SHORT RESEARCH PAPER

Seed–litter–position drives seedling establishment in grasslandspecies under recurrent droughtC. WellsteinDepartment of Biogeography, University of Bayreuth, Bayreuth, Germany

INTRODUCTION

Sudden alterations in climate and land use undoubtedly affectkey processes of plant life, such as growth and reproduction.Early developmental stages of plants are expected to be moresensitive to environmental alterations than adult stages, and assuch, represent a major bottleneck to plant regeneration(Xiong & Nilsson 1999; Walck et al. 2011). Focusing on theeffects of climate on plant regeneration, besides temperature,water supply is a critical driver for germination and plantrecruitment (Walck et al. 2011). It is widely acknowledged thatthe frequency and severity of drought during the growing sea-son will increase as current global climate change progresses(Easterling et al. 2000; Meehl et al. 2000; IPCC 2007). Reducedprecipitation has already been reported to occur across the nat-ural range of many European grassland species (IPCC 2007).

With respect to changes in land use during the last dec-ades, alterations in grassland management have been mainlydue to intensification or abandonment. Both alter defoliationregimes and subsequently the amount and timing of litterdeposition. For example, in abandoned grasslands a lack ofbiomass export leads to litter accumulation with subsequentincrease in thickness of the litter layer (Kahmen et al. 2002;Ruprecht et al. 2010). With respect to timing, freshly shedseeds can be situated in the top layer of litter carpets or,when litter input occurs after seed rain, seeds will be posi-

tioned beneath a covering layer. Field and experimental stud-ies provide general evidence for the significance of a litterlayer for seedling germination and recruitment. Largeamounts of overlaying litter cover were found to exert nega-tive effects on seedling recruitment, while moderate amountsof litter may facilitate regeneration from seeds (Xiong & Nils-son 1999; Rotundo & Aguiar 2005; Donath & Eckstein 2010).In the absence of litter, i.e. on bare soil, negative effects suchas water stress may prevail (Eckstein & Donath 2005).

The fate of seeds might also be influenced by their mobil-ity. Seed size might modulate seed mobility, as small seedsmore easily ‘percolate’ through a litter layer (Chambers et al.1991; Donath & Eckstein 2010, 2011), have a higher tendencyto be incorporated into the soil and consequently to form asoil seed bank (Bekker et al. 1998). In contrast, large-seededspecies are less mobile, are therefore more likely retained inthe litter, and show a lower tendency to accumulate seeds inthe soil. Generally, seeds that do not germinate immediatelycan persist beyond one season, with higher longevity in thesoil seed bank than on the soil surface.

There is general evidence for the impact of water supplyand litter on seedling emergence, but their interactive effectshave received little attention so far. Despite this, field studieson seedling germination and recruitment of plant speciespoint towards the importance of such interactive effects oflitter cover, as well as drought. During & van Tooren (1990)

Keywords

Climate change; facilitation; germination;

land-use change; seed mass; seedling

recruitment.

Correspondence

C. Wellstein, Department of Biogeography,

University of Bayreuth, Universitatsstr. 30,

95447 Bayreuth, Germany.

E-mail: camilla.wellstein@uni-bayreuth.de

Editor

D. Byers

Received: 23 January 2012; Accepted: 3 May

2012

doi:10.1111/j.1438-8677.2012.00635.x

ABSTRACT

Changes in land use and climate interfere with grassland ecosystem processes. HereI experimentally investigated the combined effects of land-use change related littercover and contrasting water supply on seedling emergence. In this context, the roleof the initial relative position of seeds, i.e. seeds on top of the litter versus seedsbeneath the litter in interaction with water supply has not been investigated so far.I hypothesised that facilitative effects of litter on seedling emergence occur whenseeds are covered by litter and deteriorate when litter covers the ground and seedsfall on it (seeds on top of the litter). Further, I hypothesised that the importance ofseed position for seedling emergence will increase under conditions of recurrentdrought. I performed a controlled pot experiment on seedling emergence of threecommon European grassland species (Pimpinella saxifraga, Leontodon autumnalis,Sanguisorba officinalis) by experimental manipulations of litter and water availabil-ity. Seedling emergence under moist conditions showed no significant differencesbetween each litter position compared to the control across species. In contrast,under recurrent drought, seedling emergence was significantly higher below the lit-ter compared to seeds on top of the litter and the control (i.e. no litter). In aban-doned land, seedling emergence may be limited when seeds fall on ground-coveringlitter. In contrast, in grasslands with regular low-intensity land use, seedling emer-gence may be enhanced when a moderate level of litter covers seeds at the end ofthe growing season. Protective mechanisms that occur with seeds positionedbeneath litter are particularly important under recurrent drought.

Plant Biology ISSN 1435-8603

Plant Biology ª 2012 German Botanical Society and The Royal Botanical Society of the Netherlands 1

report that the role of a protective layer is of major impor-tance in communities affected by temporal desiccation; this isin line with the findings of Ryser (1993) in dry grasslands.Furthermore, Hobbs & Mooney (1991) show that inter-annual variation in precipitation can influence the colonisa-tion of gaps, i.e. seedling recruitment on bare soil.

The role of the seed position with respect to litter in thecontext of recurrent drought is an aspect that has been dis-cussed but not experimentally tested in previous studies (e.g.Donath & Eckstein 2010). I suggest that the combined effectsof seed position and water supply are of particular relevancein evaluating the outcome of climate and land-use change onseedling recruitment. This evaluation is urgently needed togain a better understanding of species survival and vegetationdynamics in times of climate instability (e.g. Walck et al.2011). Here, I performed randomised pot experiments to testfor the combined effects of seed position, i.e. seeds on top oflitter versus seeds beneath litter, versus. seeds without litter,and contrasting water supply, i.e. recurrent drought versusfrequent watering, on seedling emergence of three commonEuropean grassland species (Pimpinella saxifraga L., Leonto-don autumnalis L., Sanguisorba officinalis L.).

The development of the hypotheses follows. As describedabove, experimental evidence has revealed facilitative effectsof moderate amounts of litter versus no litter. H1 – litterpresence: I hypothesise that in contrast to no litter, a moder-ate litter cover increases seedling emergence, irrespective ofseed position. Experimental evidence demonstrated that facil-itative effects for germination appear below a litter layer.Here, I test for this general effect in a new context with addi-tional changes in water supply. H2 – position toward litter:specifically, I expect that seedling emergence will be higherunder litter than above the litter. As indicated in field andexperimental research on germination of grassland species(Fowler 1986; Xiong et al. 2003; Eckstein & Donath 2005),the interactive effects of litter with water shortage may beparticularly important. H3 – water availability: in this con-text, I hypothesise that the effect of seed–litter–position forseedling emergence will increase under recurrent drought.

METHODS

Study species

I selected the perennial herbs Pimpinella saxifraga L., Leonto-don autumnalis L. and Sanguisorba officinalis L. as referencespecies, since they are very common in European grasslandsand predominantly occur in managed grasslands (Oberdorfer1994) as polycarpic perennials, a common plant strategy inthis vegetation type (Grime et al. 1988). Their seed mass cov-ers a range from intermediate to high: L. autumnalis0.59 ± 0.03 mg (±SD); P. saxifraga 0.87 mg ± 0.03 mg; S. of-ficinalis 2.2 ± 0.2 mg; data are estimates from this seed col-lection; per species 130 replicates of 100 seed samples wereweighed and divided by 100 to obtain mean values for singleseeds. The seed forms of P. saxifraga and S. officinalis haveno appendages, the pappus of L. autumnalis easily detachesfrom the seed. Seed collections were made from a mountainregion with traditionally managed, species-rich grasslands(mean ranging from 32 to 41 species 25 m)2; Wellstein et al.2007), with abundant occurrence of the study species (Lahn-

Dill highlands, Germany; for details of the study region seee.g. Wellstein et al. 2007). These species can be used as indi-cators on the consistency of germination in response to keyenvironmental modifications expected due to climatechanges, such as drier summers, and land-use changes, suchas abandonment.

For each species, seeds were collected between July and Sep-tember 2003 (depending on their degree of ripeness) from fivedifferent populations and at least 20 individuals of each popu-lation in grasslands of the Lahn-Dill highlands. Seeds wereair-dried, manually cleaned and stored in darkness at roomtemperature (ca. 20 �C) but exposed to ambient conditions inJanuary 2004 in order to allow for cold stratification duringearly spring, which enhances germination in several of thestudy species (Baskin & Baskin 2001; Holzel & Otte 2004).Before sowing, i.e. after storage, I applied a tetrazolium chlo-ride test on additional seed batches (n = 100 seeds per species)of the collection, which indicated high viability of the seeds(L. autumnalis 90%, P. saxifraga 86%, S. officinalis 89%).

Experimental design

I used a completely randomised experimental design to studythe effects of species (S, factor levels (k) = 3: P. saxifraga;L. autumnalis; S. officinalis), water supply (W, k = 2: con-stantly moist; recurrent drought) and seed position (P, k = 3:above litter cover; beneath litter cover; without litter cover,i.e. bare soil) on seedling emergence. For each study species,72 pots of 1 dm3 volume (10 · 10 · 10 cm) were used; eachwater supply · seed position combination (six combinations)was replicated 12 times. Each experimental pot was sownwith 50 seeds of one species at the beginning of June. Potswere filled with commercial potting soil (Fruhstorfer Erde�,Type P; Industrie-Erdenwerke Archut GmbH, Lauterbach,Germany) composed of a mixture of peat, clay and humus(pH-CaCl2: 6.1; 142 mgÆl)1 N, 97 mgÆl)1 P2O5 and 180 mgÆl)1

K2O). A grass litter level of 400 g litter m)2 (i.e. 4 g per potof 0.01 m2 area) was used; this moderate amount of litter isknown to be suitable for germination below a litter layer ingrassland species (Eckstein & Donath 2005). The grass litterwas dried clippings that originated from a mesic unfertilisedgrassland site dominated by Poa pratensis, Agrostis stolonifera,Arrhenaterum elatius and Dactylis glomerata that harbourednone of the study species. All pots were exposed to ambientconditions and covered with a polythene sheet in order tocompletely control for watering regime in a common gardenclose to Giessen, Germany (50�32¢N, 8�41.3¢E, 172 m a.s.l.).Experimental pots were arranged at random and surroundedby an additional row of pots filled with soil that served as abuffer for the outermost row of the experimental pots.

For the water supply treatment, pots were: (i) kept con-stantly moist by regular watering (80 cm3 of water addedeach time when watering, watering twice a week at the begin-ning of the experiment and once a day later in the growingseason from mid-June when evaporation had increased); or(ii) kept intermittently dry (i.e. recurrent drought) byreduced watering frequency (80 cm3 of water added eachtime when watering, watering with a reduced frequency oncea week at the beginning and two times per week from mid-June, leading to intermittently dry soil between the wateringevents). Additionally, 12 pots without any species were added

Seedlings under drought Wellstein

2 Plant Biology ª 2012 German Botanical Society and The Royal Botanical Society of the Netherlands

for each combination of water supply and seed position tocontrol for bias from seed rain that was theoretically possibleduring opening of rain-out shelters for watering. However,no germination of the target species was detected withinthese control pots.

Emerged seedlings, i.e. those that successfully penetratedthrough the cover, were counted 13 times: all 6 days between07 July 2004 (start of germination) and 04 November 2004(end of germination), and were removed whenever they wereobserved.

Statistical analysis

Three-way anova was used to analyse for effects of species(S), water supply (W), seed position (P) and their interac-tions on relative seedling emergence. Since the dependentvariables were proportion data, they were square root arc-sine-transformed (Quinn & Keough 2002). As a post-hoc testfor differences between groups, I used Tukey HSD test. Allstatistical analyses were done using statistica 6.0 (StatSoft2002).

RESULTS

In a three-way anova all main effects, i.e. species identity(S), water supply (W) and seed position (P), were highly sig-nificant (Table 1). In general, across all other factors, thethree study species were significantly different from eachother, with seedling emergence of (mean ± SE) 46.5 ± 1.1%

in L. autumnalis, 42.6 ± 1.1% in P. saxifraga and21.9 ± 1.1% in S. officinalis, according to the Tukey HSDpost-hoc test. With respect to the general effect of seed posi-tion, seedling emergence was significantly lower when seedswere sown without litter or sown on top of litter (on baresoil) compared to the position beneath litter: 32.1 ± 1.1%without, 27.4 ± 1.1% above, 51.5 ± 1.1% beneath litter. Seed-ling emergence was significantly lower when seeds experi-enced recurrent drought as opposed to constantly moistconditions (23.3 ± 0.9% versus 50.7 ± 0.9%). Both factors (P,W) and their interaction differed significantly among species(S) (Table 1).

Tukey HSD test after significant three-way interaction ofthree-way anova (S · P · W: df = 4, F = 7.7, P < 0.0001;Table 1) revealed significantly higher seedling emergence forthe seed position beneath litter cover compared to the posi-tion above litter and the position on bare soil (i.e. withoutlitter) for all three study species under recurrent drought(Fig. 1, Table 2). At the same time, these higher numbers ofemerged seedlings beneath litter equalled those achievedunder constantly moist conditions, irrespective of seed posi-tion, with the exception of S. officinalis, where seeds beneathlitter had higher emergence than seeds above litter, evenunder the moist conditions (Fig. 1, Table 2). Therefore, theeffect of seed–litter–position (i.e. difference in mean percent-age emergence between the beneath litter position and eachof the other positions) increased under recurrent drought forP. saxifraga and L. autumnalis but not for S. officinalis(Fig. 1, Table 2). Under recurrent drought, furthermore, seedposition above litter and on bare soil were not significantlydifferent (Fig. 1).

Under the continuously moist conditions, seed positionabove litter led to significantly lower seedling emergencecompared to seed position beneath litter in P. saxifraga andS. officinalis. In L. autumnalis seed position caused no signifi-cant differences under continuously moist conditions; how-ever, seedling emergence was still lower with seed positionabove litter (Fig. 1).

DISCUSSION

Across all study species, the position beneath the litter layerclearly provides benefits for seedling emergence; the positionabove litter or on bare soil reduces seedling emergence. Since

Table 1. Results of a three-way ANOVA on the effect of species identity (S),

water supply (W), and seed position (P) on percentage emergence.

source of variation df MS F P

intercept 1 82.07 5459.5 <0.0001

species (S) 2 1.79 119.2 <0.0001

water supply (W) 1 6.50 432.2 <0.0001

seed position (P) 2 1.81 120.2 <0.0001

S · W 2 0.25 16.6 <0.0001

S · P 4 0.12 8.2 <0.0001

W · P 2 0.75 50.0 <0.0001

S · W · P 4 0.12 7.7 <0.0001

Error 198 0.02

df = degrees of freedom; MS = mean sum of squares.

(a) Pimpinella saxifraga (b) Leontodon autumnalis (c) Sanguisorba officinalis

Constantlymoist

Recurrentdrought

Constantlymoist

Recurrentdrought

Constantlymoist

Recurrentdrought

80

70

70

60

50

50

40

30

20

10

0

40

30

20

10

0

60

50

40

30

20

10

0

a

d,e,f e,ff

d,e

a

b,c

d,e,f

e,f e,f

d,e

b

aa

b

b,c

c,d

b

Fina

l per

cent

age

emer

genc

e

Fig. 1. Bar plots of the final percentage seedling emergence of the study species sown without (light grey), beneath (grey) or above (dark grey) 400 gm)2

of grassland litter. Data are untransformed values, means and SD are given. Letters indicate significant differences of the three-way-interaction (spe-

cies · seed position · water supply) according to Tukey HSD test after three-way ANOVA (S · P · W: df = 4, F = 7.7, P < 0.0001).

Wellstein Seedlings under drought

Plant Biology ª 2012 German Botanical Society and The Royal Botanical Society of the Netherlands 3

the facilitative effect of litter depends on seed position, H1 –stating a positive effect of litter irrespective of seed position –must be rejected. The fact that the beneath-litter positionleads to significantly higher emergence compared to theabove-litter position confirms H2: the facilitative effect ishighest when seeds are covered by litter and is lower whenlitter covers the ground and seeds fall on top of it. Soon aftergermination, seedlings must rapidly connect with the soilthrough their primary root to avoid lethal desiccation, andthey must also reach sufficient light resources. While rootingmay be substantially hampered in positions above a litterlayer (Facelli & Pickett 1991; and references therein), theposition below a moderate amount of litter, as applied in thisexperiment, may still provide enough available light for seed-ling growth. This could be combined with additional positiveeffects, such as attenuated temperature extremes and reducedwater stress (Jensen & Gutekunst 2003; Eckstein & Donath2005; Donath & Eckstein 2010). In the above-litter position,resource investment for elongation of the radicle may be cru-cial and reduce the overall fitness of a seedling (Hamrick &Lee 1987). However, the same holds true for elongation ofthe hypocotyl in the beneath-litter positions. A seed posi-tioned on bare soil does not need to invest additionalresources in the radicle and ⁄ or hypocotyl. In this context myresults, i.e. low emergence in the no-litter treatment, indicatethat with moderate amounts of litter, these investments donot outweigh possible positive litter effects.

It is not the importance of a litter layer per se as much asthe protective function of moderate amounts of litter cover-ing seeds that is a particularly important factor for seedlingrecruitment under conditions of recurrent drought. This wasshown by a significant interaction between water supply (W)and seed position (P) in three-way anova, providing supportfor H3. The higher importance of seed position under condi-tions of recurrent drought can be explained as follows: ger-mination on top of a litter layer may be highly problematicsince the radicle of the germinating seedling must establishsoil contact as soon as possible in order to obtain access towater resources. Under recurrent drought, seedlings on topof the litter may suffer lethal desiccation during radicle estab-lishment before reaching and penetrating the soil surface.Germination on bare soil under conditions of recurrentdrought also proved to be a drawback. Even without beinghampered from making contact with the soil surface, drought

may immediately desiccate the soil surface, limiting penetra-tion of the radicle and leading to lethal desiccation of theseedling. In contrast, germinating beneath a litter layerproved to be favourable, particularly under conditions ofrecurrent drought. Most likely, a combination of reducedwater stress due to lower evaporation and attenuated temper-ature extremes protects seedlings from desiccation and allowsradicle establishment (Jensen & Gutekunst 2003; Eckstein &Donath 2005; Rotundo & Aguiar 2005; Donath & Eckstein2010).

Plant traits such as seed mass have been reported to becrucial for germination, particularly in relation to light andwater availability (Grime et al. 1981; Buckley 1982; Gross1984; Pakeman et al. 2008). In fact, larger seed mass generallyleads to larger seedling size, resulting in a better ability toestablish under shady conditions and to penetrate a litterlayer (Jurado & Westoby 1992; Leishman et al. 2000). Fur-thermore, reserve resources are needed for drought-resistancemechanisms (Leishman & Westoby 1994). The experiment inmy study revealed that species profit differently in the posi-tion below litter compared to positions above litter and inbare soil under conditions of recurrent drought (mean differ-ences: 50% and 55%, 30% and 25%, 15% and 20% higheremergence for P. saxifraga, L. autumnalis, and S. officinalis,respectively). However, none of the study species producessmall seeds, but intermediate or larger-sized seeds (0.59–2.24 mg). This may explain why all of the study species areable to profit from favourable conditions below a litter layerin situations of recurrent drought. This stresses the advantageof intermediate and larger-sized seeds with stored energy(reserve effect; Westoby et al. 1996). With respect to themoist conditions, the larger-seeded species (S. officinalis) maybe able to grow faster and therefore probably profit fromhigher levels of soil moisture below litter under the continu-ous watering regimes. Furthermore, none of the speciesshowed dependence on gap regeneration (bare soil), as wouldhave been expected for small seeds (Gross 1984; Krenova &Leps 1996; Westoby et al. 1996). Dependence on gap regener-ation is most likely related to a strong selection against darkgermination in small-seeded plant species (Milberg et al.2000). The average seed mass of plant species of grasslandcommunities in the study region, in which the three studyspecies commonly occur, is 1.82 mg (151 species, see Well-stein et al. 2007). With respect to the C-S-R strategy (Grime1988) spectra of these grasslands, an average seed mass of2.20 mg can be attributed to the C-strategy, 1.26 mg to theS-strategy and 1.62 mg to the R-strategy (calculation basedon calibrated C-S-R radii, for method see Wellstein et al.2007). S. officinalis (a CSR strategist (Grime 1988), seed mass2.24 mg, transient seed bank; Wellstein et al. 2007) mighttherefore be most representative for competitive species inthese grasslands. P. saxifraga (S ⁄ SR strategist, 0.87 mg, tran-sient seed bank) might be more representative for species fol-lowing a stress strategy, and L. autumnalis (R ⁄ CSR strategist,0.59 mg, persistent seed bank) might be more representativefor species following a ruderal strategy. Due to their low seedmass, ruderal species have higher mobility and a tendency tobe incorporated in the soil and thus to build a persistent seedbank (Holzel & Otte 2004; Donath & Eckstein 2010, 2011).This attribute might represent a buffer toward adverse effectsof climate and land-use change.

Table 2. Comparison of mean values (untransformed data) and standard

error (SE) of percentage seedling emergence within and between species

(S: P. saxifraga, L. autumnalis, S. officinalis), and experimental manipula-

tions of water supply (W: recurrent drought, constantly moist), and seed

position (P: without litter, beneath litter, above litter).

W P

Pimpinella

saxifraga

Leontodon

autumnalis

Sanguisorba

officinalis

Mean SE Mean SE Mean SE

recurrent drought without 2.8 2.8 28.2 2.9 3.8 2.8

beneath 58.7 2.8 53.7 2.8 23.23 2.8

above 9.0 2.8 22.7 2.8 7.8 2.8

constantly moist without 65.5 2.8 62.5 2.7 29.5 2.8

beneath 69.3 2.8 60.7 2.8 43.5 2.8

above 50.3 2.8 51.2 2.8 23.7 2.8

Seedlings under drought Wellstein

4 Plant Biology ª 2012 German Botanical Society and The Royal Botanical Society of the Netherlands

With respect to land-use and climate change these resultslead to the following conclusions. With the abandonment ofland use (fallow land), the amount and consequently thethickness of the litter layer increases. In this case, seedlingemergence may be limited because seeds fall on ground-cov-ering litter, i.e. in an unfavourable position. In contrast, inregularly used, low-intensity grasslands, seedling emergencemay be enhanced when a moderate amount of litter coversthe seeds at the end of the growing season. The protectivemechanisms triggered by the position of the seed beneath thelitter are likely to be of major importance under climaticfluctuations, such as recurrent drought.

ACKNOWLEDGEMENTS

The experiment was conducted while C.W. was at theDepartment of Landscape Ecology and Landscape Planning,Justus-Liebig-University, Giessen, Germany. I thank theDeutsche Forschungsgemeinschaft (DFG) for funding withinthe project ‘Land-use Options for Peripheral Regions (SFB299)’, Annette Otte for institutional support, and Josef Scholzvom Hofe for assistance with the experiment. I am indebtedto Diane Byers, Tobias W. Donath and two anonymousreviewers who gave many suggestions for improvement of aformer version of this paper.

REFERENCES

Baskin C.C., Baskin J.M. (2001) Seeds – Ecology, Bio-

geography, and Evolution of Dormancy and Germi-

nation. Academic Press, San Diego, CA.

Bekker R.M., Bakker J.P., Grandin U., Kalamees R.,

Milberg P., Poschlod P., Thompson K., Willems

J.H. (1998) Seed size, shape and vertical distribu-

tion in the soil: indicators of seed longevity. Func-

tional Ecology, 12, 834–842.

Buckley R.C. (1982) Seed size and seedling establish-

ment in tropical arid dunecrest plants. Biotropica,

14, 314–315.

Chambers J.C., MacMahon J.A., Haefner J.H. (1991)

Seed entrapment in alpine ecosystems: effects of

soil particle size and diaspore morphology. Ecology,

72, 1668–1677.

Donath T.W., Eckstein R.L. (2010) Effects of bryo-

phytes and grass litter on seedling emergence vary

by vertical seed position and seed size. Plant Ecol-

ogy, 207, 257–268.

Donath T.W., Eckstein R.L. (2011) Litter effects on

seedling establishment interact with seed position

and earthworm activity. Plant Biology, 14, 163–170.

During H.J., van Tooren B.F. (1990) Bryophyte inter-

actions with other plants. Botanical Journal of the

Linnean Society, 104, 79–98.

Easterling D.R., Evans J.L., Groisman P.Y., Karl T.R.,

Kunkel K.E., Ambenje P. (2000) Observed variabil-

ity and trends in extreme climate events: a brief

review. Bulletin of the American Meteorological Soci-

ety, 81, 417–425.

Eckstein R.L., Donath T.W. (2005) Interactions

between litter and water availability affect seedling

emergence in four familial pairs of floodplain spe-

cies. Journal of Ecology, 93, 807–816.

Facelli J.M., Pickett S.T.A. (1991) Plant litter: its

dynamics and effects on plant community struc-

ture. The Botanical Review, 57, 2–31.

Fowler N.L. (1986) Microsite requirements for germi-

nation and establishment of three grass species.

American Midland Naturalist, 115, 131–145.

Grime J.P. (1988) The C-S-R model of primary plant

strategies – origins, implications and tests. In: Gott-

lieb L.D., Jain K. (Eds), Plant Evolutionary Biology.

Chapman & Hall, London, pp 371–393.

Grime J.P., Mason G., Curtis A.V., Rodman J., Band

S.R., Mowforth M.A.G., Neal A.M., Shaw A. (1981)

A comparative study of germination characteristics

in a local flora. Journal of Ecology, 69, 1017–1095.

Grime J.P., Hodgson J.G., Hunt R. (1988) Compara-

tive Plant Ecology. A Functional Approach to Com-

mon British Species. Unwin Hyman, London.

Gross K.L. (1984) Effects of seed size and growth

form on seedling establishment of six monocarpic

perennial plants. Journal of Ecology, 72, 369–387.

Hamrick J.L., Lee J.M. (1987) Effect of soil surface

topography and litter cover on the germination,

survival, and growth of musk thistle (Carduus nu-

tans). American Journal of Botany, 74, 451–457.

Hobbs R.J., Mooney H.A. (1991) Effects of rainfall

variability and gopher disturbance on serpentine

annual grassland dynamics. Ecology, 72, 59–68.

Holzel N., Otte A. (2004) Ecological significance of

germination characteristics in flood-meadow spe-

cies. Flora, 199, 12–24.

IPCC (2007) Climate Change 2007: The Physical Sci-

ence Basis. Intergovernmental Panel on Climate

Change, Geneva, Switzerland.

Jensen K., Gutekunst K. (2003) Effects of litter on

establishment of grassland plant species: the role of

seed size and successional status. Basic and Applied

Ecology, 4, 579–587.

Jurado E., Westoby M. (1992) Seedling growth in

relation to seed size among species of arid Austra-

lia. Journal of Ecology, 80, 407–416.

Kahmen S., Poschlod P., Schreiber K.F. (2002) Con-

servation management of calcareous grasslands.

Changes in plant species composition and response

of functional traits during 25 years. Biological Con-

servation, 104, 319–328.

Krenova Z., Leps J. (1996) Regeneration of a Gentiana

pneumonanthe population in an oligotrophic wet

meadow. Journal of Vegetation Science, 7, 107–112.

Leishman M.R., Westoby M. (1994) The role of seed

size in seedling establishment in dry soil conditions

– experimental evidence from semi-arid species.

Journal of Ecology, 82, 249–258.

Leishman M.R., Wright I.J., Moles A.T., Westoby M.

(2000) The evolutionary ecology of seed size. In:

Fenner M. (Ed.), Seeds – the Ecology of Regeneration

in Plant Communities. CAB International, Walling-

ford, pp 31–57.

Meehl G.A., Karl T., Easterling D.R., Changnon S.,

Pielk R., Jr, Changnon D., Evans J., Groisman P.Y.,

Knutson T.R., Kunkel K.E., Mearns L.O., Parmesan

C., Pulwarty R., Root T., Sylves R.T., Whetton P.,

Zwiers F. (2000) An introduction to trends in

extreme weather and climate events: observations,

socioeconomic impacts, terrestrial ecological

impacts, and model projections. Bulletin of the

American Meteorological Society, 81, 413–416.

Milberg P., Andersson L., Thompson K. (2000)

Large-seeded species are less dependent on light for

germination than small-seeded ones. Seed Science

Research, 10, 99–104.

Oberdorfer E. (1994) Pflanzensoziologische Ex-

kursionsflora. 7th edn. Ulmer, Stuttgart.

Pakeman R.J., Garnier E., Lavorel S., Ansquer P., Cas-

tro H., Cruz P., Dolezal J., Eriksson O., Freitas H.,

Golodet C., Kigel J., Kleyer M., Leps J., Meier T.,

Papadimitriou M., Papanastasi V.P., Quested H.,

Quetier F., Rusch G., Sternberg M., Theau J.P.,

Thebault A., Vile D. (2008) Impact of abundance

weighting on the response of seed traits to climate

and land use. Journal of Ecology, 96, 355–366.

Quinn G.P., Keough M.J. (2002) Experimental Design

and Data Analysis for Biologists. Cambridge Univer-

sity Press, Cambridge.

Rotundo J.L., Aguiar M.R. (2005) Litter effects on

plant regeneration in arid lands: a complex balance

between seed retention, seed longevity and soil–

seed contact. Journal of Ecology, 93, 829–838.

Ruprecht E., Jozsa J., Olvedi T.B., Simon J. (2010)

Differential effects of several ‘‘litter types’’ on the

germination of dry grassland species. Journal of

Vegetation Science, 21, 1069–1081.

Ryser P. (1993) Influences of neighbouring plants on

seedling establishment in limestone grassland. Jour-

nal of Vegetation Science, 4, 195–202.

Statsoft (2002) STATISTICA 6.0 for windows. Statsoft

Inc., Tulsa, USA.

Walck J.L., Hidayati S.N., Dixon K.W., Thompson K.,

Poschlod P. (2011) Climate change and plant

regeneration from seed. Global Change Biology, 17,

2145–2161.

Wellstein C., Otte A., Waldhardt R. (2007) Seed bank

diversity in mesic grasslands and their relation to

vegetation, management and site conditions. Jour-

nal of Vegetation Science, 18, 153–162.

Westoby M., Leishman M., Lord J. (1996) Compara-

tive ecology of seed size and dispersal. Philosophical

Transactions of the Royal Society of London Series B-

Biological Sciences, 351, 1309–1317.

Xiong S., Nilsson C. (1999) The effects of plant litter

on vegetation: a meta-analysis. Journal of Ecology,

87, 984–994.

Xiong S., Johansson M.E., Hughes F.M.R., Hayes A.,

Richards K.S., Nilsson C. (2003) Interactive effects

of soil moisture, vegetation canopy, plant litter and

seed addition on plant diversity in a wetland com-

munity. Journal of Ecology, 91, 976–986.

Wellstein Seedlings under drought

Plant Biology ª 2012 German Botanical Society and The Royal Botanical Society of the Netherlands 5