Germination Characteristics of Mountain Celery 'Aciphylla Glacialis' (F. Muell.) Benth. (Apiaceae)

2009·2.5 MB·English

by  Susanna E, Morgan Venn

Research Reports
Germination characteristics ofMountain Celery
Aciphyllaglacialis (F. Muell.) Benth. (Apiaceae)
W
Susanna E Venn^’^ and John Morgan^
^DepartmentofBotany,LaTrobeUniversity,Bundoora,Victoria3086
-Currentaddress;NewSouthWalesNationalParksandWildlifeService,POBox2228,lindabyne,NSW2627
Email:[email protected]
Abstract
The germination characteristics of Mountain Celery Aciphylla glacialis (Apiaceae) were investigated. This
species failed to germinate in conditions that nine other alpine species from the same region found favour-
able. Seeds were collected from eight mountain peaks in the Victorian alpine zone. These peaks formed an
altitudinal gradient of302 m. The seeds were subjected to a cold-wet stratification treatment in an attempt
to overcome what appeared to be seed dormancy. After approximately 15 weeks ofthe cold-wet stratifica-
tion treatment, A. glacialis seedsbegan to germinate. Germination rates were fastest among seeds from the
mid to high altitude sites, which also had the greatest percentage of seeds to germinate. Mt Speculation
(1668 m) and Mt Bogong (1970 m) had the lowestpercentage ofgermination from seeds collected in 2004.
Among seeds collected in 2005, the lowestvalues were from Mt Bogong. Comparing seed collected in 2004
and 2005, there was no significant difference between final per cent germination from any site. There were
no significant relationships between altitude and germination characteristics. There were several significant
differences in the probabilityofseed germinationbetween sites. However, there was no overall trend in the
probabilityofseedgerminationbetweenhighandlowaltitudesites.Theroleofseeddormancymechanismsin
relationtothealpineenvironmentisdiscussedandtheroleofaltitudinalgradientsinrelationtoseedgermina-
tioninthisspeciesisspeculatedupon. (TheVictorianNaturalist126(1),2009,4-12)
Keywords: seedgermination,dormancy, altitudinal gradient, alpine, coldstratification
Introduction
Seedgerminationy dormancyandenvironment Cold-wet stratification, as experienced over
Seed germination and seedling emergence winterin an alpine landscape, maybe required
under natural conditions often are highly sea- to break seed dormancy in the few alpine
sonal(BaskinandBaskin 1988). Inalpineareas, species that use such internal physiological
wherethechangesingroundleveltemperatures mechanisms to restrain germination (Baskin
are predictable due to the timing ofsnowmelt, and Baskin 1988; Probert 1992). Thisstratifica-
temperature can provide a reliable seasonal tion may also be responsible for the timing of
germinationcueforseedslocatedatornearthe seedling emergence in these species. Addition-
soil surface. Previously, it was found that nine ally, photoperiodism affects the seeds ofmany
speciesfromtheVictorianalpinezonehadhigh alpine species, ensuring that growth does not
rates ofgermination (Venn 2007), as do many commence in periods which, by evolutionary
speciesinalpineareasfromthenorthernhemi- experience’ would be fatal for an emerging
sphere (Amen 1966; Bliss 1971). Optimum seedling (Korner 1999). Hence, there are vari-
temperatures for seed germination among ousstrategiestoensurethatgerminationoccurs
alpine species worldwide are often above 20°C in springorearlysummerwhen conditions are
(Amen 1966; Bell and Bliss 1980; Schiitz 2002). mostfavourableforthesurvivalofseedlings in
Many seeds also germinate opportunistically the alpine zone.
whenever suitable ambient conditions are Relationships between germination syn-
present (Bell and Bliss 1980; Korner 1999). drome, habitat and geographical distribution
Seed dormancy is therefore considered rare in have been difficult to ascertain (Baskin and
these ecosystems (Amen 1966; Bliss 1971) and Baskin 1988),evenatthespecieslevel(Meyeret
germination immediately following snowmelt, al 1990),asgermination characteristics can, in
coinciding with abundant soil moisture and part, be a function oflife-historytraits (Baskin
warmerambienttemperatures,isverycommon and Baskin 1971; Grime et al. 1981; Washitani
(Bliss 1962; Billings and Mooney 1968; Korner and Masuda 1990). However, ifthe aspects of
1999). life-history, habitat and distribution are rela-
4 TheVictorianNaturalist
Research Reports
tively similar, different adaptations to environ- inflorescences(PickeringandArthur2003)and
mental factors within species can be discerned nearlyall female flowersproduce fruit.
more clearly (Meyer et al 1990), as natural The great investment in reproduction by
selection should favour seed germination pat- female Aciphylla plants is assumed to decrease
terns that promote seedling establishment their longevity and restrict their growth, es-
(CavieresandArroyo2000).Germinationchar- pecially at higher alpine sites where the con-
acteristics, therefore, are expected to vary pre- ditions for the accumulation of biomass are
dictablyacrossanaltitudinalandenvironmental limited. The costs associated with female func-
gradient. Evidence suggests that seed viability tion at high altitudes, therefore, could result in
andseedlingvigour maydecreasewithin aspe- females flowering less frequently (Pickering
cies asaltitude increases, dueto environmental and Arthur 2003), producing fewer seeds and
constraints such as short growing seasons, low possibly producing non-viable seeds. At lower
temperatures andharsh conditions athigh alti- altitudes, environmental conditions are more
tudes (Amen 1966; Billings andMooney 1968). favourable (longer snow-free period, warmer
Therefore,germinationstrategiesmaybepartly temperatures), which could result in an equal
afunctionoflife-historytraitsandpartlydueto flowering ratio between males and females, as
habitatcharacteristics. well as greater seed production and viability.
Phenotypic differences in relation to environ-
CharacteristicsoftheAciphylla mental effects also are common in this genus,
This study investigated the germination be-
making A. glacialis a good model species for
haviour of a locally common, but regionally
investigating the differences in germination
rVaircet,orailapiannedfoNrebwthSaotutghroWwasleisn.aTlTpiienefoacraleasspei-n across an altitudinal gradient. There are few, if
any, published studies dealing with germina-
cies is the Mountain Celery Aciphylla glacialis
tionfor-anyAciphyllaspeciesinAustralia.
(E Muell.) Benth. (Apiaceae) (Fig 1 see back
cover). It is one ofonlythree Aciphylla species Researchfocus
foundinAustralia,allofwhicharerestrictedto This research examines three questions: Can
alpineand sub-alpineregions.A. simplicifoliais seed dormancy ofA. glacialis be overcome by
uncommon inVictoriaandmostlyrestrictedto cold-wet stratification? Which dormancy syn-
thehighalpineareasofNewSouthWales (Cos- drome does thebehaviourofA.glacialisexhib-
tin etal 2000), and the Tasmanian endemicA. it? Does the altitude ofseed origin or the year
procumbensis restrictedtobolsterheath, heath of seed collection influence the germination
and alpine sedge-land (Kirkpatrick 1997), The characteristics?
other40speciesofthisgenus are found inNew
Methods
Zealand, where they occur over a wide range
of habitats from lowland grasslands to alpine Seedcollectionandstudysites
areas (Raven 1973). Seeds from A. glacialis (Fig. 1 see back cover)
Aciphylla is adioecious, obligateout-crossing populations were collected during 2004, be-
genus dependent on pollinators for fertilisa- tween January and March, from seven alpine
tion. It is, therefore, likely to experience some peaks in Victoria, and again from three ofthe
competition both within and between species peaks in2005. Seedwaskeptin dry,dark,stable
for the few pollinators that are active at high conditions until the germination trialsbegan in
altitudes (Inoiiye and Pyke 1988; Pickering mmid2005. Thepeaksrepresentagradientof302
2000). Thesex-ratioofpopulationsofA.glacia- in altitude from Mt Speculation (1668 m) to
tliosimsamlaelpelbainatsseod,fw1:i1t.h5.aMnaolveesracllanraatlisooopfrfoedmuaclee sHioteoskaerre Pwliattheianu40(1k97m0omf)eaocnh oMttherB,owgiotnhg.moAsltl
fourtimesmoreflowersthanfemales(Pickering lessthan 10 kmapart (Table 1).
2000). However, thereisgenerallynochangein Meanannualrainfallofthisregionisrelatively
the number of flowers with altitude for both high cmommpared to adjacent lowland areas, over
male and female plants among the same sized 1800 (LCC 1982), with much ofthis pre-
populations(Pickeringand Hill2002).Thepro- cipitation falling as snow during winter. Aver-
portion ofbiomass that females allocate to re- age airtemperatures follow seasonal variations
production in A. simplicifolia can be 80% more withtypicaldailyminimumsand maximumsof
than males based on the average weight ofthe 2"’Cand27°C in summer, and -6°Cand 12°Cin
Vol 126 (1) 2009 5
Research Reports
Table1.Thesevenstudysitelocationsandaltitude.
Studysite Location Altitude(masl)
MtSpeculation 37°07:301S 146°38’40”E 1668
KingBilly(2) 37n245i^ 146°36T5”E 1696
MtMagdala 37n5l30”S 146°37T5”E 1725
MtHowitt 37°10:30”S 146°38’50”E 1738
MtBuller(WestKnob) 37°08;45”S 145°26T5”E 1762
MtHotham 36°58:35”S 147°07’30”E 1860
MtBogong(HookerPlateau) 36°4427”S 147°17T5”E 1970
winter (Bureau ofMeteorology 2007 unpubl.). had become more frequent. The final values
Frosts are frequent and can occur at any time ofgermination were adjusted to represent per
of year (Williams 1987). The growing season cent germination of the viable seeds present,
spansapproximatelysevenmonths,fromsnow- i.e. those that had not succumbed to fungus
melt (usually the end ofSeptember) until late attack.
April,whenthefirstsubstantialsnowfallsofthe
Dataanalysis
winter begin. All sites are classed as ‘Steep Al- Mean seed mass was compared across the sites
pine Mountains’ after Costin (1957), are above of seed origin with a factorial ANOVA and
the natural tree line and are dominated by tall
alpineherbfieldvegetation.Thelowerpeaksalso Bonferroni post-hoc tests. Final per cent ger-
mination was expressed as a percentage ofvi-
contain species typical ofadjoining sub-alpine
ableseedsthatgerminatedduringthecourseof
woodland(Costin 1957).Soilsinthisregionare
considered alpine humus and are acidic (pH 4 the trial. Germination rate (Tso) was measured
to 5) (Costin 1962; LCC 1982). agsertmhienattiimoen.inTdhaeysfitnaalkepnertocernetacgher5m0in%atoifofninoalf
Germination trial seedswas comparedbetween sites usingANO-
Filled and viable seeds were selected by visu- VAand Bonferronipost-hoctests.
ally assessing each seed using a light box and The shape of each species’ per cent germi-
bygentlysqueezingeachseed.A.glacialisseeds nation curve was compared by modelling the
did not germinate under conditions that nine germination curves as a probability of germi-
other alpine species {Brachyscome rigidula, nation, usingthe Kaplan-Meier (K-M) product
Brachyscomespathulata,Leucochrysumalbicans limit estimation function (Kaplan and Meier
albicans, Microseris sp. 2, Oleariaphlogopappa 1958;Lee 1992;Gimenez-Benavidesetal 2005)
var. subrepanda, Xerochrysum subundulatum, and comparing those models. This method
Luzula acutifolia, Trisetum spicatum and Ry- gives a mean probability of germination at
tidosperma nudiflora) found favourable (cool every sampling time and allows whole curves
treatment: 12/10°Cwith 14/10hoursdaylength tobecomparedacrosssiteswithineachspecies,
andwarmtreatment: 20/10°Cwith 12/12hours rather than only at the sampling times as with
daylength).Thus,inthisexperiment,Aciphylla more conventional statistics. The K-M product
seedswere pre-treatedwith cold-wet stratifica- limitestimationfunctionusesgerminationdata
tion. Twenty replicates often seeds from each from every seed, in the form.of elapsed time
site were arranged in petri dishes on a bed of (days) to germination (right censored data)
moistcottonwoolwithananti-fungaltreatment and also data for seeds that did not germinate
(10mlofMancozebsolution).Petridisheswere (non-censoreddata) (Gimenez-Benavidesetal
sealed with Parafilm to prevent moisture loss 2005).Thevaluesintheprobabilityofgermina-
andkeptat2°Cinthedark.Seedswereexposed tioncurvesarethenconstructedbysubtracting
to daylight for a few minutes whilst they were the K-Mvalues (0to0.9) from 1. Comparisons
beingmonitored forgermination, everytwo to ofpairedgerminationprobabilitycurvesacross
threeweeks. Petri disheswere monitored every sites were then tested by non-parametric Log-
one to two weeks after the first germination of ranktests, which provide a Mantel’s x" statistic
seedsoccurred.Theexperimentwasterminated and P-value.
at205 dayswhen germination rates across sites Linear regression, and in some cases Pearson/
had stabilised and incidences offungal attack Spearman rank correlations, were performed
TheVictorianNaturalist
6
Research Reports
across populations between combinations offi- parisons of seeds collected in 2004 revealed
nal per cent germination, lag-time, germination thatthosefromMtBogongandMtSpeculation
speed,altitudeofseedoriginandseedmass. (the highest and lowest sites respectively) had
SYSTAT version 10 (Copyright SPSS Inc., significantly lower final mean per cent germi-
2000) was usedforall statisticalanalyses. nation (P<0.05) than seeds from other sites
(Fig. 3). Ofthe seeds collected in 2005, those
Results
from Mt Bogong also had significantly lower
Germinationcharacteristics mean finalper cent germination (P<0.05) than
aSenedds200f5rosmhoawlledelseovmateiognesrmcionlalteicotnedunidner20t0h4e ttihoenootfhesreesdistefsr(oFimg.Mt3).BoFignoanlg,peMrtceHnottghearmmianna-d
experimental conditions (Fig. 2). The final mean Mt Bullershowednosignificantdifferencesbe-
percent germinationrangedfrom 39 to 95% for tweenyears(P>0.05). Linearregressionshowed
seedscollectedin2004,and30to95%forseeds
there was no significant relationship between
cgHoeolrtlmehicantaemtdihioannd2in0th0be5othrhiegsyhpeeeascrttsifv(ie9nl5ay.l5Sm±eeead1.ns53fp).reroCmocemMn-tt smieteTashnweafistniaamlpepperuronxtciielmnattthgeeelryfmir1is5ntawgteeierokmnsianfnartdoimaolnttihtfeuodsret.aarltl
0 50 100 150 200
Time (days)
Fig. 2. Mean per cent germination (±1 standard error) ofAciphyllaglacialis seed collected from the seven
alpinesitesin(a)2004and(b)2005,overthedurationoftheexperiment(days).
Vol 126(1) 2009
7
Research Reports
2004 2005
Year
Fig. 3. Mean finalper centgermination (±1 standard error) ofAciphyllaglacialisseed collected in 2004 and
2005 from different elevations (m). Different labels (a-d) above columns signify the Bonferroni significant
differenceswithineachyear.
ofthe experiment; the exact date fell between Discussion
observation periods. Germination rate {T50) Generalinterpretationofresults
varied across the altitudinal gradient (Table 2). Aciphyllaglacialis seeds did not germinate un-
However, there was no significant relationship der the laboratory conditions that seeds from
between T50 and the altitude ofseed origin. T50 nine co-occurringalpine species found favour-
did, however, show a strong relationship with able(Venn2007).ThisindicatesthatA.glacialis
finalpercentgermination(R^=0.81,P<0.001) seedsmayexperienceinnateprimaryseed dor-
indicating that fast germination rates correlate mancyInthecurrentstudyseeddormancywas
with high final germination numbers. Linear broken bywet-cold stratification, which repre-
regression showedthattherewasno significant sents anatural dormancybreakingmechanism
relationship between T50 and altitude of seed (Probert 1992) and impliesthatthisspeciesisa
origin (P>0.05). strict spring germinator. Furthermore, the few
The mean probability of germination over observations ofA. glacialis seedlings emerging
time (Fig. 4) indicates there are few trends in the field are restricted to spring and early
with altitude ofseed origin; however, there are summer(Venn 2007).
several significant differences between curves Thelength oftimeforA.glacialis seeds from
ofdifferent sites (Table 3). Ofthe 23 pairwise all populations to germinate was around 15
comparisons ofgermination probabilitybysite weeks, during which conditions were kept at a
over both years, nine showed that the higher constant 2°C. This length oftime roughly cor-
altitude site of the pair had a greater overall relates with the length of a typical Victorian
probability of germination during the course winter, taking into account inter-annual varia-
ofthe experiment. Eleven showed that a lower bilityin snowfall (Hennesseyetal 2003). Seed
sitehad ahigh germination probability. Hence, dormancy and the timing ofseed germination
there were no overall trends in seed germina- in A. glacialis under natural conditions, there-
tion probabilitywith altitudeofseedorigin. fore,maybearesultofevolutionaryexperience
Between years, the probability of germina- in this species growing in the Victorian alpine
tion was significantly different for Mt Bogong, zone.
Mt Hotham and Mt Buller. Seeds collected in In this study, considerable variation in ger-
2004 from Mt Bogong and Mt Hotham were mination characteristics between populations
more likelyto germinatethan seedcollected in fromdifferentaltitudeswasobserved. However,
2005 (Mantel’s = 16.264, P < 0.001; Mantel’s finalmeanpercentgerminationshowednosig-
= 34.267, P< 0.001 respectively). Incontrast, nificant trends with altitude, nor did seed ger-
seeds from Mt Buller collected in 2005 had a minationprobability.Therefore,germinationin
higher probability of germination than those thisspeciesislikelytobeafunctionofinherited
from2004 (Mantel’sX" = 34.025; P < 0.001). seed traits (Grime et al. 1981; Washitani and
Masuda 1990) and does notshowstrong adap-
8 TheVictorianNaturalist
Research Reports
Table2.Germinationrate(Tso)ofAciphyllaglacialisseedsfromsevensites,atvariousaltitudes.
Siteofseedorigin Altitude(m) Yearofcollection Tso(days)
MtBogong 1970 2004 157
MtHotham 1860 2004 70
MtBuller 1762 2004 96
MtHowitt 1738 2004 122
MtMagdala 1725 2004 131
KingBilly 1696 2004 70
MtSpeculation 1668 2004 205
MtBogong 1970 2005 205
MtHotham 1860 2005 122
MtBuller 1762 2005 70
1.0 a) 2004
-©•••Kilt Bogong(1970m)
-G- Mt Hcrtham(I860m)
0.8 i X Mt Buller(1762m)
—
-# Mt Hounttt(1738m)
MtMagdala(1726m)
0.6 --M——KingBitly(1696m)
Mt Speedation(1668m)
0.4
03 0.2
-Q
O
0
1.0 b) 2005
1
-I
0.8 - ...f-
0.6
0.4
$
0.2 ff- -$
.m-
50 100 150 200
Time (days)
Fig. 4. Mean probability ofgermination (±1 standard error) ofAciphylla glacialis seed collected from the
sevenalpinesitesin(a)2004and(b)2005basedonKaplan-Meiermodelsoverthedurationoftheexperiment
(days).
Vol 126 (1)2009 9
Research Reports
Table 3. Pairwise comparisons oftheprobabilityofAciphyllaglacialisseed germinationbetweensites,based
ontheprobabilityofgerminationcurves,explainedbyMantel’s fromseedcollectedin2004and2005. Sig-
nificantdifferencesareindicatedbyanasterisk(s). denotesP<0.001,**denotesP<0.01,^denotesP<0.05
Overall greater probability ofgermination between each pair is indicated in brackets with greater than (>)
symbolbetweensitenames.Sitenameabbreviationsareasfollows:BO,MtBogong(1970m);HO,MtHotham
(1860m);BU,MtBuffer(1762m);HW,MtHowitt(1738m);MA,MtMagdala(1725m);KB.KingBilly(1696
Year/Site
2004 HO BU HW MA KB SP
BO 153.986^’^’^ 30.323^^=^ 52.243’^^’^ 53.297*** 80.893*** 4.540***
(HO>BO) (BU>BO) (HW>BO) (MA>BO) (KB>BO) (BO>SP)
HO 33.978*'^* 28.733*** 35.362*** 7.264** 164.768***
(HO>BU) (HO>HW) (HO>MA) (HO>KB) (HO>SP)
BU 0.838 0.685 10.988*** 50.047***
(KB>BU) (BU>SP)
HW 0.201 5.811* 67.260***
(KB>HW) (HW>SP)
MA 6.203* 70.067***
(KB>MA) (MA>SP)
HO BU
2005
BO 160.512*^^ 195.719'^*’^
(HO>BO) (BU>BO)
HO 41.276^^=^
(BU>HO)
tationstothe changesin environmentalfactors pableofrapidgerminationhaveanearly-season
foundacrosssites.A.glacialisseedsalso appear survivaladvantage.
ttohusacttheintdiempienngdeonftgleyrmoifnaatmiboinenmtaycobnedirteiloinasn,t beTthweeevnariyaetairosnfirnomgeMrmtinBaotgioonng,chaMrtactHeoritshtiacms
on seed dormancy alone. This restrictive ger- and Mt Buffer were statistically undetectable.
mination pattern mayprotectA. glacialis seeds Hence, the 2005 seeds followed the same pat-
from the hazards ofearly germination (Meyer terns in per cent germination as those from
aeotlvleyarlenairg1lh9y,9t0ls)ep,arvieinsngpgefcrpiolasaltnlstysari‘feousctnoiomnwmtohmenelct(osBladnunnwiushsteuen-r e2Ss0ut0c4fh.incSaoelnegsdiessrftmerinnocamytMibotentHwooefteahnllaymseiatsrehssoiiwsneubdnouttshheuayhleiagrfhso-.r
tethiaslst2u0d0y5,).70T-h9e0%h,igahrerawtietshionftgheermriannagteioofnailn- atlurpeisnedsupreicniges,thaessseleigdhtmaitncurreaatsieosnipnetreimopdercaa-n
pinespeciesfromaroundtheworld (Sayersand lead to improved seed germination potential
Ward 1966; Bell and Bliss 1980; Mariko et al (Korner 1999). The timing of snowfall, snow-
19H9i3g;hKofrinnaelrm1e99a9n).per cent germination was tmieclutlaarlnyduanmubsiueanltctoemmppearraetdurteostwheerelonnogt-tpearrm-
sSiugcnhifiicnahnetrleyntrleylaftaesdttgoerfmasitnagteiromninraatteisonforlaltoews-. tsirmeinldasriitnieesitihnerg2e0r0m4inoart2i0o0n5maanyd,bteheraefroerseu,ltthoef
ing snowmelt may be important during the the predictable environmental conditions that
early phases of seedling growth (Sayers and prevailatthese sites.
Ward 1966). This ensures that seedlings estab- Dormancymechanisms
lish during periods offavourable weather and Dormancymakesintuitivesenseinalandscape
ample soil moisture and, therefore, species ca- wherewinters are coldand severe. Billings and
TheVictorianNaturalist
10
Research Reports
Mooney(1968)suggestthattheprotectivelayer Altitudinalgradients
ofsnowoverwintermayservetoinsulateseeds A relationship between germination charac-
sufficiently;hence, theyneedneverfullyevolve teristics and the altitude ofseed origin was not
specific seed dormancy syndromes. Hence, found in this study, possiblybecause thealtitu-
most alpine species lack a seed dormancy dinal differences between sites which support
mechanism and the elapsed year or more be- A. glacialis populations was small. Elsewhere,
tween seed production and germination is an thevariation ingermination characteristicsbe-
environmentally imposed cue (Amen 1966). tweenpopulationsacrossenvironmentalgradi-
However, non-dormantspecies maygerminate entsiscommon,especiallyamongstspeciesthat
in mid-winter ifthe snow melts unexpectedly, respondtovariationinhabitat.Forexample,the
and therefore experience high mortality rates shrubArtemisia tridentata, which growsacross
(Amen 1966). awide environmental range in North America
Innate seed dormancyis most frequentlyim- from desert to montane sites, showed strong
posedbyseedcoatinhibitionamongalpinespe- habitat related differences in seed germination
cies (Amen 1966; Urbanska and Schiitz 1986). patterns (Meyeretal 1990). Thompson (1973)
This mechanism can spread germination over compared the germination requirements for
an extended period and thereby encompass Silenevulgarisacrossawidegeographicalrange
several periods of favourable growth. Thus, and found that the temperature range for seed
germination of some innately-dormant spe- germination was broad across several habitats.
cies mayoccur intermittently over manyyears, However,withinahabitat,germinationcharac-
even ifgrowing season conditions are favour- teristics between populations were remarkably
able for seedling emergence (Mark 1965; Bill- similar (Thompson 1973). Some researchers
ings and Mooney 1968). Seed coat inhibition have suggested that the highest per cent ger-
can be broken by scarification ofseeds. Under mination and shortest germination rate may
natural conditions, abrasion of buried seeds come from seeds which originate from the
by frost heave or strong winds at the soil sur- highest elevations, when compared to seed of
face, blowing seeds over rocky substrates, may the same species from lower elevations (Ma-
induce the required scarification (Amen 1966; riko et al. 1993; Holm 1994; Vera 1997). Vera
Billings and Mooney 1968). In future studies, (1997) showed Calluna vulgaris seeds that ger-
deliberate scarification ofA. glacialis seeds, in minated in the high mountains in Spain also
combination with a chilling treatment, could had the highest survival rate as seedlings at
confirmwhetherseed coat inhibition occurs in the highest altitudes. Large seeds at the high-
thisspecies. est altitudes also mayhave the highest survival
Cold stratification and cold-wet stratification rates as seedlings (Holm 1994). No significant
aretechniques used tobreakseed dormancyof relationshipbetweenA.glacialisseedmassand
many species from cold climates (Amen 1966; altitude has been found (Venn 2007), nor has
Sayers and Ward 1966; Cavieres and Arroyo there been any suggestion ofsignificant differ-
2000). In this experiment, the length of the ences between seed mass from high altitude or
cold-wet stratification period required for ger- low altitude sites. Seed germination rates were
minationwasaround 15weeks. However,other significantlyloweratMtBogongandMtSpecu-
studies clearly show that seeds originating lationwhencomparedtotheothersites.Hence,
from high elevations require longer stratifica- A.glacialismaybenearitsupperandlowerdis-
tionperiods (BillingsandMooney 1968; Dome tributional limits at these sites. Pickering and
1981; Cavieresand Arroyo2000). Cavieresand Arthur (2003) speculated that the growth and
Arroyo (2000) showed that Phacelia secunda fecundity offemale A. simplicifolia plants may
seed from 1600 m needed only one month of be limited at higher altitudes due to the high
cold stratification, whereas seed from 3400 m maternal investments made during seed pro-
needed three months. This pattern may not be duction,incombinationwithstressfulenviron-
apparent in A. glacialis seeds as the altitudinal mental conditions. However, no general trends
distribution of this species is relatively small betweenaltitudeandfinalpercentgermination
and/or adaptations to annual changes in the orgerminationprobabilityforA.glacialisseeds
snowseasonlengthacrossthealtitudinalgradi- were found in thisstudy. Therefore, anylimita-
entmaybe minimal orundetectable. tions that the female plants may be experienc-
Vol 126 (1)2009 11
Research Reports
ingacrossthealtitudinalgradientdonotappear Holm S-0 (1994) Reproductive patterns ofBetulapendula
tobecausingreducedseedgermination. andB.pubescenscoll,alongaregionalaltitudinalgradient
Conclusions InionuyneorDthWernanSdwePdyekne.GEcHog(r1a9p8h8y)1P7o,l6l0i-n7a2t.ionbiologyin the
SnowyMountainsofAustralia:comparisonswithMontane
In laboratory experiments, A. glacialis seeds Colorado,USA.AustralianJournalofEcology13,91-210.
showed germination responses typical ofseeds Kaplan EL and Meier P (1958) Non parametric estimation
with an innate primary dormancy syndrome. fromincompleteobservations.JournaloftheAmericanSta-
tisticalAssociation53,457-481.
This is an effective way to ensure that spring KirkpatrickJB (1997)Alpine Tasmania. An illustratedguide
germination is cued at the end of the winter to thefloraandvegetation. (OxfordUniversityPress: Mel-
snow season in an alpine landscape. Final per KobronuerrneC)(1999)AlpinePlantLife.(Springer:Berlin)
cent germination was high, up to 95%, and no LCC(1982) Reportonthealpinestudyarea. (LandConser-
dcioflfleercetnecdesinindigfefremrienntatyieoanrsw.erFeasftoguenrdmiinnasteieodns Lev(eaJtoEihTonnW(Ci1ol9ue9ny2c:)ilNS:teaMtwieslYtobicroakul)rmneet)hods ofsurvivaldata analysis.
rates weresignificantlyrelatedto high finalper MarikoS,KoizumiH,SuzukiJandFurukawaA(1993)Alti-
cent germination. These germination charac- tudinalvariationsingerminationandgrowthresponsesof
ReynoutriajaponicapopulationsonMt.Fujitoacontrolled
teristics ofA. glacialis did not show significant thermalenvironment.EcologicalResearch8,27-34.
trends across the altitudinal gradient of sites MarkAF(1965)Vegetationandmountainclimate.InCentral
within theVictorian alpine zone. Otago,pp69-91.RGListerandRPHargreaves.(NewZea-
landGeographicalSociety:Dunedin,NZ)
References MeyerSE, Monsen SB and McArthur ED (1990) Germina-
Amen RL (1966) The extent and role ofseed dormancy in atinodncrheisllp:onpsaettoefrnAsrtoefmibseitawetreind-epnotpatuala(tAisotneravcaeriaaet)iotno.liBgoh-t
alpineplants.QuartelyReviewofBiology41,271-281. tanicalGazette151, 176-183.
BaAn,niLsotrerdJP,M,MaMeaglrikTA,FDiacnkdinSspoennsKeJrMK,LH(a2l0l0o5y)SWRiPl,lKlnosisghotf PiacknedrirnegsoCurMce(a2l0l0o0c)atSieoxn-sipnecAiufsitcradliifafenraelnpcienseindifoleorcailoduissApclia-y
snowcoverduringclimaticwarmingexposeNewZealand phyllaglacialis(Apiaceae).AustralianJournalofBotany48,
alpine plants to increased frost damage? Oecologia 144, 81-91.
Ba2s4k5i-n25C6.C and Ba.skin JM (1988) Germination ecophysi- PilcokceartiinognCinMthaenddiAoretchiuoursJaMlp(i2n0e0h3e)rPbatAtceirpnhsylolfarseismopulircciefsolaila-
BaAoslmkoeignryiJcoMafnahJneodrubrBanacaselkooiufnsBCopltCaann(ty19s77p51e,)ci2Ge8se6-ri3mn0i5na.atteimopnereactoelorgeygiaonnd. Pit(chAkepeireaifcnfeegacteC)o.MfAaaultsnitdtruaHdlielElocnoWlsoe(gx2yr0a20t82i,)os5R6ie6np-rt5ho7ed4.udcitoievceioeucsohleorgbyAacnid-
adaptation to nature in Leavenworthia spp. (Cruciferae). phyllasimplicifolia(Apiaceae).AustralianJournalofBotany
AmericanMidlandNaturalist85,22-35. 50,289-300.
BellKLandBlissLC(1980)Plantreproductioninahigharc- Probert RJ (1992) The role oftemperature in germination
BiltlicinegnsvWirDonmaenndt.MAoroctniecyanHdAAl(p1i9n6e8)ReEsceoarlcohgy12o,f1a-r1c0t.icand pelcaonpthycsoimomluongiyt.ieIsn.SEededMs:Ftehneneerc.ol(oCgAyBa:nWdalrleignegnfeorradt)ion in
alpineplants.BiologicalReviews43,481-529. Raven PH (1973) Evolution ofsubalpine and alpine plant
BlissLC0962)Adaptationsofarcticandalpineplantstoen- groups in New Zealand. New ZealandJournal ofBotany
vironmentalconditions.Arctic15. 11, 177-200.
Bliss LC (1971) Arctic and alpine plant life cycles. Annual Sayers RL and Ward RT (1966) Germination responses in
CaRvieevrieeswoLfAEcaolnodgyAarnrdoySoystMeTmaKtic(s220,004)05-S4e3e8d. germination SchailiptizneWsp(e2ci0e0s2.)BDootarnmiacanlcGyaczheatrtaect1e2r7i,st1i1c-s1a6.ndgermination
response to cold stratification period and thermal timingin twoCarexspecies.BasicAppliedEcology3, 125-
regimeinPhaceliasecunda(Hydrophyllaceae).PlantEcol- 134.
Coosgtyin1A49B,(1-189.57)ThehighmountainvegetationofAustralia. ThloogmipcaslonanPdAg(e1o9g7r3a)phiSceaelddgiesrtmriinbuattiioonn. iInnrTelaaxtoionnomtoyeacnod-
CoAsutsitnraAlBia(n1J9o6u2r)naSloiolsfBoofttahneyH5i,g1h73P-l1ai8n9s..Proceedingsofthe ELcoonldogoyn,)pp. 93-119. Ed VH Heywood. (Academic Press:
CoRsotyianlASBo,cieGtryaoyfMV,ictToortitaer7d5e,ll29C1J-2a9n9.d Wimbush DJ (2000) UrablapnisnekaplKanMtsaannddScrheivietgzeMtat(i1o9n86r)esReeaprrchodaubcotvieontibmybesreleidnei.n
KosciuszkoAlpineFlora(CSIRO:Melbourne) BotanicaHelvetica96,43-60.
Dome A (1981) Variation in seed germination inhibition Venn SE (2007) Plant recruitment across alpine summits
ofChenopodium bonus-henricus in relation to altitude of in south-eastern Australia. (Unpublished PhD thesis, La
plantgrowth.CanadianJournalofBotany59, 1893-1901. TrobeUniversity:Bundoora)
Gimenez-Benavides L, Escudero A and Perez-Garci'a F Vera ML (1997) Effects ofaltitude and seedsizeon germi-
(2005)SeedgerminationofhighmountainMediterranean nation andseedlingsurvival ofheathlandplantsin north
asbpielciitey.s:EcaolltoitguidcianlalR,esienatrecrhpo2p0u,l4a3t3i-o4n44a.nd interannual vari- WaSsphaiitna.nPila1natnEdcoMlaosgyud1a33,M10(11-919006).A comparative studyof
GrfiomrtehJMP,AMGa,soNneaGl,ACMurtaisndAVS,haRwodSm(a1n98J1,)BAandcomSRp,arMatoiwv-e gtrhaesgselarnmdincaotmimounnichtayr.acFtuernicsttiiocnsalofEcsoeleodgsyf4r,o5m43a-5m5o7i.sttall
studyofgerminationcharacteristicsinalocalflora.Journal Williams RJ (1987) Patterns ofair temperature and accu-
HeonfnEecsosleoygyK6,9,Wh1e0t1t7-o1n05P9,.Smith I. BatholsJ, HutchinsonM mounltahteioBnogoofnsgnoHwigihnPsluabianls,piVniecthoeraiat.hlAaunsdtsraalniadngJroausrsnlaalndosf
andSharpiesJ(2003)Theimpactofclimatechangeonsnow Ecology12, 153-163.
conditionsinmainlandAustralia.CSIROAtmosphericRe-
search,Aspendale,Victoria. Received29November2007;accepted17June2008
12 TheVictorianNaturalist