Response of Resprouting Shrubs to Repeated Fires in the Dry Sclerophyll Forest of Gibraltar Range National Park

2006·2.9 MB·English

by  Kirsten Je Knox

Response of Resprouting Shrubs to Repeated Fires in the Dry
Sclerophyll Forest of Gibraltar Range National Park
KiRSTEN J. E. Knox''^ and Peter J. Clarke^
'Botany, School ofEnvironmental Sciences andNatural Resources Management, University ofNew England,
Armidale, 2351 ([email protected]),
^current address: Department for Environment and Heritage, PO Box 822, Clare, SouthAustralia 5453.
Knox, K.J.E. andClarke, RJ. (2006). Response ofresprouting shrubs to repeatedfires in thedry
sclerophyll forestofGibraltarRangeNational Park. Proceedings oftheLinneanSocietyofNewSouth
Wales 121, 49-56.
Fire regimes affect survival and reproduction ofshrub species infire-prone vegetation such as occurs in
Gibraltar Range National Park. The influence offire regimes on resprouting shrubs is known for a range
ofspecies in coastal regions ofAustralia but is poorly known in montane sclerophyll communities. The
fire responses ofthree Proteaceae shrubs {Banksia spinulosa, Hakea laevipes, Petrophile canescens) and
a grasstree {Xanthorrhoeajohnsonii) were measured after the wildfire of2002 to determine whether: 1)
storageorgansizewasrelatedtopost-firegrowthandfloweringresponse, 2)firefi-equencyinfluencespost-
fire mortality and ifsurvival was related to the size ofplant; 3) fire fi-equency influences the resprouting
ability ofplants, and4) fire frequency affects pyrogenic flowering in the post-fire environment. We found
the size ofstorage organs was positively relatedto post-fire sprouting in the three shrubs andto flowering
inthegrasstree.However,highfirefi-equencyonlyaffectedthesurvivalofBanksiaspinulosaanddecreased
flowering in Xanthorrhoeajohnsonii. Survival in all species ranged between 83 and 99% and it appears
thatthe intervalsbetweenfires (7-22 years) hadbeensufficientformostadultplantsto regainthe abilityto
resprout. The abilityofjuvenileplantstodevelopthe abilitytoresproutneedstobetestedonseedlingsthat
establishedafterrecentfires.
Manuscriptreceived IMay2005, acceptedforpublication 7 December2005.
KEYWORDS:
Fire frequency, fireregime, persistence, pyrogenicflowering, resource allocation
INTRODUCTION
BondandVanWilgen 1996). Some specieshavebeen
The fire response of species is often simplified found to have greater resprouting potential in larger
into resprouters and obligate seeders, but in reality size-classes (e.g. Morrison 1995); in contrast, some
a continuum fi-om 0-100% mortality of individuals species have been found to have greater resprouting
within a population exists amongst species potential in smaller size-classes (e.g. Burrows 1985).
(Bellingham and Sparrow 2000; Vesk and Westoby Frequent fires with short inter-fire intervals may
2004; Clarke et al. 2005). Characteristics of a result in the exhaustion of buds or carbohydrates
particular fire, distribution of size-classes and the stored in the lignotuber, resulting in the mortality of
physiologicalandanatomicalfeaturesofaspecieswill resprouters (Zammit 1988; Bowen and Pate 1993).
affect the percentage mortality in a population after The intensity of a particular fire can influence what
fire. Shrub species capable of resprouting generally proportion of a population survives. Some obligate
resprout from subterranean buds (lignotubers and seeders may survive a low-intensity fire if 100%
roots suckers), but also occasionally from epicormic leafscorch does not occur (Gill 1981; Bond and van
buds on aerial stems. Grasstrees on the other hand Wilgen 1996). Onthe otherhandaveryhigh-intensity
resprout via apical buds. The ability ofan individual fire may result in the death of a large number of
toresproutfollowingfiredepends onhavingadequate individuals within apopulation that usually resprouts
dormant buds and carbohydrate storage to facilitate following fire. The minimum fire-tolerant stem size
resprouting (Bell 2001; Knox and Clarke 2005). of resprouters often increases with fire intensity for
Variation in mortality has been observed for different some species (Bradstock and Myerscough 1988;
size-classes within populations (Morrison 1995; Morrison 1995; Morrison and Renwick 2000).
RESPONSE OF RESPROUTING SHRUBS TO REPEATED FIRES
METHODS
Resproutersthatrecruitseedlingsintopopulations
followingfire generallyhave seed storedinthe soil or
in the canopy in woody fruits. Hence, understanding Fire regime maps of Gibraltar Range National
the post-fire growth and reproductive response Park were examined and dry sclerophyll forest areas
of resprouting shrubs is critical in determining that had been burnt twice, four and five times since
appropriate fire regimes in landscapes dominated 1964wereidentified.AllsiteswereburntinNovember
by resprouting species. Resprouting shrubs typically 2002 by an intense wildfire that removed most ofthe
have greater growth and reproductive vigour in the tree leaf canopy but did not totally incinerate the
yearfollowing afire (e.g.Auld 1987; BowenandPate fruits of the target species. The minimum interval
2004). In many resprouting shrubs flowering occurs between fires was approximately seven years and the
predominantly, or exclusively, following fire, e.g. maximum 22 years. Fire records showedthat all fires
Telopeaspeciosissima(Pyke 1983),Lomatiasilaifolia burnt in spring/summer, suggesting that all fires were
(Denham and Whelan 2000), Xanthorrhoea preissii of high intensity. All observations were at the same
(Lamont et al. 2000), and Stirlingia latifolia (Bowen time since the last fire (8 months). In areas ofeach of
and Pate 2004). Little is known about the factors that the fire frequency regimes two patches were chosen
influence reproductive output ofpyrogenic flowering thatwereatleast 1 kmapart. Ineachpatchthree500m
plants, although season of fire is known to strongly transects were established and the post-fire response
influence flowering in some Western Australian of three species of Proteaceae shrubs with canopy-
species (Lamont et al. 2000; Bowen and Pate 2004). held seed banks {Banksia spinulosa, Hakea laevipes,
One important component offire regime that is likely and Petrophile canescens) were measured. These
to influence post-fire flowering is the frequency of species were selected because they are ubiquitous,
bumsasthismayinfluencethestarchstoragecapacity easyto identifywhendead, andtheyonlyrecruitafter
ofplants. fire, hence their minimum age can be estimated. For
Little quantitative work has been conducted to eachspeciesthenumberofshootsresproutedfi^omthe
determine the effects of frequent fires on post-fire lignotuber, the length ofthe longest shoot resprouted
performance and mortality of shrubs that resprout andthebasalgirthofthelignotuberweremeasuredfor
following fire. The fire regime in Gibraltar Range the first 20 (approx.) individuals encountered in each
National Park provided an opportunity to examine transect. Dead plants were also recorded and their
these questions because records date back to the basal girth measured. Individuals were identified by
1960s and the number and extent of subsequent their 'skeletal' remains and their canopy-held woody
fires have spatially explicit records. Gibraltar Range finits. In addition, the post-fire flowering of the
National Park also has widespread and abundant grasstree Xanthorrhoeajohnsonii was also recorded
populations of resprouting shrubs occurring within along eachtransect. VoxXanthorrhoeabasal girthand
physiographically similar landscapes. In late 2002 an height ofthe caudex were measured andthe presence
intensive crown fire burntmost ofthe dry sclerophyll and length ofthe inflorescence were recorded for the
forest in the National Park. This event afforded an first 20 individuals encountered along each transect.
opportunityto studythepost-fire response ofspecies, Data for this study were mainly collected by
which have experienced differentfire firequencies. undergraduate students. All students collected the
Evidence of an effect of fire fi-equency would equivalent amount of data fi-om each of the fire
show that more frequently burnt sites had more dead frequency areas. This was important so that the
plants and surviving plants with reduced grow1:h and patterns in post-fire resprouting and flowering could
reproduction. If, however, these sites had smaller beattributedtothedifferentfirefrequencies,andnotto
plants,thenthesemayappeartoshowreducedsurvival, variationinsamplingamongdifferentstudents.Ineach
growthandreproductionpurelyforallometricreasons. ofthe three species ofshrubs, to test the relationship
Hence we asked whether: 1) storage organ size was between storage organ size and post-fire response,
related to post-fire growth and flowering response, basal girths were regressed against the number of
2) fire fi:equency influences the resprouting ability shoots resprouted, height of shoots resprouted and
We
ofplants 3) fire firequency and/or size ofthe storage size of inflorescence as independent variables.
organ influences post-fire mortality 4) fire frequency then used analysis of covariance (ANCOVA) to test
affects pyrogenic flowering (flower or inflorescence ifnumber of shoots or stem height were reduced by
production) in the post-fire environment. fire firequency, with lignotuber size as the covariate.
We alsousedANCOVAinXanthorrhoeato test ifthe
inflorescence length was reduced by fire firequency,
with the caudex size as a covariate. Plots ofresiduals
50 Proc. Linn. Soc. N.S.W., 127, 2006
— —
I I I 1 I 1 I
KNOXAND CLARKE
K.J.E. P.J.
established that no transformations ofraw data were analysis the response variable was the number of
necessary. Homogeneity of slopes was determined plants flowering ornot flowering.
by testing the interaction between the covariate and
the main factors. We next tested the hypothesis that RESULTS
storage organ size and/or fire frequency affects post-
fire survival by logistic regression using likelihood Of the four species sampled, only the grasstree
ratio tests. In these analyses the response variable is Xanthorrhoea johnsonii was observed to flower in
the number ofplants alive or dead. InXanthorrhoea, the immediate post-fire period (August 2003), whilst
we also tested the hypothesis that caudex volume the other species began to flower in the following
and/or fire fi-equency affects post-fire flowering by year (August2004). Allresprouting Proteaceae shrub
logistic regression using likelihood ratio tests. In this species had a positive and significant {P < 0.05)
a) Banksia
120 -
100 -
80 -
60 -
40
20 1
r^""—I—"""I—"""T—"—1——1—"-T——1—""T •—I -I 1—1 1— 1— r— 1—1 1— 1— 1—
10 20 30 40 50 60 70 80 90 10 20 30 40 50 60 70 80 90
Girth (cm) Girth (cm)
o
o b) Hakea i 60 1
e
o 50 -
•
c
o c40-
e
o
% 30 -
o
20-
t
O)
X
|io-
1—"—I—f—I—>—1—"—I <—\—— 1 oJ iiijiifmnpuo ''^o ' " 1 1 • 1 1 1
10 20 30 40 50 60 70 80 90
20 30 40 50 60 70 80 90
Girth (cm)
Girth (cm)
c) Petrophile
30 40
Girth
Figure 1. Regression ofshoot height and numberwith lignotuberbasal girth with 18 months after fire
for a)Banksia spinulosa, r^ = 0.77, 0.59; b) Hakea laevipes, r^ = 0.77, 0.70; and c)Petrophile canescens
r^ = 0.76, 0.77, across all fire frequencies offire at Gibraltar Range National Park.
Proc. Linn. Soc. N.S.W., 127, 2006 51
RESPONSE OF RESPROUTING SHRUBS TO REPEATED FIRES
Table 1. Summary results for analysis of covariance for height (Ht) and numbers
of shoots (Shoot) resprouted for Banksia spinulosa, Hakea laevipes, Petrophile ca-
nescens and length of inJBorescence (Infl.) in Xanthorrhoea johnsonii. The size co-
variate for the three shrubs was basal girth and for the grasstree it was the cau-
dex volume. *** indicates P<0.05, ** indicates P<0.01, and * indicates P<0,001
Factors B. spinulosa H. laevipes P. canescens X.johnsonii
Ht Shoot Ht Shoot Ht Shoot Infl.
Fire frequency NS NS NS NS NS NS
Size (covariate) NS
Fire fq. x size NS NS NS NS NS * NS
b) Hakea
24 - a) Banksia 24 - n 2 Fires
H
20 - 20 - 4 Fires
H
5 Fires
£ 16 - £ 16 -
o
12 - 12 - 1
b
8 - 8 - 1
4 - 4 -
-' -
Dead Alive Dead Alive
c) Petrophile 30000 1 d) Xanthorrhoea
24
D
25000 2 Fires 1
20 1 n J
T T
, £ 4 Fires
o 20000 H
16
£ 5 Fires
o Of
£
J= 12 - 1 5000 -
>
b
8 - oXOf 10000 -
4 - O 5000 -
Dead Alive
Not flowering Flowering
Figure2.Mean(+se)basalgirthoflignotubersineachofthreefirefrequenciesinGibraltarRangeNational
Parkfora)5fl«^s/as/7/«H/os«,b)jyaA^ea/aevipes,andc)Pefro/?^//ecflnesce«s.Meanvolumeofthecaudex(+se)
for%««?Aorr^o^flyo^«so«Hforeachfirefrequencywheresmallerplantsflowerinsiteswithlessfrequentfires.
52 Proc. Linn. Soc. N.S.W., 127, 2006
KNOX AND CLARKE
K.J.E. P.J.
Table 2. Number ofplants recorded in each fire frequency category and the results oflogistic
regression for fire frequency and size ofstorage organ from likeUhood ratio tests. *** indicates
P<0.05, ** indicates P<0.01, and * indicates P<0.001
Chi- Chi-squared
Logistic response
Species Fire frequency squaredFire Basal girth/
variable
frequency volume
B. spinulosa Dead 10 21 24
Alive 96 99 108 10.7* 16.6***
H. laevipes Dead 4 5 3
Alive 116 115 116 0.5 NS 8.7=*
P. canescens Dead 9 2 4
Alive 120 121 117 0.8NS 7 9**
X.johnsonii Notflowering 103 93 110
Flowering 17 7 8 8.4* 67.3***
relationshipbetweenbasal girth ofthe lignotuber and species had an increased likelihood of mortality as
post-fire response of shoots (numbers and height) 8 lignotubersize decreased(Table 3, Fig. 2.). However,
months after fire (Fig. labc). In addition, the volume fire firequency only influenced mortality in Banksia
of the caudex in Xanthorrhoea johnsonii was also spinulosa where increased fire frequency increased
positivelyrelatedto the length ofthe inflorescence (r^ the likelihood ofmortality (Table 2).
= 0.71). Finally, we examined whether fire fi-equency
Fire frequency did not reduce the height and and/or size of the caudex influenced flowering in
number of shoots resprouting when basal girth was Xanthorrhoea johnsonii. Both size of caudex and
usedasacovariate(Table 1).Firefi-equency,however, fire frequency influenced whether plants flowered or
appeared to increase the height of Hakea laevipes not with increased proportions of plants flowering
which is not consistent with the hypothesis that fire when the caudex was large (Fig. 2) and increased
fi'equencywouldreduce height. In shrub species, size proportions ofplants flowering when fire frequency
and number of shoots resprouting were significantly was low (Table 2). In addition only larger plants
related to basal girth. Hence, the apparent reduction tended to flower in populations with a high fire
in size and number ofresprouted shoots in Banksia fi-equency (Fig. 2).
simply reflects the decreased size of lignotubers
with fire frequency (Fig. 2). Fire firequency did DISCUSSION
not affect the length of the inflorescence in the
grasstree Xanthorrhoeajohnsonii and the length of Inthisstudyweexaminedtheinfluenceofstorage
the inflorescence was not significantly related to the organ size and fire fi-equency on post-fire mortality,
caudex volume (Table 1). resprouting vigour and flowering. Generally, we
Next we ask if fire fi-equency and/or size of found that larger storage organ sizewas relatedto: (i)
storage organ affect the survival of species. Only greaterpost-fire survival, (ii) more resprouting shoots
two of 358 Xanthorrhoea johnsonii plants were (iii) faster grov/ing resprouting shoots, and (iv) the
killed by fire, hence it was not possible to examine presence, andsize ofinflorescences ioxXanthorrhoea
the relationship between survival and caudex size. johnsonii. Fire fi-equency influenced the post-fire
Mortality was sufficiently high in the shrub species survival ofBanksiaspinulosa and also influenced the
to examine the effects of fire fi-equency and size presence ofinflorescences ofXanthorrhoeajohnsonii
on post-fire survival using logistic regression. All following fire (Table 2).
Proc. Linn. Soc. N.S.W., 127, 2006 53
RESPONSE OF RESPROUTEMG SHRUBS TO REPEATED FIRES
Table 3. Summary ofthe results on the effect offire frequency and size ofstorage organ on post-fire
survival and resprouting offour species with canopy-held seed banks.
Does increased
Are storage organ size Does increasedfire Does increasedfire
storage organ size
Species andpost-fire growth fi-equency affect firequency affect
affectpost-fire
response correlated? storage organ? post-fire survival?
survival?
Banksiaspinulosa Yes+ive Yes -ive Yes +ive Yes -ive
Hakea laevipes Yes +ive No evidence Yes +ive No evidence
Petrophile canescens Yes +ive No evidence Yes+ive No evidence
Does increased
Does storage organ Does fire frequency Does fire frequency
storage organ size
size affectpresence of affect size of affectpresence of
affectpost-fire
inflorescence? inflorescence? inflorescence?
survival?
Xanthorrhoea
Yes +ive No evidence No evidence Yes -ive
johnsonii
Post-fire survival for the species studied resprouting shoots. Although this pattern was not
ranged from 99% for Xanthorrhoea johnsonii to imexpected, we had hoped to be able to determine
82% for Banksia spinulosa, although we may have whether carbohydrate storage or ntxmber ofdormant
underestimated mortality if small dead plants were buds was the limiting factor when it came to the
overlooked. YorXanthorrhoeajohnsonii, individuals ability of individuals to resprout, but we did not
were found to survive fire irrespective of storage detect any trends. Rather, it would appear that
organ size. However, forthe three shrubs, individuals larger storage organs have more dormant buds
with small storage organs were more likely to be available for resprouting and greater carbohydrate
killedbyfirethanthose thathadlargerstorageorgans. stores. Whether this greater resprouting vigour for
This pattern is consistent with the idea that many individuals with larger storage organs translates to
species develop greater fire-tolerance as the storage a greater reproductive output remains to be tested at
organ increases in size and age (Morrison 1995; this site, but other studies have shown a relationship
Keith 1996). This increased tolerance is likely to be a between storage organ size and reproductive output
result ofa larger dormant bud bank and more stored (Auld 1987; Bowen and Pate 2004). Furthermore,
carbohydrate in less frequently burnt plants (Knox there was clear evidence supporting this idea in the
and Clarke 2005). We do, however, acknowledge grasstree {Xanthorrhoeajohnsonii) where the length
that the sampling technique used might have of the inflorescence was positively correlated with
underestimated the mortality of some populations the volume ofthe caudex. We also found that plants
if the fire intensity was great enough to incinerate that lacked or had a short caudex did not flower in
fhiits. This could have resulted in an under-sampling the firstyear following fire. This result contrasts with
of dead individuals, as retained woody fruits were the findings ofLamont et al. (2000) who foimd that
used to help identify individuals. There could have for a Western Australian grasstree, plant size was
also been an underestimation ofthe mortality ofpre- not positively related to the proportion of plants
reproductive individuals, as there would not have flowering.
been fi:iiits present to identify the species. Clearly Firefrequencydidnotaffectthepost-firesurvival
ifthis occurred then the percentage mortality ofthe ofHakea laevipes or Petrophile canescens, but high
populations would have been underestimated. While fire fi-equency increased the mortality of Banksia
this may have influenced the recorded percentage spinulosa. Individuals ofBanksiaspinulosa in higher
mortality ofthepopulation, we feelthatitwouldhave fire frequency sites were more likely to be killed by
little influence on our general findings. fire, were generally smaller in size when comparedto
When examining the relationship between the less fi-equentlyburnt sites. Interestingly, we foundno
size of the storage organ and resprouting vigour evidence that this increased mortality was a result of
we found that larger storage organ size was related a depletion ofthe bud bank, as the number ofshoots
to more resprouting shoots and faster growing perplant did not differ among sites with differentfire
54 Proc. Linn. Soc. N.S.W., 127, 2006
KNOX AND CLARKE
K.J.E. P.J.
frequencies. Similarly, we found no evidence that ACKNOWLEDGEMENTS
the mortality was directly related to a depletion of
carbohydrate reserves, as the length of the longest WethanktheNSWNationalParksandWildlifeService
resprout did not differ among sites with different fire (now Department of Environment and Conservation) for
frequencies. Rather, it appears that the cohort that allowing us to measure the post-fire responses ofplants in
recruited since the previous fire (1990) had not had Gibraltar Range National Park and for logistic support for
the research. The 2003 class ofthe Ecology ofAustralian
an opportunityto reachfire tolerance and it was these
Vegetationisthankedfortheirdiligentcollectionofthedata
individuals that contributed to the higher mortality
that have enabled this paper to be produced. Ian Simpson
in the more frequently burnt site. A synthesis of providedexcellentfieldassistanceforthestaffandstudents,
post-fire survival ofjuvenile resprouting species by and the University ofNew England supportedthe research
Keith (1996) suggests that some Proteaceous shrubs by extendingthebudget ofthe studentexcursion.
{Banksiaoblongifolia,Telopeaspeciosissima)develop
the ability to resprout at around five years, but others
{Banksia serrata, Isopogon anemonifolius) may take REFERENCES
more than 10 years to develop a strong resprouting
ability. Whetherjuvenile plants that recruit after fire Auld, T.D. (1987). Post-fire demography intheresprouting
are able to develop persistence in less than 10 years shrubAngophorahispida (Sm.) Blaxell: flowering,
needs to be tested on seedlings that established after seedproduction, dispersal, seedling establishment
recent fires. and survival. Proceedings officeLinneanSocietyof
Xanthorrhoea johnsonii individuals were less NewSouth Wales 109, 259-269.
Bell, D.T(2001). Ecologicalresponse syndromes in
likely to flower in the higher fire frequency site,
thefloraofsouthwesternWesternAustralia: fire
but the length of the inflorescence was not affected
resprouters vs reseeders. BotanicalReview 67, 417-
by fire frequency. This is surprising given the large
440.
investment of resources in post-fire flush flowering Bellingham, P.J. and Sparrow,A.D. (2000). Resprouting
in Xanthorrhoea johnsonii. The reduced flowering as a life history strategy in woodyplantcommunities.
in the high fire frequency sites may be a result of 0/7co5 89, 409-416.
previous fires depleting carbohydrate reserves. Knox Bond, W.J. andVanWilgen, B.W (1996). FireandPlants.
andMorrison (2005) found a similarpattern for some Chapman & Hall, London.
resprouting shrubs where individuals in high fire Bowen, B.J. andPate, J.S. (1993). The significance ofroot
starchinpost-fire shootrecovery ofthe resprouter
frequency sites had lowerreproductive outputthan in
Stirlingia latifoliaR. Br (Proteaceae).Annals of
less frequently burnt sites. Interestingly, Taylor et al.
Botany12, 7-16.
(1998) found individuals ofXanthorrhoeafulva were
Bowen, B.J. andPate, J.S. (2004). Effectofseason of
more likelytoflowerinareas withhighfire frequency bumon shootrecovery andpost-fireflowering
than areas with less frequentfires. While this appears performance intheresprouterStirlingia latifoliaR.
to contradict our findings, the intervals in that study Br (Proteaceae). AustralEcology29, 145-155.
weremuchshorterthanthose inthe currentstudy, and Bradstock, R.A. andMyerscough, RJ. (1988). The
hence it is difficultto draw comparisons. survival andpopulationresponse to frequentfires of
Previous studies that have examined the effects two woody resproutersBanksiaserrata andIsopogon
anemonifolius.AustralianJournalofBotany36,415-
offire frequency in dry sclerophyll vegetation have
431.
oftenfoundresproutersto decline inabundanceunder
Burrows,N.D. (1985). Reducingthe abundance of
very short inter-fire intervals (e.g. Gary andMorrison Banksiagrandis inthe Jarrahforestbytheuse of
1995). In the current study, the shortest interval controlledfire.AustralianForestry48, 63-70.
betweenfireswassevenyearsandatthisfirefrequency Gary, G. J. andMorrison, D.A. (1995). Effects offire
two of the four species examined were adversely frequency onplant species composition ofsandstone
affected in the higher fire frequency sites. This is an communities in the Sydneyregion: combinations of
important finding because the current Guidelines for inter-fire intervals.AustralianJournalofEcology20,
NSW 418-426.
Ecologically Sustainable Fire Management in
(Kenny et al. 2003) indicate that a lower minimum Clarke(,20R0J5.).KLnaonxd,sKc.aJp.eE.p,aWtitlelrsn,sKo.fEw.oaonddyCpalmapnbterlels,poMn.sLe.
threshold between fires for dry sclerophyll shrub to crownfire: disturbance andproductivity influence
forest is sevenyears, and the results from this current sprouting ability. JournalofEcology93, 543-555.
study indicate that such an interval may be too short Denham,A.J. andWhelan, R.J. (2000). Reproductive
forthese particular forests. ecology andbreeding system ofLomatiasilaifolia
(Proteaceae) followingfire.AustralianJournalof
Botany'X^, 261-269.
Proc. Linn. Soc. N.S.W., 127, 2006 55
RESPONSE OF RESPROUTING SHRUBS TO REPEATED FIRES
Gill,A.M. (1981).Adaptiveresponses ofAustralian
vascularplant speciestofire. InFire in theAustralian
Biota. Gill,A. M., Groves, R. H. andNoble, I. R.
(Eds).AustralianAcademy ofScience,ACT,pp. 243-
272.
Keith,D.A(1996). Fire-driven extinctionofplant
populations: asynthesis oftheory andreview of
evidence fromAustralianvegetation. Proceedings of
theLinneanSocietyofNewSouth Wales. 116, 37-78.
Kenny, B., Sutherland, E., Tasker, E. andBradstock, R.
(2003). Guidelines forEcologically SustainableFire
NSW
Management, Government, Sydney.
Knox, K.J.E. andClarke, RJ. (2005). Nutrient availability
induces contrasting allocation and starch formation in
resprouting and obligate seeding shrubs.Functional
Ecology 19, 690-698
Knox, K.J.E. andMorrison, D.A. (2005). Effects of
inter-fire intervals onthereproductive outputof
resprouters andobligate seeders inthe Proteaceae.
AustralEcology30, 407-413
Lamont, B.B., Swanborough RW. andWard, D. (2000).
Plant size and season ofbum affectflowering and
Suiting ofthe grasstreeXanthorrhoeapreissii.
AustralEcology25, 268-272.
Morrison, D.A. (1995). Some effects oflow-intensityfires
onpopulations ofco-occurring smalltrees inthe
Sydneyregion. Proceedings oftheLinneanSocietyof
NewSouth Wales 115, 109-119.
Morrison, D.A. andRenwick, J.A. (2000). Effects of
variation infire intensity onregenerationofco-
occurring species ofsmall trees inthe Sydneyregion.
AustralianJournalofBotany48, 71-79.
Pyke, G.H. (1983). Relationships betweentime since last
fire andflowering in TelopeaspeciosissimaR. Br. and
Lambertiaformosa Sm.AustralianJournalofBotany
31, 293-296.
Taylor, J.E., Monamy, V. andFox, B.J. (1998). Flowering
ofXanthorrhoeafulva: the effect offire andclipping.
AustralianJournalofBotany46, 241 - 251
Vesk,P.A. andWestoby, M. (2004). Sprouting ability
across diverse disturbances andvegetationtypes
worldwide. JournalofEcology92, 310-320.
Zammit, C. (1988). Dynamics ofresprouting inthe
lignotuberous shrubBanksiaoblongifolia.Australian
JournalofEcology 13, 311-320.
56 Proc. Linn. Soc. N.S.W., 127, 2006