| dc.description.abstract | The future utility value of seeds of Euphorbia, Kalanchoe and Edithcolia pose
challenges as none has been investigated for storage longevity under seed bank
conditions yet their wild populations are threatened by over-exploitation During
storage, seed moisture content is one of the factors known to affect seed longevity in
general, but no study has quantified the effect of moisture content and other factors
implied in literature. Therefore in this study, these challenges are addressed through
the objectives; I) to determine comparative longevity in seeds from single populations
of 24 species from three succulent genera, 2) to determine seed -longevity differences
between three populations of four species within the genus Euphorbia, 3) to quantify
moisture relations of longevity _in six Euphorbia species, and 4) to determine how
various factors such as moisture sorption properties, oil content, climate and seed
structure affect longevity in at least five Euphorbia species. Seed-lots collected from
diverse field sites in Kenya and representative for species were dried in a Gene bank
dry room at 20 degrees celcius -(0C) and 15% relative humidity (rh) prior to subsamples
being tested for comparative longevity at 45°C and 60% rh using a defined
protocol. A viability model was used to determine moisture effects on longevity
following controlled ageing tests for seed sub-samples held at six moisture contents
(range 4 - 9%) at 45°C. Water sorption properties were assessed in an automated
IGAsorp at 45°C and oil content determined using supercritical carbon dioxide
extraction performed by a fat analyser. Seed structure was assessed after longitudinal
sections of seeds were photographed using a colour Carl Zeiss camera and area
occupied by the embryo and endosperm traced thus enabling calculation of the
respective seed area. Dry weights were assessed gravimetrically after seeds were dried
for 17 hours at 103°C according to the International Seed Testing Association (1STA)
procedure for oily seeds. Using.geographical information system, climate data for six
months post seed dispersal were derived from Wordclim against latitude and
longitude data taken at the point of seed collection. Results revealed greater intragenus
differences for species. from the genus Euphorbia compared with species from
the genus Kalanchoe. Edithcolia grandis indicated longevity between these two other
genera. Contrary to the supposition of the viability model, the standard deviation of
the frequency distribution of seed deaths in time (0) varied for some seed-lots
collected from different years and / or populations. Determined species viability
constants Cw, KEand Crh did not show sensitivity to species life-forms. Further, this
study revealed that the number of water binding sites, seed size, embryo size, seed dry
weight, oil content and climate do not correlate with longevity for these species.
However, the ratio of seed embryo to seed area was significantly related to seed
longevity. It is concluded that longevity differences in these species are attributed to
the initial quality of seed-lots (Ki) and the frequency distribution of seed deaths in
time (0). In view of reasonable seed longevity predicted in four species for which
viability constants were derived, re-testing in seed banks should be relaxed from the
typical 10 years to 15-20 years for seed-lots with high initial viability and probably
every five years for accessions with lower initial viability. This study recommend that
further research towards developing viability constants of economically important
species is vital to enable seed longevity prediction for species found in different agroecological
zones in Kenya Nevertheless, ambient seed storage is not recommended
for seeds stored under field conditions similar to those for Kisumu or Nairobi
investigated in this study. | en_US |
