Journal of Experimental Botany, Vol. 58, No. 5, pp. 1185–1196, 2007
doi:10.1093/jxb/erl286 Advance Access publication 5 February, 2007
RESEARCH PAPER
Water properties in fern spores: sorption characteristics
relating to water affinity, glassy states, and storage stability
Daniel Ballesteros1 and Christina Walters2,*
1
Banco de Germoplasma, Jardı´Botànic-ICBiBE, Universitat de València, C/Quart, 80, E-46008 València, Spain
2
USDA-ARS National Center for Genetic Resources Preservation, 1111 So. Mason Street, Fort Collins,
CO 80521, USA
Received 2 August 2006; Revised 8 November 2006; Accepted 28 November 2006
Abstract
Ex situ conservation of ferns may be accomplished by
maintaining the viability of stored spores for many years.
Storage conditions that maximize spore longevity can
be inferred from an understanding of the behaviour of
water within fern spores. Water sorption properties
were measured in spores of five homosporeous species
of ferns and compared with properties of pollen, seeds,
and fern leaf tissue. Isotherms were constructed at 5,
25, and 45 C and analysed using different physicochemical models in order to quantify chemical affinity
and heat (enthalpy) of sorption of water in fern spores.
Fern spores hydrate slowly but dry rapidly at ambient
relative humidity. Low Brunauer–Emmet–Teller monolayer values, few water-binding sites according to the
D’Arcy–Watt model, and limited solute–solvent compatibility according to the Flory–Huggins model suggest that fern spores have low affinity for water.
Despite the low water affinity, fern spores demonstrate
relatively high values of sorption enthalpy (DHsorp).
Parameters associated with binding sites and DHsorp
decrease with increasing temperature, suggesting temperature- and hydration-dependent changes in volume
of spore macromolecules. Collectively, these data may
relate to the degree to which cellular structures within
fern spores are stabilized during drying and cooling.
Water sorption properties within fern spores suggest
that storage at subfreezing temperatures will give longevities comparable with those achieved with seeds.
However, the window of optimum water contents
for fern spores is very narrow and much lower than
that measured in seeds, making precise manipulation
of water content imperative for achieving maximum
longevity.
Key words: Ex situ conservation, germplasm, relative
humidity, temperature, water sorption isotherms.
Introduction
Pteridophytes (ferns) are associated with ecosystems that
are particularly sensitive to degradation, and some taxa are
in peril and require strict protection. Ex situ conservation provides an important backup strategy, as pteridophytes are very sensitive to environmental perturbations.
Germplasm banks provide ex situ conservation of many
plants by preserving seeds for decades or centuries
(Gómez-Campo, 2001; Smith et al., 2003; Guerrant et al.,
2004). However, methodologies for long-term preservation of pteridophytes in germplasm banks are not well
established. Maintaining pteridophyte spore viability, genetic
integrity, and capacity for normal growth are key research
objectives to enable their effective ex situ conservation
(Page et al., 1992).
Long-term viability of pteridophyte spores depends on
spore type or taxonomic group (Lloyd and Klekowski,
1970). The storage behaviour of the two types of spores
recognized—green and non-green—is very different. Green
spores are chlorophyllous and usually lose viability within
weeks (e.g. Equisetum sp.) or months (e.g. Osmunda
regalis), although survival for almost a year has been reported for a few genera stored at room temperature (e.g.
Onoclea and Matteuccia) (Lloyd and Klekowski, 1970;
Whittier, 1996; Lebkuecher, 1997) or at liquid nitrogen
* To whom correspondence should be addressed. E-mail: [email protected]
Abbreviations: BET, Brunauer–Emmet–Teller; RH, relative humidity; DHsorp, enthalpy of sorption; DHsorp(T), enthalpy of sorption calculated at a specific
temperature; DHsorp(w), enthalpy of sorption calculated at a specific water content; wc, water content.
ª The Author [2007]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.
For Permissions, please e-mail: [email protected]
1186 Ballesteros and Walters
temperatures (Pence, 2000; Ballesteros et al., 2005).
Longevity of non-green spores under ambient room conditions ranges from a few months (e.g. Gleicheniaceae,
Thyrsopteris elegans) to about a decade (most species),
with some species surviving several decades (e.g. Pellaea
sp., Asplenium serra, Marsilea sp.) (Lloyd and Klekowski,
1970; Dyer, 1979; Page, 1979; Windham et al., 1986;
Lindsay et al., 1992; Page et al., 1992). Though these
studies record survival times, they do not provide quantitative accounts of the effects of storage temperature and
relative humidity (RH) on the loss of spore viability or the
negative effects on gametophyte development and genetic
integrity (Smith and Robinson, 1975; Beri and Bir, 1993;
Camloh, 1999).
Early recommendations for fern spore conservation
suggested using the same dry, low-temperature conditions
used for seeds (Dyer, 1979). Routine methods for conservation of orthodox seeds in germplasm banks are based
on models that recommend dehydration to 562% water
(or equilibration to 15–20% RH) and storage at 5 C or
25 C (Roberts and Ellis, 1989; FAO/IPGRI, 1994;
Walters, 1998, 2004; Gómez-Campo, 2001). However, application of these models to existing data using fern spores
suggests that methods used for seeds may not be effective
for long-term conservation of fern spores. Refrigerated storage at about 5 C maintains high viability for about 1–6
years, an improvement over viability achieved by storage at
room temperature (Smith and Robinson, 1975; Simabukuro
et al., 1998b; Camloh, 1999; Pence, 2000; Quintanilla
et al., 2002; Aragón and Pangua, 2004; Ballesteros et al.,
2004). However, deterioration was faster for some spores
stored in the freezer (25 C) compared with the refrigerator (Simabukuro et al., 1998b; Constantino et al., 2000;
Quintanilla et al., 2002; Aragón and Pangua, 2004;
Ballesteros et al., 2004). Spores of some fern species did
not tolerate initial exposure to 25 C (Simabukuro et al.,
1998b; Quintanilla et al., 2002; Aragón and Pangua, 2004;
Ballesteros et al., 2004). Other results show no differences
in viability of three spore species following 75 months of
storage at 4 C or 20 C (Pence, 2000).
Hydrated storage of fern spores is becoming increasingly recommended because viability can be maintained
for 12–24 months at either room or refrigerated temperatures (Lindsay et al., 1992; Simabukuro et al., 1998b;
Quintanilla et al., 2002; Aragón and Pangua, 2004).
Imbibed spores maintain remarkable tolerances, having
higher survival after 48 h at 70 C than spores dried to
ambient RH (Simpson and Dyer, 1999). Germination of
hydrated spores is prevented by continuous darkness
and, analogous to seeds in secondary dormancy (Villiers,
1974), maintenance of viability is believed to result from
ongoing cellular repair. The continuous metabolism is
likely to affect storage reserves within the spore and, eventually, negatively impact germination and early gametophyte development.
The efficacy of dry storage has not been fully explored
for fern spores, either because partial drying (to ambient
RH) does not maintain longevity as well as hydrated
storage (Lindsay et al., 1992; Quintanilla et al., 2002), or
because of reported sensitivity to drying treatments. Drying with silica gel has given inconsistent results. Spores of
the tree ferns Dicksonia sellowiana and Cyathea caracasana
were reportedly damaged by silica gel drying (Constantino
et al., 2000). However, longevity and tolerance to 12 C
or 25 C were improved in spores of Cyathea dalgadii,
Lophosoria quadripinnata, Polystichum lonchitis, and
Ceterach officinarum dried over silica gel compared with
spores dried to ambient RH (Simabukuro et al., 1998b;
Constantino et al., 2000; Ballesteros et al., 2004). The
conflicting results of drying spores suggest that precise
control of water content is necessary to implement dry,
frozen storage in fern spore conservation programmes.
Viability of dry biological systems has long been
attributed to the properties of water or the removal of
water with specific properties (Leopold et al., 1994;
Hoekstra, 2005; Walters et al., 2005). Water status within
fern spores and its behaviour at different temperatures is
not known, which makes it difficult to predict achievable
longevities of fern spores or to compare physiological
differences of spores among species. A better understanding of the relationships between water content and viability of fern spores will provide guidance for optimizing
storage conditions to maximize longevity.
This study was undertaken to identify water properties
in spores of diverse fern species and to compare them
with existing information on pollen and seeds. Here water
interactions with fern spores are studied using water sorption isotherms. Classical applications of water sorption
isotherm models reveal the chemical affinity of materials
for water (Rockland, 1969; Vertucci and Leopold, 1987a,
b; Oksanen and Zografi, 1990; Costantino et al., 1998;
Lyall et al., 2003; Nagarajan et al., 2005). More recent
interpretations have also used isotherm shape to account
for structural relaxation within amorphous solids (glasses)
during plasticization by water (Zhang and Zografi,
2000). These new applications provide a means to link
classical studies of water binding with more recent concepts of glasses in order to establish a firm theoretical
basis upon which to predict the shelf-life of biological
materials.
Materials and methods
Plant materials and water content determinations
Mature fronds of different fern species were collected from wild
populations during the summer of 2005 at the Valencian Community, Spain (Table 1). Species were collected from diverse families,
environments, and cytological traits to test future hypotheses about
the influence of these factors on water properties and shelf-life.
Fronds were pressed onto glossy paper and allowed to dry. After
Water sorption properties of fern spores 1187
Table 1. Information about the species of ferns used in this study
Species
Population
Order and family
Ecology
Spore type and ploidy
Dryopteris filix-mas
Maset del Zurdo, Vistabella del Maestrat,
Castellon, Spain (1380 m)
Font del Tilde, Vistabella del Maestrat,
Castellon, Spain (1200 m)
Barranco del Juncaret, Ahin,
Castellon, Spain (650 m)
Barranco de la Safor, Villalonga,
Valencia, Spain (260 m)
Ullals Riu Verd, Massalaves,
Valencia, Spain (35 m)
Aspidiales Aspidiaceae
Pteridales Pteridaceae
Forest understorey
shade
Forest understorey
shade, moist
Forest understorey
shade, moist
Tropical moist
Monolete polyploid
(2n¼164)
Monolete polyploid
(2n¼164)
Monolete diploid
(2n¼82)
Trilete polyploid
Aspidiales Thelypteridaceae
Moist
Monolete diploid
(2n¼70)
Polystichum aculeatum
Polystichum setiferum
Pteris vittata
Thelypteris palustris
sporangial dehiscence, spores were collected from the sheets,
sieved, and subsequently stored at 80 C until used. Spores were
mailed to Fort Collins, CO, USA using expedited post and arrived
within 3 d. There was no discernible difference in the viability of
spores after collection and shipment (data not given).
Water content of spores was manipulated by equilibrating them at
5, 25, and 45 C and different RH or water vapour pressures,
adjusted using water or the saturated salt solutions listed in Table 2
(Winston and Bates, 1960; Rockland, 1969; summarized by
Vertucci and Roos, 1993). Spores were placed in open Petri plates
that were held above water or the saturated salt solutions at 5, 25,
and 45 C in screw-cap jars. Spores were also equilibrated to low
RH at 15 C to confirm temperature anomalies observed within the
5 C and 25 C isotherms. Periodically, subsamples weighing between 5 mg and 15 mg were removed from RH chambers and
hermetically sealed into preweighed 20 ll aluminium pans and
weighed using an electronic balance. Subsampling was frequent
during the first day of hydration or dehydration, then after 24 h and
48 h, then after 1 week and 2 weeks, and finally in monthly
intervals for 3–6 months. After fresh weight measurements, pans
were punctured and heated at 95 C for 36 h to obtain the sample
dry weight (DW). Water content values were calculated from the
difference in fresh and dry weight and are expressed on a dryweight basis as g H2O g1 DW. Within-replicate averages were
calculated from at least four separate subsamplings taken between
7 d and 6 months (i.e. water contents reached steady-state) after the
spores were placed in the RH chamber. Standard deviation of these
equilibrated water content measurements ranged from 8% to 10% of
the average value, probably as a result of the accuracy of the electronic
balance (61 lg) and rapid re-equilibration of the small subsamples
to room conditions during measurement. The entire equilibration
process was repeated at least once for each species, RH, and temperature treatment or until variation among replicates was <5% of
the average measurement. Similar data for pollen were reported by
Buitink et al. (1998), and are used here for comparative purposes.
Sorption characteristics of fern spores were evaluated using water
sorption isotherms constructed at 5, 25, and 45 C for RH between
0.5% and 97%, and at 15 C for RH between 0.5% and 15%. Isotherms were analysed using three different sorption models as well as
van’t Hoff analysis to determine heats of sorption at specific temperatures (isotherm models) and water contents (van’t Hoff analysis).
Application of isotherm models
Isotherm models quantify the tendency to absorb water using two
types of parameters. The hydrophilic environment, or chemical affinity for water, is quantified by the number of binding sites. The
nature of interactions between adsorbent surface and water molecules is described by the enthalpy of sorption. Applications of the
Brunauer–Emmet–Teller (BET), D’Arcy–Watt, and Flory–Huggins
models were used to quantify the chemical affinity for water of fern
Aspidiales Aspidiaceae
Aspidiales Aspidiaceae
Table 2. The relative humidity obtained from various saturated
salt solutions incubated at different temperatures (as cited by
Vertucci and Roos, 1993)
Salt
Temperature (C)
P2O5
ZnCl2
LiCl
K. acetate
MgCl2
K2CO3
Ca(NO3)2
Mg(NO3)2
NH4NO3
NaNO2
NaCl
KCl
KNO3
KH2PO4
K2SO4
5
15
25
45
0.5
5.5
15
26
33.5
47
61
–
–
–
75.5
88
93
–
–
0.5
5.5
13
–
–
–
–
–
–
–
–
–
–
–
–
0.5
5.5
13
25
33
43
50.5
53
62
64
75
85
91
–
–
0.5
5.5
11
19
31.5
39
40
–
50
60
75
81
87
92.8
96
spores, and BET and D’Arcy–Watt models to quantify the enthalpy
of sorption in fern spores.
The BET model builds from Langmuir isotherms, which saturate
at high partial pressures, using the additional assumption that adsorbate molecules cluster upon each other in progressive layers to
allow infinite adsorption (Brunauer et al., 1938; Atkins, 1982). BET
models were adapted to the special case of water adsorption by
using RH/100 to measure the partial pressure of water and water
content (wc) to describe the volume of adsorbate (assumes density
of 1 g ml1):
RH=100
1
RH=100 ðc 1Þ
þ
¼
wcð1 RH=100Þ wcmon 3c
wcmon 3c
ð1Þ
BET parameters were calculated from isotherms of fern spores by
linear regression of RH/100 and (RH/100)/[wc(1–RH/100)], where
wc is the water content of the spores for each RH/temperature
combination. The y-intercept of the regression line is equal to
1
1
wc1
and the slope is equal to (c1)wc1
mon c
mon c , where wcmon
1
(in g H2O g DW) is the BET monolayer, that describes the water
content at which sites at the adsorbent surface are filled, and c is
related to the sorption enthalpy [DHsorp(T)] of water binding at
monolayer sites at the isotherm temperature:
c ¼ exp½ð DHsorp DHvap Þ=RT
ð2Þ
and represents the excess energy released when absorbated molecules condense on the absorbent surface at a given temperature.
1188 Ballesteros and Walters
The BET equation becomes non-linear (and undefined) at RHs between
25% and 50%, depending on the properties of the adsorbent. Here,
regression analyses were constrained to RH values <40%, and the
1
slope of the BET plot [i.e. (c1)3 wc1
mon c ] was constrained to
<400 to ensure that at least six RH/wc data points were used in each
regression and that the r2 of all regressions was greater than 0.90.
The D’Arcy–Watt equation was developed to describe water
sorption onto heterogeneous materials by making provisions for
differing enthalpies between three types of sorption sites (D’Arcy
and Watt, 1970). The equation is a composite of three terms, each
representative of a different type of water binding:
wc ¼
K K 0 ðRH=100Þ
k k 0 ðRH=100Þ
þ h ðRH=100Þ þ
1 þ KðRH=100Þ
1 kðRH=100Þ
ð3Þ
The first term describes sites where individual water molecules
bind strongly (region 1), the second describes sites where water
binds weakly (region 2), and the third describes sites where water
condenses as a collection of molecules (multimolecular sorption;
region 3). The parameters K# and h represent the amount of water
associated with strong and weak binding sites, respectively, in
g H2O g1 DW, k# is related to the number of multimolecular sites,
the natural logarithm of K is related to the enthalpy of sorption at
the isotherm temperature, and k is the water activity for multimolecular sites. The value for h was calculated from linear regressions
of RH and wc, initially for RH between 30% and 70% and
subsequently for RH <90%. Values for K and K# were calculated
from linear regressions of 100/RH and 1/wc for RH between 0%
and 25%. The value of k# was calculated by constraining k¼1 and
regressing 100/RH and 1/wc for RH between 75% and 100%.
Regressions for terms were performed iteratively using an Excel
spreadsheet to achieve the highest correlation between modelled and
experimental data (Vertucci and Leopold, 1987a).
The Flory–Huggins model considers absorption as a process of
dissolution, and uses the parameter v to describe the interaction
between the adsorbate (molecules in fern spores) and the solvent
(water) (Flory, 1953). A simplified equation was used that expresses
the amount of water as water content, rather than volume, by
assuming a constant specific gravity for water and large molecular
mass of the absorbate:
lnðRH=100Þ ffi lnðwcÞ þ wc þ ð1 wcÞ2
ð4Þ
The value of v is low for highly soluble material and increases as
solubility decreases (e.g. v¼0.65 for dextran in water; Zhang and
Zografi, 2000). Values for v in fern spores were determined by
fitting the Flory–Huggins model to RH–wc data at RH >75%.
for temperatures between 5 C and 45 C and water contents
ranging from 0.005 to 0.2 g H2O g1 DW at 0.005 g H2O g1 DW
intervals. For each water content, the slope of the regression between ln(RH/100) and T1 was used to calculate DHsorp(w) according to equation 5. Correlation coefficients (r2) and percentage error
of slope (standard error of slopeOslope) were usually >0.95 and
<0.10, respectively, for water contents >0.03 g H2O g1 DW, and
support the assumption of a constant DHsorp(w) with temperature.
When these criteria were not met (often for spores at water contents
<0.03 g H2O g1 DW), the assumption that DHsorp(w) is constant
with temperature could not be supported. In these cases, two lines
were drawn to maximize r2 and minimize percentage error. Isotherms at 18 C could be constructed by extrapolating the linear
relationship of the isochores. These extrapolations assume that no
phase changes that affect the aqueous domain of the cytoplasm (i.e.
no molecular denaturation or water freezing events) occur between
5 C and 18 C.
Results
The initial water content of fern spores from five species
placed at ambient room conditions in Fort Collins, CO,
USA (about 30% RH and 22 C) was between 0.02 and
0.04 g H2O g1 DW. The water content of fern spores
placed over water vapour increased to a maximum value
within about 16–48 h depending on the species (Fig. 1;
Table 3). Spores swell as they take up water and diameters
of spores placed within liquid water or over water vapour
were similar after 72 h (data not shown), indicating that
imbibition was complete and that fern spores absorbed
similar amounts of water when imbibed in either vapour
or liquid water. Maximum water contents ranged from
0.14 (Pteris vittata) to 0.52 (Polystichum aculeatum) g
H2O g1 DW depending on species (Fig. 1; Table 3). By
contrast to spores, Typha latifolia pollen absorbed more
water in a shorter time and reached a steady-state water
content of about 0.88 g H2O g1 DW in about 30 h
(Fig. 1).
van’t Hoff analyses
Application of van’t Hoff analyses (Atkins, 1982) to pollen and
seed sorption isotherms were described previously (Vertucci and
Roos, 1993; Vertucci et al., 1994; Walters, 1998; Eira et al., 1999).
Van’t Hoff isochores describe the temperature dependence of h, the
equilibrium constant for adsorption and desorption at a specific
water content according to
ð@lnhÞ=½@ð1=TÞ ¼ DHsorpðwÞ =R
ð5Þ
where hffiRH/100 at equilibrium, T is temperature (in Kelvin), R
is the ideal gas constant (8.3143 J K1 mol1) and DHsorp(w) is the
sorption enthalpy at a given water content. By contrast to BET and
D’Arcy–Watt models, where sorption enthalpy is assumed constant
for a particular binding site but may vary with temperature, sorption
enthalpies calculated from van’t Hoff analyses are considered a
function of water content [hence the subscript (w)] and presumed
constant with temperature. Relative humidity–temperature combinations were interpolated from isotherms and used to draw isochores
Water content (g H2O g-1 dw)
0.8
pollen
0.6
P. aculeatum
P. setiferum
0.4
T. palustris
D. filix-mas
0.2
P.vittata
0
0
20
40
60
80
Time (hours)
Fig. 1. Hydration time-courses of spores from five species of ferns and
of Typha latifolia pollen initially equilibrated to ambient room
conditions (approximately 22 C and 30% RH) and then placed in
sealed jars over water.
Water sorption properties of fern spores 1189
Table 3. Hydration characteristics and coefficients for isotherm model parameters determined for spores of different species of fern
and other germplasm
Parameter
Temperature Fern species
Typha Fern
(C)
pollena frondb
P. aculeatum P. setiferum D. filix-mas T. palustris P. vittata Average
Maximum wc over water
vapour (g H2O g1 DW)
Time to maximum wc (h)
Time to equilibrate to
low RH (d)
BET monolayer value (wcmon)
(g H2O g1 DW)
25
25
5, 25, 45
5
25
45
5
BET c [c}exp(DHsorp)]
25
45
5
D’Arcy–Watt: strong sites (K#)
(g H2O g1 DW)
25
45
5
D’Arcy–Watt: weak sites (h)
25
(g H2O g1 DW)
45
D’Arcy–Watt: multimolecular
5
sites (k’)
25
45
5
D’Arcy–Watt: strong+weak
(g H2O g1 DW) (K#+h)
25
45
D’Arcy–Watt K [K}exp(DHsorp)] 5
25
45
Flory–Huggins v
5
25
45
48
5
0.035
0.027
0.028
308
80
23
0.035
0.029
0.026
0.080
0.034
0.012
0.0048
0.0044
0.0051
0.115
0.063
0.038
96
45
27
1.33
1.67
1.83
0.44
48
5
0.040
0.034
0.031
401
54
35
0.042
0.029
0.027
0.058
0.055
0.022
0.0063
0.0036
0.0046
0.100
0.084
0.049
144
62
46
1.30
1.50
1.74
0.28
0.34
30
5
20
5
0.032
0.026
0.022
110
63
39
0.030
0.024
0.023
0.056
0.033
0.032
0.0063
0.0033
0.0006
0.087
0.057
0.056
98
72
26
1.43
1.74
2.02
0.14
0.34
16
5
0.037
0.028
0.023
77
81
58
0.035
0.022
0.021
0.051
0.039
0.031
0.0078
0.0045
0.0030
0.086
0.061
0.052
114
48
52
1.38
1.68
1.88
32.4
5
0.027
0.022
0.017
100
79
71
0.028
0.020
0.017
0.036
0.029
0.015
0.0044
0.0020
0.0040
0.063
0.049
0.032
87
64
83
1.68
1.97
2.23
0.034
0.027
0.024
199
71
45
0.034
0.025
0.023
0.056
0.038
0.022
0.0059
0.0036
0.0035
0.090
0.063
0.045
108
58
47
1.42
1.71
1.94
0.88
30
4
0.050
0.045
0.073 0.048
0.045
23
17
19
57
12
0.024
0.017
0.027 0.037
0.021
0.117
0.094
0.217 0.101
0.090
0.0200
0.0280 0.019 0.0110
0.0330
0.141
0.111
0.244 0.138
0.111
91
74
134
154
31
0.78
0.83
0.55
0.97
0.73
From Buitink et al. (1998), Vertucci and Leopold (1987b), Vertucci and Leopold (1987a), respectively.
Hydrated fern spores dried quickly when exposed to
low RH. Water contents of fully hydrated spores decreased to <0.03 g H2O g1 DW (over 90% of the initial
water removed) in about 8 h (Fig. 2). After the initial
water losses depicted in Fig. 2 (first data point in Fig. 3),
water contents of spores equilibrated to the RH of the
chamber and there were no significant changes in water
content after about 7 d (Fig. 3). The water content at
equilibrium depended on the RH (Fig. 3A). For example,
P. aculeatum spores at 25 C equilibrated to 0.005, 0.019,
and 0.033 g H2O g1 DW at 0.5, 5.5, and 11% RH, respectively. The water content at equilibrium also depended on temperature, with lower temperatures generally
yielding higher water contents for a given RH (Fig. 3B).
In addition to these factors, the equilibrium water content
depended on species. Water contents of P. vittata were
usually much lower than those of other species at a given
RH–temperature combination (Fig. 3C). The kinetics of
dehydration were similar for Typha pollen, although the
duration of the initial rapid drying phase was longer, probably as a result of higher initial water contents (Fig. 2).
Water sorption isotherms, representative of the equilibrium relationship between RH and water content, were
0.6
Water content (g H2O g-1 dw)
a, b, c
0.52
Soybean
axesc
0.5% RH 5°C
pollen
0.4
T. palustris
P. aculeatum
0.2
P. vittata
0
0
100
200
300
400
500
Time (min)
Fig. 2. Drying time-courses of spores from three species of ferns
(T. palustris, closed triangles; P. aculeatum, closed squares; P. vittata;
closed circles) and of Typha latifolia pollen (open circles) hydrated for
24 h in water vapour and then placed over P2O5 (0.5% RH) at 5 C.
Initial drying rates presented here are representative of data collected
for all species.
constructed for spores from each species of fern at 5, 25,
and 45 C (Fig. 4). Isotherm shape followed the reversesigmoidal pattern typically observed for proteins and
1190 Ballesteros and Walters
0.30
0.05
P. aculeatum
A
A
25°C
25°C
0.25
0.04
11%
0.20
0.03
0.15
5.5%
0.01
Water content (g H2O g-1 dw)
0.02
0.5%
Water content (g H2O g-1 dw)
0
T. palustris
B
5.5% RH
0.04
5°C
0.03
25°C
T. palustris
P. setiferum
0.10
0.05
P. vittata
0
P. aculeatum
B
0.25
0.20
0.15
-18
0.02
5
0.10
45°C
25
45
0.01
0.05
0
0
C
0.5% RH
0
5°C
25
50
75
100
Relative Humidity (%)
0.04
0.03
T. palustris
P. aculeatum
0.02
0.01
P. vittata
0
0
10
20
30
40
50
Fig. 4. Water sorption isotherms of spores of three fern species (T.
palustris, closed squares; P. setiferum, open triangles; P. vittata, closed
diamonds) constructed at 25 C (A) and of P. aculeatum spores
constructed at 5, 25, and 45 C (B). The points represent the average
water content of spores calculated from measurements taken after 12 d
equilibration (see representative data in Fig. 3). Standard deviations of
average water content measurements were <5% of the average. Curves
are calculated from water content, RH, and temperature relationships
determined from van’t Hoff analyses (Fig. 6), and the dashed curve is
an isotherm constructed at 18 C by extrapolating van’t Hoff
isochores for P. aculeatum spores to 18 C (as seen in Fig. 7 for
spores of T. palustris).
Time (days)
Fig. 3. Changes in water content of spores from the indicated fern species
during equilibration to different RH (A) and temperature (B), and
differences in equilibrated water contents among species (C; same symbols
as in Fig. 2). Similar data were acquired for all species–RH–temperature
combinations for equilibration times up to 6 months (not shown).
polymers (Costantino et al., 1998; Zhang and Zografi,
2000) and orthodox seeds and pollen (Vertucci and
Leopold, 1987a; Vertucci and Roos, 1993; Buitink et al.,
1998). The relationship between RH and water content
depended on species, with spores of P. vittata and
Polystichum setiferum giving the lowest and highest, respectively, water contents for any given RH and temperature combination (Fig. 4A). The relationship between RH
and water content also depended on temperature, with
water contents at a given RH typically increasing with
decreasing isotherm temperature (Fig. 4B).
Water sorption isotherms of fern spores at 5, 25, and
45 C were fitted to various isotherm models to calculate
parameters relating to chemical affinity (i.e. number of hydrophilic binding sites) and enthalpy of sorption (DHsorp).
Model parameters calculated from isotherms of pollen
(Typha latifolia), embryonic axes of soybean (Glycine
max), and fronds of the desiccation-tolerant fern Polypodium polypodioides are given for comparative purposes
(data taken from Vertucci and Leopold, 1987a, b; Buitink
et al., 1998). Values for parameters of the BET monolayer
(wcmon) and strength of sorption (c) (Brunauer et al.,
1938; Atkins, 1982) were calculated for all isotherms
(Table 3). At 25 C, monolayer values averaged
0.02760.004 g H2O g1 DW among spores of all species,
and ranged from 0.022 to 0.034 g H2O g1 DW for spores
of P. vittata and P. setiferum, respectively. BET monolayer values for fern spores were less than those calculated
for pollen, embryonic axes, or desiccation-tolerant leaves.
Monolayer values for spores tended to decrease as temperature increased (Table 3).
Water sorption properties of fern spores 1191
The D’Arcy–Watt (1970) model builds from assumptions in the BET model, with an additional provision for
sorption sites with different binding strengths. The sum of
water adsorption by each type of site produced a composite
isotherm with good agreement to measurements of water
content at given RH–temperature combinations (Fig. 5). The
amount of water associated with strong binding sites was
calculated for fern spores using the D’Arcy–Watt model,
and was similar to the average amount of water calculated
as the BET monolayer (Table 3). At 25 C, the average
amount of strongly bound water was 0.02560.004 g H2O
g1 DW among species and ranged from 0.020 for spores
of P. vittata to 0.029 g H2O g1 DW for spores of P.
aculeatum and P. setiferum. Pollen and fern leaf tissue had
comparable, and soybean embryonic axes had more, water
on strong binding sites compared with fern spores. The
average amount of water associated with weak binding sites
(h) in fern spores was 0.03860.010 g H2O g1 DW, and
ranged from 0.029 g H2O g1 DW for spores of P. vittata
to 0.055 g H2O g1 DW for spores of P. setiferum (Table
3). The number of weak binding sites was two to five
times greater in soybean axes, typha pollen, and fern leaf
tissue. Similarly, soybean axes, typha pollen, and fern leaf
tissue had considerably more multimolecular sorption sites
(k#) than were calculated for fern spores. The number of
strong and weak sorption sites (K#+h) calculated from the
D’Arcy–Watt model decreased with increasing isotherm
temperature. There was no consistent relationship between
temperature and the number of multimolecular sorption
sites.
The Flory–Huggins model (Flory, 1953) considers
sorption as the initial steps of the dissolution process. The
Water content (g H2O g-1 dw)
0.3
25°C
Typha pollen
0.2
0.1
P. aculeatum
strong
weak
multi
0
0
20
40
60
80
Relative Humidity (%)
Fig. 5. Water sorption isotherms for spores of P. aculeatum fitted to the
D’Arcy–Watt (1970) model. Data points used to calculate coefficients
for this model are the same as those used in Fig. 4B at 25 C. The
D’Arcy–Watt model expresses total water content as the sum of water
associated with strong, weak, and multimolecular sorption sites
(shown). The isotherm for Typha latifolia pollen constructed at 25 C
and fitted to the D’Arcy–Watt model is also given for comparative
purposes.
rising slope of the isotherm at high RH is characterized by
the parameter v which describes adsorbent–solvent interactions. Low values of v indicate highly soluble substances and high values indicate substances that do not
tend to dissolve. Values of v calculated from fern spores
isotherms drawn at 5, 25, and 45 C were greater than 1.3
(Table 3) and tended to increase with increasing temperature. Values of v calculated from isotherms at 25 C for
soybean axes, pollen, and desiccation-tolerant fern leaves
were <1 (Table 3) and approached values for watersoluble polymers (Zhang and Zografi, 2000). At 25 C,
calculated values of v were largest for spores of P. vittata
(1.97) and least for spores of P. setiferum (1.50).
The apparent enthalpy of water sorption [DHsorp(T)] by
fern spores can also be calculated from sorption isotherm
models. The average enthalpy of water-adsorbent associations is approximated by the natural logarithm of the
BET parameter c and the D’Arcy–Watt parameter K. The
average value for the BET parameter c was 71612 for
fern spores at 25 C, which was comparable with the
value calculated for soybean axes and considerably higher
than values calculated for pollen and fern leaf tissue. The
average value for the D’Arcy–Watt parameter K was
58612 for fern spores at 25 C, which was similar to the
value calculated for pollen and less than values calculated
for soybean axes and fern leaf tissues. Values of c or K
tended to decrease with increasing temperature, with large
changes observed for P. aculeatum, P. setiferum, and
Dryopteris filix-mas, and only minor changes with temperature observed for Thelypteris palustris and P. vittata.
van’t Hoff analyses of isotherms allowed adsorption
enthalpies to be calculated at specific water contents
[DHsorp(w)]. Based on the equilibrium partitioning between
water within the vapour phase and condensed on the
adsorbate surface, high values of [DHsorp(w)] are expected
at low water contents because of the high (very negative)
free energy driving this exothermic reaction. High values
of DHsorp(w) for fern spores containing <0.05 g H2O g1
DW were apparent from the steep slopes when the natural
logarithm of RH/100 is plotted against temperature1 (in
Kelvin) (Figs 6, 7). The slope of these plots became
progressively shallower as water content increased above
0.05 g H2O g1 DW. For the most part, van’t Hoff plots
of fern spores were linear, suggesting fairly predictable
sorption–desorption equilibrium coefficients. However,
at very low water contents, van’t Hoff plots for spores of
all species tested except D. filix-mas appeared non-linear
(plots of P. setiferum appeared to be non-linear, but were
treated as linear because isotherms were constructed at
only three temperatures). The non-linear behaviour resulted
from higher than expected water contents measured in
spores equilibrated at 25 C and 45 C and RH <20%.
Assuming the linear relationship between T1 and ln(RH/
100) continues as temperature decreases below 5 C,
isotherms at 18 C can be predicted [dashed lines in
1192 Ballesteros and Walters
0
0.175 g g-1
0.125 g g-1
0.100 g g-1
0.080 g g-1
0.050 g g-1
-2
0.030 g g-1
-4
0.020 g g-1
0.015 g g-1
0.010 g g-1
ln(RH/100)
-6
0.005 g g-1
A P. aculeatum
-8
0.175 g g-1
0.125 g g-1
0.100 g g-1
0.080 g g-1
0.030 g g-1
0
0.050 g g-1
-2
-4
-6
0.020 g g-1
0.015 g g-1
Discussion
0.010 g g-1
Survival of desiccated cells is often attributed to the
association of water with biomolecules and its effects on biomolecule stability and mobility within the aqueous
domains of cells (Walters, 1998; Hoekstra, 2005; Walters
et al., 2005). Here, water–biomolecule associations in fern
spores are investigated through the relationships between
RH, water content, and temperature. These relationships
are compared among five homosporeous species (nonchlorophyllous spores) and other plant germplasm: pollen
from Typha latifolia, embryonic axes of Glycine max, and
leaf tissue from the desiccation-tolerant fern Polypodium
polypodioides (Vertucci and Leopold, 1987a, b; Buitink
et al., 1998). It is shown that fern spores tend to
equilibrate rapidly to environmental RH, approaching
steady-state water content within a few days of imbibition
or drying. By contrast with the other germplasm studied,
fern spores absorb very little water (Table 3; Figs 1, 5).
Differences were detected in the adsorption properties of
fern spores related to temperature and species and it is
hypothesized that these differences reveal information
about the ecophysiology (Table 1) or storage physiology
of the spore. Long-term studies of the life spans of fern
spores are underway (Ballesteros et al., 2004, 2005), and
the studies presented here on water properties provide
a predictive tool until that information is available.
The chemical affinity of fern spores for water was lower
than that of the other germplasm studied. This low affinity
is demonstrated by low monolayer values in the BET
model (wcmon), low amounts of water on weak and
multimolecular binding sites in the D’Arcy–Watt model
(h and k#), and high values of v in the Flory–Huggins
dissolution model (Table 3). These parameters were
highly correlated among each other (r2¼0.99, 0.72, and
0.95, respectively, for regressions between BET wcmon
and D’Arcy–Watt h, D’Arcy–Watt k#, and Flory–Huggins
v at 25 C). The BET monolayer and water associated
with D’Arcy–Watt strong binding sites, K#, were not
0.005 g g-1
B P. vittata
-8
0.0032
0.0034
0.0036
0.0038
1/Temperature (K-1)
Fig. 6. van’t Hoff analysis of water sorption isotherms of P. aculeatum
(A) and P. vittata (B) spores. Points represent the RH interpolated for
the given water content (numerals to the right) and temperature (x-axis)
from isotherms similar to those given in Fig. 4. The lines are the least
squares best fit calculated for three points or, if a discontinuity was
indicated, from two points.
1
ln(RH/100)
0.175 g g-1
0.125 g g-1
0.100 g g-1
0.080 g g-1
-1
0.050 g g
-1
-3
0.030 g g-1
-5
0.020 g g-1
-7
0.015 g g-1
-9
0.010 g g-1
0.005 g g-1
T. palustris
-11
0.0033
0.0035
0.0037
as water content increased (Fig. 8). For all species, sorption enthalpy was between 20 and 30 kJ mol1 at the
BET monolayer and between 7 and 13 kJ mol1 at
water contents corresponding to the sum of D’Arcy–Watt
strong and weak binding sites (Table 3). The break in the
van’t Hoff relationship between low and high temperature
ranges also occurred between 20 and 30 kJ mol1
(Fig. 8). The value of DHsorp(w) was close to 0 at the
higher temperature range in spores of T. palustris and
P. vittata at absolute dryness and approached 20 kJ
mol1 with increasing water content. By contrast, values of
DHsorp(w) were nearly 40 kJ mol1 for completely dry
spores of D. filix-mas, P setiferum, and P. aculeatum at
the higher temperature range and approached 20 kJ
mol1 with further hydration (Fig. 8).
0.0039
0.0041
1/Temperature (K-1)
Fig. 7. van’t Hoff analysis of water sorption isotherms of T. palustris.
Data points and lines are as described for Fig. 6. The dashed lines
represent extrapolation of the linear relationship to 18 C so that an
isotherm at that temperature can be predicted.
Fig. 7 show the extrapolation required to construct the
18 C isotherm in Fig. 4 (dashed line)].
Generally, DHsorp(w) was high (very negative) at low
water contents and exponentially decreased (approached 0)
Water sorption properties of fern spores 1193
Differential enthalpy of sorption (-kJ mol-1)
100
P. aculeatum
80
P. setiferum
5-15°C
60
40
15-45°C
D.filix-mas
20
0
T. palustris
80
P. vittata
5-15°C
5-15°C
60
40
25-45°C
25-45°C
20
0
0
0.05
0.10
0.15
0.05
0.1
0.15
Water content (g H2O g-1 dw)
Fig. 8. Differential enthalpy of sorption for fern spores as a function of the water content. Sorption enthalpy was calculated from the slopes of van’t
Hoff isochores similar to those given in Figs 6 and 7. The different values at high and low temperature ranges are indicative of discontinuities in the
linear relationships.
strongly correlated (r2¼0.26 at 25 C), and calculations of
water on strong binding sites were similar across diverse
tissues. Among the fern species studied, P. vittata had the
lowest chemical affinity and P. setiferum had the highest
chemical affinity (Table 3; compare values for wcmon in
the BET model; K#, h, and k# in the D’Arcy–Watt model,
and v in the Flory–Huggins model).
Chemical affinity for water has been linked to stress
tolerance in plant tissues (Levitt, 1980; Rascio et al.,
1992; Vertucci and Stushnoff, 1992) and seed storage
physiology (Vertucci and Leopold, 1987b; Vertucci and
Roos, 1990; Vertucci et al., 1994; Sun et al., 1997; Eira
et al., 1999; Dussert et al., 1999; Lyall et al., 2003;
Pukacka et al., 2003; Hor et al., 2005; Nagarajan et al.,
2005). Indeed, the mechanistic understanding of protection from dehydration and freezing stresses often invokes accumulation of highly hydrophilic substances such
as sugars and heat-soluble proteins (Close, 1997; Bryant
et al., 2001) that may increase the chemical affinity of
water within cells. Chemical affinity is likely to be a
complex function of the concentration of hydrophilic
residues, the degree to which macromolecules are folded,
and the interaction of small hydrophilic molecules on
macromolecular surfaces. The surprising decrease in the
BET monolayer value of proteins with the addition of
sucrose or trehalose (Constantino et al., 1998) confirms
that chemical affinity is not simply the sum of the number
of hydrophilic sites. Assessments of chemical affinity in
complex mixtures within cells are also confounded by the
amount of hydrophobic constituents that are sequestered
outside the aqueous domain of the cytoplasm (e.g. triacylglycerols and polyphenolic compounds; Vertucci and
Leopold, 1987a; Vertucci and Roos, 1990; Hor et al.,
2005). The presence of these types of molecules in fern
spores is rarely documented (Simabukuro et al., 1998a),
but needs to be considered in future studies that investigate
the potential links between water affinity and longevity.
Some studies suggest that seed longevity may be inversely correlated with the water associated with D’Arcy–
Watt weak binding sites or the proportion of water in
weak compared with strong binding sites (Sun et al., 1997;
Nagarajan et al., 2005). If true, this would suggest that
fern spores are relatively long-lived compared with other
germplasm and that spores of P. setiferum have the
shortest life span of the fern species studied. Consistent
with this prediction, fern spores appear to have longer life
spans than Typha latifolia pollen; however, they appear
to have similar or shorter life spans than many seed or
desiccation-tolerant leaf tissues (Lloyd and Klekowski,
1970; Windham et al., 1986; Lindsay et al., 1992;
Hoekstra, 2005; Walters et al., 2005). Low affinity for
water at high RH will limit the amount of water that enters
cells, preventing large changes in cell volume and slowing
metabolic activity. This feature may provide protection
against damaging hydration–dehydration cycles and thereby
promote longevity under less optimal storage environments. Perhaps this also explains the unusual longevity of
fern spores under hydrated storage conditions (Lindsay
et al., 1992).
The enthalpy of sorption also provides insights into
water associations that may influence storage physiology
of desiccated systems. Enthalpy of sorption (DHsorp) is
usually regarded as the amount of heat released when
water vapour condenses on a molecular surface (Atkins,
1194 Ballesteros and Walters
1982). However, associated reactions also contribute to
enthalpic changes. These reactions include dissolution of
the molecule as water content increases (Flory, 1953) and
structural changes to the molecule, such as swelling, contraction of hydrophobic regions, and denaturation. The
effect of these associated reactions on the magnitude of
DHsorp depends on whether they are exothermic or endothermic. Recently, DHsorp was also related to the concept
of restricted structural mobility in amorphous (i.e. noncrystalline) solids by Zhang and Zografi (2000). These
authors suggested that the value of DHsorp, derived from
the parameter c in the BET model, relates to non-ideal
changes in structure arising when glassy materials are
initially hydrated. Their hypotheses provide a rationale for
the temperature dependency of DHsorp calculated from
both isotherm models and van’t Hoff analyses of spores at
low water contents. Hence, the relative value of DHsorp
depends on whether the material is in a glassy state or not.
In isotherm models, the value of DHsorp is high at temperatures that promote glasses, and an abrupt decrease
(towards 0) in DHsorp with increasing isotherm temperatures reflects a glass transition and can be used to estimate Tg (glass transition temperature) roughly (Zhang and
Zografi, 2000).
The idea that the calculated enthalpy of sorption is
related to molecular mobility and structural stability
within amorphous solids supports an important possible
link between isotherms and viability of dried cells. These
relationships were initially noted by observations that the
magnitude of DHsorp was low (close to 0) in desiccationintolerant tissues (Vertucci and Leopold, 1987b; Vertucci
et al., 1994; Eira et al., 1999) and that DHsorp(w) could be
used as a first approximation of viscosity in seeds
(Vertucci and Roos, 1990). Studies subsequent to these
showed that changes in viscosity were linked to glass
transitions and corresponded to changes in ageing kinetics
(Leopold et al., 1994; Buitink et al., 1998, 2000; Walters,
2004). Building on the idea that the magnitude and
change of DHsorp reflects structural restrictions and
relaxations within a glass, it can be hypothesized that high
values of c in the BET model or K in the D’Arcy–Watt
model reflect a condition of high molecular stability.
Thus, fern spores that are stored at 5 C will survive
longer than their counterparts that are stored at 25 C or
45 C (Table 3). Furthermore, decreases in c or K with
increasing temperature may be reflective of imminent
glass transition and relaxation of the molecular structure
that was ‘frozen in’ when the glass was formed during
drying. Accordingly, fern spores that demonstrate a large
decrease in the values of c or K with increasing temperature can be predicted to store poorly at those elevated
temperatures. From these arguments and the temperature
dependency of c and K values given in Table 3, it can
be predicted that spores of P. aculeatum and P. setiferum
will be prone to deterioration at ambient temperatures,
while those of P. vitatta and T. palustris would be relatively stable.
High c and K values from BET and D’Arcy–Watt
models, respectively, calculated for P. vitatta and T.
palustris spores at 25 C or 45 C, arise from high water
contents measured at low RH at those temperatures. The
high water contents also result in non-linear van’t Hoff
plots and low values of DHsorp(w) (closer to 0) at low
water contents and high temperature ranges (Fig. 8). The
abrupt difference in DHsorp in van’t Hoff analyses probably
reflects resistance to volumetric changes and indicates that
the glassy state is maintained despite a rise in temperature.
Reversion to linear van’t Hoff isochores with increasing
water content may, therefore, indicate plasticization of the
glass by water, and a loosening of ‘frozen-in’ structure
within the temperature range studied. Non-linear van’t
Hoff plots may also indicate a change (rather than resistance to change) in structure or phase (Atkins, 1982).
This hypothesis and the steeper slope at lower temperatures (Figs 6–8) suggest structural changes are induced
by reducing the temperature, an event that is more indicative of a phase change such as crystallization. According
to this hypothesis, changes in lipid structure or condensation of macromolecules might possibly be involved and, if
so, would lead to speculation about potential damaging
effects of extreme drying and cooling on spore viability.
In addition to new applications of water sorption isotherms described above that may predict relative longevity, traditional applications of water sorption isotherms
have always been useful to predict suitable moisture ranges
for storage at various temperatures (Vertucci and Roos,
1990, 1993; Vertucci et al., 1994; Buitink et al., 1998;
Walters, 1998; Eira et al., 1999; Hor et al., 2005). The
optimum RH for storing spores of Woodwardia sp. (fern
genus from family Blechnaceae) corresponded to about
20% RH (Walters et al., 2005). Assuming a similar optimum for the species studied here, recommended water
contents would range from 0.039 to 0.025 g H2O g1 DW
for spores of P. setiferum and P. vittata stored at 25 C.
Reducing the storage temperature to 18 C would result
in an increase in optimum water content to 0.059 and
0.038 g H2O g1 DW, respectively, according to isotherms produced through extrapolation using van’t Hoff
analyses (Figs 4 to 7). Storing seeds and pollen at water
contents less than the BET monolayer value has been
linked to more rapid deterioration (Walters, 1998). If
a similar trend exists for fern spores, detrimental effects
would be expected if spores are stored at <7–12% RH,
depending on species and temperature (see water contents
for BET monolayer in Table 3). Also poor longevity of
spores stored at 75% RH would be expected since the transition from glassy to rubbery phase would be complete.
Water contents corresponding to this RH are surprisingly
low in fern spores compared with other germplasm (Fig.
5), and range from about 0.08 to 0.05 g H2O g1 DW for
Water sorption properties of fern spores 1195
spores of P. setiferum and P. vittata, respectively, stored
at 25 C. The isotherms presented here suggest that the
range of allowable water contents for safe storage is quite
narrow for fern spores.
Conclusions
Water sorption isotherms describe the process of hydration in terms of chemical interaction of water on molecular surfaces, relaxation of molecular structure as
mobility increases, and volumetric changes as molecules
unfold. It has been shown that fern spores from several
species have low affinity for water but a high degree of
structural relaxation when water is added. These findings
suggest that fern spores from the species studied are
stabilized when dry and are therefore amenable to dry
storage even at subfreezing temperature if water contents
are manipulated precisely. Comparison of glassy behaviour among species of fern spores and among other
germplasms will help to elucidate further the role of
water, particularly in glasses, in preservation of biological
materials, and it may ultimately provide predictive tools of
shelf-life during storage.
Acknowledgements
DB was supported by a FPU grant from the Spanish Ministry of
Science and Technology. We thank Lisa Hill for her excellent
technical assistance.
References
Aragón CF, Pangua E. 2004. Spore viability under different
storage conditions in four rupicolous Asplenium L. taxa. American Fern Journal 94, 28–38.
Atkins PW. 1982. Physical chemistry. San Francisco: W.H.
Freeman and Company.
Ballesteros D, Ibars AM, Estrelles E. 2004. New data about
pteridophytic spore conservation in germplasm banks. Plantaeuropa IVth Conference. València: España. http://www.nerium.
net/plantaeuropa/Download/Procedings/Ballesteros_daniel.pdf.
Ballesteros D, Ibars AM, Estrelles E. 2005. Conservación de
esporas verdes de pteridofitos en bancos de germoplasma. II
Congreso de Biologı́a de la Conservación de Plantas. Gijón:
España.
Beri A, Bir SS. 1993. Germination of stored spores of Pteris vittata
L. American Fern Journal 83, 73–78.
Brunauer S, Emmet PH, Teller F. 1938. Adsorption of gases in
multimolecular layers. Journal of the American Chemical Society
60, 309–319.
Bryant G, Koster KL, Wolfe J. 2001. Membrane behaviour in
seeds and other systems at low water content: the various effects
of solutes. Seed Science Research 11, 17–26.
Buitink J, Walters C, Hoekstra FA, Crane J. 1998. Storage
behaviour of Typha latifolia L. pollen at low water contents:
interpretation on the basis of water activity and glassy concepts.
Physiologia Plantarum 103, 145–153.
Buitink J, Leprince O, Hemminga MA, Hoekstra FA. 2000.
Molecular mobility in the cytoplasm: an approach to describe and
predict lifespan of dry germplasm. Proceedings of the National
Academy of Sciences, USA 97, 2385–2390.
Camloh M. 1999. Spore age and sterilization affects germination
and early gametophyte development of Platycerium bifurcatum.
American Fern Journal 89, 124–132.
Close TJ. 1997. Dehydrins: a commonalty in the response of plants
to dehydration and low temperature. Physiologia Plantarum 100,
291–296.
Costantino HR, Curley JG, Wu S, Hsu CC. 1998. Water sorption
behavior of lyophilized protein–sugar systems and implications
for solid-state interactions. International Journal of Pharmaceutics 166, 211–221.
Constantino S, Santamaria LM, Hodson E. 2000. Storage and
in vitro germination of tree fern spores. Botanic Gardens Micropropagation News 2, 58–60.
D’Arcy RL, Watt IC. 1970. Analysis of sorption isotherms of nonhomogeneous sorbents. Transactions of the Faraday Society 66,
1236–1245.
Dussert S, Chabrillange N, Engelmann F, Hamon S. 1999.
Quantitative estimation of seed desiccation sensitivity using
a quantal response model: application to nine species of the genus
Coffea L. Seed Science Research 9, 135–144.
Dyer AF. 1979. The culture of fern gametophytes for experimental
investigation. In: Dyer AF, ed. The experimental biology of ferns.
London: Academic Press, 253–305.
Eira MTS, Walters C, Caldas LS. 1999. Water sorption properties
in Coffea spp. seeds and embryos. Seed Science Research 9,
321–330.
FAO/IPGRI. 1994. Genebank standards. Rome: Food and Agricultural Organization of the United Nations/International Plant
Genetic Resources Institute.
Flory PJ. 1953. Principles of polymer chemistry. New York, NY:
Cornell University Press.
Gómez-Campo C. 2001. La práctica de la conservación de semillas
a largo plazo. In: Conservación de especies vegetales amenazadas en la región mediterránea occidental. Ed. Centro de Estudios
Ramón Areces: Madrid, 255–266.
Guerrant EO, Havens K, Maunder M, (eds). 2004. Ex situ plant
conservation: supporting species survival in the wild. Covelo,
CA: Island Press.
Hoekstra FA. 2005. Differential longevities in desiccated anhydrobiotic plant systems. Integrative and Comparative Biology 45,
725–733.
Hor YL, Kim YJ, Ugap A, Chabrillange N, Sinniah UR,
Engelmann F, Dussert S. 2005. Optimal hydration status for
cryopreservation of intermediate oily seeds: Citrus as a case
study. Annals of Botany 95, 1153–1161.
Lebkuecher JG. 1997. Desiccation-time limits of photosynthetic
recovery in Equisetum hyemale (Equisetaceae) spores. American
Journal of Botany 84, 792–797.
Leopold AC, Sun WQ, Bernal-Lugo I. 1994. The glassy state
in seeds: analysis and function. Seed Science Research 4,
267–274.
Levitt J. 1980. Responses of plants to environmental stresses. Vol.
1. Chilling, freezing and high temperature stresses. New York,
NY: Academic Press.
Lindsay S, Williams N, Dyer AF. 1992. Wet storage of fern
spores: unconventional but far more effective! In: Ide JM, Jermy
AC, Paul AM, eds. Fern horticulture: past, present and future
perspectives. Andover: Intercept, 285–294.
Lloyd RM, Klekowski Jr EJ. 1970. Spore germination and
viability in Pteridophyta: evolutionary significance of chlorophyllous spores. Biotropica 2, 129–137.
1196 Ballesteros and Walters
Lyall TW, Ellis RH, John P, Hedley CL, Wang TL. 2003. Mutant
alleles at the rugosus loci in pea affect seed moisture sorption
isotherms and the relations between seed longevity and moisture
content. Journal of Experimental Botany 54, 445–450.
Nagarajan S, Pandita VK, Joshi DK, Sinha JP, Modi BS. 2005.
Characterization of water status in primed seeds of tomato
(Lycopersicon esculentum Mill.) by sorption properties and NMR
relaxation times. Seed Science Research 15, 99–111.
Oksanen CA, Zografi G. 1990. The relationship between the glass
transition temperature and water vapour absorption by poly
(vinylpyrrolidone). Pharmaceutical Research 7, 654–657.
Page CN. 1979. Experimental aspects of fern ecology. In: Dyer AF,
ed. The experimental biology of ferns. London: Academic Press,
551–589.
Page CN, Dyer AF, Lindsay S, Mann DG. 1992. Conservation of
pteridophytes: the ex situ approach. In: Ide JM, Jermy AC, Paul
AM, eds. Fern horticulture: past, present and future perspectives.
Andover: Intercept, 269–278.
Pence VC. 2000. Survival of chlorophyllous and non-chlorophyllous
fern spores through exposure to liquid nitrogen. American Fern
Journal 90, 119–126.
Pukacka S, Hoffmann SK, Goslar J, Pukacki PM,
Wojkiewica E. 2003. Water and lipid relations in beech (Fagus
sylvatica L.) seeds and its effect on storage behaviour. Biochimica
et Biophysica Acta 1621, 48–56.
Quintanilla LG, Amigo J, Pangua E, Pajaron S. 2002. Effect of
storage method on spore viability in five globally threatened fern
species. Annals of Botany 90, 461–467.
Rascio A, Platani C, DiFonzo N, Wittmer G. 1992. Bound
water in durum wheat under drought stress. Plant Physiology 98,
908–912.
Roberts EH, Ellis RH. 1989. Water and seed survival. Annals of
Botany 63, 39–52.
Rockland LB. 1969. Water activity and storage stability. Food
Technology 23, 1241–1251.
Simabukuro EA, Carvalho MA, Felippe GM. 1998a. Reserve
substances and storage of Cyathea delgadii Sternb. spores.
Revista Brasileira de Botanica 21, 149–152.
Simabukuro EA, Dyer AF, Felippe GM. 1998b. The effect of
sterilization and storage conditions on the viability of the spores
of Cyathea delgadii. American Fern Journal 88, 72–80.
Simpson K, Dyer AF. 1999. The survival of fern dormant spores.
Pteridologist 3, 98–103.
Smith DL, Robinson PM. 1975. The effects of spore age on
germination and gametophyte development in Polypodium vulgare L. New Phytologist 74, 101–108.
Smith RD, Dickie JB, Linington SH, Pritchard HW,
Probert RJ. 2003. Seed conservation: turning science into
practice. London: Royal Botanic Gardens, Kew.
Sun WQ, Koh DCY, Ong C-M. 1997. Correlation of modified
water sorption properties with the decline of storage stability of
osmotically-primed seeds of Vigna radiata (L.) Wilczek. Seed
Science Research 7, 391–397.
Vertucci CW, Leopold AC. 1987a. Water binding in legume seeds.
Plant Physiology 85, 224–231.
Vertucci CW, Leopold AC. 1987b. The relationship between water
binding and desiccation tolerance in tissues. Plant Physiology 85,
232–238.
Vertucci CW, Roos EE. 1990. Theoretical basis of protocols for
seed storage. Plant Physiology 94, 1019–1023.
Vertucci CW, Roos EE. 1993. Theoretical basis of protocols for
seed storage. II. The influence of temperature on optimal moisture
levels. Seed Science Research 3, 201–213.
Vertucci CW, Stushnoff C. 1992. The state of water in acclimating
vegetative buds of apple and Amelanchier. Physiologia Plantarum 86, 503–511.
Vertucci CW, Crane J, Porter RA, Oelke EA. 1994. Water
properties in Zizania embryos at different moisture contents and
temperatures. Seed Science Research 4, 211–224.
Villiers TA. 1974. Seed aging: chromosome stability and extended
viability of seed stored fully imbibed. Plant Physiology 53,
875–878.
Walters C. 1998. Understanding the mechanisms and kinetics of
seed aging. Seed Science Research 8, 223–244.
Walters C. 2004. Guidelines for seed storage. In: Guerrant E,
Havens K, Maunder M, eds. Ex situ plant conservation:
supporting species survival in the wild. Covelo, CA: Island Press,
442–453.
Walters C, Hill LM, Wheeler LJ. 2005. Dying while dry: kinetics
and mechanisms of deterioration in desiccated organisms. Integrative and Comparative Biology 45, 751–758.
Whittier DP. 1996. Extending the viability of Equisetum hyemale
spores. American Fern Journal 86, 114–118.
Windham MD, Wolf PG, Ranker TA. 1986. Factors affecting
prolonged stored spore viability in herbarium collections of three
species of Pellaea. American Fern Journal 76, 141–148.
Winston PW, Bates DH. 1960. Saturated solutions for the control
of humidity in biological research. Ecology 41, 232–236.
Zhang J, Zografi G. 2000. The relationship between ‘BET’- and
‘Free Volume’-derived parameters for water vapor absorption
into amorphous solids. Journal of Pharmaceutical Sciences 89,
1063–1072.