ArticlePDF Available

Effects of sublethal attack by a sucking insect, Hyalymenus tarsatus, on Sesbania drummondii seeds: Impact on some seed traits related to fitness


Abstract and Figures

Developing seeds of Sesbania drummondii are attacked by nymphs and adults of the bug Hyalymenus tarsatus (Heteroptera: Alydidae), which kill some seeds and weaken others. Parasitism by this piercing-sucking insect reduced the resources for the future seedling and affected seed physiology, including dormancy and exudation of allelochemicals of imbibing seeds. Seeds attacked by H. tarsatus had reduced mass (20-80% reduction, depending on intensity of attack). Heavy attack led to irregular shape, changes in seed coat color, and disruption of dormancy. While intact seeds did not imbibe during a 48-hour test in water, a high proportion of bug-attacked seeds germinated, from 51 to 94%, depending on intensity of attack. Attack by H. tarsatus also affected accumulation of allelochemicals and their exudation by imbibing seeds. There were no quantitative differences in proanthocyanidin content between exudates of attacked and unattacked seeds. In contrast, concentrations of total condensed tannins were higher in exudates of attacked seeds on the third day of imbibition. This change may reflect induction of chemical defenses by herbivore attack and/or a mechanism to restore seed coat impermeability. Although difficult to quantify, effects of sublethal attack by this sucking insect on the seed bank are likely to have important consequences for the demography of S. drummondii, a short-lived perennial in habitats where conditions for recruitment are variable and unpredictable among years.
Content may be subject to copyright.
9 (1) : 28-36 (2002)
Effects of sublethal attack by a sucking insect,
Hyalymenus tarsatus, on Sesbania drummondii
seeds: Impact on some seed traits
related to fitness1
Lilian CEBALLOS2,Centre d’Ecologie Fonctionnelle et Evolutive (CNRS, UPR9056), 1919 Route de Mende, 34293
Montpellier Cedex 5, France.
Claude ANDARY, Laboratoire de Botanique, Phytochimie et Mycologie (CNRS, UPR9056), Faculté de Pharmacie,
15 Avenue C. Flahault, 34060 Montpellier Cedex 2, France.
Centre d’Ecologie Fonctionnelle et Evolutive (CNRS, UPR9056), 1919 Route de Mende, 34293 Montpellier
Cedex 5, France, e-mail:
Abstract:Developing seeds of Sesbania drummondii are attacked by nymphs and adults of the bug Hyalymenus tarsatus
(Heteroptera: Alydidae), which kill some seeds and weaken others. Parasitism by this piercing-sucking insect reduced the
resources for the future seedling and affected seed physiology, including dormancy and exudation of allelochemicals of
imbibing seeds. Seeds attacked by H. tarsatus had reduced mass (20-80% reduction, depending on intensity of attack).
Heavy attack led to irregular shape, changes in seed coat color, and disruption of dormancy. While intact seeds did not
imbibe during a 48-hour test in water, a high proportion of bug-attacked seeds germinated, from 51 to 94%, depending on
intensity of attack. Attack by H. tarsatus also affected accumulation of allelochemicals and their exudation by imbibing
seeds. There were no quantitative differences in proanthocyanidin content between exudates of attacked and unattacked
seeds. In contrast, concentrations of total condensed tannins were higher in exudates of attacked seeds on the third day of
imbibition. This change may reflect induction of chemical defenses by herbivore attack and/or a mechanism to restore seed
coat impermeability. Although difficult to quantify, effects of sublethal attack by this sucking insect on the seed bank are
likely to have important consequences for the demography of S. drummondii, a short-lived perennial in habitats where
conditions for recruitment are variable and unpredictable among years.
Keywords: seed predation, seed size, physical dormancy, seed bank, seed exudation, seedling defenses, condensed tannins,
induced plant defenses, Sesbania drummondii, legume, Hyalymenus tarsatus, Alydidae.
Résumé :Les graines en développement de Sesbania drummondii sont attaquées par les nymphes et les adultes de
Hyalymenus tarsatus (Hétéroptère : Alydidae), qui peut soit les détruire soit les altérer partiellement. Le parasitisme des
graines par cet insecte réduit la quantité de ressources disponibles aux plantules mais influe aussi sur la physiologie des
graines, comme la dormance et la production de substances allélochimiques. L’attaque des graines par H. tarsatus entraîne
une réduction de leur masse (20-80 % de réduction selon le degré d’attaque). Les attaques répétées entraînent des changements
dans la forme et la couleur des graines et provoquent la rupture de dormance. Alors que les graines intactes restent
imperméables durant des tests d’imbibition de 48h, une grande proportion des graines attaquées germent rapidement (entre
51 et 94 % des graines, selon le degré d’attaque). Les attaques par H. tarsatus peuvent aussi affecter l’accumulation de
substances allélochimiques et leur exsudation au cours de l’imbibition des graines. Il n’y a cependant pas de différence dans
le contenu en proanthocyanidines entre les exsudats des graines attaquées et des graines saines. Par contre, la concentration
en tannin condensés totaux est plus élevée chez les graines attaquées lors du troisième jour de l’imbibition. Ce changement
dans la teneur en tanins condensés totaux peut refléter l’induction de défenses chimiques suite à l’attaque par un phytophage
et / ou un mécanisme pour restaurer l’imperméabilité du tégument. Bien que difficiles à quantifier, les effets de ces
attaques sublétales par des insectes suceurs sur le stock de graines au sol doivent être importants sur la démographie de
S. drummondii, espèce pérenne à cycle de vie court occupant des habitats où les conditions pour le recrutement fluctuent
entre années.
Mots-clés : Prédation des graines, taille des graines, dormance, banque de graines, protection des plantules, tannins conden-
sés, défense induites, Sesbania drummondii, légume, Hyalymenus tarsatus, Alydidae.
1Rec. 2001-03-19; acc. 2001-12-29.
2Alternate address: Laboratoire de Botanique, Phytochimie et Mycologie (CNRS,
UPR9056), Faculté de Pharmacie, 15 Avenue C. Flahault, 34060 Montpellier Cedex,
3Present address: Laboratoire de Génétique et de Biologie des Populations Végétales,
Bat. SN2, Université de Lille I, 59855 Villeneuve d’Ascq Cedex, France.
4Present address: Laboratoire d’Ecologie Terrestre (UMR 5552), Université Paul
Sabatier, 118 Route de Narbonne, Bât 4R3, 31062 Toulouse Cedex 4, France.
5Author for correspondence.
Écoscience, Volume 9(1),2362-Ceballos Proofs
ÉCOSCIENCE, VOL. 9 (1), 2002
In plants, various critical life history functions are inte-
grated within seeds, and several seed traits interact to reduce
the effects of environmental variability. Dormancy and dis-
persal of seeds allow them to escape unfavorable conditions
in time and space (Venable & Brown, 1988). These risk-
reducing traits are substitutable under certain conditions,
and trade-offs between dispersal and dormancy have been
found (Templeton & Levin, 1979). Seed banks associated
with dormancy buffer the negative effects of bad years,
limit fluctuations of population size, decrease the risk of
extinction, and potentially increase the genetic diversity of
populations (Venable, 1989; Levin, 1990; Baskin & Baskin,
For many plants, most seeds do not germinate immedi-
ately when mature, but remain dormant for long periods.
Dormancy may be imposed by physiological inhibition of
germination of the embryo and/or by physical mechanisms
(such as impermeable seed coats). Physical dormancy is
found in many families and has evolved many times inde-
pendently (Baskin & Baskin, 1998). Persistence of seeds in
soil seed banks makes them an attractive and predictable
resource for various predators. Accordingly, selection has
favored the evolution of various resistance mechanisms in
seeds capable of long periods of dormancy. In legume
seeds, for example, dormancy is caused by progressive
dehydration of seed tissues and impermeabilization of the
seed coat, preventing water uptake by dormant seeds. Seed
coats of many legumes contain high concentrations of tan-
nins, which deter granivores and contribute to making seed
coats impermeable to water and oxygen (Egley et al., 1985).
They thus physically impede the entry of pathogens (Kulik
& Yaklich, 1991; Helsper et al., 1994) and also protect seed
lipid and protein reserves from oxidative degradation. These
protective mechanisms have been studied extensively in
seeds of legumes because of the economic interest of this
family. Some legumes have seeds capable of prolonged
periods of dormancy (e.g., 158 years for Cassia multijuga;
Baker, 1989).
Among the threats to seed coat integrity are attacks by
insects that pierce and suck developing or mature seeds.
Attacks by such insects, if sufficiently severe, can kill
seeds. Those that survive will have reduced nutrient
reserves. Piercing-sucking insects may also introduce
pathogens that kill or weaken dormant seeds or germinat-
ing seedlings (Kremer & Spencer, 1989). To the best of our
knowledge, few studies have examined the effects of
attacks by such insects on permeability of the seed coat and
thus on the seed’s capacity for dormancy (Karban and
Lowenberg, 1992).
This study examines the effects of attack of seeds of
Sesbania drummondii (Rydb.) Cory (Fabaceae: Papil-
ionoideae) by a piercing-sucking insect, Hyalymenus tarsat-
us (F.) (Heteroptera: Alydidae, subfamily Alydinae). Most
members of this subfamily feed on seeds, principally of
legumes, but their hosts also include other plant families
(Schaefer, 1980). Many species are polyphagous on
legumes, and some are important pests on pulse crops
(Yonke & Medler, 1968; Aina, 1975). As in other alydines,
adults and nymphs of H. tarsatus pierce immature pods
with the proboscis and suck the reserves of developing
seeds. The fate of the attacked seed depends on the intensity
of attack and on the rapidity of seed response. The bugs can
kill seeds when repeated attacks deplete seed nutrients or
kill the embryo (predation). When intensity of attack is
lower, seeds can often mature. However, these non-lethal
attacks might have negative effects on seeds (parasitism).
Because seed size can influence seedling establishment
(Leishman & Westoby, 1994), a reduction in the reserves
stored in the seed should reduce fitness. Little is known
about how developing seeds tolerate attacks by potential
seed predators (Rosenthal & Kotanen, 1994; Mack, 1998).
Not only can reserves of S. drummondii seeds be reduced by
bug attack, but damage to the seed coat could facilitate
transmission of pathogens (Harman, 1983; Mills, 1983).
Alydids have been implicated in transmission of seed-borne
diseases (Yonke & Medler, 1968). Furthermore, damage by
bugs could increase seed coat permeability and disrupt seed
dormancy (Van Staden, Manning & Kelly, 1989).
In this study, we characterize alterations of physiological
mechanisms in attacked seeds and attempt to understand
how seed parasitism affects the physiology of young
seedlings. First, we determine seed mass as a function of the
class of attack described by a semi-quantitative scale.
Secondly, we compare seed coat morphology and rates of
spontaneous germination in attacked and unattacked seeds.
This comparison suggests that parasitism affects the devel-
opmental and physiological processes involved in imperme-
abilization of the seed coat. Thirdly, we describe plant
response to bug attack, focusing on profiles of the polyphe-
nolic compounds that may explain the observed differences
between attacked and intact seeds in both seed coat color
and permeability. Finally, we discuss the potential impact of
sublethal attacks on seed fitness.
Sesbania drummondii is a perennial shrub that grows in
humid areas (lakeshores, riversides, rice fields, and similar
habitats) along the Gulf coastal plain, from Florida to Texas
and south to the state of Veracruz in Mexico. It is often con-
sidered a weed. It can reach 4 m in height and survive for
10 years. Seeds of S. drummondii are spherical, about
110 mg in mass (Marshall, 1982), and can stay dormant for
10 years or more in the soil, where they form a seed bank.
Seeds are protected by an indehiscent pod with the consis-
tency of tough paper, which contains 3 to 8 seeds (mean 6).
The pods possess a winged pericarp. They float and are dis-
persed by water, blown by wind like sailboats. Fruits and
seeds used in this study were collected in December 1995
on parent plants from natural populations in the coastal
plain of Texas. Dry fruits were stored in paper bags in dark-
ness at 10°C until their use in this study, conducted in
April-May 1998.
Three populations were studied: LBJ (for Lyndon
Baines Johnson) State Park (population 1: 33° 18’ N,
100° 25’ W); Gerhardt ranch (population 2), near Vienna
(29° 45’ N, 96° 94’ W); and Lake Texana (population 3),
near Edna (28° 58’ N, 90° 38’ W). To analyze inter-individ-
ual variation in the intensity and effects of bug attack, we
sampled 10 plants in each population. For each plant, 10
pods, each containing 6 seeds, were sampled (1,800 seeds in
total) to take into account within-individual variation. We
chose only pods containing 6 seeds in order to control for a
possible effect of seed number per pod in our observations
and experiments.
Hyalymenus tarsatus is a wide-ranging species and has
been recorded on Sesbania spp. and on undetermined
mimosoid legumes in Texas and Mexico (Schaefer, 1980)
but feeds principally in infructescences of Asteraceae in
Brazil (Oliveira, 1985). We observed H. tarsatus only on S.
drummondii in this study, but other local hosts are not
excluded. It does not, however, feed on S. vesicaria (Jacq.),
which occurs abundantly in the study area. Adults are rapid
fliers. Life history, duration of development of immature
stages, and movements of adults appear to be unknown.
Both adults and the ant-mimetic nymphs (Oliveira, 1985)
feed on developing seeds of S. drummondii. Seeds in mature
fruits, with lignified seed coats and pod walls, are not fed
upon. We never observed H. tarsatus feeding on seeds or
pods on the ground beneath the plant. Field observations
show that bugs arrive in numbers at the time when young
pods first begin to appear in large numbers and that bug
density varies among individual host plants. However, noth-
ing is known about host orientation and the cues determin-
ing host recognition and host preference.
This analysis, conducted in April-July 1998, aimed at
characterizing the impact of parasitism of H. tarsatus on
easily defined seed morphological traits. Pods with six
seeds were carefully opened, and the position of each seed
within the pod was noted from 1 (closest to the peduncle) to
6 (farthest from the peduncle). We defined different classes
of attack as follows: 0 (aborted seeds), 1 (unattacked seeds),
2 (slightly attacked seeds; bearing usually 1-2 puncture
marks, visible as tiny brown spots), 3 (moderately attacked
seeds; usually with 3-5 punctures), 4 (heavily attacked
seeds; usually 6 punctures), and 5 (seeds completely
decomposed). For each seed, we noted the position, class of
attack, color of the seed coat (cream or yellow-green,
brown, red) and the shape (normal or misshapen). Seeds
were weighed individually to the nearest mg.
Variation in the intensity of seed attack was analyzed
using a type III ANOVA with multinomial error (GEN-
MOD procedure of SAS [SAS Institute, 1999]). The full
model included as explanatory variables the effects of
source population, of individual (nested within population),
of pod (nested within individual), and of position of seeds
(1 to 6). Backwards selection was used to identify the fac-
tors with the most significant effects in the full model.
The same type of ANOVA was performed on the color
and shape of seeds as a function of intensity of attack. This
analysis was restricted to population 2, the only population
in which seeds of all attack classes were present in suffi-
cient number to carry out the analysis. The full model
included the effects of individual, of pod (nested within
individual), and of seed position, as well as the intensity of
attack of each seed. As in the previous ANOVA, backwards
selection was used to identify the factors with the most sig-
nificant effects in the full model.
Data on seed mass were analyzed using type III
ANOVA (GLM procedure of SAS [SAS Institute, 1999]) in
a backwards selection model using the same explanatory
variables as in the analyses of seed color and shape.
Comparisons of mean seed mass among populations and
among different degrees of attack were performed using a
Tukey-Kramer test (P<0.05).
Three lots of 100 seeds from population 2 (10 replicates
of 10 seeds for each population) were set to imbibe at room
temperature (20°C) in petri dishes (100% humidity): unat-
tacked seeds (class 1), attacked seeds (classes 2 and 3), and
heavily attacked seeds (class 4). After 48 hours, the number
of imbibing seeds, indicated by protrusion of the radicle,
was counted. Dormancy in seeds of S. drummondii is physi-
cal, and once the radicle emerges, the seed is committed to
germination. The effect of attack level on permeability of
seeds was analyzed using a type III ANOVA with multino-
mial error (GENMOD procedure of SAS [SAS Institute,
1999]) and comparisons of mean number of seeds germinat-
ed were performed using the Tukey-Kramer test (P<0.05).
For this experiment, from each of six different plants
we chose a lot of 120 intact seeds and another lot of 120
attacked seeds, giving a total of 12 lots. For attacked seeds,
moderately attacked seeds (classes 2 and 3) were used
because seeds from classes 4 and 5 were heavily infested by
fungi. Fungal infection could have had confounding effects
on exudation of allelochemicals.
Each of the 12 lots was arbitrarily divided into four 30-
seed subsamples, and each of these subsamples was further
divided into three replicates of 10 seeds each. Seeds were
mechanically scarified with a razor blade next to the lens,
the part where papilionoid seeds usually become permeable
(Baskin & Baskin, 1998). The exact mass of each seed and
the total mass of each 10-seed replicate were noted.
Scarified seeds were then soaked in water (10 seeds per 100
ml distilled water). To examine exudation over the time
course of imbibition and to avoid any effect of removal of
exudate on the exudation process, exudate from each of the
four arbitrary subsamples was collected only once, on day
1, 2, 3, or 4 of imbibition. Extracts from each replicate were
freeze-dried and weighed separately (results expressed in
mg exudates g dormant seeds-1).
The rate of exudation was analyzed using a type III
ANOVA (GLM procedure of SAS [SAS Institute, 1999]) in
a backwards selection model with the time (day of imbibi-
tion), the degree of attack, and the individual plant as
explanatory variables. Comparisons of the dry mass of the
exudates released by seed at different time of imbibition
were performed using a Tukey-Kramer test (P<0.05).
Imbibition extracts were dissolved in 70% acetone to
the final concentration of 1% (Weight/Volume), then cen-
trifuged at 10,000 rpm. Extracts were analyzed using a
spectrophotometer (PYE/UNICAM/Vis®) to quantify con-
tents of both total condensed tannin and proanthocyanidins.
Condensed tannins, as measured by our analyses, are consti-
Ville et province (ou état) de la
ÉCOSCIENCE, VOL. 9 (1), 2002
tuted of both monomers (catechins) and polymers.
Proanthocyanidins correspond to the polymerised portion of
the condensed tannins. These oligopolymers account for
much of the activity of condensed tannins. It is useful, how-
ever, to take into account all condensed tannins, including
monomers, because the latter are sometimes predominant
and have particular biological effects.
A 0.1-ml aliquot of acetonic extract was mixed with
0.5 ml of a solution of 0.1% DMCA (dimethylaminocin-
namaldehyde) in a methanol and 12N HCl (3:1) mixture.
DMCA reacts specifically with catechin monomers and oli-
gopolymers of catechins to yield a complex, blue-green in
colour, whose maximum absorbance is measured at 640 nm,
after 10 minutes of contact (Treutter, Freucht & Santos-
Buelga et al., 1994). The control was an acetone blank.
Concentrations were expressed as mg of catechin equiva-
lents per g dormant seeds.
A 0.15-ml aliquot of acetonic extract was mixed with
0.9 ml of butanol-HCl (90:10) with 50 mg of Fe3+ added as
a catalyst, in sealed tubes. After being shaken vigorously
(Vortex), tubes were heated for 50 seconds in a microwave
oven (power output 900W). After 10 minutes of cooling,
PAs (in acid solution and in the presence of Fe as a catalyst)
hydrolyze to give molecules of anthocyanins, absorbance of
which is measured at 545 nm and compared to an acetone
blank (Treutter, Freucht & Santos-Buelga et al., 1994).
Concentrations were measured as mg cyanidin equivalents
per g dormant seeds.
Solubility in water and biological activity of exuded
polyphenols depend partly on their molecular weight, i.e.,
their degree of polymerization. PD was obtained by divid-
ing PA concentrations (BuOH/HCl assay) by CT
concentrations (DMCA assay). The higher the PD, the
larger the size of PAs in the extract. For each sample (N= 6
replicates sample-1), concentrations of CT and PA, and
polymerization degree, were compared and differences
within samples were tested for statistical significance, using
a two-factor (degree of attack, individual from which seeds
were collected) ANOVA (GLIM, 1987) and LSD tests
(P<0.05, STATISTIX, 1994). Data on polymerization
degree were analyzed after normalizing them by arcsin
square root transformation.
Usually, seeds were attacked by H. tarsatus during
intermediate stages of seed development, i.e., when seed
coats were not yet lignified but seeds were large enough to
contain substantial amounts of nutritive reserves. The
immediate response of individual seeds to attack involved
secretion of a whitish latex that formed a plug. In the fol-
lowing minutes, this plug turned tan and hardened, sealing
the opening. The greater the intensity of attack on develop-
ing seeds, the more pronounced the alteration of their mor-
phology was. Seeds attacked were small, misshapen, and
fragile. Nevertheless, many of them gave positive reac-
tions in tetrazolium tests for viability (Ceballos, unpubl.
data), and a proportion of such seeds are capable of giving
healthy seedlings. When a developing seed was submitted
to numerous and/or prolonged attacks, its seed coat turned
red (possibly due to production of quinones) and lignified
precociously, at a stage when seed filling was not yet com-
plete. Parasitism thus affected at least some biochemical
processes in developing seeds, those responsible for pig-
mentation and hardening of the seed coat.
Intensity of seed attack by H. tarsatus varied signifi-
cantly among populations studied (Genmod, ANOVA:
χ22= 1,182.16, P< 10-4; Table I). Population 1 had very lit-
tle attack, while population 2 had a very high level, and
population 3 a moderate level of attack (Figure 1). Also,
within each population, there was great variation in intensi-
ty of attack among individuals (χ218 = 258.71, P< 10-4;
Table I) and among pods within individual plants
(χ290 = 427.40, P< 10-4; Table I). Position of seeds within
the pod showed no variation in attack. None of the interac-
tions between factors tested were significant.
The ANOVA performed on data for population 2
showed that intensity of attack significantly influenced the
color (χ24= 726.30, P < 10-4) and shape (χ24= 383.49,
P<10-4). Unattacked seeds (class 1) were creamy-white or
yellow-green in color and never misshapen; slightly
attacked seeds (class 2) were mostly (about 95%) brown in
color and 7% were misshapen; moderately attacked seeds
(class 3) were brown (66%) or red (34%), with 39% mis-
shapen; heavily attacked seeds (class 4) were almost all red
(97%) and misshapen (84%); seeds that had been heavily
attacked and obviously fungus-infected (class 5) were with-
out exception red in color, and almost all (97%) were mis-
The effect of attack on seed mass was highly signifi-
cant (F4,1670 = 32.58, P < 10-3; Table II), with mean mass
of seeds decreasing as intensity of attack increased. Mean
mass of unattacked seeds (class 1) was 104.7 ±28.3 mg.
Mean mass of seeds of attack class 2 (102.7 ±52.0 mg) was
not significantly different from that of class 1, but was more
variable. Mean mass progressively decreased in seeds of
classes 3 (80.7 ±23.1 mg) and 4 (37.9 ±23.1 mg). Finally,
mean mass of heavily attacked and fungus-infected seeds
(class 5) was 22.2 ±16.1 mg. The effects of population and
of individual (within population) were also significant, par-
tially due to significant interactions of these variables with
the degree of attack (Table II). Mass of seeds from popula-
tion 2 (mean = 47.0 ±47.6 mg) was much lower than in the
other populations (population 1: mean = 102.4 ±17.0 mg;
Literature cited:1985.
TABLE I. Results of the ANOVA of degree of attack of seeds of
S. drummondii by H. tarsatus (multinomial error, GENMOD
procedure of SAS [SAS Institute, 1999]).
Source df χ2P
Population 2 1,182.96 < 10-4
Individual (population) 18 258.71 < 10-4
Pod (individual) 90 427.40 < 10-4
population 3: mean = 103.9 ±42.8 mg). This difference was
due to the fact that level of attack was higher in population
2. When only unattacked seeds (class 1) were compared
among these three populations, mass was not significantly
different (F2,713 = 1.41, P = 0.245).
The effect of parasitism on permeability of seeds was
highly significant (χ22= 65.92, P< 10-4). None of the intact
seeds imbibed water during the 48 hours of the test, whereas
an average of 51.0 ±7.4% of slightly attacked seeds (class 2
and 3) and 94.0 ±8.4% of heavily attacked seeds (class 4)
imbibed spontaneously, triggering their germination. Table
III shows that differences were significant in each pairwise
comparison of attack classes.
Mass of exuded matter did not differ significantly
among seeds from the six maternal parents (F5,44 = 0.54,
P= 0.75) or between intact and attacked seeds
(F1,44 = 2.61, P = 0.11). Thus, attack caused no quantitative
change in the mass of exudates. In contrast, there was a
strong time effect on exudation of seeds (F3,47 = 12.81,
-5) (Figure 2). The difference between successive his-
tograms indicates the mean additional mass exuded during
each day. Comparison of means (LSD test, P< 0.05)
showed significant differences in mass of matter exuded
between the different days. Rate of exudation was about
equal on days 1 and 2, but a smaller mass of material was
exuded between days 2 and 3, and an even smaller amount
between days 3 and 4.
Condensed tannin concentrations in the exudates varied
significantly over time (F3,15 = 19.74, P= 2. 10-5), increas-
ing from day 1 to day 3 of imbibition, then decreasing on
day 4 (Figure 3a). Overall, attacked seeds exuded signifi-
cantly more tannins than did unattacked seeds (F1,15 = 5.2,
P= 0.04). This difference was consistently observed over
the entire time course of imbibition, but for each day taken
singly, only the difference on day 3 was significant. Seeds
produced by different maternal parents did not differ in con-
centrations of CT. None of the interaction terms (between
individual, time and degree of attack) were significant.
As for condensed tannins, the effect of time on exuda-
tion of proanthocyanidins (PA) was highly significant
(Figure 3b; F3,15 = 60.21; P< 10-5). PA concentration con-
tinued to increase in the exudates over the 4 days, but
only the difference between days 1 and 4 (the beginning
and end of the experiment) was significant. In contrast to
the results for CT, seeds produced by different individual
plants differed significantly (F5,15 = 5.34; P= 0.01) in
proanthocyanidin (PA) concentrations, contributing to the
much greater standard deviations for PA (Figure 3b) than
for CT (Figure 3a). While degree of attack did not signifi-
cantly affect concentrations of proanthocyanidins exuded,
two interaction terms (time•individual and attack•individ-
ual) were marginally significant, showing that individual
plants varied in the time course of PA exudation and in
their response to attack.
Polymerization degree (PD) is the ratio PA/CT, and
patterns for this variable thus reflected those of its two com-
Population 1
Class of attack
Population 2
020 40 60 80 100
Population 3
60 80 100
Percentage of seeds
FIGURE 1. Frequency of different classes of attack of seeds by Hyalymenus tarsatus in the three studied populations of Sesbania drummondii. Class
1=unattacked; 2 = slightly attacked; 3 = moderately attacked; 4 = heavily attacked; 5 = seeds heavily attacked and obviously fungus-invested.
TABLE II. Results of the ANOVA (Type III) of seed mass of
Sesbania drummondii (GLM procedure of SAS [SAS Institute,
Effect FP
Attack F4, 1670 = 32.58 < 10-4
Population F2, 1670 = 15.50 < 10-4
Individual (population) F27, 1670 = 1.66 0.018
Population•attack F8, 1670 = 2.29 0.020
Individual•attack F72, 1670 = 1.55 0.003
TABLE III. Effect of attack by Hyalymenus tarsatus on permeabili-
ty of seeds. Values given are mean (±SD) percentages of seeds of
three attack classes (see text for definition of classes) that imbibed
and germinated when placed in water.
Seeds Mean (%) Tukey-Kramer test*
Unattacked 0 a
Moderately attacked 51.00 (±7.38) b
Attacked 94.00 (±8.43) c
*Different letters in the rightmost column indicate significant differences
(comparison of means, Tukey-Kramer test, P<0.05).
ÉCOSCIENCE, VOL. 9 (1), 2002
ponents (Figure 3c). There were significant differences
among individuals (F5,15 = 3.7, P= 0.02) and the time
effect was very strong (F3,15 = 70.99, P< 10-5). PD changed
little during the first three days, but increased abruptly on
the fourth day (Figure 3c). Tannins were apparently subject-
ed to marked polymerization on the fourth day of imbibi-
tion. Effect of attack on PD of exudates was not significant,
and none of the interaction terms were significant.
A large proportion of developing seeds of S. drum-
mondii that were fed on by adults and nymphs of H.
tarsatus survived these attacks. However, feeding by this
bug affected several fitness-relevant traits of seeds. Bug
attack was associated with reduced mass of mature seeds,
and the strength of the effect depended on the intensity of
attack. Severe bug attack on developing seeds also led to
loss of seed dormancy. Finally, seed attack was associated
with changes in the metabolism of polyphenolic compounds
of seed coats, resulting in changes in seed coat color and in
increased concentration of condensed tannins exuded by
seeds during imbibition. Reduced seed mass is likely to
have strong negative effects on fitness. While loss of dor-
mancy could conceivably benefit some seeds, its conse-
quences overall are also likely to be strongly negative. The
effect of sublethal attack on seeds by this insect can thus be
described as parasitism. The observed effects on metabolism
of polyphenolics suggest that the plant may have a limited
capacity for adaptive response to bug attack.
Parasitism of developing seeds of S. drummondii by H.
tarsatus resulted in substantial reduction in the amount of
stored reserves, which should reduce the probability of
establishment of seedlings upon emergence. The mecha-
nisms behind this effect may be complex. In addition to
removing reserves already stored in the seed, seed-feeding
bugs may disrupt development and reduce further storage of
reserves (Bates et al., 2001). Parasitism by H. tarsatus also
results in misshapen seeds. Attack appears to lead to preco-
cious lignification of the seed coat, which may physically
limit the volume available for stored reserves. Whatever the
mechanisms behind it, reduced seed mass leads to decreased
fitness under competition (Winn, 1988). Reduced seed mass
is also likely to have a strong negative effect when a fluctu-
ating and unpredictable abiotic environment leaves only a
narrow window of time for establishment. Seedlings of S.
drummondii establish in habitats characterized by strongly
Time (days)
Dry mass of exudates (mg)
bc c
FIGURE 2. Cumulative dry mass of exudates released by seeds of
Sesbania drummondii after 1, 2, 3, and 4 days of imbibition. Means and
SD of 12 lots of 120 seeds each. Different letters above histograms indi-
cate significant differences (comparison of means, Tukey-Kramer tests,
P< 0.05).
FIGURE 3. Total concentrations of phenolic compounds exuded by
attacked and unattacked imbibing seeds of Sesbania drummondii after 1,
2, 3, and 4 days of imbibition: (a) total condensed tannins (CT), concen-
tration expressed as catechin equivalents, in mg g of seeds-1; (b) proantho-
cyanidins (PA), concentration expressed as cyanidin equivalents, in
mg g of seeds-1; (c) degree of polymerization (PD). For each day, means
and standard deviations for 6 individuals are given (30 seeds per individual
per day for each of the two attack classes). Different letters above his-
tograms indicate significant differences (comparison of means, LSD
tests, P< 0.05).
Mean concentration
Unattacked seeds
Attacked seeds
a) Total condensed tannins
Mean concentration
b) Proanthocyanidins
Time (days)
Degree of polymerization
c) Polymerization degree
Not in Literature cited.
fluctuating water level and must grow quickly to a size
that enables them to survive flooding. This rapid initial
growth is supported by protein and energy reserves stored
within the seeds (and by the photosynthetic cotyledons of
seedlings). Substantial losses of seed reserves could thus
greatly reduce seedling growth and survival.
Parasitism of seeds of S. drummondii by H. tarsatus
also had a striking effect on seed coat permeability. In
legumes, dormancy results from development of imperme-
ability of the seed coat. Seeds whose coats were pierced by
this bug became permeable to water. The extent of this
effect, in terms of rupture of dormancy, depended on inten-
sity of attack. Even at moderate intensity of parasitism,
however, physical dormancy was disrupted, resulting in
spontaneous imbibition by 51% of seeds. In the most heavi-
ly attacked classes, this proportion exceeded 90%, meaning
a near absence of dormancy for these seeds. Our findings
are similar to those of Karban and Lowenberg (1992), who
found that attack by lygaeid and scutellerid seed bugs
removed the physical dormancy of seeds of Gossypium spp.
(Malvaceae). The ecological and evolutionary consequences
of such enhancement of germination are difficult to pre-
dict and may not necessarily be negative (Karban and
Lowenberg, 1992). More rapid germination may sometimes
confer an advantage. However, some effects are likely to be
decidedly negative. Attacks on the integrity of the seed
coat may facilitate pathogen infection or initiate biochemi-
cal changes that reduce seed fitness. Impermeability of
dormant seeds to oxygen protects lipid and protein reserves
from oxidative degradation (peroxidation, conformational
changes of proteins, etc.). Oxygen entering pierced seed
coats leads to deterioration, characterized by reduction in
levels of proteins, lipids, ATP pools, RNA, and polysaccha-
rides (Anderson & Baker, 1983). As a consequence, the
metabolism of seedlings is seriously disrupted.
Loss of seed dormancy is also likely to reduce fitness in
this plant. Conditions for seedling establishment vary great-
ly among seasons and years as a consequence of unpre-
dictable fluctuations of water level in the temporarily flood-
ed habitats where this plant grows. Dormancy allows seeds
to survive the large proportion (in most sites) of years that
are either too dry or too wet for seedling establishment.
While our study did not address effects of bug attack on
biochemical processes affecting seed contents, it showed
that attack affected metabolic processes in the developing
seed coat. Seed coat color at maturity was conspicuously
affected. The higher the intensity of attack, the redder the
seed coat was. This pattern suggests that attack on develop-
ing seeds led to oxidation of phenols to quinones. Werker,
Marbach and Mayer (1979) reported the presence of quinones
in the palisade layer of mature legume seeds and suggest-
ed that reticulation of tannins to polysaccharides of
macrosclereid cell walls involved the oxidation of phenols
to quinones. The response of developing seeds to intense
bug attacks suggests that attack may trigger the early
expression of a normal developmental program that renders
seeds impermeable. Such an effect could be interpreted in
two ways. It might simply reflect damage-caused disruption
of development. Alternatively, it could be an adaptive repair
response to restore impermeability. If the second alternative
holds, then the mechanism appears to be of limited effec-
tiveness when levels of attack are high.
Parasitism of seeds by H. tarsatus influenced biochemi-
cal profiles of seed exudates, being associated with an
increase in the amount of condensed tannins exuded during
imbibition, an increase that became significant the third day
of imbibition (Figure 3). Seed coat condensed tannins are
known to be involved not only in defense (Graham, 1991;
Kulik & Yaklich, 1991; Helsper et al., 1994), but also in
processes of impermeabilization (Marbach & Mayer, 1975;
Egley et al., 1985). They also affect seed coat color (Powell,
Oliveira & Matthews, 1986; Cabrera & Martin, 1989), and
may have contributed to the color effect noted above.
Our data support the hypothesis that parasitism of
developing seeds results in the induction of increased pro-
duction of condensed tannins in the seed coat. Because
seed-coat tannins are regarded as constitutive components
of these organs, induction of their synthesis in response to
environmental stresses or signals has been little examined.
Our data suggest that it would be interesting to examine this
question in other plant species.
Proanthocyanidin concentration in the exudates, in con-
trast, was not significantly correlated with level of seed par-
asitism, although there was a non-significant trend in this
direction. Polymerization of catechins began on the third
day and experiments ran only four days; a longer duration
of imbibition might have revealed an effect. PA concentra-
tion appeared to respond differently to attack in different
individuals, with two of the six individuals tested containing
lower concentrations of PA in attacked lots.
Finally, parasitism did not affect PD, which reflects the
size of oligopolymeric PAs. PD (and thus the size of tan-
nins) increased over the time course of imbibition (Figure 3c).
Mirroring variation in one of its components (PA), PD was
variable among individuals. Exudation of tannins by imbib-
ing seeds is a dynamic process, with continual release and
oxidative condensation of tannins over the course of
imbibition (L. Ceballos, unpubl. data). This polymerization
of exuded tannins should enhance their overall biological
activity against pathogens. Antifungal activity of S. drum-
mondii exudates remains significant over 9 days (Ceballos
et al., 1998). The fact that polymerization during imbibition
was not affected by parasitism of seeds suggests that the
mechanisms behind it were neither negatively affected by
damage to developing seeds nor enhanced by induction.
In contrast to seed predation, the impact that seed-
attacking insects have on fitness when they do not kill seeds
outright has been relatively little studied. Compared to the
effects of chewing insects, the effects of sucking insects on
plants are in general difficult to quantify, and the effects of
piercing-sucking insects such as H. tarsatus on seeds that
survive attack will be more difficult to quantify than those
of chewing insects such as bruchids and weevils (Janzen,
1976). For unknown reasons, intensity of bug attack var-
ÉCOSCIENCE, VOL. 9 (1), 2002
ied greatly among the three populations of S. drummondii
we studied and among individuals within populations.
Observations also suggested variation in attack among fruits
of the same individual plant. This variation is likely to
affect the relative importance of seed predation and seed
parasitism as outcomes of the interaction, with parasitism as
the predominant effect under lower attack intensities. The
effects of seed parasitism are likely to be cryptic unless
studies are specifically designed to incorporate effects of
seed mass and dormancy on seedling survival.
We thank J., H., and L. McKey for help in collecting pods, J.
and H. McKey and P. and F. Gerhard for allowing access to S.
drummondii populations on their property, A. Caizergues for
help in statistical analysis, J. Schaffner (Department of
Entomology, Texas A&M University, emeritus) for identifying
H. tarsatus; and the French Government (MENRT) for a doctoral
grant to L. Ceballos.
Literature cited
Aina, J. U., 1975. The life history of Riptortus dentipes F.
(Alydidae, Heteroptera), a pest of growing cowpea pods.
Journal of Natural History, 9: 589-596.
Anderson, J. D. & J. E. Baker, 1983. Deterioration of seeds during
aging. Phytopathology, 73: 321-325.
Baker, H. G., 1989. Some aspects of the natural history of seed
banks. Pages 15-27 in M. A. Leck, V. T. Parker & R. L.
Simpson (ed.) Ecology of Soil Seed Banks. Academic Press,
San Diego, California.
Baskin, C. C. & J. M Baskin, 1998 (ed.). Seeds: Ecology,
Biogeography, and Evolution of Dormancy and Germination.
Academic Press, San Diego, California.
Cabrera, A. & A. Martin, 1989. Genetics of proanthocyanidin con-
tent and its relationship with flower and testa colours in Vicia
faba L. Journal of Agricultural Science, 113, 93-98.
Ceballos, L., M. Hossaert-McKey, D. McKey & C. Andary, 1998.
Rapid deployment of allelochemicals in exudates of germinat-
ing seedsof Sesbania (Fabaceae): Roles of seed anatomy and
histolocalization of polyphenolic compounds in anti-pathogen
defense of seedlings. Chemoecology, 8: 141-151.
Egley, G. H., R. N. Paul Jr, S. O. Duke & K. C. Vaughan, 1985.
Peroxidase involvement in lignification in water-impermeable
seed coats of weedy leguminous and malvaceous species. Plant,
Cell and Environment, 8: 253-260.
Graham, T. L., 1991. Flavonoids and isoflavonoids distribution in
developing soybean seedling tissue and in seed and root exu-
date. Plant Physiology, 95: 594-603.
GLIM, 1985. Royal Statistical Society, London.
Harman, G. E., 1983. Mechanisms of seed infection and pathogen-
esis. Phytopathology, 73: 326-329.
Helsper, J. P. F. G., A. Van Norel, K. Burger-Meyer & J. M.
Hoogendijk, 1994. Effect of the absence of condensed tannins
in faba beans (Vicia faba) on resistance to foot rot, Ascochyta
blight and chocolate spot. Journal of Agricultural Sciences,
123: 349-355.
Janzen, D. H., 1976. Reduction of Mucuna andreana
(Leguminosae) seedling fitness by artificial seed damage.
Ecology, 57: 826-828.
Karban, R. & G. Lowenberg, 1992. Feeding by seed bugs and
weevils enhances germination of wild Gossypium species.
Oecologia, 92: 196-200.
Kremer, R. G. & N. R. Spencer, 1989. Impact of seed-feeding
insect and microorganisms on velvetleaf (Abutilon theophrasti)
seed viability. Weed Science, 37: 211-216.
Kulik, M. M. & R. W. Yaklich, 1991. Soybean seed coat structure:
Relationship to weathering resistance and infection by the fun-
gus Phomopsis phaseoli. Crop Science, 31: 108-113.
Leishman, M. R. & M. Westoby, 1994. Hypotheses on seed size:
Tests using the semiarid flora of western New South Wales,
Australia. American Naturalist, 143: 890-906.
Levin, D. A., 1990. The seed bank as a source of genetic novelty
in plants. American Naturalist, 135: 563-572.
Mack, A. L., 1998. Advantage of large seed size: Tolerating rather
than succumbing to seed predators. Biotropica, 30: 604-608.
Marbach, I. & A. M. Mayer, 1975. Changes in catechol oxidase
and permeability to water in seed coats of Pisum elatius during
seed development and maturation. Plant Physiology, 56: 93-96.
Marshall, D. L., 1982. The nature and consequence of variation in
the components of yield in Sesbania macrocarpa, S. drum-
mondii and S. vesicaria. Ph.D. Thesis, University of Texas.
Austin, Texas.
Mills, J. T., 1983. Insect fungus associations influencing seed dete-
rioration. Phytopathology, 73: 330-334.
Oliveira, P. S., 1985. On the mimetic association between nymphs
of Hyalymenus spp. (Hemiptera: Alydidae) and ants.
Zoological Journal of the Linnean Society, 83: 371-384.
Powell A. A., M. Oliveira & S. Matthews, 1986. Seed vigour in
cultivars of dwarf French beans (Phaseolus vulgaris) in relation
to the colour of the testa. Journal of Agricultural Science, 106:
Rosenthal, J. P. & P. M. Kotanen, 1994. Terrestrial plant tolerance
to herbivory. Trends in Ecology and Evolution, 9: 145-148.
SAS Institute, 1999. SAS System, Version 8. Cary, North Carolina.
Schaefer, C. W., 1980. The host plants of the Alydinae, with a note
on heterotypic feeding aggregations (Hemiptera: Coreoidea:
Alydidae). Journal of the Kansas Entomological Society, 53:
Stabell, E., M. K. Upadhyaya & B. E. Ellis, 1996. Development of
seed coat-imposed dormancy during seed maturation in
Cynoglossum officinale. Physiologia Plantarum,97: 28-34.
STATISTIX, 1994. Version 4.0 Analytical software. Tallahassee,
Templeton, A. R. & D. A. Levin, 1979. Evolutionary conse-
quences of seed pools. American Naturalist, 114: 232-249.
Treutter, D., W. Freucht & C. Santos-Buelga, 1994. Determination
of catechins and procyanidins in plant extracts: A comparison
of methods. Acta Horticulturae, 381: 789-796.
Van Staden, J., J. C Manning & K. M. Kelly, 1989. Legume seeds:
The structure:function equation. Pages 417-449 in C. H. Stirton
& J. L. Zarucchi (ed.). Advances in Legume Biology.
Monographs in Systematic Botany of the Missouri Botanical
Garden, 29. Missouri Botanical Garden, St. Louis, Missouri.
Venable, D. L., 1989. Modeling the evolutionary ecology of seed
banks. Pages 67-87 in M. A. Leck, V. T. Parker & R. L.
Simpson (ed.). Ecology of Soil Seed Banks. Academic Press,
San Diego, California.
Venable, D. L. & J. S. Brown, 1988. The selective interactions of
dispersal, dormancy and seed size as adaptations for reducing
risks in variable environments. American Naturalist, 131:
Écrire au long s.v.p.
Ecrire au long s.v.p. Changer le texte en con-
séquence de donner le titre ou nom du produit.
Werker, E., I. Marbach & A. M. Mayer, 1979. Relation between
the anatomy of the testa, water permeability and the presence of
phenolics in the genus Pisum. Annals of Botany, 43: 765-771.
Winn, A. A., 1988. Ecological and evolutionary consequences of
seed size in Prunella vulgaris. Ecology, 69: 1537-1544.
Yonke, T. R. & J. T. Medler, 1968. Biologies of three species of
Alydus in Wisconsin. Annals of the Entomological Society of
America, 61: 526-531.
TABLE IV. Results of the ANOVA of concentration of proantho-
cyanidins (PA) in seed exudates (GLM procedure of SAS [SAS
Institute, 1999]).
Effect FP
Time F3, 15 = 60.21 < 10-5
Individual F5, 15 = 5.34 < 0.01
Attack F1, 15 = 0.078 ns
Individual•attack F5, 15 = 2.95 0.047
Individual•time F15, 15 = 2.58 0.038
Ce tableau n’a pas été cité!
Date : Signature : Node téléphone / Phone No. :
Numéro de commande :
Purchase order number:
Auteur (s) / Author (s):
Titre / Title:
Numéro du manuscrit :
Manuscript number :
Tel qu’ indiqué dans les directives aux auteurs, des frais de publication de 50 $CAD par page sont facturés aux auteurs à
la suite de la publication de leur manuscrit dans Écoscience. À ces frais, s’ajoutent ceux qui correspondent à la publica-
tion de plus de trois tableaux et plus de trois figures par article, à raison de 50 $CAD par tableau et figure excédentaire.
As indicated in the Guide to authors, publication fees of CAD$50 per page will be charged to authors following the
publication of their article in Écosience. These fees are in addition to those for tables and figures in excess of three per
article, at the rate of CAD$50 per table ans figure.
Ce formulaire ne constitue pas une facture officielle, mais plutôt un avis de facturation à signer et à retourner en guise
d’acceptation des frais présentés ci-dessous.
This form is not the official invoice but an invoicing note to sign and return in acceptance of the charges detailed below.
Écoscience, Pavillon Marchand, Université Laval, Sainte-Foy, Québec, G1K 7P4,
À retourner immédiatement à:
Please return immediately to:
Avis de facturation
Invoicing note
Frais de publication / Publication fees
Nombre de pages / Number of pages
Frais pour les tableaux / Fees for tables
Nombre de tableaux excédant 3 / Number of tables in excess of 3
Frais pour les figures / Fees for figures
Nombre de figures excédant 3 / Number of figures in excess of 3
Frais pour les figures couleur / Fees for color figures
Note : Les taxes applicables seront ajoutées lors de la facturation officielle.
Note: Applicable taxes will be added on the official invoice.
Total CAD$
× 50 =
× 50 =
× 50 =
Effects of sublethal attack by a sucking insect, Hyalymenus tarsatus, on Sesbania
drummondii seeds: Impact on some seed traits related to fitness
Écoscience, Pavillon Marchand, Université Laval, Sainte-Foy, Québec,
G1K 7P4, Canada, tel. (418) 656-3188, fax (418) 656-2346, e-mail:
Date limite pour commander:
Deadline to submit orders:
Auteur (s) / Author (s):
Titre / Title:
Les tirés à part peuvent être commandés avec ce formulaire, par lettre ou par bon de commande officiel acheminé à
l’adresse ci-dessus. La commande doit parvenir à Écoscience au plus tard à la date indiquée au haut du présent for-
mulaire. Veuillez mentionner le numéro de la commande si le paiement doit être effectué par une université ou le ser-
vice de comptabilité d’un organisme. Les paiements par carte de crédit ne sont pas acceptés.
Reprints may be ordered on this form, by letter or by official purchase order sent to the address given above. The order
must reach Écoscience no later than by the date indicated in the upper part of this form. Please enter purchase order
number if payment is to be processed through a university or institute accounting office. Payment by credit card is not
Adresse d’expédition / Sent reprints to:
Adresse de facturation / Send invoice to:
Numéro du manuscrit :
Manuscript number :
Remarques /Remarks:
Date : Signature : Node téléphone / Phone No. :
S.V.P. retourner à l’adresse suivante:
Please return to:
Nombre de tirés à part commandés
Number of reprints ordered Numéro de commande
Purchase order number
Formulaire de commande de tirés à part
Reprints order form
Eco-2362 2002-02-25
Effects of sublethal attack by a sucking insect, Hyalymenus tarsatus, on Sesbania
drummondii seeds: Impact on some seed traits related to fitness
Prix des tirés à part / Prices of reprints
En vigueur en janvier 2002 / Effective in January 2002
Nombre de pages / Quantité commandée / Quantity ordered
Number of pages 50 100 200 300
1-4 138 $ 177 $ 206 $ 233 $
5-8 244 $ 281 $ 316 $ 352 $
9-12 338 $ 376 $ 434 $ 529 $
13-16 416 $ 462 $ 561 $ 575 $
17-20 503 $ 591 $ 689 $ 850 $
21-24 581 $ 637 $ 817 $ 1009 $
25-28 660 $ 744 $ 944 $ 1170 $
29-32 738 $ 844 $ 1072 $ 1331 $
33-36 817 $ 892 $ 1199 $ 1488 $
Les tirés à part sont tous brochés à
plat ; ils sont fournis dans le même
format que celui de la revue ; ils ne
comprennent pas de couverture.
Ces prix sont sujets à changements
si la commande est reçue après la
date limite indiquée au verso.
Les frais de poste et les taxes
applicables s’ajoutent aux prix men-
Le délai accordé pour le paiement est
de 30 jours.
Les prix sont indiqués en dollars
Reprints are all side-wire stappling; they
are in the same size and format than
the journal; all are without cover.
Prices are subject to changes when
order is received after the deadline indi-
cated on reverse side.
Postage charges and applicable taxes
will be added to prices quoted above.
Payment term is 30 days.
Prices are given in Canadian dollars.
À noter Please note
... In addition to their impacts on germination, the morphological characteristics and chemical content of seed coats also affect seed predation (Ceballos et al. 2002;Islam et al. 2003;Caldas and Blair 2009;Tigist et al. 2020). Janzen (1980Janzen ( , 1983 reported that two species of weevil, namely, Stator pruininus and Acanthoscelides griseolus, attacked the seeds of S. emerus in northwestern Costa Rica. ...
... The adult beetles deposit their eggs in the mature pods, and the larvae hatch and bore into the mature seeds ( Janzen 1983). In another study, Ceballos et al. (2002) showed that developing seeds of Sesbania drummondii are attacked by nymphs and adults of the bug Hyalymenus tarsatus. They argued that the massive attack by H. tarsatus causes irregular shape, changes in seed coat color, and disruption of dormancy. ...
... However, Ceballos et al. (2002) reported that the concentrations of total condensed tannins released by seeds attacked by Hyalymenus tarsatus were higher than those of intact seeds. They proposed that this difference in tannin content is a response mechanism to produce a more potent chemical defense, restore seed coat impermeability, or both. ...
... Seed-sucking insects feed on seeds by piercing the coat and sucking out some of the contents. Such attacks are not necessarily lethal, but they may influence the fitness of seeds and seedlings by decreasing the amount of nutrients available for germination, damaging the embryo, increasing water uptake, or introducing pathogens (Ceballos et al., 2002). Seed-sucking insects are found worldwide, but their effects on seed survival, germination, and establishment have only been investigated a few times (Ceballos et al., 2002). ...
... Such attacks are not necessarily lethal, but they may influence the fitness of seeds and seedlings by decreasing the amount of nutrients available for germination, damaging the embryo, increasing water uptake, or introducing pathogens (Ceballos et al., 2002). Seed-sucking insects are found worldwide, but their effects on seed survival, germination, and establishment have only been investigated a few times (Ceballos et al., 2002). ...
... Higher densities of Nysius groenlandicus produced a higher germination speed (Table I). This increase in germination speed is similar to that found in seeds of Gossypium species attacked by seed bugs and weevils, in which the introduction of holes in the seed coat increased the seed's uptake of water and speed of germination (Karban & Lowenberg, 1992;Sallabanks & Courtney, 1992;Ceballos et al., 2002). In the present study, the increased speed of germination after feeding by bugs is most likely a result of piercing of the seed (either the hole itself or the fact that the hole enables water flow into the seed). ...
Full-text available
Seeds of Arctic plants face numerous threats prior to dispersal. The growing season varies across years in terms of degree days, and herbivory, predation, and pathogens are critical threats. In this study the results of different densities of the Arctic seed bug, Nysius groenlandicus (Heteroptera), piercing the seed coat of Silene acaulis and sucking out nutritional content were observed. In order to study the effect of Nysius groenlandicus on seed mass and germination of Silene acaulis, seeds were placed in Petri dishes with different densities of seed bugs. The herbivory affected the seed mass, leading to an average mass loss of 3.0% in fed-upon seeds compared to non-fed-upon seeds. However, the average seed mass lost seemed independent of the densities of seed bugs. A significant negative correlation between seed mass loss and number of germinations for seeds exposed to seed bugs was found. Furthermore, the germination speed of the seeds increased with increasing density of Nysius groenlandicus. The significance of this interaction is discussed, and we hypothesize that feeding might benefit the establishment of seedlings.
... It is known that cecidogeous insects can avoid and alter chemical defenses of those plants that lodge them for their benefit and that of their species. In this case, condensed tannins are known due to their fungicidal activity (Scalbert, 1991;Lattanzio et al, 2006), even in trichomes (Aziz et al., 2004;Kelsey et al., 1984;Li et al., 1996), and defense against herbivory (Forkner et al., 2004;Ceballos et al., 2002). Some authors claim that the digestion of tannins in the alkaline pH in the midgut of the insects may produce hydrogen peroxide. ...
Chañar ( Geoffroea decorticans - Fabaceae) is a tree from South America that is normally infected with galls originated by insects. One of its parasites is Allodiplosis crassa (Cecidomyiidae, Diptera) which produces globular galls with sticky prolongations. Since this plant has medicinal uses in Argentina, its infestation could alter the quality of the plant drug. The surface of insect-induced galls usually contains defensive features such as trichomes, increased hardness and an increase in the content of polyphenolic compounds. The objective of this research is to assess the structural and histochemical features of the gall and to compare the content of polyphenolic metabolites in the gall, in the healthy leaf and in lignified stems of G. decorticans . The methanolic extract from the galls showed the highest amount of polyphenolic and proanthocyanidins and the lowest amount of hydroxycinnamic derivatives and flavonoids compared to the methanolic extract of the leaves. The photographs taken from the external surface of the gall showed that some prolongations have heads. The histochemical analysis showed that the prolongations have a high amount of proanthocyanidins and flavonoids; and that the heads are reactive to Sudan III. These phytochemical and histological characteristics may have a defensive role against harmful fungi and parasites that attack the larvae of the A. crassa . The results of this study show the presence of defensive features in an insect-induced gall of a medicinal plant with potential implications in the pharmacological activity of this species. This is the first report of a histochemical and phytochemical study in G. corticans galls.
... Ninfas y adultos introducen el estilete (aparato bucal) a través de la cascara del fruto, observándose punteados en la superficie del fruto joven (Coto y Sauders 2004). Los ataques repetidos provocan cambios en la forma y el color del órgano reproductivo de la planta (Ceballos et al. 2002). En El Salvador, se encontró en cafetales ubicados entre 500 a 1100 msnm. ...
Full-text available
Los cafetales albergan gran parte de la biodiversidad, formada por especies animales y vegetales, cuya presencia hace que el ecosistema sea más estable. La multitud de cadenas alimenticias existentes, impide que un sólo organismo se multiplique aceleradamente, esto es importante ya que ayuda al mantenimiento del material genético diverso y por ende a la flora y fauna. Los cafetales de El Salvador conservan la biodiversidad de 209 especies de árboles nativos y 21 exóticas, 188 especies de aves, 101 residentes y 37 migratorias (42 de estas amenazadas y 19 en peligro de extinción a nivel local); además de 31 especies de pequeños mamíferos, 8 en peligro de extinción; unas 26 especies de reptiles y 8 especies de anfibios que poseen varias especies en peligro de extinción, entre otros (PROCAFE 2000). Se han registrado 68 especies de avispas (Hymenoptera: Ichneumonidae: Pimplinae), con 4 especies nuevas para la ciencia (Gauld et al. 2002); además, se han identificado más de 17 especies de termitas (Orden Blattaria: Isoptera) asociadas a los cafetales de El Salvador (Sermeño et al. 2003). Gracias a los cafetales bajo sombra se podrían conservar alejadas del peligro y reducir la amenaza de extinción (PROCAFE 2000).
... For example, reduced germination percentages are usually observed with increasing damage to the seeds (e.g., Vallejo-Marín et al. 2006;Peréz et al. 2008a). Additionally, the reduction or removal of seed reserves can affect germination times but in an inconsistent manner; that is, it can either accelerate or prolong germination depending on the species (e.g., Ceballos et al. 2002;Vallejo-Marín et al. 2006). Finally, removal of cotyledon reserves may also result in decreased seedling performance (Janzen 1976;Zhang and Maun 1991;Kennedy et al. 2004), as seedling size and survival are strongly related to the amount of energy reserves in the seeds (Moles and Westoby 2004). ...
Full-text available
Background Seed loss to predators is a common phenomenon across plant communities and an important determinant of plant recruitment. Although seed predators commonly destroy the entire seed, partial seed consumption has been reported for many species; however, the degree to which seed mass loss affects germination dynamics and survival of new individuals has been poorly documented. We simulated seed damage in natural conditions to examine how different levels of cotyledonary reserve removal affect germination dynamics and seedling performance of Myrcianthes coquimbensis (Myrtaceae), a threatened Atacama Desert shrub. The experiment combined two levels of seed maturity with three levels of seed mass loss. Results Removal of the cotyledon reserves and seed maturity negatively affected the odds and the temporality of seedling emergence; nonetheless, seedlings were able to emerge from seed fragments, of either mature or immature seeds, that lost up to 75% of their original mass. Removal of cotyledonary reserves had negative effects on seedling size but no effect on root:shoot ratios. Conclusions Our findings indicate that the loss of cotyledonary reserves in M. coquimbensis seeds is not necessarily lethal. Moreover, we posit that tolerance to partial seed consumption is likely a key reproductive strategy, which enables recruitment in this species.
... In other legumes, e.g. Sesbania drummondii, it has been reported that attacks by Hyalymenus tarsatus cause seed color change that also causes alteration in the seed coat physiology that triggers early germination (Ceballos et al. 2002). In the present study, there was no evidence that changes in seed color due to pea weevil attack are associated with any fitness or physiological mechanisms. ...
Full-text available
Field pea (Pisum sativum L. subsp. sativum) is an important agricultural crop worldwide, as a main source of protein in human diet and as animal fodder. In Ethiopia, it is the second most important legume crop next to faba bean (Vicia faba L.). However, the production is threatened by pea weevil (Bruchus pisorum L.), which is a rapidly spreading insect pest throughout the country. During June–October 2011, a total of 602 pea accessions from Ethiopia were screened for pea weevil resistance at three field sites in Ethiopia. From this trial, accessions with relatively low mean percent seed damage (PSD) were selected and evaluated during June-October 2012 in replicated trials. Some genotypes from the selected accessions were also studied under greenhouse conditions for up to three generations. Both in the field and greenhouse trials, a significant level of variation in PSD were observed among accessions/genotypes. However, a few of them showed relatively consistent results across sites and years. The gene bank accessions 32454 and 235002 had consistently <40 % PSD. These accessions had 17 and 33 % PSD, respectively, at a site where the highest and overall mean PSD were 92 and 75 %, respectively. Also, promising genotypes with consistently low levels of seed damage were identified in accessions 226037 and 32410. The incorporation of such promising accessions/genotypes into pea breeding programs may lead to the development of field pea varieties with enhanced resistance against pea weevil and consequently contribute to sustainable field pea production in Ethiopia and beyond.
Nysius species are one of the emerging pests of perilla crop in Korea. Seed weight loss and germination of perilla seeds caused by Nysius plebeius were studied by exposing perilla seeds cv. Deulsaem to six insect infestation levels (0, 5, 10, 15, 20, and 25 individual) for 3 and 5 days in the laboratory. The seed weight loss and germination rate were significantly affected by infestation levels and feeding durations. The feeding impacts on seeds increased with increasing in bug density and duration. The higher seed weight loss, and weak germination were recorded at higher bug density (25 individuals) for longer (5 days) feeding duration. The triphenyl tetrazolium chloride, TTC, showed lower seed viability for damaged seeds than undamaged seeds. The outcomes of the study are discussed in terms of impacts of perilla seed bugs infestation, and suggest longer field exposure of perilla seeds to seed bugs may bring lower yield with poor quality of seeds and low germination potential of seeds. Future studies should examine the changes in nutrient contents of seeds after exposed to perilla seed bugs.
Cocoa is known as an important source of flavan-3-ols, but their fate "from the bean to the bar" is not yet clear. Here, procyanidin A2 found in native cocoa beans (9-13 mg/kg) appeared partially epimerized into A2(E1) through fermentation, whereas a second epimer (A2(E2)) emerged after roasting. At m/z 575, dehydrodiepicatechin A was revealed to be the major HPLC peak before fermentation, whereas F1, a marker of well-conducted fermentations, becomes the most intense after roasting. RP-HPLC-ESI(-)-HRMS/MS analysis performed on a procyanidin A2 model medium after 12 h at 90 °C revealed many more degradation products than those identified in fermented cocoa, including the last epimer of A2, A2 open structure intermediates (m/z 577), and oxidized A-type dimers (m/z 573).
Full-text available
Seeds of most Great Basin lupine ( Lupinus spp. [Fabaceae]) species are physically dormant and thus, difficult to establish in uniform stands in seed production fields. We designed this study to examine 5 seed scarification techniques, each with 11 levels of application (including a non-scarified control), to reduce the physical seed dormancy of longspur lupine ( L. arbustus Douglas ex Lindl.), silvery lupine ( L. argenteus Pursh), hairy bigleaf lupine ( L. prunophilus M.E. Jones), and silky lupine ( L. sericeus Pursh). These 4 perennial Great Basin lupine species are of interest for both rehabilitation and restoration of degraded rangelands. We evaluated 10 treatments of each of 5 scarification methods, one mechanical, 2 thermal, and 2 chemical (sulfuric acid and sodium hypochlorite) techniques on the above-mentioned species. The sulfuric acid and the mechanical scarification treatments significantly improved germination for both silvery and silky lupine. Additionally, one thermal scarification method (60 s at 95 °C [203 °F]) was effective for silvery lupine. Both sulfuric acid and sodium hypochlorite scarification methods had treatment levels that significantly improved germination of hairy bigleaf lupine. For longspur lupine, all treatments within the 5 scarification methods either decreased or were not a significant improvement of germination as compared with the control, except for the treatment of soaking the seeds for 35 s at 95 °C (203 °F). We found scarification to be an effective tool for reducing physical dormancy in silvery lupine, hairy bigleaf lupine, and silky lupine, thus allowing for a more efficient use of limited seeds.
Aims We examined the importance of partial seed consumption (cotyledon loss) by rabbits in the early establishment of seedlings of cork oaks restricted to nutrient-impoverished soils. Methods To determine the importance of cotyledons in the growth and development of seedlings, we simulated two levels of predation [light (30 % cotyledon loss) and heavy (60 % loss) partial consumption] and two soil nutrient contents (nutrient-poor soil, nutrient-rich soil). Seedlings height, root length, dry root and shoot biomass, specific leaf mass, leaf density, gas exchange, chlorophyll fluorescence parameters and photosynthetic pigment concentrations were determined. Results Results indicated that effect of nutrient level on the growth of the oak seedlings was more important than that of cotyledon biomass. However, in nutrient–poor soils, cotyledon biomass played a major role in the early performance of cork oaks. Acorns grown in nutrient-rich substrate, despite having greater aerial vigor, were slower to develop a vertical root, and hence less likely to reach permanent moisture. Cotyledon loss caused a decrease in the biomass of roots and shoots when acorns were heavily consumed, and as a result experienced a reduction in net photosynthetic rate, stomatal conductance and chlorophyll concentration. Survival of seedlings was unaffected by either soil type or cotyledon loss. Conclusions Our results show that effects of soil type on the survival of oak seedlings were more important than those of cotyledon biomass. However, in a competitive situation, cotyledon biomass, as an indicative of growth nutrient support rather than an energy source, could be vital in a nutrient-poor environment, particularly in Mediterranean climate regions and for species with little inherent drought tolerance (as is the case of Quercus spp.), where rapid root growth is required to ensure that contact with soil moisture is maintained over the first summer.
Full-text available
Field infestations of a seed-feeding insect developed from overwintered populations reduced viability of velvetleaf seed to 17.5 and 15.5% at two locations in central Missouri, compared to 95.5 and 87.5% at insect-free sites. Insect feeding enhanced the proportion of seedborne microorganisms in seed up to 98% compared to the average fungal infection of 8% for seed not exposed to the insect. There was a strong negative correlation between fungal infection associated with insect feeding and percent velvetleaf seed viability. The insect transmits microorganisms externally just as pollen is carried by various other insect species and not by ingestion and regurgitation. The effectiveness of the insect on reducing seed viability and seed production in central Missouri is mainly limited by the time required to build up populations capable of significantly affecting early-season velvetleaf seed production.
Full-text available
Removal of 1%, 5%, and 10% of the seed weight of Mucuna andreana seeds severely reduced the ability of the seedling to withstand artificial herbivory. Seed weight was removed by drilling holes in the seeds in order to mimic bruchid damage.
Seed size varies over several orders of magnitude in any one community. We outline a number of hypotheses that could account for this variation, briefly discuss the reasoning and evidence underlying each of these hypotheses, and then test each hypothesis with a database of 248 species from the semiarid woodlands of western New South Wales, Australia. Information on seed weight, growth form, plant longevity, height, dispersal mode, dormancy, and germination season was used. We considered not only pairwise relationships between seed weight and each other variable, but also alternative hypotheses whereby relationships arose as a result of indirect correlations through other variables. The strongest associations of seed size were with plant height and growth form. The seed-size variation accounted for by growth form largely overlapped with that accounted for by plant height, but each also accounted for some further variation independently of the other. Of the five hypotheses tested, the correlative patterns were inconsistent for two. Two others showed the predicted pattern, but these patterns could alternatively be interpreted as arising from secondary correlation via the combination of plant height and growth form. Only plant height, growth form, and dispersal mode had significant associations with seed size independent of the other attributes measured.
Seed size has significant demographic consequences in Prunella vulgaris. Measurements of the effects of seed size were obtained by sowing seeds of known size at four field sites and recording seedling emergence, survival, and size. The intensity of selection on seed size was calculated from these data. Large seeds had a significantly greater probability of emergence at most study sites in each of 2 yr. The effects of seed size were expressed most strongly during the early part of the life cycle, between sowing and emergence, and much less so in later phases of the life cycle. In contrast to a prediction based on the greater relative frequency of large-seeded species in later successional habitats, the effect of seed size on percent seedling emergence did not differ significantly between an old-field and a woodland habitat. However, it is likely that the intensity of natural selection on seed size is greater in an old-field than in a woodland population because the natural distributions of seed size in old-field populations include more small seeds than do those in woodland populations. An analysis of the costs and benefits of producing large and small seeds revealed that in addition to selection favoring large seeds, there was selection favoring individuals that produced large seedlings at three of the four study sites. At the fourth site, there was no selection favoring larger seeds or parents that produced larger seeds. Substantial capacity for phenotypic plasticity in seed size suggests that there may be little opportunity for an evolutionary response despite strong selection favoring large seeds.
Seed size, dormancy, and dispersal share 3 population-dynamic functions in temporally and spatially varying environments: risk reduction of bet hedging, escape from crowding, and escape from sib competition. A model was developed to explore ways they may interact to reduce risk. The risk-reducing properties of these seed traits evolve only in response to global temporal variance. Thus, to understand how selection impinges on the seed traits, creating fitness interactions, one must understand the factors contributing to global temporal variance and how they are mitigated by the various seed traits. Since the traits interact to reduce variance, arbitarily fixing any 1 trait at different values alters the fitness-maximizing values of the others, resulting in trade-offs among traits. The authors explore how changes in the number of independent environmental patches, probability of favorable conditions, radius of dispersal, and spatial and temporal autocorrelation of environmental conditions alter selection on the interacting syndrome of seed traits. -from Authors