Comparison of germination and seed bank dynamics of ...sourcedb.ib.cas.cn/cn/ibthesis/201301/P020130104332326552685.pdf · Comparison of germination and seed bank dynamics of dimorphic

Comparison of germination and seed bank dynamics of dimorphic seeds of thecold desert halophyte Suaeda corniculata subsp. mongolica

Dechang Cao1,2, Carol C. Baskin3,4, Jerry M. Baskin3, Fan Yang1,2 and Zhenying Huang1,*1State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, Chinese Academy of Sciences, Beijing

100093, China, 2Graduate School of Chinese Academy of Sciences, Beijing, 100039, China, 3Department of Biology, Universityof Kentucky, Lexington, KY 40506, USA and 4Department of Plant and Soil Sciences, University of Kentucky, Lexington,

KY 40546, USA* For correspondence. E-mail [email protected]

Received: 31 May 2012 Returned for revision: 16 July 2012 Accepted: 7 August 2012 Published electronically: 12 September 2012

† Background and Aims Differences in dormancy and germination requirements have been documented in het-eromorphic seeds of many species, but it is unknown how this difference contributes to maintenance and regen-eration of populations. The primary aim of this study was to compare the seed bank dynamics, includingdormancy cycling, of the two seed morphs (black and brown) of the cold desert halophyte Suaeda corniculataand, if differences were found, to determine their influence on regeneration of the species.† Method Seeds of the two seed morphs were buried, exhumed and tested monthly for 24 months over a range oftemperatures and salinities, and germination recovery and viability were determined after exposure to salinity andwater stress. Seedling emergence and dynamics of the soil seed bank were also investigated for the two morphs.† Key Results Black seeds had an annual dormancy/non-dormancy cycle, while brown seeds, which were non-dormant at maturity, remained non-dormant. Black seeds also exhibited an annual cycle in sensitivity of germin-ation to salinity. Seedlings derived from black seeds emerged in July and August and those from brown seeds inMay. Seedlings were recruited from 2.6 % of the black seeds and from 2.8 % of the brown seeds in the soil, andonly 0.5 % and 0.4 % of the total number of black and brown seeds in the soil, respectively, gave rise to seedlingsthat survived to produce seeds. Salinity and water stress induced dormancy in black seeds and decreased viabilityof brown seeds. Brown seeds formed only a transient soil seed bank and black seeds a persistent seed bank.† Conclusions The presence of a dormancy cycle in black but not in brown seeds of S. corniculata and differencesin germination requirements of the two morphs cause them to differ in their germination dynamics. The studycontributes to our limited knowledge of dormancy cycling and seed bank formation in species producing hetero-morphic seeds.

Key words: Dormancy, halophyte seeds, salinity, seed germination, seedling recruitment, seed bank dynamics,Suaeda corniculata.

INTRODUCTION

The production of two or more distinct types of seeds by asingle plant is known as seed heteromorphism (Venable,1985; Mandak, 1997; Imbert, 2002). In addition to morpho-logical differences, heteromorphic seeds also differ in disper-sal capacity (Flint and Palmblad, 1978; Venable and Levin,1985; Imbert, 1999) and germination behaviour (Baskin andBaskin, 1976; Venable and Levin, 1985; Wang et al., 2008).Imbert (2002) compiled a worldwide list of 218 heteromorphicspecies, the majority of which belong to the Asteraceae andChenopodiaceae (now included in Amaranthaceae). Studieson seed heteromorphism have focused on differences in ger-mination responses (El-Keblawy, 2003; Carter and Ungar,2003; Yao et al., 2010), seedling growth (Khan and Ungar,1984; Katembe et al., 1998; Song et al., 2008) and plasticityin production of heteromorphic seeds, i.e. the proportion ofheteromorphic seeds may change when the plants are grownin different environments (Imbert and Ronce, 2001; El-Keblawy, 2003). However, there are only a few comprehensive

studies on the contribution of heteromorphic seeds to popula-tion regeneration (Venable and Levin, 1985).

Seed banks play an important role in ensuring persistence ofplant populations (Hutchings and Booth, 1996; Leck andSimpson, 1987; Anderson et al., 2012) and contribute tofuture genetic variability of species (Tanksley and McCouch,1997; McCue and Holtsford, 1998). Seeds in seed banksmay go through annual dormancy/non-dormancy cycles(Baskin and Baskin, 1985; Baskin et al., 1987; Baskin et al.,1993), which regulate seed germination and seedling emer-gence and thus determine the optimal timing for plant estab-lishment (Baskin et al., 1993; Baskin and Baskin, 1998).However, little information is available on seed banks(Carter and Ungar, 2003; Joley et al., 2003) and even lesson dormancy cycles of the different morphs (Carter andUngar, 2003). The only other study we are aware of on dor-mancy cycling in heteromorphic species was done on theamphicarphic weedy species Emex australis (Panetta andRandall, 1993). However, these investigators apparently usedonly aerial diaspores in their study. They did not mention

# The Author 2012. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved.

For Permissions, please email: [email protected]

Annals of Botany 110: 1545–1558, 2012

doi:10.1093/aob/mcs205, available online at www.aob.oxfordjournals.org

at Institute of Botany, C

AS on January 3, 2013

http://aob.oxfordjournals.org/D

ownloaded from

mailto:[email protected]

http://aob.oxfordjournals.org/

that Emex produces aerial and subterranean achenes that aredimorphic (Evenari et al., 1977; Weiss, 1980).

Suaeda corniculata subsp. mongolica (hereafter Suaedacorniculata) is a leaf succulent halophyte belonging to theChenopodiaceae (now placed in the Amaranthaceae)(Lomonosova et al., 2008), a family in which seed heteromorph-ism is common (Mandak and Pysek, 2001; Imbert, 2002; Atiaet al., 2011). Suaeda corniculata is an annual herb widely dis-tributed in steppes and semi-desert zones of central southernSiberia, Mongolia and northern China (Lomonosova et al.,2008). It produces two distinct seed morphs: black seeds andbrown seeds (Lomonosova et al., 2008).

The purpose of this study was to determine if there are dif-ferences in seed dormancy cycles of the two seed morphsburied in the soil under natural environmental conditions andif morphs vary in their ability to persist in the soil, i.e. toform a persistent seed bank. If so, how do the differences in-fluence persistence of soil seed banks and regeneration ofthe population? To answer these questions, we tested germin-ation of each of the two seed morphs and investigated thedynamics of their soil seed banks as well as seedling emer-gence. In some species of Suaeda, brown seeds are lessdormant than black seeds (Song et al., 2008; Wang et al.,2008), and we hypothesized that this is the case forS. corniculata. Consequently, brown seeds in the soil areready to germinate whenever environmental conditions are fa-vourable for germination, while black seeds can attain highgermination only when they come out of dormancy. Inwhich case, brown seeds may be depleted from the soil seedbanks faster than black ones, and more seedlings may bederived from brown than from black seeds. These aspects ofthe hypothesis were tested in this study. In addition, salinityand drought stress are the two most important environmentalstresses facing S. corniculata, which grows in saline aridareas. Thus, germination tests were conducted on the twoseed morphs after exposure to salinity and water stress to deter-mine effects of soil salinity and drought stress on dormancyand viability of seeds in soil.

MATERIALS AND METHODS

Habitat and seed source

Mature seeds of Suaeda corniculata subsp. mongolica were col-lected on 20 November 2009 and 15 November 2010 fromseveral hundred plants growing in a saline steppe on the OrdosPlateau in Inner Mongolia, northern China (38814′N,107829′E; 1311 m a.s.l.). This area has a typical continentalsemi-arid climate with mean annual precipitation of 250–490 mm, primarily from June to August. Mean annual tempera-ture is about 6.0–9.0 8C. Brown seeds and black seeds were handseparated after harvest, and they were tested separately in allexperiments. Seeds collected in 2009 were dried at room tem-perature (10–18 8C, 17–32 % relative humidity) for 1 weekbefore initiation of germination tests for fresh seeds and burialof seeds in soil. After air-drying at room temperature and handseparation, seeds collected in 2010 were dry stored at approx.–20 8C until use, except those tested for dormancy breakingand sown in the field (see below).

Black:brown seed ratio and seed mass

Fifteen individual plants were collected in natural habitats inNovember 2009, and all seeds were obtained from each of the15 plants and mixed together. Five replicates of about 1000seeds arbitrarily chosen from the mixed seed lot weredivided into black or brown seeds based on colour of thetesta, and the black:brown seed ratio was determined. Also,black and brown seeds were hand separated from the mixedseed lot, and the thousand-seed mass of each type was deter-mined by weighing five replicates of 1000 seeds of eachmorph.

Germination tests of fresh seeds

Fresh seeds of both morphs were tested for germination onfilter paper moistened with distilled water at 15/5, 20/5, 25/15and 30/15 8C (12/12 h) in the light (12 h photoperiod, approx.100 mmol m22 s21, cool white fluorescent light) and in continu-ous darkness. The higher temperature coincided with a 12 hlight period and the lower temperature with a 12 h darkperiod. The alternating temperature regimes represent the ap-proximate mean daily maximum and minimum air temperaturesfor each month during the growing season on the Ordos Plateau:April and October, 15/5 8C; May and September, 20/5 8C; Juneand August 25/15 8C; and July, 30/15 8C. Four replicates of 25seeds each were used in each test condition. Final germinationpercentages were determined after 20 d. Emergence of theradicle was the criterion for this and all other germinationexperiments. After all germination tests, non-germinated seedswere examined under a dissecting microscope to determinewhether the embryo was firm and white, indicating that theywere viable, or soft and grey, indicating that they were non-viable. Tetrazolium tests confirmed that the firm, whiteembryos were viable and the soft, grey ones non-viable. Onlyviable seeds were used in calculating germination percentagesunless otherwise stated. Viability (%V) for each morph was cal-culated as [(NG ¼ NV)/NT] × 100, where NG is the number ofseeds that germinated, NV the number of non-germinated butviable seeds and NT the total number of seeds in each dish(i.e. 25) incubated at 30/15 8C in the light.

Germination of each seed morph was also tested at 0.2, 0.4,0.6, 0.8, 1.2 and 1.6 M NaCl at 30/15 8C, using four replicatesof 25 fresh seeds each. Final germination percentages weredetermined after 20 d. Non-germinated black seeds were testedfor viability using the methods described above, and onlyviable seeds were used in calculating germination percentages.However, salinity decreased the viability of brown seeds, sogermination percentages may be overestimated when onlyviable seeds were taken into account. Since viability ofbrown seeds was 100 % in distilled water, germination percen-tages (%G) of brown seeds in each Petri dish at each of thesalinities was calculated as (NG/25) × 100, where NG wasthe number of seeds that germinated.

Dormancy breaking of black seeds

Since fresh mature black seeds were dormant (see theResults), cold stratification, gibberellic acid (GA3) and testascarification were used to break their dormancy. Mature

Cao et al. — Germination and seed bank dynamics of dimorphic seeds of Suaeda1546




ownloaded from


black seeds that were freshly collected in November 2010 wereused in this experiment.

Cold stratification. Freshly collected black seeds were arrangedevenly on two layers of filter paper in a metal box (20 cmlength × 10 cm width × 10 cm depth) containing wet sand(10–13 % water content). The metal box was covered with alid and kept in darkness in a refrigerator at 5 8C. After 0 (CK),5, 10, 15 and 20 d cold stratification, four replicates of 25seeds each were arbitrarily chosen to test germination at 15/5,20/5, 25/15 and 30/15 8C (12/12 h) in the light (12 h photo-period, approx. 100 mmol m22 s21, cool white fluorescentlight). The higher temperature coincided with the 12 h lightperiod and the lower temperature with the 12 h dark period.Final germination percentages were determined after 20 d.

GA3 treatments. Four replicates of 25 freshly collected blackseeds each were tested for germination in 0 (CK), 0.1 and1 mmol L21 GA3 solutions at 15/5 8C (12/12 h) in the light(12 h photoperiod, approx. 100 mmol m22 s21, cool whitefluorescent light). The higher temperature coincided with the12 h light period and the lower temperature with the 12 hdark period. Final germination percentages were determinedafter 20 d.

Testa scarification. Seed coats of four replicates of 25 freshlycollected seeds were scarified using a scalpel, being carefulnot to touch the embryos. Scarified seeds were tested for ger-mination at 15/5 8C as described above.

Light requirement for germination of non-dormant black seeds

Fresh black seeds germinated to a higher percentage in lightthan in darkness (see below, Fig. 5). To determine if non-dormant black seeds also have a light requirement for germin-ation, we tested germination of buried black seeds exhumed inMay 2010, which were non-dormant (see the Results), in light(12 h photoperiod, approx. 100 mmol m22 s21, cool white fluor-escent light), and in total darkness at 15/5, 20/5, 25/15, and 30/15 8C (12/12 h). Four replications of 25 seeds each were used.Final germination percentages were determined after 20 d.

Dormancy cycling of the two seed morphs during burial

Following initial germination tests of fresh seeds in 2009,about 1000 seeds each of the black and brown morphs wereplaced in 24 nylon bags. Each of the 48 bags was buried at adepth of 2 cm in 20 cm diameter × 50 cm tall plastic pots withdrainage holes. The pots were filled with soil collected fromthe habitat from which the seeds were collected and thenburied on nearly flat ground at the Ordos Sandland EcologicalResearch Station of the Chinese Academy of Sciences on 1December 2009. Mean daily maximum and minimummonthly air temperatures for each month during the studywere obtained from data collected at the Research Station.

Seeds of the two morphs were exhumed and tested for ger-mination on the first day of each month for 24 months, begin-ning in May 2010. Germination tests were conducted at 15/5,20/5, 25/15 and 30/15 8C in light (12 h photoperiod) asdescribed above. Four replications of 25 seeds each wereused. Final germination percentages were determined after

20 d. Following the germination tests, non-germinated seedswere tested for viability as described above.

Effect of salinity on germination of buried seeds

Each time buried seeds were exhumed, they were incubatedat four salinities and in distilled water. Since almost all brownseeds had germinated before the first exhumation in May2010, germination tests were conducted only on black seeds.The buried black seeds were tested for germination in 0.2,0.4, 0.6 and 0.8 M NaCl. Seeds were placed in Petri dishes ontwo layers of filter paper moistened with 3 mL of test solution,and the dishes were wrapped with plastic film. Four replicates of25 seeds each were used. The seeds were incubated at 30/15 8Cin light (see above). Final germination percentages weredetermined after 20 d. Following the germination tests, non-germinated seeds were tested for viability as described above.

A relative germination percentage (%GR) was used to evalu-ate the effect of salinity on germination of buried seeds:%GR ¼ (GS/GW) × 100, where GS is the germination percent-age at salinity S at 30/15 8C and GW is the germination per-centage in distilled water.

Effect of salinity and water stress on dormancy and viability of thetwo seed morphs

Black and brown seeds that had been stored dry at –20 8Cfor 5 months were exposed to 1.0 M NaCl and isotonic poly-ethylene glycol (PEG)-8000 solutions for different periods oftime in April 2011, after which they were tested for germin-ation in distilled water and for viability. The osmotic potentialof 1.0 M NaCl solution was determined by the Van’t Hoffequation p ¼ cRT, where c is osmolality in mol L21, R thegas constant (8.31 J K21 mol21) and T temperature (K). Theisotonic PEG-8000 solution was prepared according toMichel (1983).

For both black and brown seeds, 24 samples of 25 seedseach were placed in 5 cm Petri dishes; 3.5 mL NaCl solutionwas added to each of 12 of the dishes, and 3.5 mL ofPEG-8000 solution was added to each of the other 12. Thedishes were placed in an incubator at a constant temperatureof 20 8C. This temperature was chosen to replicate theaverage temperature of early summer (before the beginningof the rainy season) when seeds were most likely to experiencesalinity and drought stress. A constant temperature rather thanalternating temperatures was used to avoid changes in osmoticpotential of PEG-8000 solutions. Four dishes were selectedrandomly after 20, 40 and 60 d pre-treatment in each of theNaCl and PEG-8000 solutions, and the seeds were transferredto Petri dishes and moistened with distilled water at 30/15 8C.Seeds that had germinated during the pre-treatment werecounted and removed, and the remaining seeds were washedin running water twice before being transferred to distilledwater. The seeds transferred to distilled water were incubatedfor 20 d before determination of final germination percentages,and non-germinated seeds were tested for viability as describedabove. Recovery germination percentage (%RG) was deter-mined as [(a – b)/(c – b)] × 100 %, where a is the numberof seeds germinated in solutions plus those that germinatedafter being transferred to distilled water, b is the number of

Cao et al. — Germination and seed bank dynamics of dimorphic seeds of Suaeda 1547




ownloaded from


seeds germinated during the pre-treatment with NaCl orPEG-8000 solution, and c is the total number of seeds in adish (Pujol et al., 2000; Guma et al., 2010). To detect theeffect of salinity and water stress on the viability of black andbrown seeds, the percentage of viable seeds (%V) after exposureto stress was determined as [(a – b + d)/(c – b)] × 100 %,where a, b and c are the same as described above and d is thenumber of non-germinated viable seeds after 20 d incubationin distilled water.

The control consisted of 16 dishes of 25 seeds withoutpre-treatment with NaCl or PEG-8000 solution dry stored inthe 20 8C incubator. Four dishes of seeds were incubated indistilled water at 30/15 8C after 0, 20, 40 and 60 d of drystorage at 20 8C. Final germination percentages and viabilityof non-germinated seeds were determined after 20 d.

Dynamics of soil seed banks of the two seed morphs

Soil seed banks of S. corniculata were investigated monthlyfrom April to November in 2011 in the natural habitat wherethey were collected. Ten soil cores (5 cm diameter × 10 cmdepth) were collected in a monospecific stand of S. corniculata.The soil cores were sub-divided into three layers: 0–2 cm,2–5 cm and 5–10 cm. The soil samples were washed withwater through a 0.5 mm mesh sieve and seeds were hand sortedfrom the residue (roots and vegetative parts). The number ofviable black and brown seeds each was determined, as describedabove. Only viable seeds were used in calculating seed density inthe soil seed bank.

Seedling emergence and survivorship of sown dimorphic seeds

Ten samples each of 100 black and brown seeds collected in2010 were sown in ten (20 cm × 20 cm) plots in the naturalhabitat of S. corniculata in November 2010. Soil to a depthof 10 cm was removed from the plots, and then the plotswere refilled with soil from the natural habitat that had beensieved by a 0.5 mm mesh sieve to remove seeds ofS. corniculata. A nylon cloth (0.2 mm mesh) was used toisolate the soil in each plot from soil outside the plot and toprevent transport of seeds into and out of each plot. Theedge of the cloth extended about 0.5 cm above the soilsurface. To prevent escape of seeds from, or entry into, theplots, we also had a buffer zone around each of them. Soilin a 10 cm wide × 10 cm deep zone around each plot wasreplaced by the sieved soil mentioned above. In each plot,100 seeds were planted evenly at a depth of about 1 cm. Allnewly emerged seedlings were labelled and recorded eachmonth from April to September 2011, To avoid missing anyseedlings that emerged but soon died, investigation of newlyemerged seedlings was conducted twice each month (on the15th and the last day), and the number of emerging seedlingswas summed each month. The number of seedlings that sur-vived until the end of the experiment was counted (see below).

The experiment was terminated on 30 September 2011when the plants had turned red, indicating the end of thegrowing season. On this date, the number of living seedlingsthat had emerged in each month was recorded, and the soilin the plots was retrieved to determine the number of remainingviable seeds. The number of viable seeds remaining in the soil

was determined by the same method used to determine thenumber of seeds in soil seed banks. The percentage of seedlingemergence (%PE) was determined as (NES/NT) × 100, whereNES is the number of seedlings that emerged during the experi-ment and NT the total number of seeds sown at each plot (100seeds). Survival percentage (%SP) of seedlings was determinedas (NLS/NES) × 100, where NLS is the number of living seedlingson September 30 and NES the same as described above.

All plants that survived until the end of the experimentproduced some seeds. Success of reproduction (%PSR) wasdetermined as (%PE × %SP), where %PE is the percentageof seedling emergence and %SP the survival percentage ofseedlings as described above. During the experiment, soilwater content (SWC), soil salinity, precipitation and tempera-ture at the study site were monitored. Eight soil cores 0–5 cmdeep were collected randomly within the study population ofS. corniculata on the 15th day of each month from April toSeptember 2011. The moisture content of the soil sampleswas determined by oven-drying at 105 8C for 48 h. The salinityof soil samples was analysed by the residue drying qualitymethod (Bao, 2000). Monthly mean temperature andmonthly mean precipitation data were obtained from a nationalweather station 10 km away from the field site.

Statistical analyses

Means of germination percentages of fresh seeds and ger-mination percentages and percentage of viable seeds aftervarious periods of pre-treatments of salinity and droughtstress were compared by analyses of variances (ANOVAs).Fisher’s l.s.d. test was performed for multiple comparisonsto determine significant (P , 0.05) differences betweenmeans. An independent sample t-test was performed to deter-mine significant differences between germination percentagesof fresh seeds in the light and in continuous darkness andbetween germination percentages and percentages of viableseeds after the same period of exposure to NaCl andPEG-8000 solutions. Variables were arcsine square-root trans-formed to meet assumptions of ANOVA for normality andhomogeneity of variance when necessary. Statistical analyseswere performed using SPSS Version 13.0 for windows(SPSS, Chicago, IL, USA).

RESULTS

Black:brown seed ratio and seed mass

The ratio of black to brown seeds collected from 15 individualswas 4:1, thus, there were four times more black than brownseeds produced. Black and brown seeds had a thousand-seedmass of 240.35+ 3.86 mg (mean+ s.e., n ¼ 5) and315.78+ 1.75 mg, respectively.

Germination of fresh seeds

The viability of fresh black and brown seeds was 96.0+ 2.8and 100 %, respectively. In distilled water, fresh black seedsgerminated to significantly higher percentages at high (25/15and 30/15 8C) than at low (15/5 and 20/5 8C) temperatures inboth the light and continuous darkness (Fig. 1A). Light





ownloaded from


promoted germination of fresh black seeds at all temperatureregimes (Fig. 1A). Brown seeds germinated to .97 % at allfour temperature regimes in both light and continuous dark-ness (Fig. 1B). Salinity had detrimental effects on germinationof both black and brown seeds, but black seeds were more sen-sitive to all salinities tested than brown seeds (Fig. 2).Germination of black seeds was ,12 % at 0.4 M NaCl,whereas germination of brown seeds did not decrease to,12 % unless they were incubated at 1.6 M NaCl.

Dormancy breaking of black seeds

Cold stratification gradually increased germination percentageof black seeds at 15/5 8C (Fig. 3). Black seeds germinated to.70 % at 15/5 8C after 15 d cold stratification. Throughout thecold stratification period, black seeds maintained high germinationpercentages at 20/5, 25/15 and 30/15 8C (Fig. 3). Treatmentswith GA3 and testa scarification also significantly increased thegermination percentage of black seeds at 15/5 8C (Fig. 4).

Light requirement for germination of non-dormant black seeds

Buried black seeds exhumed in May 2010 germinated to .75%at all temperatures tested in the light (Fig. 5). However, darknessprohibited seed germination significantly at all temperatures(P , 0.05). Germination of black seeds exhumed in May 2010had decreased to ,30 % at 15/5 8C in darkness (Fig. 5).

Dormancy cycling of the two seed morphs during burial

Black seeds exhumed in May 2010 germinated to a higher per-centage at 15/5 and 20/5 8C compared with fresh black seeds,

and, regardless of temperature, seeds germinated to ≥70 %(Fig. 6B). However, germination of exhumed black seedsdecreased dramatically in summer, and only 5 % of them germi-nated at the most favourable temperature regime (30/15 8C) inAugust and September 2010. Seeds exhumed in October toDecember 2010 exhibited an increase in germination percent-age, and those exhumed in December germinated to high percen-tages at all temperature regimes. Seeds exhumed monthly fromDecember to May had high germination percentages, but insummer germination decreased, as it did in 2010 (Fig. 6B).

When the seeds were first exhumed in May 2010, scarcelyany brown seeds were viable, as most of them had germinatedduring burial.

Effect of salinity on germination of buried seeds

Dormancy cycling varied quantitatively with the salinity atwhich germination was tested (Fig. 6C), but the pattern wassimilar to that of seeds incubated in distilled water at 30/15 8C.When the black seeds were first exhumed in May 2010, theyhad higher relative germination percentages at salinities of0.2, 0.4 and 0.6 M NaCl than fresh seeds (Fig. 6C). Relativegermination percentages at salinities of 0.2, 0.4 and 0.6 M

NaCl were high in spring, autumn and winter, but low insummer (Fig. 6C). Throughout the experiment, only one seedgerminated at 0.8 M NaCl (in January 2011).

Effect of salinity and water stress on dormancy and viability of thetwo seed morphs

Exposure of black seeds to a salinity of 1 M NaCl for 20 dincreased seed germination, but germination decreased after

100

75

50

25

0

100

75

50

25

0

LightDark

Aa

Aa Aa Aa Aa Aa Aa Aa Aa

Aa

BaA Black seeds

B Brown seeds

Ba

Ab

20/5 25/15

Temperature (°C)

Ger

min

atio

n (%

)G

erm

inat

ion

(%)

ABbBCb

Cb

30/1515/5

FI G. 1. Germination percentages of fresh black (A) and brown (B) seeds of Suaeda corniculata at various temperatures in the light (12 h photoperiod) and incontinuous darkness. Bars with the same uppercase letters are not significantly different (P . 0.05) across all temperature regimes. The same lowercase lettersindicate no significant difference (P . 0.05) between germination percentages (means+ s.e.) in light and continuous darkness with the same thermoperiod.





ownloaded from


exposure for 40 or 60 d (Fig. 7A). Exposure to the isotonicPEG-8000 solution for 20 and 40 d did not affect germinationof black seeds, but 60 d of exposure decreased germinationsignificantly (P , 0.05, Fig. 7A). Exposure to a salinity of 1M NaCl and to isotonic PEG solutions did not decrease the via-bility of black seeds within 60 d (Fig. 8A).

The viability of brown seeds that were not pre-treated in 1 M

NaCl or in isotonic PEG-8000 solution was 100 % (Fig. 8B),and they had 100 % germination in distilled water (Fig. 7B).Viability of brown seeds exposed to both 1 M NaCl and isotonicPEG solution for 20 d decreased with time (Fig. 8B). Therewere 12.8+ 2.8 % and 15.0+ 3.0 % brown seeds that wereviable after 60 d pre-treatment in 1 M NaCl and isotonicPEG-8000 solution, respectively. Thus, about 85 and 87 %of the brown seeds died following exposure to salinity and

water stress for 60 d. No viable non-germinated brown seedswere found after 20 d germination tests in distilled waterafter they had been transferred from NaCl or PEG-8000 solu-tion. Thus, after being pre-treated in 1 M NaCl and isotonicPEG-8000 solution, the recovery germination percentage(%RG) of brown seeds changed in the same way as did the per-centage of viable seeds (%V).

Dynamics of soil seed banks for dimorphic seeds

Both black and brown seeds were most abundant in the 0–2 cm level of soil, and seed density decreased with an increasein soil depth (Fig. 9). No seeds, except for one black seed(found in a soil core obtained in August 2011), were foundat the 5–10 cm soil depth.

100

80

60

40

20

0

Ger

min

atio

n (%

)

15/5

Aa

Ba Ba Ba

Bb

Cb

Cb

Ab

20/5 25/15 30/15

Temperature (°C)

LightDark

FI G. 5. Germination percentages of black seeds of Suaeda corniculataexhumed in May 2010 and incubated at various temperatures in the light(12 h photoperiod) and in continuous darkness. Bars with the same uppercaseletters are not significantly different (P . 0.05) across all temperature regimes.The same lowercase letters indicate no significant difference (P . 0.05)between germination percentages (means+ s.e.) in the light and continuous

darkness with the same thermoperiod.

100

75

50

25

00 0·2 0·4 0·6

Salinity (mol L–1)

0·8 1·2 1·6

BrownBlack

Aa Aa Aa

Da

Ea

Ba

Ca

Ab

Bb

Db Db Db Da

Ger

min

atio

n (%

)

Cb

FI G. 2. Germination percentages of fresh black and brown seeds of Suaedacorniculata at various salinities at 30/15 8C in a 12 h photoperiod. Bars withthe same uppercase letters indicate no significant difference (P . 0.05)between germination percentages (means+ s.e.) of the same seed morphsacross all salinities. The same lowercase letters indicate no significant differ-ence (P . 0.05) between germination percentages (means+ s.e.) of the two

seed morphs at the same salinity.

100

75

50

25

00 5 10 15

15/5 °C20/5 °C25/15 °C30/15 °C

20

Duration of cold stratification (d)

Ger

min

atio

n (%

)

FI G. 3. Germination percentages of black seeds of Suaeda corniculata incu-bated at various alternating temperatures in a 12 h photoperiod after 0, 5, 15

and 20 d of cold stratification at 5 8C in continuous darkness.

Pretreatment

100

75

50

a

bb b

25

0

Ger

min

atio

n (%

)

CK TS 0·1 GA 1 GA

FI G. 4. Germination percentages of fresh black seeds of Suaeda corniculatain 0.1 (0.1 GA) and 1 mmol L21 GA3 (1 GA) and after testa scarification (TS)at 15/5 8C (12 h/12 h) in a 12 h photoperiod. CK is the untreated control.Values are means+ s.e. Bars with the same letters indicate no significant dif-

ference (P . 0.05) between germination percentages.





ownloaded from


There were 3464+ 803 (mean+ s.e.) black seeds m22 at the0–10 cm soil depth in April 2011 (Fig. 9) and the numberdeclined month by month, reaching 662+ 153 seeds m22 inSeptember. Seed density began to increase in October 2011,and it reached 3311+ 829 seeds m22 in November (Fig. 9).

The number of brown seeds in the soil showed similar tem-poral changes to those of black seeds (Fig. 9). The density ofbrown seeds was 509+ 170 seeds m22 in April 2011, whichdeclined to 153+ 78 seeds m22 in June and to 0 in July,August and September. As a result of input of brown seedsinto the soil in autumn (mostly October) 2010, the densityreached 1120+ 354 seeds m22 in November 2011 (Fig. 9).

Seedling emergence and survivorship of sown seed morphs

Seedlings from black seeds emerged in July and Augustwhen the primary precipitation event of the year occurred in

2011 (Fig. 10). Only 2.6+ 0.7 % of the black seeds recruitedseedlings (%PE), of which approx. 85 % (2.2/2.6) emerged inJuly (Fig. 10). None of the seedlings that emerged in Augustsurvived until the end of the growing season (30 September).About 19 % (0.5/2.6) of the seedlings that emerged survived(%SP) and produced seeds. Thus, 0.5 % of the seeds sown pro-duced plants that reproduced (%PSR); however, at the end ofthe experiment 35.1+ 3.4 % of the black seeds were stillalive in the soil.

Seedlings derived from sown brown seeds appeared in thefield plots in May (Fig. 10). Seedling emergence percentage(%PE) was 2.8+ 0.6 %, and about 14 % (0.4/2.8) of the seed-lings survived to reproduce successfully (%SP). No brownseeds persisted in soil at the end of the experiment. Thus, al-together, only 0.4+ 0.2 % of the brown seeds sown produceda plant that reproduced (%PSR).

20

A

B

C

0

MaxMax–20

100

75

50

25

0N D J F M A M J J A S O N D

Date exhumed (month)J F M A M J J A S O N D

Rel

ativ

e ge

rmin

atio

n (%

)

100

75

50

25

0

Ger

min

atio

n (%

)Te

mpe

ratu

re (

°C)

15/5 °C20/5 °C

25/15 °C30/15 °C

0·2 mol L–1

0·4 mol L–1

0·6 mol L–1

0·8 mol L–1

FI G. 6. (A) Mean monthly maximum and minimum air temperatures at the seed burial site. (B) Germination percentages (means+ s.e.) of black seeds ofSuaeda corniculata incubated at various temperature regimes for a 12 h photoperiod following 0–24 months of burial in soil in the field. (C) Relative germinationpercentages of black seeds of S. corniculata incubated at 30/15 8C in a 12 h photoperiod at salinities of 0.2, 0.4, 0.6 and 0.8 mol L21 NaCl solutions following

0–24 months of burial in soil.





ownloaded from


DISCUSSION

Black seeds of S. corniculata have a dormancy mechanism atmaturity that prevents germination at low temperatures.Germination of black seeds can be improved by testa scarifica-tion and GA3, which are a characteristic of seeds with physio-logical dormancy (Baskin and Baskin, 1998). Also, a veryshort period of cold stratification (15 d) increased germinationof black seeds remarkably. Taken together, the effectiveness ofthese three treatments on dormancy break indicates that blackseeds have non-deep physiological dormancy (Baskin andBaskin, 1998). Artificial cold stratification failed to increasegermination of black seeds to .75 % at low temperatures.However, exposure to low winter temperatures in the fieldresulted in .95 % germination of buried black seeds at alltemperature regimes tested. This difference may be due to dif-ferences in conditions between artificial and natural coldstratification. Wetson et al. (2008) found that differences intemperature and salinity during dormancy affected subsequentgermination of S. maritima seeds.

Black seeds of S. corniculata have less mass than brownones and need light to germinate, while the brown seeds donot require light. A light requirement for germination generallydecreases with an increase in seed mass in many species intemperate areas (Milberg et al., 2000; Schutz et al., 2002),and it increases the ability of a species to form a persistentsoil seed bank (Baskin and Baskin, 1989; Pons, 1991).Black seeds of S. corniculata maintain a persistent soil seedbank, while brown seeds form only a transient soil seedbank. Thus, black seeds persist for a longer period in thesoil than brown seeds. The light requirement for germinationin black seeds may also act as a depth-sensing cue.

Fresh black seeds of S. corniculata can germinate to highpercentages at high temperatures and to low percentages atlow temperatures, thus the seeds are conditionally dormant.Seeds are non-dormant in late winter–spring of the next yearand they can germinate to high percentages at both high andlow temperatures (Fig. 6). By July or August, most of theseeds had entered (2010) or re-entered (2011) dormancy viaconditional dormancy (CD). A small percentage of the seeds,

100

75

50

25

0

100

Aa Aa Aa

Aa Aa Aa Aa AaAa

Bb

BCb

Duration of pretreatment (d)

0 20 40 60

Cb

Cb Cb

Bb

Aa

Bb

Aa Aa Aa

Ca

Bb

AaAa

75

50

25

0

Ger

min

atio

n (%

)G

erm

inat

ion

(%)

CKNaClPEG

A Black

B Brown

FI G. 7. Germination percentages (means+ s.e.) of black (A) and brown (B) seeds of Suaeda corniculata in a 12 h photoperiod at 30/15 8C after 0, 20, 40 and 60d of exposure to 1 mol L21 NaCl or isotonic PEG-8000 solutions for a 12 h photoperiod at 20 8C. CK seeds were dry stored at 20 8C for 0, 20, 40 and 60 d. Barswith the same uppercase letters indicate no significant difference (P . 0.05) between germination percentages of seeds after various periods of exposure to NaCl,PEG-8000 solution or dry storage. The same lowercase letters indicate no significant difference (P . 0.05) between germination percentages of seeds pre-treated

with NaCl or PEG-8000 solution for the same periods.





ownloaded from


especially in the second year of burial (2011), remained condi-tionally dormant and retained the ability to germinate at 25/15and 30/15 8C in August and September. Thus, it generally canbe stated that when black seeds are first produced they are in astate of CD, and they become non-dormant by the followinglate winter–spring. Further, those that do not germinate inthe first spring subsequently cycle between dormancy (D)and non-dormancy (ND): CD � ND ↔ D. This pattern ofannual dormancy changes fits that of summer annuals shownin Baskin and Baskin (1998; panel b of fig. 4.4).

In addition to dormancy cycles, black seeds of S. corniculataalso exhibit an annual cycle of sensitivity of germination to sal-inity. The cycle of sensitivity to salinity corresponds closely todormancy cycles of seeds in the soil seed bank. Inhibitoryeffects of salinity on germination decrease as black seedscome out of dormancy and increase as non-dormant seedsre-enter dormancy. Baskin and Baskin (1998) concluded thatthe temperature range for seed germination and sensitivity ofseeds to light, substrate moisture and hormones is correlatedwith the continuum of dormancy states. In our study, sensitivity

of seed germination to salinity for black seeds of S. corniculatais also shown to be a correlate with the dormancy continuum.Thus, the increase in salt tolerance as seeds come out of dor-mancy and the decrease as seeds enter dormancy adds a newcorrelate to those formulated by Baskin and Baskin (1998) forseeds as they cycle back and forth through a dormancycontinuum.

In contrast, brown seeds of S. corniculata do not form a per-sistent soil seed bank, and of course they cannot have dor-mancy cycles. Brown seeds were non-dormant at maturity,and they germinated to high percentages across the range ofhigh to low temperatures and in continuous darkness. Thus,it is not surprising that the buried brown seeds had germinatedin the spring when they were exhumed. Not unexpectedly, lowwinter temperature did not induce non-dormant brown seeds ofthis summer annual into dormancy (Baskin and Baskin, 1998).Carter and Ungar (2003) also found different dormancy char-acteristics in dimorphic seeds of the two summer annual halo-phytes Atriplex prostrate and Salicornia europaea(Amaranthaceae). Small seeds of both A. prostrate and

100

75

50

25

0

100

Aa

Aa Aa Aa Aa AaAa

Bb

BCb

Duration of pretreatment (d)

0 20 40 60

Cb

Cb Cb

Bb

Aa Aa

Aa Aa Aa Aa Aa AaAa Aa AaAa Aa Aa

75

50

25

0

Via

ble

seed

s (%

)V

iabl

e se

eds

(%)

CKNaClPEG

A Black

B Brown

FI G. 8. Percentages (means+ s.e.) of viable black (A) and brown (B) seeds of Suaeda corniculata in a 12 h photoperiod at 30/15 8C after 0, 20, 40 and 60 d ofexposure to 1 mol L21 NaCl or isotonic PEG-8000 solutions for a 12 h photoperiod at 20 8C. CK seeds were dry stored at 20 8C for 0, 20, 40 and 60 d. Bars withthe same uppercase letters indicate no significant difference (P . 0.05) between percentages of viable seeds (%PV) after various periods of exposure to NaCl,PEG-8000 solution or dry storage. The same lowercase letters indicate no significant difference (P . 0.05) between percentages of viable seeds pre-treated by

NaCl or PEG-8000 solution for the same periods.





ownloaded from


S. europaea had dormancy cycling, whereas large seeds werenon-dormant (S. europaea) or conditionally dormant(A. prostrate) at maturity and were non-dormant in the follow-ing spring (Carter and Ungar, 2003).

The different performance of black and brown seeds afterexposure to salinity and drought stress may explain to someextent the mechanism of the difference in their dynamics ofdormancy. Black seeds were induced into dormancy during ex-posure to salinity and water stress, while brown seeds remainednon-dormant. It can be inferred that non-dormant black seedsthat have not germinated in a germination season may becomedormant and persist in the soil seed bank when they experienceenvironmental stress. The viability of brown seeds is largelyreduced by salinity and drought stress, and very few seedsmaintain viability after exposure to salinity and droughtstress. However, the seeds that remain viable germinatereadily after removal of environmental stress. Brown seedscan therefore form only a transient soil seed bank. The differ-ences in responses of dimorphic seeds of S. corniculata to sal-inity are similar to those of S. splendens (Redondo-Gomezet al., 2008), which also produces brown and black seeds.Black seeds of S. splendens maintain viability for a longerperiod of time under high salinities than brown seeds(Redondo-Gomez et al., 2008). However, it is unclear

whether dimorphic seeds of S. splendens differ in persistencein the soil seed bank.

Dormancy cycles of seeds have been shown in many speciesgrowing in humid temperate regimes (Baskin and Baskin,1980, 1985, 1989), and they usually have been considered tobe driven primarily by seasonal temperature changes(Vanlerberghe and Van Assche, 1986; Van Assche andVanlerberghe, 1989; Bouwmeester and Karssen, 1992). Inour study, a 60 d exposure to 1.0 M salinity or to an isotonicPEG solution induced black seeds of S. corniculata into dor-mancy, which suggests that soil salinity and drought mayalso play a role in regulating dormancy cycles of black seedsof S. corniculata in their cold desert habitat. In this study,seeds were exposed to soil salinity (1.0 M) in the field foronly 1 month (June) in summer, and soil salinity wasdecreased by precipitation in July when seedlings appearedin the plots. Thus, it seems unlikely that soil salinity induceddormancy in black seeds of S. corniculata. However, itshould be noted that soil salinity in May is also very high,and the temperature may be around 20 8C. Also, in Julythere is high variability in time when substantial precipitationwill occur. Thus, black seeds of S. corniculata could experi-ence high salinity in the soil in May, June and July, and thisperiod would be long enough for high salinity to promote

4000

3000

2000

1000

0

See

d de

nsity

(se

eds

m–2

)S

eed

dens

ity (

seed

s m

–2)

See

d de

nsity

(se

eds

m–2

)

1500

1000

500

0

4000

3000

2000

1000

0

1500

1000

500

0

4000

3000

2000

1000

0Apr May Jun Jul Aug

Month

Sep Oct Nov Apr May Jun Jul Aug

Month

Sep Oct Nov

1500

1000

500

0

Black (0–2 cm) Brown (0–2 cm)



FI G. 9. Densities of black (left) and brown (right) seeds at different soil depth in their natural habitat from April to November 2011. Values are means+ s.e.Only one black seed (in August) and no brown seeds were present at the 5–10 cm soil depth.





ownloaded from


dormancy induction in black seeds of S. corniculata. Althoughsoil salinity may play a role in regulating dormancy inductionin black seeds, it only induces a small proportion of them intodormancy (Fig. 7). However, a high percentage of the buriedblack seeds are induced into dormancy in summer (Fig. 6),leading us to conclude that high temperatures in summermay also be involved in regulating dormancy cycling ofblack seeds of S. corniculata.

Black seeds of S. corniculata have a ‘cautious’ strategy of ger-mination (Gutterman, 1993). They maintain viability after aperiod of exposure to salinity stress and can commence germin-ation once this stress is removed, which is a typical response ofhalophyte seeds to salinity stress (Ungar, 1962; Khan and Ungar,1997; Hanslin and Eggen, 2005; Song et al., 2005). A long ex-posure to salinity and drought stress induced some black seedsinto dormancy, which allows the species to withstand unfavour-able periods of environmental stress. Brown seeds can germin-ate at high salinity of 1 M NaCl, which makes them distinct

from seeds of glycophytes (non-salt-tolerant plants). However,most of them cannot maintain viability during salinity stress,which makes them different from other halophyte seeds.Brown seeds of S. corniculata show an ‘opportunistic’ strategyof germination (Gutterman, 1993), and they are ready to germin-ate whenever there are favourable conditions for germination,which gives the population a competitive advantage via earlygermination (Ross and Harper, 1972).

The ecological literature is sometimes confusing on the im-portance of seed dormancy and the maintenance of persistentsoil seed banks. Cavieres and Arroyo (2001) state that forma-tion of a persistent seed bank depends upon seed dormancy.However, Thompson et al. (2003) did not find a close relation-ship between seed dormancy and persistence in the soil bycomparing the longevity index and dormancy states of 599species. In our study, retention of viability of seeds ofS. corniculata played a big role in maintenance of a seedbank. Exposure to high salinity and drought stress led to a

Pre

cipi

tatio

n (m

m)

Soi

l wat

er c

onte

nt (

%)

Num

ber

of s

eedl

ings

Num

ber

of s

eedl

ings

150

20

15

10

5

0·9

0·6

0·3

0·0

100

50

0

25

20

15

10

5

0

2

1

0

2

3

1

0

Temperature

Precipitation

SWC

Emergence

Survival

SS

D Brown

C Black

B

A

Emergence

Survival

Soil salinity (m

ol L–1)

Apr May Jun Jul Aug

Month

Sep

Temperature (°C

)

FI G. 10. (A) Mean air temperature and precipitation, (B) soil water content (SWC) and salinity (SS), and seedling emergence and survival from (C) black and(D) brown seeds of Suaeda corniculata.





ownloaded from


dramatic decrease in viability of brown seeds. Furthermore,since brown seeds germinate to high percentages in darkness,even if they could remain viable the seed bank would bedepleted by in situ germination. The black seeds maintain via-bility and enter dormancy after exposure to high salinity anddrought stress, and they have a light requirement for germin-ation, which delays depletion of the soil seed bank.Differences in the ability to maintain viability and in dor-mancy characteristics of the two seed morphs ofS. corniculata lead to difference in their persistence in soilseed banks.

A conceptual model of the seed dynamics of black andbrown seeds of S. corniculata is shown in Fig. 11. Blackseeds are conditionally dormant when they mature inautumn, and they become non-dormant after exposure tocold temperature during early winter and maintain ND inlate winter and spring. Black seeds of S. corniculata are differ-ent from type C seeds of Salsola affinis (Amaranthaceae),which came out of dormancy in the field (Junggar Basin,NW China) in late winter or early spring (Wei et al., 2007).Although daily minimum temperature in early winter(November) is below 0 8C, daily maximum temperature (8.6

8C) in the natural habitat during this time is in the temperaturerange for cold stratification for seeds of many summer annualsin temperate regions (Baskin and Baskin, 1998). A 15 d coldstratification at 5 8C can break dormancy of black seeds.Thus, dormancy release of black seeds can occur in earlywinter. High temperature and high salinity in the soil mayinduce dormancy in black seeds in summer, and the seedsbecome conditionally dormant after they are exposed to highsalinity. High precipitation in summer decreases soil salinities,which lead to 2.6 % germination of black seeds. However, only0.5 % of the black seeds had the probability to recruit seed-lings that survived to produce seeds the first year. At the endof the growing season, 35.1 % of black seeds entered the per-sistent soil seed bank, and they had an annual dormancy cycleof D ↔ ND.

Brown seeds of S. corniculata perform differently fromblack seeds (Fig. 11). Brown seeds are non-dormant whenthey are shed from the mother plant and enter the transientsoil seed bank in autumn. They are non-dormant throughwinter and in the spring of the following year, and their ger-mination can be triggered by the relatively small amount ofprecipitation in spring. Thus, brown seeds can result in

Seed production and dispersalAdult plants Adult plants

Black seeds (CD) Brown seeds (ND)

Winter: low temperature

100 % 100 %

Spring: low temperature, low precipitation, moderate salinity

Germination

Early summer: high temperature, high salinity

NDND

NDND

CD

D

Mid- and late-summer: high temperature, high precipitation, low salinity

Autumn: low temperature

CD

CD

CD

DND Dormancycycling

The transient soil seed bank

The persistent soil seed bank

Germination

Ger

min

atio

nE

mer

ged

seed

lings

Non

-em

erge

d se

edlin

gsE

merged seedlings

Non-em

erged seedlings

2·6

%

62·3

%

Die85·0 %

2·8 %12·2 %

0·5 % 0·4 %

35·1 %

FI G. 11. Conceptual model of changes in dormancy status of black (in black) and brown (in red) seeds of Suaeda corniculata and soil seed bank dynamics andseedling regeneration (in blue) of the population. The percentages indicate the proportions of seeds in a transient soil seed bank that recruit seedlings and enter thepersistent soil seed bank. Abbreviations: CD, conditional dormancy; D, dormancy; ND, non-dormancy. The dashed-dotted lines represent possible fates of black

seeds in the persistent soil seed bank.





ownloaded from


establishment of the population in earlier summer than blackseeds. About 2.8 % of brown seeds in the transient soil seedbank recruited seedlings, but only seedlings derived from 0.4% of seeds produced seeds. Apparently, high soil salinity inearly summer leads to a dramatic decrease in viability ofbrown seeds, and no brown seeds were found in the soil inJuly, August or September. Thus, it can be inferred that allbrown seeds were non-viable in the field by July.

Differences in performance of the two seed morphs ofS. corniculata in the soil seed bank increase fitness of thespecies in unpredictable saline environments. Brown seedscan germinate under a wider range of temperatures and sali-nities than black seeds. Seed germination and seedling emer-gence is triggered by small precipitation events in earlyspring. Early germination gives plants a selective advantage,since they are less likely to be attacked by predators and patho-gens than late germinating individuals (Seiwa, 1998). Earlygermination also endows early germinating individuals witha competitive advantage to capture resources (Ross andHarper, 1972), and thus they are also likely to grow muchlarger and thus to produce more seeds than those that germin-ate late. Germination of black seeds is regulated by an annualdormancy cycle of D ↔ ND, and non-dormant black seedshave stricter germination requirements than brown seeds.They cannot germinate in natural saline habitats until soil sal-inity is decreased by precipitation. Thus, seedlings derivedfrom black seeds of S. corniculata emerge mainly in thesummer rainy season. Although seedlings of late-germinatingblack seeds have a shorter growing season than those of early-germinating brown seeds, young seedlings are exposed to morefavourable moisture conditions for growth than those frombrown seeds. However, if seedlings/plants from brown seedscan survive until the summer rainy season they wouldalready be established and could respond rapidly to improvemoisture conditions for growth.

ACKNOWLEDGEMENTS

This work was funded by the Key Basic Research andDevelopment Plan of China (2010CB951304) and theNational Natural Science Foundation of PR China(30872074, 30970461 and 31170383).

LITERATURE CITED

Anderson TM, Schutz M, Risch AC. 2012. Seed germination cues and theimportance of the soil seed bank across an environmental gradient inthe Serengeti. Oikos 121: 306–312.

Atia A, Rabhi M, Debez A, Barhoumi Z, Abdelly C, Smaoui A. 2011.Factors controlling germination and dormancy processes in dimorphicfruits of Atriplex inflata (Chenopodiaceae). Plant Ecology andEvolution 144: 307–312.

Bao SD. 2000. Soil chemistry and agriculture analysis. (In Chinese.) Beijing:China Agriculture Press.

Baskin CC, Baskin JM. 1998. Seeds: ecology, biogeography, and evolution ofdormancy and germination. San Diego, CA: Academic Press.

Baskin CC, Chesson PL, Baskin JM. 1993. Annual seed dormancy cycles intwo desert winter annuals. Journal of Ecology 81: 551–556.

Baskin JM, Baskin CC. 1976. Germination dimorphism in Heterotheca sub-axillaris var. subaxillaris. Bulletin of the Torrey Botanical Club 103:201–206.

Baskin JM, Baskin CC. 1980. Ecophysiology of secondary dormancy inseeds of Ambrosia artemisiifolia. Ecology 61: 475–480.

Baskin JM, Baskin CC. 1985. The annual dormancy cycle in buried weedseeds: a continuum. BioScience 35: 492–498.

Baskin JM, Baskin CC. 1989. Physiology of dormancy and germination inrelation to seed bank ecology. In: Leck MA, Parker VT, Simpson RL.eds. Ecology of soil seed banks. San Diego, CA: Academic Press.

Baskin JM, Baskin CC, McCormick JF. 1987. Seasonal changes in germin-ation responses of buried seeds of Portulaca smallii. Bulletin of theTorrey Botanical Club 114: 169–172.

Bouwmeester HJ, Karssen CM. 1992. The dual role of temperature in theregulation of the seasonal changes in dormancy and germination ofseeds of Polygonum persicaria L. Oecologia 90: 88–94.

Carter CT, Ungar IA. 2003. Germination response of dimorphic seeds of twohalophyte species to environmentally controlled and natural conditions.Canadian Journal of Botany 81: 918–926.

Cavieres LA, Arroyo MTK. 2001. Persistent soil seed banks in Phaceliasecunda (Hydrophyllaceae): experimental detection of variation alongan altitudinal gradient in the Andes of central Chile (33 8S). Journal ofEcology 89: 31–39.

El-Keblawy A. 2003. Effects of achene dimorphism on dormancy and progenytraits in the two ephemerals Hedypnois cretica and Crepis aspera(Asteraceae). Canadian Journal of Botany 81: 550–559.

Evenari M, Kadouri A, Gutterman Y. 1977. Eco-physiological investiga-tions on the amphicarpy of Emex spinosa (L.) Campd. Flora 166:223–238.

Flint SD, Palmblad IG. 1978. Germination dimorphism and developmentalflexibility in the ruderal weed Heterotheca grandiflora. Oecologia 36:33–43.

Guma IR, Padron-Mederos MA, Santos-Guerra A, Reyes-Betancort JA.2010. Effect of temperature and salinity on germination of Salsola vermi-culata L. (Chenopodiaceae) from Canary Islands. Journal of AridEnvironments 74: 708–711.

Gutterman Y. 1993. Seed germination in desert plants. Berlin:Springer-Verlag.

Hanslin HM, Eggen T. 2005. Salinity tolerance during germination of sea-shore halophytes and salt-tolerant grass cultivars. Seed ScienceResearch 15: 43–50.

Hutchings MJ, Booth KD. 1996. Studies on the feasibility of re-creatingchalk grassland vegetation on ex-arable land. I. The potential roles ofthe seed bank and the seed rain. Journal of Applied Ecology 33:1171–1181.

Imbert E. 1999. The effects of achene dimorphism on the dispersal in timeand space in Crepis sancta (Asteraceae). Canadian Journal of Botany77: 508–513.

Imbert E. 2002. Ecological consequences and ontogeny of seed heteromorph-ism. Perspectives in Plant Ecology, Evolution and Systematics 5: 13–36.

Imbert E, Ronce O. 2001. Phenotypic plasticity for dispersal ability in theseed heteromorphic Crepis sancta (Asteraceae). Oikos 93: 126–134.

Joley DB, Maddox DM, Schoenig SE, Mackey BE. 2003. Parameters affect-ing germinability and seed bank dynamics in dimorphic achenes ofCentaurea solstitialis in California. Canadian Journal of Botany 81:993–1007.

Katembe WJ, Ungar IA, Mitchell JP. 1998. Effect of salinity on germinationand seedling growth of two Atriplex species (Chenopodiaceae). Annals ofBotany 82: 167–175.

Khan MA, Ungar IA. 1984. The effect of salinity and temperature on the ger-mination of polymorphic seeds and growth of Atriplex triangularis Willd.American Journal of Botany 71: 481–489.

Khan MA, Ungar IA. 1997. Effects of thermoperiod on recovery of seed ger-mination of halophytes from saline conditions. American Journal ofBotany 84: 279–283.

Leck MA, Simpson RL. 1987. Seed bank of a freshwater tidal wetland: turn-over and relationship to vegetation change. American Journal of Botany74: 360–370.

Lomonosova M, Brandt R, Freitag H. 2008. Suaeda corniculata(Chenopodiaceae) and related new taxa from Eurasia. Willdenowia 38:81–109.

Mandak B. 1997. Seed heteromorphism and the life cycle of plants: a litera-ture review. Preslia 69: 129–159.

Mandak B, Pysek P. 2001. Fruit dispersal and seed banks in Atriplex sagit-tata: the role of heterocarpy. Journal of Ecology 89: 159–165.

McCue KA, Holtsford TP. 1998. Seed bank influences on genetic diversity inthe rare annual Clarkia springvillensis (Onagraceae). American Journalof Botany 85: 30–36.





ownloaded from


Michel BE. 1983. Evaluation of the water potentials of solutions of polyethyl-ene glycol 8000 both in the absence and presence of other solutes. PlantPhysiology 72: 66–70.

Milberg P, Andersson L, Thompson K. 2000. Large-seeded species are lessdependent on light for germination than small-seeded ones. Seed ScienceResearch 10: 99–104.

Panetta FD, Randall RP. 1993. Variation between Emex australis populationsin seed dormancy/non-dormancy cycles. Australian Journal of Ecology18: 275–280.

Pons TL. 1991. Induction of dark dormancy in seeds: its importance for theseed bank in the soil. Functional Ecology 5: 669–675.

Pujol JA, Calvo JF, Ramirez-Diaz L. 2000. Recovery of germination fromdifferent osmotic conditions by four halophytes from southeasternSpain. Annals of Botany 85: 279–286.

Redondo-Gomez S, Mateos-Naranjo E, Cambrolle J, Luque T, FigueroaME, Davy AJ. 2008. Carry-over of differential salt tolerance in plantsgrown from dimorphic seeds of Suaeda splendens. Annals of Botany102: 103–112.

Ross MA, Harper JL. 1972. Occupation of biological space during seedlingestablishment. Journal of Ecology 60: 77–88.

Schutz W, Milberg P, Lamont BB. 2002. Seed dormancy, after-ripening andlight requirements of four annual Asteraceae in south-western Australia.Annals of Botany 90: 707–714.

Seiwa K. 1998. Advantages of early germination for growth and survival ofseedlings of Acer mono under different overstorey phenologies in decidu-ous broad-leaved forests. Journal of Ecology 86: 219–228.

Song J, Fan H, Zhao YY, Jia YH, Du XH, Wang BS. 2008. Effect of salinityon germination, seedling emergence, seedling growth and ion accumula-tion of a euhalophyte Suaeda salsa in an intertidal zone and on salineinland. Aquatic Botany 88: 331–337.

Song J, Feng G, Tian CY, Zhang FS. 2005. Strategies for adaptation of Suaedaphysophora, Haloxylon ammodendron and Haloxylon persicum to a salineenvironment during seed-germination stage. Annals of Botany 96: 399–405.

Tanksley SD, McCouch SR. 1997. Seed banks and molecular maps: unlock-ing genetic potential from the wild. Science 277: 1063–1066.

Thompson K, Ceriani RM, Bakker JP, Bekker RM. 2003. Are seed dor-mancy and persistence in soil related? Seed Science Research 13:97–100.

Ungar IA. 1962. Influence of salinity on seed germination in succulent halo-phytes. Ecology 43: 763–764.

Van Assche JA, Vanlerberghe KA. 1989. The role of temperature on the dor-mancy cycle of seeds of Rumex obtusifolius L. Functional Ecology 3:107–115.

Vanlerberghe KA, Van Assche JA. 1986. Dormancy phases in seeds ofVerbascum thapsus L. Oecologia 68: 479–480.

Venable DL. 1985. The evolutionary ecology of seed heteromorphism.American Naturalist 126: 577–595.

Venable DL, Levin DA. 1985. Ecology of achene dimorphism in Heterothecalatifolia: I. Achene structure, germination and dispersal. Journal ofEcology 73: 133–145.

Wang L, Huang Z, Baskin CC, Baskin JM, Dong M. 2008. Germination ofdimorphic seeds of the desert annual halophyte Suaeda aralocaspica(Chenopodiaceae), a C4 plant without Kranz anatomy. Annals ofBotany 102: 757–769.

Wei Y, Dong M, Huang Z. 2007. Seed polymorphism, dormancy and germin-ation of Salsola affinis (Chenopodiaceae), a dominant desert annual inha-biting the Junggar Basin of Xinjiang, China. Australian Journal of Botany55: 464–470.

Weiss PW. 1980. Germination, reproduction and interference in the amphicar-pic annual Emex spinosa (L.) Campd. Oecologia, 45: 244–251.

Wetson AM, Cassaniti C, Flowers TJ. 2008. Do conditions during dormancyinfluence germination of Suaeda maritima? Annals of Botany 101:1319–1327.

Yao SX, Lan HY, Zhang FC. 2010. Variation of seed heteromorphism inChenopodium album and the effect of salinity stress on the descendants.Annals of Botany 105: 1015–1025.





ownloaded from


Documents

Comparison of germination and seed bank dynamics of ...sourcedb.ib.cas.cn/cn/ibthesis/201301/P020130104332326552685.pdf · Comparison of germination and seed bank dynamics of dimorphic