Kilimokwanza.org 2026- R2

1. Introduction: The Global and Regional Imperative

The final decade of the 20th century marked a pivotal transition in the global agricultural research architecture, particularly regarding the role of root and tuber crops in the developing world. Between 1990 and 1998, the International Potato Center (CIP), operating within the Consultative Group on International Agricultural Research (CGIAR), initiated what has been retrospectively termed the “science phase” of its engagement in Sub-Saharan Africa. This period was characterized by a fundamental shift from the ad-hoc distribution of germplasm to a rigorous, data-driven program of physiological modeling, multi-locational evaluation, and participatory varietal selection intended to modernize the potato (Solanum tuberosum) and sweetpotato (Ipomoea batatas) sectors in Tanzania.¹

During this era, a comprehensive study published by FAO and CIP economists underscored a historical underestimation of root and tuber crops, revealing that potato production in the developing world had increased at a faster rate over the preceding thirty years than any other food crop except wheat.³ The data suggested that by the turn of the century, developing countries would produce more than a third of the world’s potatoes. However, this expansion was driven largely by increases in cultivated area rather than intensification or yield improvement. In Tanzania, despite a conducive agro-ecological environment in the Southern Highlands and Northern Zone, national potato yields stagnated at approximately 5 to 7 tons per hectare, drastically below the crop’s genetic potential of over 40 tons per hectare.⁴

The “science phase” was therefore launched as a deliberate intervention to close this yield gap. It operated under the premise that science, with comparatively small investments in research, could raise average yields to at least 30 tons per hectare through the introduction of adapted, disease-resistant, and early-maturing genotypes.³ This report provides an exhaustive analysis of this period, detailing the agronomic trials, breeding strategies, and institutional collaborations that introduced clones such as Kikondo (CIP 720050), Tigoni (CIP 381381.13), and Asante (CIP 381381.20) to the Tanzanian farming system. It explores how researchers navigated the biological pressures of late blight (Phytophthora infestans) and bacterial wilt (Ralstonia solanacearum) while attempting to meet the socio-economic demand for early maturity—a critical trait for smallholder farmers navigating the erratic rainfall patterns of the East African highlands.

2. Institutional Framework and the “Science Phase” Paradigm

2.1 Defining the “Science Phase”

The term “science phase” refers to a strategic reorientation within CIP and its national partners during the early 1990s. Prior to this period, international cooperation often focused on the broad dissemination of global germplasm with limited local adaptation data. The 1990s, however, introduced a regime of “impact-oriented research,” where breeding objectives were strictly aligned with rigorous environmental modeling and socio-economic feedback loops.²

In 1992, CIP refined its global strategy through extensive partnerships with National Agricultural Research Systems (NARS), conducting a priority-setting exercise to identify specific development bottlenecks. This exercise utilized quantitative scoring to assess the potential impact of research projects, leading to a consolidation of efforts into six core program thrusts, including Production Systems, Germplasm Management, and Disease Management.⁵ For Tanzania, this meant that germplasm introduction was no longer a passive receipt of seeds from Lima or Nairobi but an active scientific inquiry involving “Genotype by Environment” (GxE) interaction studies, physiological growth modeling, and onsite selection.

2.2 The Tanzanian Research Nodes: Uyole and Tengeru

The execution of this mandate in Tanzania relied on two primary research institutions that served as the operational hubs for CIP’s regional activities.

Uyole Agricultural Centre (UAC): Located in the Southern Highlands (Mbeya region), UAC served as the primary testing ground for the country’s most productive potato belt. The Southern Highlands, accounting for approximately 90% of national production, offered a diverse range of altitudes from 1,800 to 2,800 meters above sea level (masl), providing an ideal laboratory for testing chilling tolerance, late blight resistance, and yield stability.⁴

HORTI-Tengeru: Situated in the Northern Zone near Arusha, HORTI-Tengeru focused on the volcanic agro-ecologies of Kilimanjaro and Arusha. This institution played a pivotal role in seed multiplication and the maintenance of germplasm for the northern corridor, linking Tanzania to the potato economies of Kenya and Uganda. The station was critical for evaluating early-maturing varieties suitable for the bimodal rainfall patterns of the north.⁷

2.3 Regional Networks: PRAPACE and SARRNET

The scientific work in Tanzania was not isolated but integrated into broader African networks. The PRAPACE network (Programme Régional de l’Amélioration de la Culture de la Pomme de Terre et de la Patate Douce en Afrique Centrale et de l’Est) and SARRNET (Southern Africa Root Crops Research Network) facilitated the movement of elite clones between countries. This cross-border collaboration was instrumental in the late 1990s, allowing Tanzania to test and eventually adopt varieties like Tigoni and Asante that had been simultaneously developed and released in Kenya.⁹ These networks provided the mechanism for “spillover effects,” where research investments in one national program could benefit neighbors, a core tenet of the CGIAR’s efficiency model during the donor-fatigued 1990s.⁵

3. Physiological Modeling and Yield Gap Analysis

A distinguishing feature of the “science phase” was the application of crop growth models to understand the physiological limits of potato production in tropical highlands. Researchers utilized tools such as the LINTUL-POTATO model to simulate potential yields based on solar radiation, temperature, and crop physiology, independent of biotic constraints.¹¹

3.1 The LINTUL-POTATO Simulations

Modeling conducted during this period, including trials in analogous environments in Burundi and the Tanzanian highlands, sought to quantify the “yield gap”—the difference between water-limited potential yield and actual farm yield. The LINTUL model calculates daily dry matter increment as a function of photosynthetically active radiation (PAR) interception and light use efficiency (LUE).

The simulations revealed that in the cool highlands (e.g., Kitulo, 2800 masl), the potential yield for well-adapted CIP clones exceeded 60 tons per hectare. However, the model also highlighted a sharp physiological penalty in warmer, lower-altitude zones (800–1200 masl). High temperatures were found to accelerate crop senescence, reduce the harvest index (the ratio of tuber weight to total plant weight), and increase respiration losses. The simulation confirmed that “early maturity” was not merely a market preference but a physiological necessity for warmer zones; varieties needed to bulk rapidly before high temperatures or terminal drought curtailed the growing season.¹¹

3.2 The Reality of Farm Yields

Against this high theoretical potential, actual farmer yields in Tanzania during the 1990–1998 period averaged between 5 and 7 tons per hectare.⁴ This massive discrepancy was attributed to a complex of biotic and abiotic constraints:

• Genetic Erosion: Farmers relied on degenerate seed of old European varieties like Sasamua and Baraka, which had lost resistance to local disease strains.
• Late Blight Pressure: The continuous cropping systems in the highlands allowed Phytophthora infestans to survive year-round, creating a high-inoculum environment that overwhelmed susceptible varieties.
• Seed Degeneration: The accumulation of viruses (PLRV, PVY) in clonally propagated material reduced vigor significantly over successive generations.

The scientific mandate, therefore, was to introduce germplasm that could bridge this gap by combining the high harvest index predicted by models like LINTUL with durable resistance to the biological threats prevalent in the field.

4. Evaluation of Solanum tuberosum Clones (1990–1998)

The core activity of the 1990–1998 period was the systematic evaluation of advanced breeding lines introduced from CIP headquarters in Lima, Peru. These populations were bred specifically for the “lowland tropics” and “highland tropics,” designated as Populations A and B in CIP’s breeding strategy.¹² The objective was to identify clones that combined early maturity (90–110 days) with horizontal resistance to late blight.

4.1 Multi-Locational Yield Trials

Extensive trials were conducted by UAC across an altitudinal gradient, testing clones at Uyole (1800 masl), Igeri (2300 masl), and Kitulo (2800 masl). These sites represented the varying agro-ecological zones of the Southern Highlands. The performance data generated during these trials provided the empirical basis for variety release recommendations.

The following table synthesizes yield data from UAC reports covering the mid-1990s, comparing advanced CIP clones against the standard local check, Sasamua.

Table 1: Comparative Yield Performance (t/ha) of Advanced CIP Clones vs. Local Check in Southern Highlands (Mean of Two Years)

Genotype / Clone ID	Uyole (1800 masl)	Kitulo (2800 masl)	Mean Yield (t/ha)	Yield Gain vs. Check (%)
CIP 382169.4	13	43	28.0	+115%
CIP 382171.3	26	35	30.5	+135%
CIP 382150.2	23	27	25.0	+92%
CIP 382135.2	18	24	21.0	+61%
CIP 382124.2	13	32	22.5	+73%
CIP 382174.6	9	37	23.0	+77%
Sasamua (Local Check)	10	23	16.5	–

Source: Derived from Uyole Agricultural Centre Progress Reports.⁴

Analysis of Performance:

• High-Altitude Adaptation: Clone CIP 382169.4 demonstrated exceptional adaptation to the highest altitude site (Kitulo), achieving a yield of 43 t/ha. This performance indicated a strong tolerance to frost and an ability to utilize the long, cool growing season effectively.
• Stability and Plasticity: Clone CIP 382171.3 emerged as the most stable high-yielder, performing best at the mid-altitude Uyole site (26 t/ha) while maintaining very high yields at Kitulo. This plasticity—the ability to perform well across diverse environments—was a key selection criterion for resource-poor farmers who farm across heterogeneous micro-climates.
• Genetic Gain: The top performing clones consistently yielded double the tonnage of the local variety Sasamua. This confirmed that the yield constraints were not solely agronomic (fertility/water) but heavily genetic; the local varieties simply lacked the physiological efficiency and disease resistance of the new germplasm.⁴

4.2 Pathology Screening: Late Blight and Bacterial Wilt

The biological environment of the 1990s was dominated by the spread of aggressive A2 mating types of Phytophthora infestans. CIP’s breeding strategy shifted from vertical resistance (single R-genes), which was easily overcome by the pathogen, to horizontal resistance (polygenic), which slowed infection rates.³

Pathology trials conducted at UAC evaluated clones on a 0–9 scale (where 0 is immune and 9 is highly susceptible).

Table 2: Late Blight (LB) Resistance Ratings and Yield Correlations (Mean of Six Years)

Variety / Clone	LB Rating (0-9)	Mean Yield (t/ha)	Resistance Class
CIP 720050 (Kikondo)	2.9	23.3	Moderately Resistant
Tana (Roslin Tana)	2.1	23.3	Resistant
K-110-C(8)	2.5	15.0	Resistant
Baraka	3.1	14.6	Moderately Susceptible
Sasamua	3.4	19.0	Susceptible
Kennebec	4.4	17.3	Highly Susceptible

Source: UAC multilocational trial data.⁴

The data highlights the superiority of CIP 720050, later released as Kikondo. With an LB rating of 2.9, it offered a significant improvement over the popular Sasamua (3.4) and Kennebec (4.4). Crucially, Kikondo combined this resistance with high yield potential, unlike clone K-110-C(8), which was more resistant (2.5) but yielded poorly (15.0 t/ha). This balance of traits led to Kikondo becoming the dominant variety in Tanzania, eventually occupying more than 50% of the potato area by the post-2000 period.¹³

4.3 The Introduction of Tigoni and Asante

Toward the end of the reporting period (1998), the collaborative synergy between Tanzania and Kenya, facilitated by CIP, bore significant fruit. Two varieties developed at the Tigoni Potato Research Centre in Kenya using CIP germplasm were introduced for testing in Tanzania: Tigoni and Asante.

• Tigoni (CIP 381381.13): A white-skinned, tall, vigorous variety suited for processing (chips/fries). It demonstrated high dry matter content and resistance to late blight. However, farmers noted it had a very short dormancy period, leading to rapid sprouting and greening if not marketed immediately. Agronomically, it was an early-to-medium maturer (3–4 months) with yields ranging from 35–45 t/ha under optimal conditions.¹⁴
• Asante (CIP 381381.20): A red-skinned variety with round tubers, highly preferred for the fresh market due to consumer preference for red potatoes in Tanzania. Like Tigoni, it was bred for early maturity (110 days) and high yields (35–45 t/ha). Asante was particularly valued for its moderate tolerance to both Late Blight and PLRV (Potato Leaf Roll Virus).¹⁶

These introductions marked a diversification of the genetic base, offering farmers specific options for processing (white skin) versus table consumption (red skin), and reinforcing the “early maturity” trait required for double-cropping systems.

5. Sweetpotato Evaluation and the “Early Maturity” Imperative

While “potato” often refers to Solanum tuberosum, CIP’s mandate expansion in 1986 to include sweetpotato (Ipomoea batatas) led to a parallel and equally rigorous evaluation program in Tanzania during the 1990s. This component of the “science phase” was driven by the urgent food security needs of the semi-arid lowlands and the Lake Zone.

5.1 Participatory Varietal Selection (PVS)

Unlike the breeder-centric potato trials, sweetpotato research employed Participatory Varietal Selection (PVS). Surveys conducted between 1990 and 1995 across major production zones revealed that early maturity was the second most prioritized trait by farmers (cited by 88% of respondents), ranking just below high yield.¹⁸

Farmers defined “early maturity” as the ability to scavenge a harvest within 3 to 4 months of planting. This trait was essential for:

1. Drought Escape: Avoiding the terminal drought of the short rainy season.
2. The Hungry Gap: Providing food availability during the critical months before the maize harvest.
3. Piecemal Harvesting: The ability to harvest individual roots progressively without destroying the plant.

5.2 Release of Elite Clones

Based on these criteria, CIP and the Tanzanian national program released several landmark varieties during this period:

Table 3: Characteristics of CIP/National Sweetpotato Releases (1990–1998)

Variety Name	Clone Code / Origin	Key Agronomic Attributes	Socio-Economic Fit
SPN/O (Simama)	Tanzania Landrace / CIP Selection	Moderate yield stability, drought tolerant, wide adaptability.	Became the benchmark variety across East Africa (known as Kemb 10 in Kenya).
Sinia	Local Selection	High yield, good root shape, acceptable taste.	Preferred for commercial fresh root sales in urban markets.
Mavuno	SP/93/2	Early maturity, dual-purpose (roots + leaf vegetable).	Critical for food security; leaves used as relish.
Juhudi	SP/93/34	Weed smothering ability, high root yield.	Reduces labor requirement for weeding (gender benefit).
Vumilia	SP/93/23	“Vumilia” means “tolerate”; extreme drought tolerance.	Targeted for semi-arid agro-ecologies.

Source: Kapinga et al. (1995); Rees et al. (1998).¹⁹

Trials in 1996/97 confirmed the superiority of these varieties. For instance, SPN/O yielded 23.57 t/ha at Uyole, significantly outperforming the trial mean of 14.12 t/ha.¹⁹ Post-harvest studies further validated their suitability, showing moderate weight loss during storage, a crucial factor for transport to distant urban markets like Dar es Salaam.¹⁹

6. Seed Systems and Technology Transfer

A major finding of the 1992 CIP priority-setting exercise was that superior genetics were failing to reach farmers due to the absence of a formal seed system.⁵ The “science phase” thus extended beyond breeding to the science of seed physiology and multiplication.

6.1 Diffuse Light Storage (DLS)

During the 1990s, CIP introduced and validated the use of Diffuse Light Storage (DLS) technology in Tanzania. Traditional practice involved storing seed tubers in dark, unventilated heaps, leading to etiolated (long, weak) sprouts and rapid physiological aging. DLS technology utilized indirect natural light and ventilation to suppress apical dominance.

Research conducted at Tigoni and replicated in Tanzanian on-farm trials demonstrated that DLS storage resulted in:

• Green, robust sprouts: Short, sturdy sprouts that did not break during transport/planting.
• Multiple sprouting: Breaking apical dominance to encourage multiple stems, which correlates directly with higher tuber numbers per plant.
• Extended shelf-life: Allowing farmers to store seed for 5–7 months, bridging the gap between seasons effectively.²⁰

6.2 The “Informal” Formal System

With the formal seed certification agency (TOSCA, later TOSCI) still in its nascent stages regarding potato standards, the research centers themselves (UAC and HORTI-Tengeru) acted as the primary engines of seed dissemination. This “project-mode” distribution, funded by donors like the Swiss Development Cooperation (SDC), allowed for the rapid deployment of varieties like Kikondo and Asante directly to farmer groups and NGOs, bypassing the stalled formal commercial sector. This period saw the establishment of in vitro tissue culture units at research stations, enabling the cleaning of viral-infected stocks and the production of disease-free minitubers for the first time in Tanzania.²²

7. Socio-Economic Impact and Gender Dynamics

The scientific interventions of the 1990s had profound socio-economic ripple effects, particularly regarding gender roles and market dynamics.

7.1 Gendered Trait Preferences

The sweetpotato trials revealed significant divergence in trait preference between genders. While male farmers and researchers often prioritized total tonnage and market appearance (root shape), female farmers—who were responsible for household food security and processing—prioritized early maturity, cooking quality (starchiness/flouriness), and leaf palatability. The success of varieties like SPN/O and Mavuno was largely due to their alignment with these female-prioritized traits. Mavuno (SP/93/2), for instance, was explicitly valued for its ability to produce abundant edible leaves for relish, a critical nutritional supplement for women and children.¹⁸

7.2 The Economic Cost of Disease

The evaluation programs quantified the economic toll of reliance on old varieties. With Sasamua and Baraka succumbing to late blight, farmers were forced to spray fungicides up to 12 times per season to secure a harvest. The introduction of moderately resistant varieties like Kikondo and Asante allowed farmers to reduce spray frequencies to 2–3 times per season. This reduction in input costs, combined with the yield potential of >20 t/ha, significantly increased the gross margins for smallholder adopters, transforming potato from a subsistence crop to a lucrative cash crop in the Southern Highlands.⁴

8. Challenges and the Climatic Shock of 1997/98

The resilience of the new germplasm was severely tested during the 1997/1998 El Niño event, which brought extreme weather variability to East Africa. In some zones, excessive rains led to catastrophic late blight epidemics, wiping out susceptible local landraces and validating the resistance of clones like Kikondo. Conversely, in the Eastern Zone (Dakawa/Kibaha), the subsequent La Niña drought caused massive crop failures. Sweetpotato trials in 1997/98 saw yields plummet, with some varieties failing to establish entirely. However, SPN/O and Sinia demonstrated superior survival rates compared to other clones, reinforcing the importance of the “drought tolerance” breeding objective.¹⁹

Furthermore, the adoption of Tigoni faced hurdles due to its processing characteristics. While high-yielding, its tendency for rapid sprouting (short dormancy) made it risky for farmers who lacked immediate market access, illustrating the disconnect between “biological success” (yield) and “market success” (storability).¹⁵

9. Conclusion

The period from 1990 to 1998 represented a foundational “science phase” for root and tuber crops in Tanzania. Through the strategic intervention of the International Potato Center and its collaboration with UAC and HORTI-Tengeru, the country moved from a passive recipient of foreign seeds to an active participant in global breeding networks. The rigorous evaluation of physiological traits, disease resistance, and culinary quality led to the identification and release of transformative varieties.

The potato variety Kikondo (CIP 720050) and the sweetpotato variety SPN/O stand as the enduring legacies of this era. Kikondo provided the necessary resistance to late blight to sustain production in the highlands, while SPN/O offered the early maturity and drought tolerance required for food security in the lowlands. The introduction of Tigoni and Asante from Kenya further diversified the genetic landscape, setting the stage for the commercialization of the sector. By grounding germplasm introduction in rigorous science—from LINTUL yield modeling to participatory taste panels—CIP and its Tanzanian partners successfully laid the genetic and institutional groundwork that continues to support the livelihoods of millions of farmers in the region today.

Works cited

1. World science report, 1996 – UNESCO Digital Library, accessed January 6, 2026, https://unesdoc.unesco.org/ark:/48223/pf0000102819
2. Coupled Performance and Change in the Making – WUR eDepot, accessed January 6, 2026, https://edepot.wur.nl/264764
3. CIP in 1995 The International Potato Center Annual Report – CGSpace, accessed January 6, 2026, https://cgspace.cgiar.org/bitstreams/6192c820-94cf-433d-b9d3-d57774bc4de9/download
4. SOUTHERN HIGHLANDS OF TANZANIA, accessed January 6, 2026, http://www.tzonline.org/pdf/progressonpotato.pdf
5. CIP Program Report 1992 | PDF | Subtropics | Potato – Scribd, accessed January 6, 2026, https://www.scribd.com/document/253412617/CIP-Program-Report-1992
6. CHAPTER 1 – BACKGROUND, accessed January 6, 2026, https://www.fao.org/4/x5808e/x5808e08.htm
7. Tanzania Sets Record in Potato Research, to Release Improved …, accessed January 6, 2026, https://www.potatopro.com/zh-hans/news/2018/tanzania-sets-record-potato-research-release-improved-varieties
8. (PDF) International research on vegetable improvement in East and …, accessed January 6, 2026, https://www.researchgate.net/publication/318551216_International_research_on_vegetable_improvement_in_East_and_Southern_Africa_Adoption_impact_and_returns
9. PO2.4.2 – Seed technologies and business models – MEL, accessed January 6, 2026, https://mel.cgiar.org/overview/activity/id/504
10. CV Prof. Patrick Ayiecho Olweny.docx – Maseno university, accessed January 6, 2026, https://safs.maseno.ac.ke/sites/default/files/2021-08/CV%20Prof.%20Patrick%20Ayiecho%20Olweny.docx
11. (PDF) Growth analysis and modelling of CIP potato genotypes for …, accessed January 6, 2026, https://www.researchgate.net/publication/290149463_Growth_analysis_and_modelling_of_CIP_potato_genotypes_for_their_characterization_in_two_contrasting_environments_of_Burundi
12. (PDF) Potato and sweetpotato breeding at the International Potato …, accessed January 6, 2026, https://www.researchgate.net/publication/376647589_Potato_and_sweetpotato_breeding_at_the_international_potato_center_approaches_outcomes_and_the_way_forward
13. CIP Annual Report 2008 | PDF | Potato | Food Security – Scribd, accessed January 6, 2026, https://www.scribd.com/document/253404120/CIP-Annual-Report-2008
14. POTATO PRODUCTION – JICA, accessed January 6, 2026, https://www.jica.go.jp/project/english/kenya/015/materials/c8h0vm0000f7o8cj-att/materials_25.pdf
15. Hard times for potato farmers – Infonet-Biovision, accessed January 6, 2026, https://infonet-biovision.org/file/3590/download?token=TvgoEuUg
16. Stacking three late blight resistance genes from wild species directly …, accessed January 6, 2026, https://pmc.ncbi.nlm.nih.gov/articles/PMC6523587/
17. Changes in Phytochemical Content During Different Growth Stages …, accessed January 6, 2026, https://scispace.com/pdf/changes-in-phytochemical-content-during-different-growth-2v3pimx1eg.pdf
18. Farmer criteria for selection of sweetpotato varieties – GOV.UK, accessed January 6, 2026, https://assets.publishing.service.gov.uk/media/57a08ceeed915d3cfd0016ea/R7520l.pdf
19. Tanzanian sweetpotato varieties and their associated acceptable …, accessed January 6, 2026, https://assets.publishing.service.gov.uk/media/57a08d78ed915d3cfd001a4c/R6507e.pdf
20. Prolonging the shelf-life of seed potato tubers at farm level: Cold …, accessed January 6, 2026, https://www.researchgate.net/profile/Jane-Mbugua-2/publication/326352366_Prolonging_the_shelf-life_of_seed_potato_tubers_at_farm_level_Cold_storage_or_Diffused_Light_Store/links/5b474c2caca272c609389955/Prolonging-the-shelf-life-of-seed-potato-tubers-at-farm-level-Cold-storage-or-Diffused-Light-Store.pdf?origin=scientificContributions
21. (PDF) Yield Performance of Potato Seed Tubers After Storage in a …, accessed January 6, 2026, https://www.researchgate.net/publication/272922775_Yield_Performance_of_Potato_Seed_Tubers_After_Storage_in_a_Diffuse_Light_Store_DLS
22. GENERAL SUMMARY, accessed January 6, 2026, http://library.procurementmonitor.org/backend/files/BUDGET%202001CAPITALS%202.pdf
23. International Potato Center (CIP) – The World Bank, accessed January 6, 2026, https://thedocs.worldbank.org/en/doc/245321620043744244-0240021974/original/WorldBankGroupArchivesFolder1760494.pdf
24. 253412692-CIP-Program-Report-1993-1994.pdf, accessed January 6, 2026, http://cipotato.org/wp-content/uploads/1993/01/253412692-CIP-Program-Report-1993-1994.pdf