This preview shows page 1. Sign up to view the full content.
Unformatted text preview: as calculated in 4year
increments (see Fig. 2). In a future paper, time courses
for individual accessions will be analysed separately
to provide information on intraspeciﬁc and harvest
year variability.
A second set of seed ageing data was available
through an experiment initiated in 1977 by Dr Phillip
Stanwood, which compares the longevity of 42 species
stored at 5, 2 18 and 2 1968C [data for cryogenic
storage are reported in Walters et al. (2004) and data
for 2 188C are not presented]. Unlike the NPGS
collection described above, seeds used in this
experiment were commercially available varieties
donated by seed companies or breeders. The
experiment included 1–21 accessions per species
(median ¼ 3 accessions per species), harvested in 1977
or 1978 (see Table 2). Seeds, with water contents
ranging from 3.5 to 9.5% (depending on species), were
stored in envelopes, plastic vials or cans at a constant
temperature (58C), and germination was assessed 3 periodically. As with the NPGS dataset, germination
data were averaged among accessions within a species
for each monitoring time.
Longevity parameters for each species in both
datasets were calculated by solving for the coefﬁcients
of the Avrami equation using a least squares linear
ﬁt of the double logarithmic expression:
ln½lnðN 0 =N Þ ¼ n½lnðtimeÞ þ y0 , where n is the slope
and y0 is the yintercept. So that the double ln value of
N0/N could be calculated, the constant N0 was
assigned as 0.5 þ the maximum germination percentage of the averaged time course (usually the average
initial germination percentage). Values for N were the
average germination percentage for the time (t) in
years (solid circles in Fig. 2). The coefﬁcient f and the
exponential factor n of the Avrami equation were
calculated from the coefﬁcients of the regression line
ðf ¼ eð2y0 =nÞ and n ¼ slope). The time for germination
to drop to 50% (P50) was determined by interpolating
(if the ﬁnal average germination percentage was
, 50%) or extrapolating (if the ﬁnal average germination percentage was . 50%) the Avrami equation,
using the coefﬁcients calculated from ﬁtted time
courses. For the few species with initial germination
percentages , 70%, P50 was calculated as the time for
germination to decrease to half the initial value.
Values of P50 for species represented in this study
were also calculated using the Ellis viability equations
if coefﬁcients were reported in the literature. The
water content used for this calculation was the
average water content for the species in the NPGS
collection; the storage temperature was assigned as
58C; and the initial germination was assigned as the
maximum germination percentage of the averaged
time course of the NPGS collection (i.e. N0 2 0.5).
Longevity in the genebank was also compared to
longevity in the soil. Results from Duvel’s buried seed
experiment (Toole and Brown, 1946) were used as a
measure of seed persistence in the soil, expressed as
the s...
View
Full
Document
 Spring '13
 sf

Click to edit the document details