, at an average of 2 months and
e.9 months after planting respectively. Second season, the first weeding was
done by 38% at 1.7 months and the second by none; third-season weeding were
doneby 58% and 35%. The Highlands and Akwaya weeded earlier first season,
and Mamfe Forest second. Orly 18% weeded by one month first season, and 27%
second or third. Harvesting -egan at an average 3.4 months after planting
first season, lasting an average 3 weeks, and began and ended earlier in se-
TLU IRA-Ekona p.35
cond and third seasons.
Varieties: Farmers preferred their maize soft, except for minorities in the
Highlands (40%) and Akwaya (33%), and white, except for 15% (highest in Ak-
waya) that preferred Bamenda yellows or a local mixed-color composed of mixed
purple, blue, black and white grains. The dominant local in the lowlands was
Calabar, a soft, white, floury, short, early-maturing, weevil-prone variety
with small ears originated in Eastern Nigeria. Local Highland varieties were
flinty, hard, and white or pale yellow. All preferred sweet maize, and many
mentioned large ears, weevil resistance and brightness (shiny grains) as desi-
rable characteristics. 72% had planted only white maize, 25% yellow (or
mixed-color) and white (most in Akwaya, Mamfe West and Eyumojock), and 4% only
yellow (or mixed-color). At least 26% of the seed was Calabar, 13% Bamenda,
and 2% a probable Bamenda improved variety ("Ntui"); the rest was identifiable
only by color. 34% was bought in the market, usually from other local far-
mers. 67% used part or all of their own seed, but fully 84% claimed to store
seed to plant, indicating substantial losses to weevils.
Field Problems: Farmers reported losing 20% of their crop, 22% of first-sea-
son production and 16% of second-season, with the worst losses occurring in
Eyumojock (34%) and Mamfe Forest (28%). Only the Highlands reported more
losses in second season. Losses were often attributed to multiple causes, but
primarily to stem borers (58% of complaints), which appeared equally in both
season, and to wild animal destruction (33%, or over half in Mamfe Forest and
Mamfe West). Borer infestations ranging from 2-67%, but usually under 5%,
were seen in 42% of the fields visited that were over a month old. Highlands
and Mamfe Forest farmers also complained of stunting and yellowed leaves,
sometimes identified as streak virus and once, tar-dot. Streak virus attack
was seen in 83% of the fields visited, from 5-67% infestation, averaging 26%.
Eyumojock had a major problem with termites destroying seed and young shoots,
and every zone had some damage from leaf-cutters. Poor grain formation and
drought were also mentioned. Drought and early bird damage were common in
1988, and were combatted by replanting until the farmer ran out of seed.
Asked why they did not grow more maize than at present, 16% said they would
not want to, 10% gave excuses of lack of land, labour or capital, 20% `d
they had run out of seed, 18% complained of animal damage, 13% of field insect
damage and 10% of weevils, 11% mentioned poor yields and 3% poor land, and 13%
the lack of a market in which to sell it.
Processing and marketing: Over three-quarters of the households ate mai-e
almost daily during the harvest months, but none were able to eat it all ye-r
round. Only one never ate it at all. Paradoxically, considering their j-
duction, the Highlands ate maize the least, over a third eating it less th:.i 4
times a week. 8% said they bought maize most months of the year, and 27%
bought it in months of scarcity, Mamfe Forest farmers being the most common
Maize was most often consumed.green by 97% as roasted ears, 92% as boiled
ears, and 73% as chaff (boiled grains, usually fresh, with beans) but ~. o
dried by 74% as koki (ground, mixed with pepper and leaves, tied in plar..
leaf and boiled), 41% as fufu (coarse flour porridge) and 34% as pap (i...
ground flour porridge). Fufu was rarely consumed in Mamfe West and Eyumojoc
TLU IRA-Ekona p.36
Two households fried maize, one pounded it, one ate it raw and one made beer.
35% of maize produced was sold, about half in Mamfe West and Akwaya and a
fifth in the Highlands, by 52% of producers (highest in Mamfe West and Mamfe
Forest and least in Akwaya). 99% sold maize green (sometimes already boiled)
and 77% dried, usually still on the cob. 63% sold maize for seed (only a
fifth in Akwaya) and a third or more in Mamfe Forest and the Highlands sold it
for animal feed. Most sold in their village market (56%) or the nearest lar-
ger village (30%); 15% sold at roadside to neighbors and passers-by. The
primary selling months were June to August and November in Mamfe Forest, July
to August and November in Mamfe West, July and August in Eyumojock and June
and July in the Highlands. Although heaps or basins of cobs were the most
common selling unit, some in Mamfe Forest and the Highlands sold whole jute
bags full. Average reported prices per kilo of large volumes (basins and
bags) of dried grain equivalent varied from 40-45fr/kg in Eyumojock, the High-
lands and Akwaya to 75-85 in Mamfe Forest and Mamfe West; but average prices
per cob were lowest in Mamfe West and the Highlands, where selling in small
heaps of cobs was more common.
Storage: Maize was stored by 89% of farmers for over a month, from 60% in
Eyumojock to 96% in the Highlands. Storage averaged 6.6 months, with 45%
going bad, mostly from weevils, by the end of storage. 51% put the maize in
bandase3 to dry, 43% hung it on vines over the fire, and 4% hung it away from
smoke (which controls weevils), thinking it would not germinate if smoked. 3%
bagged it before drying. After drying, most kept it on the banda or on ropes
over the reduced, normal kitchen fire; 9% bagged it and 3% put it in sealed
tins, with "country onion" or wood ash to deter weevils. 87% reported weevil
damage and 6% rat damage. Method of storage was unrelated to reported level
of loss. 84% saved for seed, and 29% for sale.
Importance: Approximately 12.400 tons of cassava were harvested in the 1987
crop year, equivalent to 770 million francs at local prices or 16.5 million
kilocalories of food energy. This made it fourth in economic importance and
first in nutritional importance in Manyu Division.
Ecozones; The most important zones were Mamfe West, with 63% of total produc-
tion, and Akwaya, with 12%. Mamfe West produced an average 2.9 tons per
household, Akwaya 2.6 tons, Eyumojock 1.4 tons, Mamfe Forest 560kg and the
Highlands only 260kg.
Associations: Cassava was planted in 29% of fields by 76% of farmers, as a
major crop in 87% of the fields. Among fields with the crop, 7% was inter-
cropped as a sole major crop with secondary crops and 71% as major or secon-
dary crop with other major crops. In Mamfe West and Mamfe Forest, it was
a2 A banda is a storage area on a low ceiling or shelf built over the
kitchen fire. Immediately after harvest the fire is built up and kept continuous.
TLU IRA-Ekona p.37
grown in association with maize, egusi, groundnuts, cocoyam and taro in more
than half its fields, and with yam in over a third. In the Highlands it was
often monocropped, and in Akwaya, planted with egusi.
Planting methods: Cassava was planted on mounds (73%) and also on beds (9%,
mostly in the Highlands) and on the flat (18%, mostly in Akwaya). Planting
distances varied from 30 to 320cm, with a median of 90cm and a mean of 96cm.
An average of 1.2 sticks were planted together in the same spot. Densities
varied around a mean of 10.800pph, ranging from a high average of 17.400pph in
Mamfe West and 11.000 in Akwaya to 7.000-8.000 in the other zones. Converted
to monocrop equivalent, the crop occupied an estimated 8% of the total field
area, or 16% of food-crop fields.
Calendar: Planting was usually done in first season (93%), generally in
April, after the first rains, but sometimes in March or May. Often the cas-
sava was planted one to two months after other crops, or later. Second-season
planting was done by 7%, in August. A first weeding was done by 90% of far-
mers growing the crop, and a second by 56%, at an average of 2.4 months and
5.4 months after planting respectively. Harvesting began at an average 12.2
months after planting, lasting an average 13.8 months, to the 21st month in
Mamfe Forest (where it began by the 10th) and to the 30th in Mamfe West and
Varieties: All cassava cuttings came from the farmers' own or neighboring
farms. 65% planted red Panya, a tall (2.5-3m), non-branching type which often
lodged. 13% planted a white variety (none improved), and many complained that
the white could only be used for processing, and made people sick. 22% plan-
ted both. Akwayans had three local varieties, and the local epidemic of cas-
sava mealy bug indicated that they were bringing planting material in from
Field Problems: Farmers reported losing 19% of their crop, with the worst
losses occurring in Akwaya (50%), Eyumojock (31%) and Mamfe Forest (28%). In
Akwaya, the worst problem was the cassava mealy bug, identified from field
samples. Farmers described twisted leaves and a white powder (the bugs) at
the tip, resulting in small, rotten tubers. Ten of the eleven farmers repor-
ting losses in Akwaya blamed mealy bug, three plus the eleventh also blaming
cutting-grass and other animals. Outside Akwaya, 68% blamed losses on wild
animals, 43% tuber rot, and one the rainbow grasshopper's leaf destruction.
Mealy bug was not seen outside Akwaya, even in Eyumojock, the other border
zone. Mosaic was in most fields, but not severely, and brown leaf spot was
common in Mamfe West. Lodging was a common problem, sometimes uprooting the
plants and sometimes causing the whole field to be covered by resprouting
Processing and marketing: Cassava was most often consumed as garri (8+%,
soaked in water several days, grated and roasted) and boiled tubers (71%).
56% ate it as pounded fufu (pounded from boiled tubers, and often mixed with
cocoyam in Akwaya), 53% as water fufu ("akpwu", soaked in water until soft
enough to pulp by hand, and kept wet), 27% as ekwan (mixed with pounded coco-
yam, wrapped and boiled, especially in Mamfe West and Mamfe Forest), 10% in
roasted tubers, and a few in porrage (large chunks in stew), kumkum (finely
TLU IRA-Ekona p.38
ground flour), accra (pounded fufu mixed with banana and fried in balls),
makumba (water fufu slices eaten with coconut) and myondo (fermented kumkum
twisted into narrow ropes). 7% (mostly in the Highlands) did not eat it.
46% said they bought garri most months of the year, and 6% roots; 1% and
10% respectively bought them in months of scarcity. 49% of production was
sold, especially in Mamfe West and Eyumojock, by 80%.of producers, least in
Akwaya. 28% sold it raw (especially in Mamfe Forest), 62% as garri and 9% as
water fufu. Most sold in their village market (53%) or in the nearest larger
village (25%); 4% went to a larger market at a greater distance, and 17% sold
at roadside. Basins of roots and garri and cups of garri were the most common
sales units. Average reported prices per kilo of roots varied from 40fr in
most zones to 30fr in Mamfe Forest and 12fr in Akwaya.
Storage: No one stored cassava up to a month.
Importance: Approximately 12.400 tons of cocoyams were harvested in the 1987
crop year, equivalent to 770 million francs at local prices or 16.5 million
kilocalories of food energy. This made them second in economic importance and
third in nutritional importance in Manyu Division.
Ecozones: The most important zone was the Highlands, with 64% of total pro-
duction. The Highlands produced an average 1.160kg per household, Eyumojock
600kg, Akwaya 415kg, Mamfe West 335kg and Mamfe Forest 320kg.
Associations: Cocoyams were planted in 44% of fields by 91% of farmers, as a
major crop in 90% of the fields. Among fields with the crop, 2% was inter-
cropped as a sole major crop with secondary crops and 98% as major or secon-
dary crop with other major crops. In Mamfe Forest and Mamfe West, it was
grown in association with taro and maize in more than half the fields, and
with plantains, yam, egusi, grounhdnuts and cassava in over a third. In the
Highlands and Akwaya the most common association was taro, and in Eyumojock
taro and plantain.
Planting methods: Cocoyams were planted on mounds (57%), beds (38%, mostly in
the Highlands), and sometimes on the flat (mostly in Akwaya). Planting dis-
tances varied from 15 to 140cm, with a median of 60cm and a mean of 59cm.
Densities varied around a mean of 28.600pph, ranging from a high average of
62.000pph in Akwaya through about 32.000 in Eyumojock and Mamfe West to about
25.000 in Mamfe Forest and the Highlands. Converted to monocrop equivalent,
the crop occupied an estimated 11% of the total field area, or 17% of food-
Calendar: Planting was always done in first season, generally in late March
and April, after the first rains, in the lowlands, and February or March in
the Highlands. It was sometimes planted a month before or after the other
crops in the field. Second-season planting was done by 9%, all in the High-
lands, in October and November. A first weeding was done by 99% of farmers
TLU IRA-Ekona p.39
growing the crop, and a second by 51%, at an average of 2.4 months and 5.6
months after planting respectively. Harvesting began at a mean 8.8 months
after plantiTng, lasting a mean 3.7, to the tenth month on average except in
the Highlands and Akwaya, where it was able to continue until the fifteenth.
Varieties: 79% of farmers replanted their own cocoyams or those from nearby
abandoned fields, but 21% bought them in local markets, usually because of
losses to rot in the field and storage. 30% of the Highlanders bought mate-
rial to plant. Despite the root rot, white was the only preferred variety.
99% planted the white, 29% the red, and 6% (in Akwaya and a few in Mamfe Fo-
rest) the tetraploid yellow. The red was most often found in Mamfe Forest
(43%) and the Highlands (38%). 42% saved cormels for seed after harvest.
Field Problems: Farmers reported losing 44% of their crop, with the least in
Akwaya (29%) and worst losses occurring in Mamfe West and Eyumojock (52%).
The usual culprit identified was the root rot, caused by the Pythium complex
of fungi (72% of complaints and in most fields visited). 12% also blamed ani-
mals, 4% boring caterpillars and 3% drought. The most puzzling complaints
were cases of tuber rot 16% -that did not seem to be attributable to root
rot, either because the rot was only discovered at harvest, the leaves remain-
ing healthy, or because Ibo taro in the same field was worse affected by the
same thing. Some of these were described as having white scabies, or spots,
on the tubers, which caused them to stop growing and rot, while the roots
remained healthy. It was said to appear at about 3-4 months. Only old plan-
ting material was available at the time of the survey in the affected regions.
Cocoyam and taro uprooted in August in one Mamfe Forest field had white in-
sects related to the mealy bug that might have left the marks seen on the old
material; but the insects, according to IRA entomologist M. Tchuanyo, could
not have caused rot, and the tubers also had dying roots, although the farmer
thought they were healthy because the enlarged surface roots still survived.
Processing and marketing: Cocoyam was eaten most often as boiled tubers (by
93%), pounded fufu (74%), ekwan (61%, grated, wrapped in vegetable leaves and
boiled), porrage (49%, stewed in large chunks), and roasted tubers (35%). A
few made Bible or ekwan (pounded and boiled in leaf-wrapped sheets or plan-
tain-leaf bundles). Only two households did not eat cocoyams.
15% said they bought the food most months of the year, and 35% bought it in
months of scarcity. 30% of production was sold, by 47% of producers, most in
Akwaya and Mamfe Forest. All sold it raw. Most sold in their village market
(62%) or in the nearest larger village (27%); 7% went to a larger market at a
greater distance, and 5% sold at roadside. The primary selling time was the
dry season through the early rains. Basins of tubers were the most common
sales unit. Average reported prices per kilo varied from 75fr in Mamfe West
and Mamfe Forest to 45fr in the Highlands and 30fr in Akwaya.
Storage: About 80% of storage was inside the farm, half in a roofed shed
(bush house), usually on the ground and often covered by leaves if in the
open. A few stored in a hole in the ground. 20% stored at home, on the
floor, usually with wood ash to deter rot and mixed with other tubers. 58% of
growers stored the crop, four-fifths for planting material and over half to
sell. Almost all reported rottening by the end of storage, averaging 30%
TLU IRA-Ekona p.40
loss, with the reported levels highest in Eyumojock and unrelated to method of
storage. Cocoyams were stored an average of four months.
Importance: Approximately 9.600 tons of taro were harvested in the 1987 crop
year, equivalent to 496 million francs at local prices or 10.8 million kilo-
calories of food energy. This made it fifth in economic importance and fourth
in nutritional importance in Manyu Division.
Ecozones: The most important zones were Mamfe Forest, with 36% of total pro-
duction, the Highlands, with 29%, and Mamfe West, with 26%. Mamfe Forest
produced an average 880kg per household, Mamfe West 450kg, the Highlands
405kg, Eyumojock 415kg, and Akwaya 215kg.
Associations: Taro was planted in 41% of fields by 92% of farmers, as a major
crop in 82% of the fields. Fields with taro were always intercropped with
other major crops. In Mamfe Forest and Mamfe West, it was grown in associa-
tion with cocoyam and maize in more than half the fields, and with yams, plan-
tains, egusi, groundnuts and cassava in over a third. The most common as-
sociation in the other zones was cocoyam.
Planting methods: Taro was planted on mounds (62%), beds (33%, mostly in the
Highlands), and sometimes on the flat (mostly in Akwaya). Planting distances
varied from 15 to 140cm, with a median of 55cm and a mean of 58cm. The mami
variety was sometimes planted closer than the Ibo. Densities varied around a
mean of 30.100pph, ranging from a high average of 71.000pph in Akwaya to
36.000 in Mamfe West and 26.000-27.000 in the other zones. Converted to mono-
crop equivalent, the crop occupied an estimated 8% of the total field area, or
14% of food-crop fields.
Calendar: Planting was usually done in first season (99%), generally in late
March and April, after the first rains, in the lowlands, and February or March
in the Highlands. Sometimes it was planted one to two months after other
crops. Second-season planting was done by 11%, in Mamfe Forest in September-
October and in the Highlands in October-November. A first weeding was done by
94% of farmers growing the crop, and a second by 42%, at an average of 2.4
months and 5.1 months after planting respectively. Harvesting began at an
average 5.8 months after planting for Ibo, until 8.2 months, and 7.5 months
for mami, until 10.0 months.
Varieties: 89% of taro farmers planted the Ibo variety, a short-cycle taro
used primarily for fufu. 47% planted "mami coco", a 8-10 month variety pre-
ferring land straight out of fallow, that produced multiple corms and cormels
and could be used for a wider variety of foods. A six-month specimen dis-
played in Mamfe Forest consisted of a cluster of five corms about 7.5cm in
diameter over 8-10 2cm-diameter cormels, the latter used for seed as well as
"* See also Section F.3. Cocoyams.
TLU IRA-Ekona p.41
food. The mami was most common nearer the Nigerian border. 86% of farmers
planted their own material, and 16% bought in the market, especially in Akwaya
and the Highlands. 42% saved cormels for seed.
Field Problems: Farmers reported losing 22% of their crop, with the smallest
losses occurring in Akwaya (12%) and the Highlands (15%). 74% attributed
their losses to tuber rot and only 7% to root rot; most of the tuber rot men-
tions included a description of white scabies or spots that caused the tubers
to stop growing, and often they said the plants' leaves looked well at har-
vest. Mami and Ibo taro seemed to be equally affected. Pythium complex was
identified in a Manyu taro brought to the Ekona tissue culture laboratory.
20%, mostly in Eyumojock, complained of animal destruction.
Processing and marketing Taro was consumed as pounded fufu (or achu) by 82%
of households, boiled tubers by 80%, porrage by 37%, ekwan by 20%, roasted
tubers by 14%, and by a few as Bible, accra, fried or kwa coco. Only one did
not eat it at all.
4% said they bought the food most months of the year, and 13% bought it in
months of scarcity. 30% of production was sold, especially in Mamfe Forest,
by 40% of producers, few in the Highlands. All sold it raw. Most sold in
their village market (60%) or in the nearest larger village (19%); 10% went to
a larger market at a greater distance, and 10% sold at roadside. The primary
selling months were from September into dry season. Basins of tubers were the
most common sales unit. Average reported prices per kilo varied from 33fr in
Akwaya to 50-60fr elsewhere; mami always had a better selling price in the
Storage: 54% of taro growers stored, 45% of them to sell and 84% for planting
material. 70% stored on the farm, 45% inside a roofed, open shed, on the
ground, a shelf or sometimes in a hole. 30% stored on the floor at home,
usually with wood ash. Almost all complained of rot, and whenever cocoyams
and taro were stored together, the farmers reported somewhat more rot in the
taro. Taro rot losses averaged 31%, being lowest in Akwaya. It was stored an
average 4.2 months, least in the Highlands.
5. Plantains and Bananas:
Importance: Approximately 666 thousand bunches of plantains and 380 thousand
bunches of bananas were harvested in the 1987 crop year, equivalent to 549 and
152 million francs respectively at local prices or 6.6 and 3.4 million kilo-
calories of food energy. This made plantains first in economic importance and
fifth in nutritional importance in the division, with bananas ninth on both
Ecozones: The most important zones for plantains were Mamfe West, with 38% of
total production, Mamfe Forest, with 34%, and Eyumojock, with 15%. Eyumojock
produced an average 680kg per household, Mamfe Forest 610kg, Mamfe West 490kg,
the Highlands 130kg, and Akwaya only 30kg. For bananas, the most important
zone was Mamfe Forest, with 65% of total production. Mamfe Forest produced an
average 670kg per household, Eyumojock 150kg, the Highlands 110kg, Mamfe West
TLU IRA-Ekona p.42
70kg, and Akwaya almost none.
Associations: Plantains and bananas were planted in 38% and 21% of fields
respectively by 86% and 46% of farmers, as a major crop in 87% and 80% of the
fields. Among fields with the crop, only 3% and 2% were intercropped as a
sole major crop with secondary crops and 94% and 98% as major or secondary
crop with other major crops. Plantains were grown in association with coco-
yams and taro in more than half the fields, and with bananas and maize in over
a third. Bananas were grown in association with plantains and cocoyams in
more than half the fields, and with taro in over a third.
Planting methods: Plantains and bananas were always planted on the flat.
Planting distances for plantains varied from 150 to 600cm, with a median of
340cm and a mean of 351cm; for bananas, from 100 to 600cm, with a median of
340 and a mean of 344. Densities varied around a mean of 810pph for plantains
and 850 for bananas, ranging from a high average of 1.000-1.100pph in most
zones to 600 in the Highlands and 900 in Mamfe West. Converted to monocrop
equivalent, the crops occupied an estimated 14% and 8% respectively of the
total field area, or 7% and 2% of food-crop fields.
Calendar: Planting was usually done in first season (97%), generally in late
March and April, after the first rains, in the lowlands, and February or March
in the Highlands. Usually plantains and/or bananas were planted several
months before tree crops, and sometimes a month or more before or after other
food. Second-season planting was done by 3%, in November in the Highlands. A
first weeding was done by 93% of farmers growing the crop, and a second by
56%, at an average of 2.5 months (later in Mamfe Forest and Eyumojock) and 5.9
months after planting respectively. Harvesting began at an average 11.6
months after planting, lasting for 21.7 months, until the 24th month in Eyumo-
jock, 30th in Mamfe Forest, 33rd in Mamfe West, 37th in the Highlands and 43rd
Varieties: 83% of growers planted their own plantains and bananas. 16%
bought local plantains and 1% improved ones; 10% bought local and 8% improved
bananas. At least some of the improved suckers came from a local man working
in a Bamenda agricultural training institute. The crops were known by a wide
range of local names, few of which could be identified: woman, man, Aza, red,
black, white, short, long, two-hand, bobi tanap, highland, long-finger, two-
hand and nine-hand plantains; long, short, small-finger, nine-hand, yellow,
America, Lacatan and black-and-white bananas.
Field Problems: Farmers reported losing 25% of their plantain crop, with the
worst losses occurring in Eyumojock (55%) and the lowest in Mamfe West (15%).
Banana losses were seldom mentioned. 70% complained of stem borer (black rot
in the stem, with lodging), 29% of nematodes (dead roots with lodging), 10% of
animals, and 6% in Mamfe Forest and Mamfe West of possible black sigatoka, an
aerial fungus which became epidemic in these zones in rainy season 1988 and
was positively identified from leaf samples and photos by P. Kofi of IRA.
Processing and marketing: Plantains were consumed boiled by 94% of house-
holds, as porrage by 73%, roasted by 65%, and by a few fried. Only two did
not eat it.
TLU IRA-Ekona p.43
19% said they bought plantains most months of the year, and 36% bought it
in months of scarcity. 37% of plantains and 24% of bananas were sold, by 73%
and 65% of producers. All sold it raw. Most sold plantains in their village
market (59%) or in the nearest larger village (21%); 13% went to a larger
market at a greater distance, and 7% sold at roadside. 70% sold bananas lo-
cally. Average reported prices per plantain bunch varied from BOOfr in Akwaya
to 1.000-1.100fr in most zones to 1.400fr in Mamfe West; per banana bunch,
230fr in Mamfe West and Eyumojock and 450fr in Mamfe Forest and the Highlands.
Importance: Approximately 3.900 tons of yams were harvested in the 1987 crop
year, equivalent to 309 million francs at local prices or 4.0 million kilo-
calories of food energy. This made them seventh in economic importance and
sixth in nutritional importance in Manyu Division.
Ecozones: The most important zones were Mamfe West, with 39% of total produc-
tion, and Mamfe Forest, with 26%. Mamfe West produced an average 275kg per
household, Mamfe Forest 260kg, Eyumojock 215kg, Akwaya 360kg, and the High-
Associations: Yams were planted in 29% of fields by 72% of farmers, as a
major crop in 55% of the fields. Among fields with the crop, all were inter-
cropped, 1% as a sole major crop with secondary crops. Yams were grown in as-
sociation with maize, taro, cocoyams, egusi, groundnuts and cassava in more
than half the fields, and with plantains and beans in over a third.
Planting methods: Yams were planted on mounds (50%), beds (27%, mostly in the
Highlands), and on the flat (about 30% in Mamfe Forest, Mamfe West and Ak-
waya). Planting distances varied from 40 to 185cm, with a median of 90cm and
a mean of 88cm. Densities varied around a mean of 12.900pph, ranging from a
high average of 17.000pph in the Highlands and Mamfe West to 8.000 in Eyumo-
jock. Converted to monocrop equivalent, the crop occupied an estimated 4% of
the total field area, or 7% of food-crop fields.
Calendar: Planting was always done in first season, generally in late March
and April, after the first rains, in the lowlands, and February or March in
the Highlands. A first weeding was done by 98% of farmers growing the crop,
and a second by 49%, at an average of 2.6 months and 5.8 months after planting
respectively. Harvesting began at an average 7.8 months after planting, las-
ting 2.1 months.
Varieties: The most popular yams were sweet yams (D. dumetorum), planted by
54% of yam farmers, especially in the Highlands; water yams (D. alata), plan-
ted by 52% and dominant in Mamfe West and Akwaya; and Ikom (white, Calabar-
D. rotundata) yams, planted by 46% and dominant in Mamfe Forest and Eyumo-
jock. A few farmers in the lowlands planted the yellow yam D. Cavensis and
TLU IRA-Ekona p.44
the red D. shipiana. Judging by price, however, these preferences reflected
ease of cultivation rather than consumer tastes. 70% of farmers planted their
own yam setts, or got them from a neighbor, and 30% (mostly in Eyumojock)
bought them in a market. 30% said they stored for seed.
Field Problems: Farmers reported losing 20% of their crop, with the worst
losses occurring in Mamfe West and Eyumojock (31%). 68% complained of tuber
rot, most occurring early and often accompanied by white scale or "powder".
14% complained of animal destruction. 14% complained of tuber-boring beetles
or other insects, and one, of vine-eating beetles.
Processing and marketing: Yams were eaten boiled by 95% of households, as
porrage by 68%, pounded fufu by 21%, roasted by 20%, in ekwan by 10%, and by a
few in chips. Two did not eat them.
13% said they bought the food most months of the year, and 18% bought it in
months of scarcity, especially in Mamfe Forest. 42% of production was sold,
especially in Mamfe Forest and almost never in the Highlands, by 53% of produ-
cers (less in Akwaya). 74% sold it raw, and 26% boiled (sweet yams, in the
Highlands and some Mamfe Forest farmers). Most sold in their village market
(58%) or in the nearest larger village (27%); 7% went to a larger market at a
greater distance, and 8% sold at roadside. The primary selling months were
the dry season, and in Mamfe Forest and Akwaya, April and May. Heaps were the
most common sales unit, so prices could not be calculated, but they seemed to
be higher in Mamfe Forest.
Storage: A third of growers stored yams, for an average 5.1 months, highest
in Mamfe West and Mamfe Forest. Highlanders did not store. Two thirds stored
on the farm, 40% in a bush house, but seldom on a shelf. The rest stored at
home. 40% stored to sell and 90% to plant. 70% complained of rots (or white
scale coatings) and one of animal damage. Average reported losses were only
Importance: Approximately 430 tons of shelled groundnuts were harvested in
the 1987 crop year, equivalent to 148 million francs at local prices or 1.8
million kilocalories of food energy. This made them tenth in economic impor-
tance and ninth in nutritional importance (tied with bananas and irish pota-
toes) in Manyu Division.
Ecozones: The most important zones were the Highlands, with 43% of total
production, and Akwaya, with 29%. Akwaya produced an average 145kg per house-
hold, the Highlands 35kg, Mamfe West and Mamfe Forest 15kg, and Eyumojock
Associations: Groundnuts were planted in 29% of fields by 80% of farmers, as
a major crop in 31% of the fields. Among fields with the crop, all were
intercropped as a major or secondary crop with other major crops. In Mamfe
Forest and Mamfe West groundnuts were grown in association with maize in al-
TLU IRA-Ekona p.45
most all cases, with egusi, cocoyams, taro, cassava and yams in more than half
the fields, and with beans in over a third. In the Highlands and Eyumojock,
the only common association was maize, and in Akwaya, maize and egusi.
Planting methods: Groundnuts were planted on beds in the Highlands, beds and
mounds in Mamfe Forest, mounds in Mamfe West, mounds and beds in Eyumojock and
on.the flat in Akwaya. Planting distances varied from 10 to 50cm, with a
median of 20cm and a mean of 22cm. An average of 1.1 seeds were planted per
hole, ranging from 1 to 4. Densities varied around a mean of 229.000pph,
ranging from a high average of 440.000pph in Eyumojock and 360.000 in Akwaya
to 200.000-225.000 in the other zones. Converted to monocrop equivalent, the
crop occupied an estimated 2% of the total field area, or 5% of food-crop
Calendar: Planting was usually done in first season (95%), generally in late
March and April, after the first rains, in the lowlands, and February or March
in the Highlands. Second-season planting was done by 15%, in August. A first
weeding was done by 92% of farmers growing the crop, and a second by 18%, at
an average of 2.0 months and 2.8 months after planting respectively. The
Highlands and Akwaya weeded earlier and more often. Harvesting began at an
average 3.6 months after planting, lasting two to three weeks.
Varieties: The most common type was Yaounde, with 70% of growers planting,
followed by Bafia (19%) and Ngaoundere/Garoua (12%). 66% planted whites and
53% reds. 65% bought from the market to plant, although 45% saved for seed.
Field Problems: Farmers reported losing 18% of their crop, with the worst
losses occurring in Akwaya and Mamfe Forest. 29% complained of rosette (most-
ly in the Highlands and Akwaya), 35% of animals (Mamfe Forest and Mamfe West),
some of pod-eating insects, lack of seed formation, mal-formed pods and seed,
pod rot and drought.
Processing and marketing: 10% said they bought the food most months of the
year, and 26% bought it in months of scarcity, especially in the Highlands.
45% of production was sold, especially in Akwaya and the least in the High-
lands, by 61% of producers. Most sold in their village market (59%) or in the
nearest larger village (18%); 10% went to a larger market at a greater dis-
tance, and 13% sold at roadside. The primary selling months were July to
August, or October to December in Akwaya. Cups were the most common sales
unit, except in the Highlands (tins) and Akwaya (jute bags sold in Nigeria).
Average reported prices per kilo varied from 60fr in Akwaya to 350-360fr in
Mamfe Forest and the Highlands.
Storage: 45% of growers stored, for an average 6.6 months, though only 3.4 in
Mamfe West and over 7 elsewhere. Most dried them in the sun and then bagged
them, but 15% dried them in containers (all experiencing extra losses), 20%
kept them in open baskets after drying and 7% in sealed tins or calabash.
Four-fifths stored for seed and almost a half to sell. The farmers reported
13% losses, least in Mamfe West, mostly to weevils but sometimes to rats or
.TLU IRA-Ekona p.46
8. Egusi melon:
Importance: Approximately 500 tons of egusi melon seed uncrackedd) were har-
vested in the 1987 crop year, equivalent to 174 million'francs at local prices
or 2.2 million kilocalories of food energyl". This made.i..eighth in economic
importance but minor in nutritional importance in Manyu..DiAision.
Ecozones: The most important zone was Mamfe West, with 65% of total produc-
tion. Mamfe West produced an average 60kg per household of egusi seed,
Akwaya 40kg, Mamfe Forest 25kg, Eyumojock 20kg, and the Highlands almost none.
Associations: Egusi melon was planted in 26% of fields by 62% of farmers, as
a major crop in 21% of the fields, always with with other major crops. Egusi
was grown in association with maize almost always, with taro, cocoyams,
groundnuts, cassava and yams in more than half the fields.
Planting methods: Egusi was planted on mounds (83%) and sometimes on the flat
in Mamfe Forest, always so in Akwaya. Planting distances varied from 20 to
400cm, with a median of 40cm and a mean of 54cm. An average of 2.3 seeds were
planted per hole, ranging from 2 to 4. Densities varied around a mean of
79.000pph, ranging from a high average of about 120.00Opph in Mamfe Forest and
Mamfe West to 40.000 in Eyumojock and 10.000 in Akwaya. Converted to monocrop
equivalent, the crop occupied an estimated 1% of the total field area, or 3Y
of food-crop fields.
Calendar: Planting was always done in first season, generally in late March
and April, after the first rains. A first weeding was done by 85% of farmerL
growing the crop, at an average of 2.3 months after planting, and a second by
9%. Harvesting began at an average 3.9 months after planting, lasting an
average three to four weeks.
Varieties: The "Ibo" egusi, Citrullus lanatus, a ground vine, was plante:
everywhere except possibly on a few Highland farms. "Country" egusi, a clim-
bing vine known in eastern Cameroon as "ngan" (Cucumeropsis mannii), was also
found on a few farms in Mamfe Forest, the Highlands and Akwaya. A quarter of
Mamfe Forest and Akwaya farmers had to purchase seed in the market, and 44%
stored seed to plant.
Field Problems: Farmers reported losing 20% of their crop, with the worst
losses occurring in Akwaya. 66% blamed the ladybug Chrysomelidae, found all
over the province in various colors but with the same voracious habit of cut-
ting the first leaves off at the stem. 7% more identified unspecified earl
death, 23% animal destruction and 10% yellowed or twisted leaves and low pro
duction. The Chief of Agricultural Post at Kembong confirmed that the local
egusi also suffered from stem wilt.
Processing and marketing: 8% said they bought the food most months of the
year, and 14% bought it in months of scarcity, especially in Akwaya. 62% of
as There is no source for the energy content of Citrullus. The figure
assumes an equivalence of cracked (shelled) egusi to shelled groundnuts.
TLU IRA-Ekona p.47
production was sold, especially in Mamfe West and Mamfe Forest, by 77% of
producers, especially in Mamfe West. Most sold in their village market (55%)
or in the nearest larger village (29%); 11% went to a larger market at a grea-
ter distance, and 5% sold at roadside. Jute bags and cups were the most com-
mon sales units. Average reported prices per kilo varied from 105fr in Akwaya
and 125fr in Eyumojock to 400-440fr in Mamfe West and Mamfe Forest.
Storage: 44% of growers stored, for a mean of 8.5 months. Most stored in
bags after rotting and washing the melons, extracting the seeds and sun-drying
them. A few used calabashes, tins or covered baskets. 53% stored to sell and
87% for seed. Only 4% losses were noted, to weevils, rats, cockroaches and
Importance: Approximately 1.1200 tons of beans were harvested in the 1987
crop year, equivalent to 111 million francs at local prices or 3.8 million
kilocalories of food energy. This made it eleventh in economic importance but
seventh in nutritional importance in Manyu Division.
Ecozones: The most important zone was the Highlands, with 84% of total pro-
duction. The Highlands produced an average 135kg per household, Mamfe West,
Mamfe Forest and Akwaya 15-20kg, and Eyumojock almost nothing.
Associations: Beans were planted in 23% of fields by 60% of farmers, as a
major crop in 25% of the fields, always with other major crops. Beans were
grown in association with maize (almost always in the Highlands), taro,
groundnuts and cocoyams in more than half its fields, and with yams, cassava
and egusi in over a third.
Planting methods: Beans were planted on mounds in Mamfe West, mounds and on
the flat in Mamfe Forest, and beds in the Highlands. Planting distances va-
ried from 20 to 200cm, with a median of 50cm and a mean of 55cm. An average
of 2.1 seeds were planted per hole, ranging from 2 to 3. Densities varied
around a mean of 69.000pph, ranging from a high average of 110.000pph in Mamfe
Forest to 73.000 in the Highlands and lower in the few other cases recorded.
Converted to monocrop equivalent, the crop occupied an estimated 2% of the
total field area, or 5% of food-crop fields.
Calendar: Planting was usually done in first season (93%), generally in late
March and April, after the first rains, in the lowlands, and February or March
in the Highlands. Second-season planting was done by 35%, in August and Sep-
tember. A first weeding was done by 92% of farmers growing the crop, and a
second by 17% (in the Highlands), at an average of 2.0 months and 3.4 months
after planting respectively. Harvesting began at an average 3.4 months after
planting, lasting an average three to four weeks, longer in Akwaya.
Varieties: 55% of bean growers planted cowpeas (Viqna unquilata) and 50%
dwarf beans (Phaseolus vulgaris). The local cowpea in Akwaya stayed long in
TLU IRA-Ekona p.48
the field, climbed rapidly and produced little seed. Two farmers in Mamfe
Forest had a variety they calle.'"Isoki", type unknown. 69% of growers plan-
ted seed from their own or a neighbor's farm, and 37% bought in a market,
especially in Mamfe West and Akwaya. Storage for seed occurred mostly in the
HiRfl1ids and Mamfe Forest.
Fiel'Problems: Farmers reported losing 11% of their crop, with the worst
loses occurring in the Highlands (16%). Outside the Highlands, losses were
blamed on animals; inside, to insects cutting: the young shoots (43%) and
blight (57%, early leaf withering).
Processing and marketing: 18% said they bought the food most months of the
ye;.r, and 19% bought it in months of scarcity, especially in the Highlands.
39% of production was sold, especially in Mamfe West, by 65% of producers,
least in Mamfe Forest. Cups, and tins in the Highlands, were the most common
sales units. The average reported price per kilo in the Highlands was 135fr.
Storage: 40% of growers stored, for a mean of 6.7 months in Mamfe Forest and
the Highlands and 3 in Mamfe West. Two-thirds'stored in,bags: or bundles, 25%
in sealed calabashes, tins or bottles, and ine in a basket." 20% added wood
ash or chemical. All stored for seed, and 60%rfor sale. .70% .reported weevil
damage, with a mean 25% loss.
Importance: Approximately 3.900 tons of irish potatoes were harvested in the
1987 crop year, equivalent to 435 million francs at local prices, or 3.0 mil-
lion kilocalories of food energy. This made them sixth in economic importance
but only ninth in nutritional importance in Manyu Division. The sweet potato
harvest was only 125 tons and was minor both-economically and nutritionally.
Ecozones: The most important zone was thi Highlands, which produced all the
irish potatoes (660kg per household, although most was :in the highest sub-
zone) and 83% of the sweet potatoes. The Highlands produced an average 15kg
of sweet potatoes and the other zones almost nothing.
Associations: Irish and sweet potatoes were planted in 6% and 5% of fields
respectively, by 16% and 12% of farmers, as a major crop in 71% and 12% of the
fields. Among fields with irish potatoes, 20% were intercropped as a sole
major crop with secondary crops and 60% as major or secondary crop with other
major crops; all sweet potatoes were grown with other major crops. There was
no pattern of associations by crop.
Planting methods: Irish and sweet potatoes were planted on beds in the High-
lands, and sweet potatoes on mounds in Mamfe Forest. Planting distances va-
ried from 20 to 250cm for irish potatoes, higher in the lower altitudes, with
a median of 35cm and a mean of 60cm. Densities varied around a mean of
28.700pph. Sweet potato planting distances varies from 30 to 120 cm, with a
median of 80 and a mean of 79, and a mean density of 6.900pph. Converted to
TLU IRA-Ekona p.49
monocrop equivalent, the irish potato crop occupied an estimated ?% of the
total field area, or 5% of food-crop fields, the sweet potatoes only 0.1% and
Calendar: Planting of irish potatoes was always done in first season, in
March. Second-season planting was done by 54%, in the highest sub-zone (where
people also planted three seasons) anytime between June and October. In the
first season the first weeding was done by 64% of farmers growing the crop,
and the second by 36%, at an average of 1.7 months and 3 months after planting
respectively. Second-season weeding was only done by 33%, at 2.5 months.
Potatoes below the highest sub-zone were seldom weeded. Harvesting began at
an average 3.1 months after planting, at 3-5 months in the highest subzone and
2-3 below, lasting an average two weeks. Sweet potatoes were planted first
season by 93% of growers, in March to April, sometimes later than the other
crops, and in August by 17%. 84% weeded a first time, at 2.0 months, and 39%
a second, at 3.9 months. Harvesting of sweet potatoes began at 5.3 months and
continued 1.3, being very early in the Highlands and Akwaya and late in Mamfe
Forest and Mamfe West.
Varieties: No variety names were available from farmers, only colours of
irish potatoes. Two growers interviewed bought irish potatoes for planting
from an agricultural missionary centre in Dschang; of the rest, half bought
from markets and half used local material. Sweet potatoes planted were usual-
ly from local farms, otherwise the local market.
Field Problems: Farmers reported losing 30% of their irish potato crop. No-
one mentioned sweet potato losses. All complaints came from the highest sub-
zone where the crop was basic to the system. 88% complained of cut-worms in
the soil, eating the new shoots, as well as their carrot, cabbage, leek and
garlic crops. 50% complained of leaf blight, and 25% of animals, complaints
also mirrored for their other crops.
Processing and marketing: 6% said they bought irish potatoes most months of
the year, and 3% bought it in months of scarcity, all in the Highlands. Only
2% bought sweet potatoes in some months. 66% of irish potato production was
sold, by 73% of producers, and all sweet potato producers sold some. The
primary selling months for irish potatoes were June, July and January; prices
were better in January. Jute bags were the most common sales unit. The ave-
rage reported price per kilo was from 82fr in the highest subzone but 171fr in
the central market of Yaounde, where one farmer took a truck.
Storage: Irish potatoes were stored an average of 3.5 months, spread on the
floor or occasionally on a shelf, for both seed and sale. All rotted, and
some germinated or were eaten by rats, for an average loss of 21%.
Importance: Approximately 450 tons of rice were harvested in the 1987 crop
year, equivalent to 9 million francs at local prices, or 1.600 million kilo-
calories of food energy. This made it minor in economic and nutritional im-
portance in Manyu Division.
TLU IRA-Ekona p.50
Ecozones: The only rice zone was Akwaya, with an average production per
household of 485kg.
Associations: Rice was planted in 5% of fields by 4% of farmers, as a major f
crop in 82% of the fields, and .always in association with other major crops.
.Planting methods Rice was planted on the flat. Planting distances varied
from 20 to 45cm, with a median and mean of 27.5cm. An average of 7.6 seeds
were planted per hole, ranging from 3 to 12. There was no transplanting, but
if none germinated the hole was replanted. Much seed was lost to birds and
termites. Densities varied around a mwan of a million plants per hectare at
planting; the one field visited-had been planted at 900.000 and contained
600.000 at four weeks. Converted.to monocrop equivalent, the crop occupied an
estimated 0.4% of the total field area in Manyu, or 1% of food-crop fields.
Calendar Planting was done in late March and April, after the first rains.
The first weeding was done by all farmers growing the crop, at 1.5 months'
after planting, and a second by 30%, at 2.2 months. Harvesting began at an
average 4.4 months after planting, lasting an average three to four weeks;
some fields were harvested at 3-4 months and some at 5-6.
Varieties Akwayan varieties were named "apapa", a white grain, and "madame",
a red one. Farmers saved seed to plant.
Field Problems: Farmers reported losing 29% of their crop, 75% reporting
losses to cutting-grass, 50% birds, and 25% premature yellowing.
Processing and marketing: 91% said they bought the food most months of the
year, practically all outside Akwaya,-and 2% bought it in months of scarcity.
72% of production was sold, by all producers. Most was taken down to Nigeria
in September to November. Jute bags were the common sales unit. The average
reported price per kilo was 20fr.
Storage: Rice was stored by 70%, for sale and seed, for an average 8.4
months. Over half sun-dried the grain and then bagged it; the rest dried it in
a banda. Over half reported weevil damage, a third rot and one rats, with an
average loss of 12%.
TLU IRA-Ekona p.51
6. Conclusions and Recommendations:
Five zones were delineated and surveyed containing about 79% of the farming
population in Manyu. The Korup Forest and most of Akwaya Sub-Division were
Arguing from the criteria of farming population, present production and
agronomic problems, Manyu Division contains two of the six priority super-
zones of South West Province designated by the TLU. One is the Highlands zone
of Fontem, and is recommended to IRA-Bambui, IRA Dschan,. and University Cen-
tre Dschang for special attention. The other comprises the three lowland
zones of Mamfe Central and Eyumojock, which share many commonalities. These
are Mamfe West, Eyumojock, and Mamfe Forest zones.
The Manyu Highlands contains 38% of the farmers in surveyed zones, produ-
cing 40% by economic value of Manyu food crops. Because of the steepness of
the terrain, soil infertility and erosion are already notable. Land is still
more available than in Dschang, whose system it closely resembles. The recent
transfer of marketing authority for Arabica coffee from UCCAO to FOFCOOP has
endangered its access to food- as well as tree-crop chemicals and fertilizers,
of which it makes the highest use in the province. Food marketing for specia-
lty and general markets in Dschang and-Mamfe is also well developed conside-
ring the transport difficulties. Coffee is a major crop. Varietal improve-
ments from IRA's existing highland stock, input access, and soil conservation
research are priorities.
The Mamfe Forest contains 22% of farmers, producing 25% of food crops.
Despite a large land area, villages are widely separated. Each family plants
some fields by preference near the village, generating short fallows and fer-
tility maintenance and animal control problems. More distant fields usually
require heavy investment in tree felling, special mounding techniques and li-
mitations as to type of crop in order to counter the very heavy granitic soil.
Marketing is almost all within the, local village area; some goes to Mamfe
Town, south to Nguti and east to Fontem. Some cocoa is also produced. Maize
and cassava varieties, maize storage and yam multiplication techniques will
improve production, but large production increases will only follow better
roads. Within the next generation, improved soil fertility maintenance tech-
niques will have to be found.
Mamfe West contains 28% of farmers, producing 24% of food crops (more by
energy value). Villages are very large and most forested land is distant from
roads and villages. Most fields are cut out of bush fallows which are main-
tained at 4-6 years, apparently sufficient for restoration of the normal (low)
level of fertility on this sandy granitic soil. Mamfe West feeds Mamfe Town
and exports some food to Nigeria. Cocoa and palms are also major crops, and a
local palm cooperative is large and active. The palm cooperative may be able
to be used to provide food-crop inputs, particularly plantain and egusi chemi-
cals and fertilizer, that individual samll farmers remote from Douala cannot
obtain. IRA cassava varieties and plantain hygienic techniques can be very
helpful. Effective recommendations for insect control in young egusi would
TLU IRA-Ekona p.52
also have a major impact on women's incomes. Increasing population will shor-
ten fallows, possibly reducing fertility levels critically in the near future.
Eyumojock zone contains 7% of farmers producing 6% of food crops. The
villages are more oriented towards hunting and Nigerian trade and less to crop
marketing, but share attributes of both Mamfe Forest and Mamfe West. Most
villages are not reachable by road, and very small in size. IRA technology
suitable for Mamfe West and Mamfe Forest will also help here, but both farming
and markets will be restricted for years to come.
Akwaya Centre accounts for only 5% of farmers, 5% of food crops and no tree
crops. The rest of the sub-division apparently shares its isolation, but not
its cropping system or marketing orientation. The soil is poor, and its major
staples are in danger of eradication, cocoyams from root rot and cassava from
mealy bug. Because of transport problems, its major cash crops are high-
value-for-weight: rice, groundnuts and egusi seed, sold to a small market in
Nigeria at low prices. Even without roads, improved cassava and maize varie-
ties and the release of the biological control against mealy bug can help to
2. Management and investments
Almost all households are native to Manyu, most to their present villages.
Thus almost all access to land is by inheritance of customary rights by vil-
Labour payments, school fees and Christmas/New Year holidays are the in-
vestments that demand the most cash from the farmers. Paid labour is used by
a priority of housPiehlds, almost entirely for clearing; almost 60.000fr CFA
ar: spent for this p'.rpose per household in a year. Labour bottlenecks occur
in the months of March and April, for land preparation and early planting. In
the busiest season of the year women work 10 hours a day, and men 6.8 hours.
Husband and wife both participate in land clearing, the husband cutting the
forest or bush and the wife preparing the mounds or beds. Weeding and harry
vesting are mostly done by women and their children, except for tree fields.
Njanggi is used more often than paid'labour, and both are used most often in
clearing and land preparation.
The extension service in Manyu is inadequate in coverage, with few agricul-
tural posts, and lack of transport means that supervision is scanty and the
outreach of each technician severely limited. Some form of transport at both
sub-delegation and technician level is badly needed, and other ways also need
to be found to encourage staff to remain in remote stations.
Technology recommendations should be tailored to reduce the heavy labour
inp'its in the busy months of land preparation. More extension personnel
should be recruited and trained to work with both food and cash cron farmers
in Manyu and constant effort should be maintained to interact directly with
the women as well as the men, to improve research understanding and extension
education. Both MINAGRI's Community Development and the Ministry of Social
TLU IRA-Ekona p.53
and Women's Affairs have a major role to play in helping the agricultural
technicians to reach groups of women farmers.
Manyu markets are small and local. Marketing problems include the lack of
good roads to evacuate produce to major cities, especially from Fontem to
Dschang and Mamfe area to Bamenda or Kumba. Akwaya has no roads at all and
has no incentives for production. With good roads the food produced would
reach more markets within and outside the division.
Management problems include land preparation costs, the incipient decrease
in land fertility as fallows begin to shorten (especially near villages), lack
of varieties and chemicals to control pests and diseases, and crop destruction
by bush and domestic animals. Control should be placed on the domestic ani-
mals. Obang Farm should be revitalized to multiply improved varieties of food
and tree crops for Manyu and the Province, and the cooperatives should be
urged to stock both tree and food crop chemicals and fertilizers and to work
together with the extension service and IRA to explain their use and the eco-
The most important Manyu food crops in economic terms are cocoyams and
maize, followed by plantains, cassava .and taro; however, plantains represent
almost half of crop sales by value, followed by maize and cassava. In energy
terms, cassava accounts for a third of the calories produced, followed by
maize, cocoyams and taro.
Cocoyams are grown primarily in the Highlands; in the other zones many
farmers have abandoned or reduced them due to the root rot. The red cocoyam
is very unpopular despite its greater resistance. There is some evidence that
a form of tuber rot, separate from root rot, is also present in much of the
division. Taro, most common in Mamfe Forest but important throughout, has
been affected by root rot and especially tuber rot, apparently identical to
that in cocoyams, in both Ibo and forest varieties. The tuber rot complica-
tion should be investigated.
Maize is a specialty of the Highlands and Mamfe Forest areas. Calabar
(soft white maize) is preferred. Stem borers and animal damage are major con-
straints, and streak virus was widespread in both 1988 and 1989. Maize is
eaten in large quantities after harvest, and then stored for seed; a minority
consume it in dry forms. About a third is sold, both green and for seed; many
farmers lose their seed in storage, owing to drying problems (most plant first
season only) and weevils. Storage improvements are as necessary as varietal
ones to the future of maize in the Division.
Plantains are most common in Eyumojock, Mamfe West and Mamfe Forest, and
bananas in Mamfe Forest. They are usually planted by the husband inside the
wife's fields, and inside his own tree fields, and given little attention
until harvest. Losses from stem borer, and sometimes nematodes, are heavy,
and in 1988 a major epidemic of black sigatoka disease hit the division. IRA
TLU IRA-Ekona p.54
has provided some advice and should continue to assist if the farmers can
combine to purchase the inputs needed. Plantains are a major cash source for
Cassava is produced predominantly by Mamfe West and Akwaya. Mamfe West
reported few problems, but lodging is common due to heavy wind storms; the IRA
cassava will lodge less. Akwaya obtains enough cassava to eat only by plan-
ting twice as large an area as they needed before the mealy bug came; they
badly need the release of the biological control. IRA cassava varieties
should be fully acceptable to these zones as they rely a lot on processing,
but in the other zones cassava is eaten more often boiled, and wide acceptance
may be difficult. The other zones report heavy wild animal destruction (ro-
dents) and some tuber rot.
Other crops of interest in the division are yams, egusi melon, dwarf beans,
irish potatoes, carrots and cabbages.
Yams are grown as a secondary staple in all zones, but least in the High-
lands. Tuber rot problems are common. Egusi is an important cash earner for
women in Mamfe West and Eyumojock, and is severely attacked in the few weeks
after emergence by beetles and grasshoppers.
Dwarf beans are grown in the Highlands, and seem to do reasonably well.
The upper highlands, around Nkongle, specialize in temperate zone crops: irish
potatoes, carrots and cabbage, which do very well but only with intensive
chemical applications to destroy cutworms and leaf fungi. The chemicals have
been promoted and supervised by a recently transferred extension agent, and
supplied by West Province. Posting of another trained technician is necessary
to the continued success of this technology-intensive system.
TLU IRA-Ekona p.55