Directory UMM :Data Elmu:jurnal:A:Agricultural Systems:Vol64.Issue1.Apr2000:
Agricultural Systems 64 (2000) 55±65
www.elsevier.com/locate/agsy
New sorghum and millet cultivar introduction in
Sub-Saharan Africa: impacts and research agenda
M.M. Ahmed a, J.H. Sanders b,*, W.T. Nell c
a
ILRI, Addis Ababa, Ethiopia
Department of Agricultural Economics, Purdue University, 11445 Krannert Building,
West Lafayette, IN 47907-1145, USA
c
Centre for Agricultural Management, University of the Orange Free State, Bloemfontein, South Africa
b
Received 16 November 1999; received in revised form 6 February 2000; accepted 8 February 2000
Abstract
In spite of substantial introduction of new sorghum and millet cultivars in semiarid SubSaharan Africa, there has been minimum aggregate impact on yields (FAO and ICRISAT,
1996: The World Sorghum Economies: Facts, Trends and Outlook. FAO, Rome, Italy and
ICRISAT, Andhra Pradesh, India) in contrast with other crops, such as cotton and maize.
Only where inorganic fertilizers and improved water retention or irrigation were combined
with new cultivars were there large yield increases. Given the low soil fertility and irregular
rainfall in semiarid regions, both increased water availability and higher levels of principal
nutrients apparently will be necessary for substantial yield increase. The cultivar-alone strategy is unlikely to have a signi®cant sustainable yield eect and therefore reduce poverty in
semiarid Sub-Saharan Africa. # 2000 Elsevier Science Ltd. All rights reserved.
Keywords: Sub-Saharan Africa; Sorghum and millet; Impact; Aggregate yields; Inorganic fertilizers
1. Introduction
Poverty alleviation has been a principal objective of technology development
strategies in Sub-Saharan Africa for crops with drought resistance, speci®cally sorghum, millet, peanuts, and cowpeas. Poverty is concentrated in rural areas and a
high proportion of the population (25%) of Sub-Saharan Africa lives in semiarid
regions (Sanders et al., 1996, p. xix). In 1990, 48% (218 million people) of the population
of Sub-Saharan Africa were living in poverty (World Bank, 1992, p. 30). Sorghum and
millet contribute to the food security of many of the world's poorest, most food-insecure,
* Corresponding author. Tel.: +1-765-494-4221; fax: +1-765-494-9176.
E-mail address: [email protected] (J.H. Sanders).
0308-521X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0308-521X(00)00013-5
56
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
agroecological zones (FAO and ICRISAT, 1996). These crops are also a principal staple
food in the urban areas of many of the poorest countries in Sub-Saharan Africa.
The major emphasis of technological change in the semiarid region over the last three
decades has been new cultivars. Our objective is to evaluate the success of new cultivar
strategies for seven case studies of new cultivar introduction of sorghum and millet in
Sub-Saharan Africa and then to suggest an alternative strategy to improve the productivity gains of the introduced cultivars. The ®rst question is: have new cultivars been
introduced and, if so, what are the eects of the new cultivars and associated technologies on aggregate yield levels? In the next section we respond to these questions, drawing
upon our ®eld studies in various countries. Next we dichotomize the new production
systems into those that have had a noticeable aggregate impact on yields and those that
have not. Then we consider the components of the successful technology components
observed. In the conclusions we oer some implications for policies and markets.
2. New cultivar introduction in semiarid Sub-Saharan Africa
Farming in drought areas has been characterized as a subsistence activity with
farmers producing a wide array of crops (including multiple cultivars of the same
crop) for their own consumption, using few purchased inputs. With low expected
pro®tability of crop production, it is a rational strategy for farmers to retreat toward
subsistence (as farmers did in the Great Depression), market little, purchase few
inputs, and produce more for their own consumption (Paarlberg, 1988, p. 8). Given
the climatic and price variation, poor soils, insect and disease problems, agriculture
is risky in semiarid regions. Rather than focusing on pro®tability and risk reduction
in agriculture, public policy has been to search for technologies involving minimal
expenditures by farmers and no foreign exchange. Hence, the emphasis in new technology introduction has been on new cultivars and increased utilization of on-farm
and within-region resources to minimize cash expenditures.
Despite lagging behind many other commodity-based research programs, such as
maize and cotton, sorghum and pearl millet research in Sub-Saharan Africa has been
successful in diusing a large number of new cultivars onto farmers' ®elds. The last
two decades of research have resulted in the release of over 40 sorghum cultivars in 23
countries and 16 millet cultivars in 12 countries (Miller et al., 1996; Rohrbach, 1996).
Among the large number of new cultivars, we studied seven success stories of diffusion (Table 1). Outside of South Africa and Sudan, most of the new cultivars are
open-pollinated and early maturing. Thus, drought escape was a principal factor in
the breeders' strategy to overcome the moisture constraint in the semiarid environments. During the last 30 years in the Sahel, rainfall has been one standard deviation
below the long-term normal (Fig. 1). In southern Africa, there were several major
droughts in the 1980s and 1990s.
Since there were substantial drought problems in the 1980s, the earliness of these
cultivars has been much appreciated. However, short-seasoned cultivars are generally
adopted by farmers as part of a portfolio with other longer-season cultivars to take
advantage of the years when rainfall is adequate or good.
57
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
Table 1
Selected characteristics of the successful new sorghum and millet cultivar introductionsa
Cultivarb
Country
Year of
release
Season
length
Current
adoptionc
Internal
rate of
return
Change in
input use
HD-1 (S)
Sudan
1983
Medium
17% (1993)
36% (low
fertilization)
Fertilization
(irrigated area)
S-35 (S)
Cameroon
1986
Short
33% (1996)
2%
No change
observed
Chad
1989
SV-2 (S)
Zimbabwe
1987
Short
36% (1995)
18%
No change
observed
Zambian
cultivarsd (S)
Zambia
1987±93
Short,
medium,
and long
35% 1996
12%
Fertilized only
on large farms
PMV-2 (M)
Zimbabwe
1991
Short
26% (1995)
31%
No change
reported
Okashana-1 (M)
Namibia
1990
Short
35% (1993)
13%
No change
reported
NK-283 (S)
South Africa
Unknown
Medium to
long
50±70% (1998)
NA
Fertilization and
water retention
27% (1996)
a
Sources: Ahmed and Sanders, 1992; Sanders et al., 1994; Anandajayasekeram et al., 1995; Ahmed, 1996;
Chisi et al., 1997; Yapi et al., 1997.
b
Sorghum and millet, respectively, are identi®ed by the letters S and M in parentheses.
c
Latest available information with the year in parentheses.
d
Include two hybrids and three open-pollinated cultivars of which Kuyuma and Sima (released in 1989) and
MMSH-928 (released in 1993) are the most widely adopted.
Earliness gives drought escape but reduces the potential of the plant to respond to
better growing conditions since the plant will not be in the ®eld long enough to take
advantage of these conditions (Shapiro et al., 1993). An example of the disadvantage of
earliness is the S-35 cultivar. S-35 was selected from ICRISAT material in northern
Nigeria and then made available in regional trials to Cameroon. In 1984, widespread
trials on farmers' ®elds were implemented. In this major drought year, the yields of the
early S-35 doubled the local and other new cultivar yields (Johnson, 1987, p. 657;
Kamuanga and Fobasso, 1992, p. 22). There was no yield advantage to S-35 in normal
and good rainfall years after 1984. For a portfolio addition to protect against dry years,
S-35 was very popular. After its release in 1986, farmers in northern Cameroon began
rapidly introducing S-35 into their mix of cultivars of dierent season lengths. Nevertheless, since S-35 was only one of a number of cultivars grown by individual farmers
and yield eects were small except in poor rainfall years, the rate of return was very
low, only 2%, a marginal investment. Thus, we would expect only small-scale or
partial adoption of early maturing cultivars and lower rates of return than would be
expected with normal or late-maturing cultivars accompanied with higher input
levels (Table 1). So there is a cost to drought-escape strategies.
58
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
Fig 1. Rainfall in the Sahel, 1922±96. Note: this ®gure graphs the number of standard deviations from
average long-term rainfall in the Sahel for each year. The base long-term mean was calculated from
rainfall over the period 1922±96. Source: Figure updated from Sanders et al., 1996, p. 8 with unpublished
data from USAID/FEWS.
Field surveys in several countries indicated that diusion of these improved sorghum and millet cultivars is delayed by the slow evolution of the seed industry. The
lack of sucient high-quality seeds was cited as the major constraint to adoption of
improved cultivars by 60, 66, and 45% of the non-adopters of PMV-2, SV-2, and the
new sorghum cultivars in Zambia, respectively. In Sudan, the public sector seed
producers were unable to respond to the rapidly increasing demand for HD-1 seeds
(Ahmed and Sanders, 1992). In the early 1990s, approximately one-third of Gezira
farmers complained about their inability to obtain HD-1 seeds and/or inorganic
fertilizers and cited this lack of availability as a principal determinant of their
inability to introduce the new sorghum hybrid (Nichola, 1994).
In southern Africa, the private sector has shown little interest in selling open-pollinated varieties to small-scale farmers. If farmers depend upon non-governmental
organization (NGO) and government handouts of seeds, the development of private
marketing institutions will be hindered unless suciently high prices are charged for
the private sector to be interested in entering the market. Private seed and other input
markets are well established in South Africa, and the private seed sector is reentering
the production of hybrid sorghum seeds in the Sudan. Outside of South Africa in the
other six cases, government programs played an important role in promoting new
sorghum and millet cultivars, including providing the seed and often other inputs.
3. Patterns of technology adoption
Sorghum area in Sub-Saharan Africa reached a record high of over 23 million
hectares in 1997 (Fig. 2) with production increasing at 2.9% annually and millet
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
59
Fig. 2. Sorghum area, production and average yield in Sub-Saharan Africa, 1980±97. Note: from 1980 to
1997, the annual rate of growth of sorghum area in Sub-Saharan Africa was 3.7%, production 2.9%, and
yield ÿ0.8%. Source: Data from FAO (1998).
production increasing even faster at 3.3%. Unfortunately, these gains were based
upon continuing area expansion while yields continued to decline at 0.8. and 0.7 rates
for sorghum and millet, respectively, for the period 1980±96 (FAO, 1998). During the
past two decades with increasing population pressure on the land, the fallow-system
method of restoring soil fertility has been disappearing. With declining soil fertility in
traditional crop production regions, sorghum and millet have been pushed into more
marginal crop areas. Due to the riskiness of fertilizer response in marginal rainfall
regions and the failure of government policy to support crop productivity increase in
the traditional production zones, there has been little substitution of higher input
use for the soil-fertility depletion. So these gains in aggregate production are not
sustainable given the continuing decline of land productivity.
In all seven cases of new cultivar diusion evaluated, there were farm-level yield
gains. However, only in the two cases of sorghum in the Sudan and South Africa are
aggregate yield increases clearly visible (Figs. 3 and 4). Fig. 3 compares sorghum
yields on the main irrigation project of the Gezira (approximately 50% of the irrigated area in Sudan is in this scheme) with those in the mechanized drylands, the
main sorghum production area the Sudan. From 1985±1996, irrigated yields
increased by 3%, whereas on the mechanized drylands they declined by 0.3% per
year. The yield increases in the Gezira resulted from a combination of a high-yielding cultivar (Hageen Dura-1), fertilization, and improved agronomic practices
(Ahmed and Sanders, 1992).
On the mechanized vertisols of Sudan, sorghum is produced extensively without
fertilization or water-retention technologies, but new cultivars are widely adopted.
Many new sorghum cultivars were introduced in the early 1970s as combinable, highyielding cultivars (e.g. Dabar-1 and Gadam Elhamam-47; see Nichola and Sanders,
1996). In the 1990s, three new cultivars were introduced (SRN-39, IS-9830 and M90393) with some diusion being reported (Rohrbach, 1996). Even with these new
cultivars, there were no aggregate yield eects on the mechanized vertisols.
60
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
Fig. 3. Sorghum yields in the irrigated Gezira and the mechanized rainfed sector of Sudan (1976±96).
Note: From 1985 to 1996, the annual rate of growth of yield in the Gezira was 3% and in the mechanized
zone was ÿ0.3%. Source: Nichola and Sanders, 1996; Sanders et al., 1996. Updated from Sanders,
Shapiro and Ramaswamy with data from Nichola and Sanders.
Fig. 4. Sorghum yields in Cameroon, Zimbabwe, Zambia, and South Africa. Source: Data from FAO (1998).
Sorghum production in South Africa is another example where high-yielding
hybrids were combined with inorganic fertilizers and water-retention technologies.
On the sandy soils of the semiarid regions of South Africa, sorghum producers
increase water storage in the soil by controlling weeds in the o season and cultivating to absorb more water. A sub-surface crust of 1±2 m in depth holds the water
and makes it accessible to the crop. With the utilization of this water-conservation
technique, the South Africans then do not need to select for earliness for drought
escape. Rather, they combine a medium- to long-season-length cultivar with moderate
to high levels of fertilization and attain yields of up to 7 Mt/ha.
Sorghum yields averaged 1.8 t/ha in South Africa since 1980 despite the six
drought years after 1980 and the 1991±92 drought. This contrasts with the 0.8 t/ha
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
61
in the rest of Sub-Saharan Africa (FAO, 1998). Sorghum yields in South Africa
show substantial variation with weather but they are substantially higher than those
of the other three countries illustrated where there was also rapid introduction of
new sorghum cultivars (Fig. 4).
The success of the intensive production pattern of sorghum in Sudan and South
Africa is similar to the success of cotton in Francophone Africa and maize technologies in the Sudano-Guinean zone of the Sahel (Sanders et al., 1996; pp. 54±58). In
these higher-rainfall regions, new cultivar introductions were combined with cropmanagement improvements, including increases in fertilization, density, and pest
control. This maize technology introduction has resulted in rapid increases in maize
yields of 1.2±4.9% annual growth rates in Mali, Ghana, and Burkina Faso. Similarly, over 60% of the threefold increases of sorghum yields in the US in 30 years
were due to improved agronomy practices, especially fertilization, herbicides, and
water control (Miller and Kebede, 1984, pp. 6, 11).
Sorghum and pearl millet production in Sub-Saharan Africa follows two distinct
paths: intensive and extensive. The observed intensive development path utilizes
hybrids and inorganic fertilizers under irrigation or with water-harvesting technology. This system characterizes sorghum production in the irrigated Gezira of Sudan
and semiarid rain-fed regions of South Africa.
The extensive system uses a mixture of traditional and improved, early maturing
cultivars. New cultivars were generally adopted as a risk-avoidance strategy without
improved agronomy. Neither fertilization nor water-retention technologies were
practiced. This system dominates the small-scale farming that produces most of the
sorghum and pearl millet in Sub-Saharan Africa. As a result, yields were low and
sometimes declined even in countries where rapid diusion of new cultivars was
reported (Table 1; Fig. 4).
Our principal explanation for the limited aggregate gains for sorghum and millet
is the over-reliance on a cultivar-alone strategy in the harsh environments where
sorghum and millet are produced. In these agroecological zones, both water stress
and inadequate soil fertility are generally encountered.
In Sub-Saharan Africa there are widespread soil-nutrient de®ciencies. The soilfertility situation is getting worse as increasing population pressure on the land
reduces eectiveness of the fallow system or eliminates this practice. In these nutrient-stressed environments, new cultivars and low external inputs can support no
more than low-productivity, subsistence type of agriculture (Crosson and Anderson,
1994, p. 16). Nevertheless, many new cultivars have been released recently and large
future impacts are possible if complementary technologies are introduced.
Is there evidence for the yield and pro®t-increasing eects of fertilization on
semiarid soils? In Burkina Faso on heavier soils there was extensive testing over 2
years in four regions, indicating that the combination of a water-retention technology and moderate inorganic fertilization increased yields by 50±100% (Sanders et
al., 1996, p. 78). On lighter soils with higher levels of sand, the crusting or runo
problem is less severe, but rapid in®ltration of water often results in water becoming
unavailable to the plant. Here fertilization and higher densities have been shown to
increase water-use eciency by apparently increasing the water-holding capacity of
62
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
these sandy soils (ICRISAT, 1987, 1988; Mokwunye and Hammond, 1992, pp. 131,
132). Reviewing the fertilization issue on sorghum and millet, Shapiro and Sanders
(1998, pp. 475, 478) showed the pro®tability on both heavy (with the water-retention
technology of tied ridges) and light soils. The combination of inorganic and organic
fertilizers on millet and sorghum were shown to substantially outyield individual use
of either in a recent review (Bekunda et al., 1997, p. 66). Organic fertilizers are
apparently an excellent water-retention device in both sandy and heavier soils.
4. Input and product markets
Higher input use to simultaneously overcome soil nutrients and moisture constraints
is necessary to obtain signi®cant yield eects from new cultivars of sorghum and millet.
There is much discussion of substituting for soluble inorganic fertilizers with organic
fertilizers, rotations, rock phosphate, and several other more exotic or region-speci®c
alternatives. However, cereal yields have nowhere been substantially increased without
moderate levels of inorganic fertilizers. These other soil-fertility measures need to be
thought of as complements to, rather than substitutes for, inorganic fertilizers (QuinÄones et al., 1997). It is unlikely that Sub-Saharan Africa will make scienti®c breakthroughs in soil fertility or change the basic economics of these would-be substitutes.
As the risk of income losses is expected to be an important determinant of adoption of inorganic fertilizers in semiarid regions, increased water availability is critical
to reduce such risks. The combined technology of inorganic fertilizers, water-retention techniques (or irrigation), and new cultivars is expected to substantially increase
yields and improve farmers' incomes and nutrition. Organic fertilizers will frequently need to complement the inorganic ones to improve the retention of water
and nutrients in the soil. Once soil moisture and nutrients are available, breeding
strategies can emphasize medium to late-maturing cultivars to respond to the
improved agronomic environment.
If sorghum and millet are not suciently pro®table because markets do not expand
or governments continue to provide poor policy support, farmers will either not use
inputs or will use them on smaller areas with sorghum and millet. In this case, regional
and international research priorities will need to shift from the traditional cereals
since adoption of cultivars alone is insucient to improve yields. A poverty-alleviation policy of introducing cultivars alone or with some soil-fertility measures but
without inorganic fertilizer will not be sustainable as soil nutrient mining continues.
The major emerging constraint, once the diusion process of new cultivars and
associated technologies has begun, is the ability of input markets (seeds, fertilizer,
and credit) to respond to rapidly increasing demands. Policy measures that reduce
the prices paid by farmers for the inputs used or increase the prices received by
farmers for the resulting outputs will strengthen farmer incentives to adopt these
innovations. Public investment in marketing and transportation infrastructure
would reduce input costs and increase producer prices by reducing transaction costs.
Attention to the evolution of new uses in the product markets should also accelerate diusion by moderating the between-season price collapses and partially
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
63
osetting the long-run price decline resulting from successful technology introduction. Market development, as for the use of domestic cereals in bread and beer, can
serve an intermediate function until domestic incomes rise suciently to begin the
rapid dietary change to animal products and therefore the shift to the rapidly
accelerating use of cereals as feed, as has already occurred in many developing
countries.
5. Conclusions
Everywhere in the world that yields have been successfully increased in semiarid
regions, including Sub-Saharan Africa, more water and inorganic fertilizers have
been required. The breeding eort with sorghum and millet has been outstanding
and needs to continue, but things need to be made easier for the breeders with an
emphasis on improving the production environment and increasing the pro®tability
of traditional cereal production. Short-term yield increases that are not sustainable
due to increasing soil-fertility de®cits will not have long-term eects on poverty
reduction.
During the last two decades, increases in sorghum and millet output in SubSaharan Africa to feed the growing population were possible only through rapid
area expansion because yields were low and declining. In the long run, the current
area expansion can be reversed by successful implementation of the combined technology. Over time, as the bene®ts of combining moderate levels of inorganic fertilizers, water retention, and improved cultivars become apparent, farmers will
gradually increase the level of input use by reinvesting part of the additional gains.
The increased production of these traditional staples can be absorbed by new markets, especially the use of feed grains for poultry and milk production as the demand
for animal products accelerates with economic growth.
An important issue for this region is whether low-input strategies have been
prompted by the poor resource base and riskiness of these agricultural systems or by
the over-emphasis in public policy on maintaining low food prices to the detriment
of the modernization process in agriculture. To transform the agricultural sector in
semiarid regions, farming needs to be pro®table and risk reduction feasible, thereby
encouraging farmers to purchase more inputs and increase land productivity. Technical options are available but farmers need a pro®table environment. An overemphasis by policymakers on maintaining low urban food prices can make the shift
to more intensive production practices less likely. So technical, economic, and public
policy support need to be combined for increasing productivity and incomes in
semiarid regions.
References
Ahmed, M.M., 1996. Costs and Returns to Development and Diusion of SV-2 and PMV-2 (unpublished
mimeo). ICRISAT Southern and Eastern Africa Region, Bulawayo, Zimbabwe.
64
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
Ahmed, M., Sanders, J.H., 1992. The economic impacts of Hageen Dura-1 in the Gezira scheme, Sudan.
In: Proceedings of the International Sorghum and Millet CRSP Conference, Royal Nueces Hotel,
Corpus Christi, TX, USA, 8±12 July, 1991. University of Nebraska, INTSORMIL Publication No. 921, Lincoln, NE.
Anandajayasekeram, P., Martella, D.R., Sanders, J.H., Kupfma, B., 1995. Report on the Impact Assessment of the SADC/ICRISAT Sorghum and Millet Improvement Program, Vol. I. SACCAR, Gaborone, Botswana.
Bekunda, M.A., Bationo, A., Ssali, H., 1997. Soil fertility management in Africa: a review of selected
research trials. In: Buresh, R.A., Sanchez, P.A., Calhoun, F. (Eds.), Replenishing Soil Fertility in
Africa, SSSA Special Bulletin No. 51, Ch. 4. Soil Society of America and American Society of Agronomy, Madison, WI, pp. 63±79.
Chisi, M., Anandajayasekeram, P., Martella, D., Ahmed, M., Mwape, M., 1997. Impact Assessment of
Sorghum Research in Zambia, 1983±2010. SACCAR, Gaborone, Botswana.
Crosson, P., Anderson, N. 1994. Achieving a sustainable agricultural system in Sub-Sahara Africa, Paper
No. 7, Post-UNCED Series, Building Blocks for Africa 2025. World Bank, Washington, D.C.
FAO (Food and Agriculture Organization of the United Nations), 1998. Agriculture database. FAO
Committee on Commodity Problems, Rome, Italy.
FAO and ICRISAT (International Crops Research Institute in Semi-Arid Tropics), 1996. The World Sorghum Economies: Facts, Trends and Outlook. FAO, Rome, Italy and ICRISAT, Andhra Pradesh, India.
ICRISAT (International Crops Research Institute for the Semiarid Tropics), 1987. ICRISAT Sahelian
Centre Annual Report, Niamey, Niger.
ICRISAT (International Crops Research Institute for the Semiarid Tropics), 1988. Annual Report, Niamey, Niger.
Johnson, J.J., 1987. The 1985 on-farm test results of sorghum varieties: comparison of S-35 with local
varieties over two years and 165 sites. In: Menyonga, J.M., Bezuneh, T., Youdeowei, A. (Eds.), Food
Grain Production in Semiarid Africa, pp. 651±683. Fourth Dimension Printing Services, Essex, UK.
Kamuanga, M., Fobasso, M., 1992. Role of Farmers in the Evaluation of an Improved Variety: The Case
of S-35 Sorghum in Northern Cameroon, unpublished mimeo. IAR, Maroua, Cameroon.
Miller, F.R., Cisse, L., Hanak-Freud, E., Sanders, J.H., 1996. Report of the Center Commissioned
External Review (CCER) of the ICRISAT Western and Central African Region. ICRISAT, Andhra
Pradesh, India.
Miller, F.R., Kebede, Y., 1984. Genetic contributions to yield gains in sorghum, 1950 to 1980. In: Fehr,
W.R. (Ed.), Genetic Contributions to Yield Gains of Five Major Crop Plants. Crop Science Society of
America and American Society of Agronomy, CSSA Special Publication No. 7, Ch. 1, Madison, WI,
pp. 1±14.
Mokwunye, U., Hammond, L.L., 1992. Myths and science of fertilizer used in the tropics. In: Myths and
Science of Soils of the Tropics, Soil Science Society of America, and American Society of Agronomy,
SSSA Special Publication No. 29, Ch. 7, Madison, WI.
Nichola, T., 1994. Adoption of Hybrid Sorghum at the Farm and Regional Levels in the Gezira Irrigated
Scheme of the Sudan, unpublished PhD dissertation. Purdue University, Department of Agricultural
Economics, West Lafayette, IN.
Nichola, T., Sanders, J.H., 1996. A probit analysis of the determinants of adoption when inputs are
rationed: the Gezira experience with hybrid sorghum. Science, Technology, and Development 14 (3),
107±119.
Paarlberg, D., 1988. Toward a Well-Fed World. Iowa State University Press, Ames, IA.
QuinÄones, M.A., Borlaug, N.E., Dowswell, C.R., 1997. A fertilizer based green revolution for Africa. In:
Buresh, R.A., Sanchez, P.A., Calhoun, F. (Eds.), Replenishing Soil Fertility in Africa, SSSA Special
Bulletin No. 51, Ch. 4, Soil Society of America and American Society of Agronomy, Madison, WI.
Rohrbach, D.D., 1996. Background: Impact of Research in Southern and Eastern Africa (unpublished
mimeo). ICRISAT Southern and Eastern Africa Region, Bulawayo, Zimbabwe.
Sanders, J.H., 1994. Economic impact of the Commodity Research Networks of SAFGRAD. In: Sanders,
J.H., Bezuneh, T., Schroeder, A.C., Assessment of the SAFGRAD Commodity Networks. USAID/
AFR, OAU/STR-SAFGRAD, Ouagadougou, Burkina Faso, pp. 1±27.
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
65
Sanders, J.H., Nagy, J.H., Ramaswamy, S., 1990. Developing new agricultural technologies for the
Sahelian countries: the Burkina Faso case. Agricultural Systems 9 (2), 97±112.
Sanders, J.H., Shapiro, B.I., Ramaswamy, S., 1996. The Economics of Agricultural Technology in Semiarid Sub-Saharan Africa. Johns Hopkins University Press, Baltimore, MD.
Shapiro, B.I., Sanders, J.H., 1998. Fertilizer use in semiarid West Africa: pro®tability and supporting
policy. Agricultural Systems 56 (4), 467±482.
Shapiro, B., Sanders, J.H., Reddy, K.C., Baker, T.G., 1993. Evaluating and adapting new technologies in
a high risk agricultural system Ð Niger. Agricultural Systems 42, 153±170.
USAID/FEWS (Famine Early Warning System), 1994. Bulletin SH-11, Nov. 30, USAID/FEWS Project,
Tulane/Pragma Group, Arlington, VA.
World Bank, 1992. Development and the Environment, World Bank Development Report. World Bank,
Washington, DC.
Yapi, A.M., Debrah, K.S., Gandaoua, G., Njomah, C., 1997. Impact of Germplasm Research Spillovers:
The Case of Sorghum Variety S-35 in Cameroon and Chad. ICRISAT Sahelian Center, Bamako, Mali.
www.elsevier.com/locate/agsy
New sorghum and millet cultivar introduction in
Sub-Saharan Africa: impacts and research agenda
M.M. Ahmed a, J.H. Sanders b,*, W.T. Nell c
a
ILRI, Addis Ababa, Ethiopia
Department of Agricultural Economics, Purdue University, 11445 Krannert Building,
West Lafayette, IN 47907-1145, USA
c
Centre for Agricultural Management, University of the Orange Free State, Bloemfontein, South Africa
b
Received 16 November 1999; received in revised form 6 February 2000; accepted 8 February 2000
Abstract
In spite of substantial introduction of new sorghum and millet cultivars in semiarid SubSaharan Africa, there has been minimum aggregate impact on yields (FAO and ICRISAT,
1996: The World Sorghum Economies: Facts, Trends and Outlook. FAO, Rome, Italy and
ICRISAT, Andhra Pradesh, India) in contrast with other crops, such as cotton and maize.
Only where inorganic fertilizers and improved water retention or irrigation were combined
with new cultivars were there large yield increases. Given the low soil fertility and irregular
rainfall in semiarid regions, both increased water availability and higher levels of principal
nutrients apparently will be necessary for substantial yield increase. The cultivar-alone strategy is unlikely to have a signi®cant sustainable yield eect and therefore reduce poverty in
semiarid Sub-Saharan Africa. # 2000 Elsevier Science Ltd. All rights reserved.
Keywords: Sub-Saharan Africa; Sorghum and millet; Impact; Aggregate yields; Inorganic fertilizers
1. Introduction
Poverty alleviation has been a principal objective of technology development
strategies in Sub-Saharan Africa for crops with drought resistance, speci®cally sorghum, millet, peanuts, and cowpeas. Poverty is concentrated in rural areas and a
high proportion of the population (25%) of Sub-Saharan Africa lives in semiarid
regions (Sanders et al., 1996, p. xix). In 1990, 48% (218 million people) of the population
of Sub-Saharan Africa were living in poverty (World Bank, 1992, p. 30). Sorghum and
millet contribute to the food security of many of the world's poorest, most food-insecure,
* Corresponding author. Tel.: +1-765-494-4221; fax: +1-765-494-9176.
E-mail address: [email protected] (J.H. Sanders).
0308-521X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0308-521X(00)00013-5
56
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
agroecological zones (FAO and ICRISAT, 1996). These crops are also a principal staple
food in the urban areas of many of the poorest countries in Sub-Saharan Africa.
The major emphasis of technological change in the semiarid region over the last three
decades has been new cultivars. Our objective is to evaluate the success of new cultivar
strategies for seven case studies of new cultivar introduction of sorghum and millet in
Sub-Saharan Africa and then to suggest an alternative strategy to improve the productivity gains of the introduced cultivars. The ®rst question is: have new cultivars been
introduced and, if so, what are the eects of the new cultivars and associated technologies on aggregate yield levels? In the next section we respond to these questions, drawing
upon our ®eld studies in various countries. Next we dichotomize the new production
systems into those that have had a noticeable aggregate impact on yields and those that
have not. Then we consider the components of the successful technology components
observed. In the conclusions we oer some implications for policies and markets.
2. New cultivar introduction in semiarid Sub-Saharan Africa
Farming in drought areas has been characterized as a subsistence activity with
farmers producing a wide array of crops (including multiple cultivars of the same
crop) for their own consumption, using few purchased inputs. With low expected
pro®tability of crop production, it is a rational strategy for farmers to retreat toward
subsistence (as farmers did in the Great Depression), market little, purchase few
inputs, and produce more for their own consumption (Paarlberg, 1988, p. 8). Given
the climatic and price variation, poor soils, insect and disease problems, agriculture
is risky in semiarid regions. Rather than focusing on pro®tability and risk reduction
in agriculture, public policy has been to search for technologies involving minimal
expenditures by farmers and no foreign exchange. Hence, the emphasis in new technology introduction has been on new cultivars and increased utilization of on-farm
and within-region resources to minimize cash expenditures.
Despite lagging behind many other commodity-based research programs, such as
maize and cotton, sorghum and pearl millet research in Sub-Saharan Africa has been
successful in diusing a large number of new cultivars onto farmers' ®elds. The last
two decades of research have resulted in the release of over 40 sorghum cultivars in 23
countries and 16 millet cultivars in 12 countries (Miller et al., 1996; Rohrbach, 1996).
Among the large number of new cultivars, we studied seven success stories of diffusion (Table 1). Outside of South Africa and Sudan, most of the new cultivars are
open-pollinated and early maturing. Thus, drought escape was a principal factor in
the breeders' strategy to overcome the moisture constraint in the semiarid environments. During the last 30 years in the Sahel, rainfall has been one standard deviation
below the long-term normal (Fig. 1). In southern Africa, there were several major
droughts in the 1980s and 1990s.
Since there were substantial drought problems in the 1980s, the earliness of these
cultivars has been much appreciated. However, short-seasoned cultivars are generally
adopted by farmers as part of a portfolio with other longer-season cultivars to take
advantage of the years when rainfall is adequate or good.
57
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
Table 1
Selected characteristics of the successful new sorghum and millet cultivar introductionsa
Cultivarb
Country
Year of
release
Season
length
Current
adoptionc
Internal
rate of
return
Change in
input use
HD-1 (S)
Sudan
1983
Medium
17% (1993)
36% (low
fertilization)
Fertilization
(irrigated area)
S-35 (S)
Cameroon
1986
Short
33% (1996)
2%
No change
observed
Chad
1989
SV-2 (S)
Zimbabwe
1987
Short
36% (1995)
18%
No change
observed
Zambian
cultivarsd (S)
Zambia
1987±93
Short,
medium,
and long
35% 1996
12%
Fertilized only
on large farms
PMV-2 (M)
Zimbabwe
1991
Short
26% (1995)
31%
No change
reported
Okashana-1 (M)
Namibia
1990
Short
35% (1993)
13%
No change
reported
NK-283 (S)
South Africa
Unknown
Medium to
long
50±70% (1998)
NA
Fertilization and
water retention
27% (1996)
a
Sources: Ahmed and Sanders, 1992; Sanders et al., 1994; Anandajayasekeram et al., 1995; Ahmed, 1996;
Chisi et al., 1997; Yapi et al., 1997.
b
Sorghum and millet, respectively, are identi®ed by the letters S and M in parentheses.
c
Latest available information with the year in parentheses.
d
Include two hybrids and three open-pollinated cultivars of which Kuyuma and Sima (released in 1989) and
MMSH-928 (released in 1993) are the most widely adopted.
Earliness gives drought escape but reduces the potential of the plant to respond to
better growing conditions since the plant will not be in the ®eld long enough to take
advantage of these conditions (Shapiro et al., 1993). An example of the disadvantage of
earliness is the S-35 cultivar. S-35 was selected from ICRISAT material in northern
Nigeria and then made available in regional trials to Cameroon. In 1984, widespread
trials on farmers' ®elds were implemented. In this major drought year, the yields of the
early S-35 doubled the local and other new cultivar yields (Johnson, 1987, p. 657;
Kamuanga and Fobasso, 1992, p. 22). There was no yield advantage to S-35 in normal
and good rainfall years after 1984. For a portfolio addition to protect against dry years,
S-35 was very popular. After its release in 1986, farmers in northern Cameroon began
rapidly introducing S-35 into their mix of cultivars of dierent season lengths. Nevertheless, since S-35 was only one of a number of cultivars grown by individual farmers
and yield eects were small except in poor rainfall years, the rate of return was very
low, only 2%, a marginal investment. Thus, we would expect only small-scale or
partial adoption of early maturing cultivars and lower rates of return than would be
expected with normal or late-maturing cultivars accompanied with higher input
levels (Table 1). So there is a cost to drought-escape strategies.
58
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
Fig 1. Rainfall in the Sahel, 1922±96. Note: this ®gure graphs the number of standard deviations from
average long-term rainfall in the Sahel for each year. The base long-term mean was calculated from
rainfall over the period 1922±96. Source: Figure updated from Sanders et al., 1996, p. 8 with unpublished
data from USAID/FEWS.
Field surveys in several countries indicated that diusion of these improved sorghum and millet cultivars is delayed by the slow evolution of the seed industry. The
lack of sucient high-quality seeds was cited as the major constraint to adoption of
improved cultivars by 60, 66, and 45% of the non-adopters of PMV-2, SV-2, and the
new sorghum cultivars in Zambia, respectively. In Sudan, the public sector seed
producers were unable to respond to the rapidly increasing demand for HD-1 seeds
(Ahmed and Sanders, 1992). In the early 1990s, approximately one-third of Gezira
farmers complained about their inability to obtain HD-1 seeds and/or inorganic
fertilizers and cited this lack of availability as a principal determinant of their
inability to introduce the new sorghum hybrid (Nichola, 1994).
In southern Africa, the private sector has shown little interest in selling open-pollinated varieties to small-scale farmers. If farmers depend upon non-governmental
organization (NGO) and government handouts of seeds, the development of private
marketing institutions will be hindered unless suciently high prices are charged for
the private sector to be interested in entering the market. Private seed and other input
markets are well established in South Africa, and the private seed sector is reentering
the production of hybrid sorghum seeds in the Sudan. Outside of South Africa in the
other six cases, government programs played an important role in promoting new
sorghum and millet cultivars, including providing the seed and often other inputs.
3. Patterns of technology adoption
Sorghum area in Sub-Saharan Africa reached a record high of over 23 million
hectares in 1997 (Fig. 2) with production increasing at 2.9% annually and millet
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
59
Fig. 2. Sorghum area, production and average yield in Sub-Saharan Africa, 1980±97. Note: from 1980 to
1997, the annual rate of growth of sorghum area in Sub-Saharan Africa was 3.7%, production 2.9%, and
yield ÿ0.8%. Source: Data from FAO (1998).
production increasing even faster at 3.3%. Unfortunately, these gains were based
upon continuing area expansion while yields continued to decline at 0.8. and 0.7 rates
for sorghum and millet, respectively, for the period 1980±96 (FAO, 1998). During the
past two decades with increasing population pressure on the land, the fallow-system
method of restoring soil fertility has been disappearing. With declining soil fertility in
traditional crop production regions, sorghum and millet have been pushed into more
marginal crop areas. Due to the riskiness of fertilizer response in marginal rainfall
regions and the failure of government policy to support crop productivity increase in
the traditional production zones, there has been little substitution of higher input
use for the soil-fertility depletion. So these gains in aggregate production are not
sustainable given the continuing decline of land productivity.
In all seven cases of new cultivar diusion evaluated, there were farm-level yield
gains. However, only in the two cases of sorghum in the Sudan and South Africa are
aggregate yield increases clearly visible (Figs. 3 and 4). Fig. 3 compares sorghum
yields on the main irrigation project of the Gezira (approximately 50% of the irrigated area in Sudan is in this scheme) with those in the mechanized drylands, the
main sorghum production area the Sudan. From 1985±1996, irrigated yields
increased by 3%, whereas on the mechanized drylands they declined by 0.3% per
year. The yield increases in the Gezira resulted from a combination of a high-yielding cultivar (Hageen Dura-1), fertilization, and improved agronomic practices
(Ahmed and Sanders, 1992).
On the mechanized vertisols of Sudan, sorghum is produced extensively without
fertilization or water-retention technologies, but new cultivars are widely adopted.
Many new sorghum cultivars were introduced in the early 1970s as combinable, highyielding cultivars (e.g. Dabar-1 and Gadam Elhamam-47; see Nichola and Sanders,
1996). In the 1990s, three new cultivars were introduced (SRN-39, IS-9830 and M90393) with some diusion being reported (Rohrbach, 1996). Even with these new
cultivars, there were no aggregate yield eects on the mechanized vertisols.
60
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
Fig. 3. Sorghum yields in the irrigated Gezira and the mechanized rainfed sector of Sudan (1976±96).
Note: From 1985 to 1996, the annual rate of growth of yield in the Gezira was 3% and in the mechanized
zone was ÿ0.3%. Source: Nichola and Sanders, 1996; Sanders et al., 1996. Updated from Sanders,
Shapiro and Ramaswamy with data from Nichola and Sanders.
Fig. 4. Sorghum yields in Cameroon, Zimbabwe, Zambia, and South Africa. Source: Data from FAO (1998).
Sorghum production in South Africa is another example where high-yielding
hybrids were combined with inorganic fertilizers and water-retention technologies.
On the sandy soils of the semiarid regions of South Africa, sorghum producers
increase water storage in the soil by controlling weeds in the o season and cultivating to absorb more water. A sub-surface crust of 1±2 m in depth holds the water
and makes it accessible to the crop. With the utilization of this water-conservation
technique, the South Africans then do not need to select for earliness for drought
escape. Rather, they combine a medium- to long-season-length cultivar with moderate
to high levels of fertilization and attain yields of up to 7 Mt/ha.
Sorghum yields averaged 1.8 t/ha in South Africa since 1980 despite the six
drought years after 1980 and the 1991±92 drought. This contrasts with the 0.8 t/ha
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
61
in the rest of Sub-Saharan Africa (FAO, 1998). Sorghum yields in South Africa
show substantial variation with weather but they are substantially higher than those
of the other three countries illustrated where there was also rapid introduction of
new sorghum cultivars (Fig. 4).
The success of the intensive production pattern of sorghum in Sudan and South
Africa is similar to the success of cotton in Francophone Africa and maize technologies in the Sudano-Guinean zone of the Sahel (Sanders et al., 1996; pp. 54±58). In
these higher-rainfall regions, new cultivar introductions were combined with cropmanagement improvements, including increases in fertilization, density, and pest
control. This maize technology introduction has resulted in rapid increases in maize
yields of 1.2±4.9% annual growth rates in Mali, Ghana, and Burkina Faso. Similarly, over 60% of the threefold increases of sorghum yields in the US in 30 years
were due to improved agronomy practices, especially fertilization, herbicides, and
water control (Miller and Kebede, 1984, pp. 6, 11).
Sorghum and pearl millet production in Sub-Saharan Africa follows two distinct
paths: intensive and extensive. The observed intensive development path utilizes
hybrids and inorganic fertilizers under irrigation or with water-harvesting technology. This system characterizes sorghum production in the irrigated Gezira of Sudan
and semiarid rain-fed regions of South Africa.
The extensive system uses a mixture of traditional and improved, early maturing
cultivars. New cultivars were generally adopted as a risk-avoidance strategy without
improved agronomy. Neither fertilization nor water-retention technologies were
practiced. This system dominates the small-scale farming that produces most of the
sorghum and pearl millet in Sub-Saharan Africa. As a result, yields were low and
sometimes declined even in countries where rapid diusion of new cultivars was
reported (Table 1; Fig. 4).
Our principal explanation for the limited aggregate gains for sorghum and millet
is the over-reliance on a cultivar-alone strategy in the harsh environments where
sorghum and millet are produced. In these agroecological zones, both water stress
and inadequate soil fertility are generally encountered.
In Sub-Saharan Africa there are widespread soil-nutrient de®ciencies. The soilfertility situation is getting worse as increasing population pressure on the land
reduces eectiveness of the fallow system or eliminates this practice. In these nutrient-stressed environments, new cultivars and low external inputs can support no
more than low-productivity, subsistence type of agriculture (Crosson and Anderson,
1994, p. 16). Nevertheless, many new cultivars have been released recently and large
future impacts are possible if complementary technologies are introduced.
Is there evidence for the yield and pro®t-increasing eects of fertilization on
semiarid soils? In Burkina Faso on heavier soils there was extensive testing over 2
years in four regions, indicating that the combination of a water-retention technology and moderate inorganic fertilization increased yields by 50±100% (Sanders et
al., 1996, p. 78). On lighter soils with higher levels of sand, the crusting or runo
problem is less severe, but rapid in®ltration of water often results in water becoming
unavailable to the plant. Here fertilization and higher densities have been shown to
increase water-use eciency by apparently increasing the water-holding capacity of
62
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
these sandy soils (ICRISAT, 1987, 1988; Mokwunye and Hammond, 1992, pp. 131,
132). Reviewing the fertilization issue on sorghum and millet, Shapiro and Sanders
(1998, pp. 475, 478) showed the pro®tability on both heavy (with the water-retention
technology of tied ridges) and light soils. The combination of inorganic and organic
fertilizers on millet and sorghum were shown to substantially outyield individual use
of either in a recent review (Bekunda et al., 1997, p. 66). Organic fertilizers are
apparently an excellent water-retention device in both sandy and heavier soils.
4. Input and product markets
Higher input use to simultaneously overcome soil nutrients and moisture constraints
is necessary to obtain signi®cant yield eects from new cultivars of sorghum and millet.
There is much discussion of substituting for soluble inorganic fertilizers with organic
fertilizers, rotations, rock phosphate, and several other more exotic or region-speci®c
alternatives. However, cereal yields have nowhere been substantially increased without
moderate levels of inorganic fertilizers. These other soil-fertility measures need to be
thought of as complements to, rather than substitutes for, inorganic fertilizers (QuinÄones et al., 1997). It is unlikely that Sub-Saharan Africa will make scienti®c breakthroughs in soil fertility or change the basic economics of these would-be substitutes.
As the risk of income losses is expected to be an important determinant of adoption of inorganic fertilizers in semiarid regions, increased water availability is critical
to reduce such risks. The combined technology of inorganic fertilizers, water-retention techniques (or irrigation), and new cultivars is expected to substantially increase
yields and improve farmers' incomes and nutrition. Organic fertilizers will frequently need to complement the inorganic ones to improve the retention of water
and nutrients in the soil. Once soil moisture and nutrients are available, breeding
strategies can emphasize medium to late-maturing cultivars to respond to the
improved agronomic environment.
If sorghum and millet are not suciently pro®table because markets do not expand
or governments continue to provide poor policy support, farmers will either not use
inputs or will use them on smaller areas with sorghum and millet. In this case, regional
and international research priorities will need to shift from the traditional cereals
since adoption of cultivars alone is insucient to improve yields. A poverty-alleviation policy of introducing cultivars alone or with some soil-fertility measures but
without inorganic fertilizer will not be sustainable as soil nutrient mining continues.
The major emerging constraint, once the diusion process of new cultivars and
associated technologies has begun, is the ability of input markets (seeds, fertilizer,
and credit) to respond to rapidly increasing demands. Policy measures that reduce
the prices paid by farmers for the inputs used or increase the prices received by
farmers for the resulting outputs will strengthen farmer incentives to adopt these
innovations. Public investment in marketing and transportation infrastructure
would reduce input costs and increase producer prices by reducing transaction costs.
Attention to the evolution of new uses in the product markets should also accelerate diusion by moderating the between-season price collapses and partially
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
63
osetting the long-run price decline resulting from successful technology introduction. Market development, as for the use of domestic cereals in bread and beer, can
serve an intermediate function until domestic incomes rise suciently to begin the
rapid dietary change to animal products and therefore the shift to the rapidly
accelerating use of cereals as feed, as has already occurred in many developing
countries.
5. Conclusions
Everywhere in the world that yields have been successfully increased in semiarid
regions, including Sub-Saharan Africa, more water and inorganic fertilizers have
been required. The breeding eort with sorghum and millet has been outstanding
and needs to continue, but things need to be made easier for the breeders with an
emphasis on improving the production environment and increasing the pro®tability
of traditional cereal production. Short-term yield increases that are not sustainable
due to increasing soil-fertility de®cits will not have long-term eects on poverty
reduction.
During the last two decades, increases in sorghum and millet output in SubSaharan Africa to feed the growing population were possible only through rapid
area expansion because yields were low and declining. In the long run, the current
area expansion can be reversed by successful implementation of the combined technology. Over time, as the bene®ts of combining moderate levels of inorganic fertilizers, water retention, and improved cultivars become apparent, farmers will
gradually increase the level of input use by reinvesting part of the additional gains.
The increased production of these traditional staples can be absorbed by new markets, especially the use of feed grains for poultry and milk production as the demand
for animal products accelerates with economic growth.
An important issue for this region is whether low-input strategies have been
prompted by the poor resource base and riskiness of these agricultural systems or by
the over-emphasis in public policy on maintaining low food prices to the detriment
of the modernization process in agriculture. To transform the agricultural sector in
semiarid regions, farming needs to be pro®table and risk reduction feasible, thereby
encouraging farmers to purchase more inputs and increase land productivity. Technical options are available but farmers need a pro®table environment. An overemphasis by policymakers on maintaining low urban food prices can make the shift
to more intensive production practices less likely. So technical, economic, and public
policy support need to be combined for increasing productivity and incomes in
semiarid regions.
References
Ahmed, M.M., 1996. Costs and Returns to Development and Diusion of SV-2 and PMV-2 (unpublished
mimeo). ICRISAT Southern and Eastern Africa Region, Bulawayo, Zimbabwe.
64
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
Ahmed, M., Sanders, J.H., 1992. The economic impacts of Hageen Dura-1 in the Gezira scheme, Sudan.
In: Proceedings of the International Sorghum and Millet CRSP Conference, Royal Nueces Hotel,
Corpus Christi, TX, USA, 8±12 July, 1991. University of Nebraska, INTSORMIL Publication No. 921, Lincoln, NE.
Anandajayasekeram, P., Martella, D.R., Sanders, J.H., Kupfma, B., 1995. Report on the Impact Assessment of the SADC/ICRISAT Sorghum and Millet Improvement Program, Vol. I. SACCAR, Gaborone, Botswana.
Bekunda, M.A., Bationo, A., Ssali, H., 1997. Soil fertility management in Africa: a review of selected
research trials. In: Buresh, R.A., Sanchez, P.A., Calhoun, F. (Eds.), Replenishing Soil Fertility in
Africa, SSSA Special Bulletin No. 51, Ch. 4. Soil Society of America and American Society of Agronomy, Madison, WI, pp. 63±79.
Chisi, M., Anandajayasekeram, P., Martella, D., Ahmed, M., Mwape, M., 1997. Impact Assessment of
Sorghum Research in Zambia, 1983±2010. SACCAR, Gaborone, Botswana.
Crosson, P., Anderson, N. 1994. Achieving a sustainable agricultural system in Sub-Sahara Africa, Paper
No. 7, Post-UNCED Series, Building Blocks for Africa 2025. World Bank, Washington, D.C.
FAO (Food and Agriculture Organization of the United Nations), 1998. Agriculture database. FAO
Committee on Commodity Problems, Rome, Italy.
FAO and ICRISAT (International Crops Research Institute in Semi-Arid Tropics), 1996. The World Sorghum Economies: Facts, Trends and Outlook. FAO, Rome, Italy and ICRISAT, Andhra Pradesh, India.
ICRISAT (International Crops Research Institute for the Semiarid Tropics), 1987. ICRISAT Sahelian
Centre Annual Report, Niamey, Niger.
ICRISAT (International Crops Research Institute for the Semiarid Tropics), 1988. Annual Report, Niamey, Niger.
Johnson, J.J., 1987. The 1985 on-farm test results of sorghum varieties: comparison of S-35 with local
varieties over two years and 165 sites. In: Menyonga, J.M., Bezuneh, T., Youdeowei, A. (Eds.), Food
Grain Production in Semiarid Africa, pp. 651±683. Fourth Dimension Printing Services, Essex, UK.
Kamuanga, M., Fobasso, M., 1992. Role of Farmers in the Evaluation of an Improved Variety: The Case
of S-35 Sorghum in Northern Cameroon, unpublished mimeo. IAR, Maroua, Cameroon.
Miller, F.R., Cisse, L., Hanak-Freud, E., Sanders, J.H., 1996. Report of the Center Commissioned
External Review (CCER) of the ICRISAT Western and Central African Region. ICRISAT, Andhra
Pradesh, India.
Miller, F.R., Kebede, Y., 1984. Genetic contributions to yield gains in sorghum, 1950 to 1980. In: Fehr,
W.R. (Ed.), Genetic Contributions to Yield Gains of Five Major Crop Plants. Crop Science Society of
America and American Society of Agronomy, CSSA Special Publication No. 7, Ch. 1, Madison, WI,
pp. 1±14.
Mokwunye, U., Hammond, L.L., 1992. Myths and science of fertilizer used in the tropics. In: Myths and
Science of Soils of the Tropics, Soil Science Society of America, and American Society of Agronomy,
SSSA Special Publication No. 29, Ch. 7, Madison, WI.
Nichola, T., 1994. Adoption of Hybrid Sorghum at the Farm and Regional Levels in the Gezira Irrigated
Scheme of the Sudan, unpublished PhD dissertation. Purdue University, Department of Agricultural
Economics, West Lafayette, IN.
Nichola, T., Sanders, J.H., 1996. A probit analysis of the determinants of adoption when inputs are
rationed: the Gezira experience with hybrid sorghum. Science, Technology, and Development 14 (3),
107±119.
Paarlberg, D., 1988. Toward a Well-Fed World. Iowa State University Press, Ames, IA.
QuinÄones, M.A., Borlaug, N.E., Dowswell, C.R., 1997. A fertilizer based green revolution for Africa. In:
Buresh, R.A., Sanchez, P.A., Calhoun, F. (Eds.), Replenishing Soil Fertility in Africa, SSSA Special
Bulletin No. 51, Ch. 4, Soil Society of America and American Society of Agronomy, Madison, WI.
Rohrbach, D.D., 1996. Background: Impact of Research in Southern and Eastern Africa (unpublished
mimeo). ICRISAT Southern and Eastern Africa Region, Bulawayo, Zimbabwe.
Sanders, J.H., 1994. Economic impact of the Commodity Research Networks of SAFGRAD. In: Sanders,
J.H., Bezuneh, T., Schroeder, A.C., Assessment of the SAFGRAD Commodity Networks. USAID/
AFR, OAU/STR-SAFGRAD, Ouagadougou, Burkina Faso, pp. 1±27.
M.M. Ahmed et al. / Agricultural Systems 64 (2000) 55±65
65
Sanders, J.H., Nagy, J.H., Ramaswamy, S., 1990. Developing new agricultural technologies for the
Sahelian countries: the Burkina Faso case. Agricultural Systems 9 (2), 97±112.
Sanders, J.H., Shapiro, B.I., Ramaswamy, S., 1996. The Economics of Agricultural Technology in Semiarid Sub-Saharan Africa. Johns Hopkins University Press, Baltimore, MD.
Shapiro, B.I., Sanders, J.H., 1998. Fertilizer use in semiarid West Africa: pro®tability and supporting
policy. Agricultural Systems 56 (4), 467±482.
Shapiro, B., Sanders, J.H., Reddy, K.C., Baker, T.G., 1993. Evaluating and adapting new technologies in
a high risk agricultural system Ð Niger. Agricultural Systems 42, 153±170.
USAID/FEWS (Famine Early Warning System), 1994. Bulletin SH-11, Nov. 30, USAID/FEWS Project,
Tulane/Pragma Group, Arlington, VA.
World Bank, 1992. Development and the Environment, World Bank Development Report. World Bank,
Washington, DC.
Yapi, A.M., Debrah, K.S., Gandaoua, G., Njomah, C., 1997. Impact of Germplasm Research Spillovers:
The Case of Sorghum Variety S-35 in Cameroon and Chad. ICRISAT Sahelian Center, Bamako, Mali.