Abstract:

Groundnut production in Uganda is rain-fed. As such, drought is a major challenge, especially when
it occurs in combination with biotic stresses such as groundnut rosette virus disease. The objectives
of this study were; (1) to assess smallholder farmers’ perceptions of drought in groundnut production,
(2) to determine farmers’ varietal selection criteria in groundnut, (3) to assess genetic diversity of the
Ugandan groundnut collection, and (4) to determine combining ability and gene action controlling
drought tolerance traits and physical quality traits under different water regimes, thereby identifying
promising parents and crosses for cultivar development. A participatory rural appraisal revealed that
groundnut was grown mainly by smallholder farmers under both mono and mixed cropping systems
depending on the region. Most of the farmers (>77%) from all regions indicated that the major impact
of drought on their farming activity was low yields. The participatory variety selection study showed
that farmers have effective methods for selecting varieties based on preferred traits. However, the
farmers’ selection methods need to be complemented by breeder methods. Results of genetic
diversity studies showed that 50% of the 2986 SNPs were polymorphic with a mean PIC of 0.15 for
all populations combined, implying low allelic diversity in the groundnut collection. A high level of
genetic relatedness was observed at population level, as individuals from all assigned populations
aggregated into three clusters based on dissimilarity analysis under principal component analysis
(PCA). However, a high level of genetic differentiation existed within individuals rather than among
populations based on analysis of molecular variance (AMOVA) and cluster analysis. Based on
structure analysis, three gene clusters were found underlying genepools for the groundnut populations
used in this study.
The study on genetic analysis for drought tolerance traits revealed that both additive and non-additive
gene actions controlled the inheritance of the traits in groundnut. The general combining ability effects
(GCA) were more important than the specific combining ability (SCA) for the SPAD chlorophyll meter
readings (SCMR) and harvest index (HI). This observation was similar under both drought stressed
and non-drought stressed environments and for all the groundnut botanical types. The SCA effects
for traits such as drought tolerance index based on pod yield, and biomass were significant. Given
such unlikely predominance for either additive or non-additive gene effects, the breeding procedure
would require selection to be conducted in the later generations in order to harness both gene effects.
In addition, significant associations between drought stressed and non-drought stressed
environments were observed for SCMR, harvest index, pod yield, number of pods, leaf length, and
leaf width. Therefore, since SCMR and HI are relatively stable and less influenced by environmental
conditions, they can thus be measured under both drought stressed and non-drought stressed
conditions. However, selection based on these traits can be done more effectively under non-drought
iii
Overall, this study has highlighted the mismatch between the release of new varieties and adoption
rates despite the fact that farmers are aware of the new improved materials. Therefore, there is a
need for various stakeholders to strengthen their role in the transfer of research outputs to end users
(farmers). Among the key recommendations, breeders, extension staff and farmers need to work
together to make the technology that has been developed to have relevancy. This study also
confirmed that the groundnut genepool in Uganda has a low genetic base. However, from the
information obtained from the study, future hybridizations avoiding parents sharing a common
ancestry can be done or diversity can be improved by enriching the collection with some introductions
from geographically different areas. Identification of genotypes with drought tolerance and/or physical
quality traits helps in the quest to develop new varieties, which are well adapted, and with farmer
preferred traits. The observation that additive gene effects are involved in governing drought and
physical traits, besides, positive correlations between these traits indicates that drought tolerant
materials can be improved by accumulating additive genes using methods that favour general
combining ability such as recurrent selection breeding procedures. Where possible, molecular
techniques such as genomic selection can be applied to shorten the entire process.
stressed environment. In addition, the best breeding approach would be the one that emphasizes
additive gene action such as recurrent selection or simple selection by pedigree breeding.
The study on gene action governing physical quality traits showed the significance of GCA, SCA and
maternal effects for various traits. The genetic analysis indicated that expression of the majority of
the physical quality traits was regulated by additive gene action suggesting possibility of early
generation selection, while non-additive gene action also played an equally important role in the
control of the physical seed traits. However, the gene control varied with botanical varieties for some
traits. Among the Virginia botanical variety types, non-additive gene action was predominant for all
the traits regardless of the test environment implying that botanical classification has a bearing on the
parental selection in improving these traits in this subgroup of groundnuts. In cases where both
additive as well as non-additive gene effects controlled the expression of the traits, the breeding
strategy has to delay selection to later generations or should involve recurrent selection that favours
both additive and non-additive gene action. Significant maternal effects for some traits signify the role
of the maternal parent in the expression of quality traits and importance of parental selection in quality
breeding. Therefore, one of the parents to be used in hybridization should have large pods and seeds
for obtaining better segregants. Low but positive or negative association of drought tolerant traits with
physical quality traits, regardless of stress regime, can be exploited for improving seed yield without
compromising seed size. Thus, the selection of high pod yield potential under no-stress or small
reduction in pod yield under drought stress should guarantee high seed size.