Mutegi-Murori, R: Towards Identifying the Physiological and

breeding
materials for drought tolerance;
5) To screen the CIAT mapping population with simple sequence repeats (SSR)
and expressed simple sequence repeat (ESSR) markers for linkage analysis.
Thirty one putative drought-tolerant and drought-susceptible African cassava
germplasm accessions from IITA were micro-propagated using direct and in-direct
techniques, at Kenya Agricultural Research Institute (KARI), Nairobi, Kenya. In
direct micro-propagation, plantlets were hardened using vermiculite and multiplied
through nodal cuttings. In in-direct micro-propagation, plantlets were first multiplied
through sub-culturing and later hardened. The direct micro-propagation method had a
higher multiplication rate. The number of plantlets obtained in 7 months using the
direct method were 1173 as compared to 722 attained using the in-direct micropropagation.
Rapid micro-propagation through nodal cuttings was cheaper in terms of
consumables and an effective alternative to enhance rates of multiplication, over the
in-direct method and the more conventional technique like the use of stem cuttings.
Agronomic and morphological evaluation of contrasting African cassava germplasm
accessions was carried out in water-stressed and well-watered environments at 5 time
points. The trial was conducted at the experimental field of KARI, Kiboko Research
Station in Makindu, Eastern Kenya, a site characterized by Acri-orthic Ferralsol soil.
Analysis of variance was performed using the agronomic and morphological data, and
broad sense heritability was estimated.
In general, significant differences were observed among the accessions, suggesting a
strong genetic basis for the phenotypic variation observed. Variation was also notable
in water-stressed and well-watered environments for a majority of traits evaluated.
This was due to the artificial water applied since, during the trial period, there was
hardly any rainfall.
At harvest, leaf length and width of certain accessions at the water-stressed site
approached that of the well-watered treatment. On average, the estimated mean
percentage leaf retention was high in the well-watered treatment. However, leaf
retention in some of the accessions assessed was almost the same in both treatments.
These accessions tended to produce higher yields. Thus, it may be desirable to select
for higher leaf retention when developing varieties adapted to dry areas. The range of
yields under stress was from 3.3 to 36.7 kg/m2, whereas, under the well-watered
treatment, it was smaller: 28.3 to 53.3 kg/m2. Differences among accessions in yield
and overall above-ground fresh biomass showed that these are important primary
traits to phenotype germplasm under favorable and water-stress conditions. In
addition, 4 accessions G26, G11, G8 and G31 were more tolerant than the rest of the
genotypes evaluated, calling for further research and their involvement in agricultural
experimentation under drought-prone conditions.
Considering relationships between traits, genotype ability for both accumulation of
above-ground fresh biomass and to partition carbon into roots (harvest index) were
among the traits most correlated with root yield. However, accessions, environment
and the interaction of both influenced the traits strongly. Thus, it is important that
agro-morphological field trials be conducted in several locations for several seasons
for effective evaluation of their influences on traits that might be relevant for
phenotypically assessing drought tolerance.
Unlike for agro-morphological traits, where a drought-tolerant accession could be
identified from the yield, which is of primary concern, selection of an outstanding
accession in a water-limited environment using metabolic traits was not achieved in
this study. Maybe the well-watered plants experienced an incipient stress due to low
humidity, symptomless diseases or nutrient deficiency, which was not ascertained in
this study. Despite this, significant differences were observed between the waterstressed
and well-watered treatments for the traits evaluated, except protein and
amylose content. Performance in individuals was variable, although insignificant
differences were observed between the different stress phases.
Changes in sugar concentration have a role in the drought-tolerance of the accessions
evaluated, although, their relative contribution to drought stress could not be
determined from the available data. Further work is imperative to identify and
quantify sugar concentrations in relation to osmotic adjustment in these accessions. In
addition, further research to determine the time course of ABA accumulation, as a
cassava plant goes from its young stage to aging, is required so as to know the stage at
which ABA data are most informative.
Two genetic linkage maps were constructed using a South American mapping
population of 228 individuals derived from a cross between a drought-tolerant and a
drought-susceptible parent. A set of 377 simple sequence repeats (SSR) and expressed
simple sequence repeats (ESSR's) were utilized for the initial polymorphism
screening. Differences in map size, interval, number and mean distance between
markers were apparent between the two maps. The female map had 14 linkage groups
as compared to 13 in the male map. Twenty seven allelic bridges were noticeable
between the two maps. In addition, 25 markers showed collineality with other
available cassava maps. Forty six markers, whose map distances had not been
determined previously, were mapped in this study. These maps form an important
platform upon which to characterize the genetic basis of drought tolerance in cassava.
Continued addition of more markers in these maps will refine the utility of the
resource for future cassava breeding efforts.
In conclusion, four African cassava accessions apparently have the ability to
withstand severe drought. However, a majority of the accessions evaluated gave poor
response in adaptability to water-limited conditions. This suggests that further agroecologically
based research is required on these materials, since they represent diverse
improved accessions from IITA breeding activities. This, coupled with markerassisted
genetic analysis, would be an appropriate approach for the identification of
drought-tolerant accessions.