Importance of physiological integration of dwarf bamboo to persistence in forest understorey: a field experiment

Journal of Ecology

2002

90

, 78–85

© 2002 British Ecological Society

Blackwell Science Ltd

Importance of physiological integration of dwarf bamboo to persistence in forest understorey: a field experiment

TOMOYUKI SAITOH†, KENJI SEIWA‡ and AYA NISHIWAKI§

†

Laboratory of Forest Ecology, Department of Biodiversity Science, Graduate School of Agricultural Science, Tohoku University, Narugo, Miyagi, 989-6711, Japan,

‡

Laboratory of Forest Ecology, Department of Biodiversity Science, Tohoku University, Narugo, Miyagi, 989-6711, Japan and

§

Restoration Ecology, Regional Agricultural Science, Faculty of Agriculture, Miyazaki University, Kihanadai, Miyazakisi, Miyazaki, Japan

Summary

1

Sasa

spp., dwarf bamboo which dominate the undergrowth of temperate forests inJapan occur as clonal fragments in which ramets in light gaps to are connected to thosein shaded understoreys by long rhizomes. We test whether persistence under shadedconditions is supported by translocation of assimilates from illuminated ramets. A densepopulation of

Sasa palmata

growing at an open site, was exposed to two light conditions(homogeneous: open–open and heterogeneous: open–shaded) and two rhizome connec-tion treatments (intact and severed) in a full factorial design.

2

Ramet mass, and the mass of many parts of the clonal fragments, was much lower inthe shade than in the open, but this effect was less marked when the rhizome connectionwas intact than when it was severed. Clone parts in shade may therefore be supported bytranslocation from connected clone parts in the open, with such physiological integra-tion enhancing persistence where light supply is heterogeneous as in the gap–understoreycontinuum.

3

Above-ground biomass was reduced sooner than that below ground. Clonal fragmentsof

S. palmata

recover via dormant buds on rhizomes, whose longer persistence wouldtherefore enhance performance of the clonal fragment.

4

Specific leaf area (SLA) was greater in shade than in the open, irrespective of rhizomeconnection, suggesting that individual leaves show morphological plasticity independ-ently of only physiological integration.

Key-words

:

clonal plants, gap, light, rhizome connection,

Sasa palmata

Journal of Ecology

(2002)

90

, 78–85

Introduction

The physiological integration between connectedramets increases net growth and survivorship moresignificantly when resources are heterogeneous ratherthan homogeneous (Hartnett & Bazzaz 1983; Pitelka& Ashmun 1985; Alpert & Mooney 1986; Slade &Hutchings 1987; Stuefer

et al

. 1994). This is becauseramets under unfavourable microenvironments aresupported by translocation of resources from morefavourably placed ramets (e.g. Friedman & Alpert 1991;Stuefer

et al

. 1994; Wijesinghe & Handel 1994). Suchbehaviour has been observed for many perennial clonalherbs growing in various habitats, such as coastal sanddunes with contrasting levels of salinity (Alpert 1991;Evans 1992; Hester

et al

. 1994) and grassland, pastureand river banks subject to frequent fine-scale disturb-

ances (Hartnett & Bazzaz 1983; Stuefer

et al

. 1994;Lötscher & Hay 1997).

For understorey species, in most types of forest, lighttends to be the limiting resource (Grime 1979; Chazdon1988). In both temperate and tropical forests, however,the frequent creation of various sizes of gaps results inabundant light for understorey plants (Nakashizuka1984; Denslow 1987; Yamamoto 1989), albeit hetero-geneously distributed. Nutrient availability is also affectedby gap creation and is consequently patchy (Lechowicz& Bell 1991; Denslow

et al.

1998).Dwarf bamboos, which are rhizomatous, perennial,

semi-woody plants, predominate in the understorey ofdeciduous broad-leaved forests in the cool temperateregion of Japan (Usui 1961; Suzuki 1978), occurring asdense patches of culms (e.g. 100 m

2

per patch) (Tadokoro& Yajima 1990). In one species,

Sasa palmata

, rametsgrowing in the understorey near the edges of gapstend to be connected by long rhizomes to other partsof the same fragment in the gap (Saitoh & Seiwa,

Correspondence: T. Saitoh, (tel. + 81-229-847311; fax + 81-229-846490; e-mail [email protected]).

JEC_631.fm Page 78 Monday, January 28, 2002 6:04 PM

79

Physiological integration in dwarf bamboo

© 2002 British Ecological Society,

Journal of Ecology

,

90

, 78–85

unpublished data). It is thus hypothesized that rametsof

S. palmata

can persist under light-limited conditionsin the forest understorey by translocating assimilatesfrom actively photosynthesizing ramets in gaps. Manyclonal perennial plants grow in forest understoreysand nutrient (water) translocation between rametswithin a clonal fragment has been investigated in homo-geneously shaded environments (DeByle 1964; Ashmun

et al.

1982). Studies in heterogeneous environmentsare however, needed to document the importance ofphysiological integration of the rhizome for persistencein the understorey.

In heterogeneous environments, plants usually changetheir biomass allocation to each organ (roots, leaves,stems) in order to optimize growth by making all resourcesequally limited (Bloom

et al

. 1985). In contrast, severalstudies of clonal perennial herbs have reported thatwhen ramets under limited light conditions are con-nected to ramets with abundant light, they allocate arelatively low proportion of biomass to leaves (Friedman& Alpert 1991; Stuefer

et al

. 1994; Stuefer

et al

. 1996;Alpert 1999). This trait of local specialization of rametssuggests that within a clonal fragment each ramet maymaximize resource acquisition in its own microenviron-ment (Friedman & Alpert 1991; Stuefer

et al

. 1994).Such functional division of labour, however, wouldoccur more slowly in semi-woody

Sasa

species than inherbaceous perennials because of the slower tissue turn-over. Below-ground parts of

Sasa

, such as rhizomesand roots, aid persistence in the forest understorey byforaging for resources as well as functioning as storageorgans (Oshima 1961a), and if maintaining them iscritical, below-ground parts would be maintained for alonger period than culms and leaves under limited lightconditions.

This study evaluated the importance of physiologicalintegration and division of labour in

S. palmata

. Adense population was subjected to four experimentaltreatments under field conditions: two light conditions(homogeneous: open–open and heterogeneous: open–shaded) were combined with two rhizome connectionstatuses (intact and severed) in a full factorial design.We considered: (i) whether ramets of

S. palmata

undersevere shading conditions were relieved by the trans-location of assimilates from illuminated ramets, and(ii) when and how the functional division of labouroccurs in response to shade or rhizome severing.

Materials and methods

Thirty-five species of

Sasa

, belonging to five sections,are widely distributed in Japan and adjacent regions(Suzuki 1978) where they often predominate in grass-land and in the understorey of many types of forests inboth temperate and boreal zones. They form large areasof dense undergrowth, but distribution in mature forestunderstorey may be patchy.

Sasa palmata

(Marlic)

Nakai (Gramineae, Bambusoideae) is a monocarpicspecies belonging to the section

Sasa

, with a generationinterval of a hundred or more years (Makita 1998).Culms are produced from nodal buds on rhizomes inApril, and the previous year’s leaves are usually shed byAugust as new ones develop. This means that

S. palmata

retains a constant number of leaves throughout the year(Oshima 1961a). Photosynthetic assimilates, gained dur-ing the summer, are translocated into the rhizome fromOctober to December, and stored until used to producenew organs in the following spring (Oshima 1961c).

The experiment was carried out along the edge of abroad-leaved deciduous forest at the experimental farmof Tohoku University in Narugo town, Miyagi prefec-ture, northern Japan (38

°

45

′

N, 140

°

45

′

E; approximately190 m a.s.l.). The mean annual rainfall and temperature,and maximum and minimum temperatures meas-ured at the meteorological station nearest to the studyarea (Kawatabi, approximately 200 m distant), were1612 mm, 10.5

°

C, 34.0

°

C and –7.4

°

C in 1997, and2082 mm, 10.9

°

C, 31.5

°

C and –10.4

°

C in 1998, respect-ively (The Japan Meteorological Agency 1997–98).Average snow accumulation in 1997 was approxim-ately 23 cm (maximum, 64 cm), and usually lasts fromDecember to March. The soil is a well-drained MelanicAndosol (FAO–UNESCO 1998). The experimental quad-rat (4

×

150 m) was set up in the open area betweenfarmland and a secondary deciduous broad-leavedforest where the tall-tree layer was dominated by

Pinusdensiflora

and also included

Carpinus laxiflora

,

Prunusgrayana

,

P. verecunda

and

Cryptomeria japonica

(plant-ing).

S. palmata

was distributed densely and evenly acrossthe experimental quadrat, where 40 plots (1

×

1 m),separated by at least 1 m (mean 2.7 m), were estab-lished on 16 June 1997. The four treatments imposedare shown in Fig. 1. The shaded treatment increasedheterogeneity in light availability between connectedramets inside and outside the plot. Shade cages (2 m inheight) were covered by shade cloth, which transmittedapproximately 1% of ambient photosynthetically activephoton flux density (PPFD). To prevent the temper-ature inside the cage from rising, shade cloth stoppedapproximately 10 cm above the ground surface. Rhizomeconnections between the ramets inside and outside ofhalf the plots were severed and wooden partitions(1 m

×

30 cm) were inserted into the periphery of theplot to prevent rhizomes from growing into or out ofthe plot. Half the plots were shaded in a random-ized block design, with 10 blocks and one of the fourtreatment combinations (open/intact, open/severed,shaded/intact and shaded/severed) in each block.

Experimental treatments were applied on 23 June1997. To estimate the above- and below-ground biomassat the start of the experiment, all

S. palmata

materialwas harvested and excavated from five additionalplots (1 m

2

) which were located randomly within the


80

T. Saitoh, K. Seiwa & A. Nishiwaki


Journal of Ecology

,

90

, 78–85

experimental quadrat (Harvest 0). At this time, wefound that most of ramets within a plot were connectedwith several ramets growing outside, indicating thatsevered treatment would strongly interfere with thetranslocation of substances (e.g. nutrient, assimilate,water). Harvest 1 was conducted in five randomlychosen blocks in mid-October 1997, when

S. palmata

had completed its annual growth (see Oshima 1961a).The remaining five blocks were harvested on 3 August1998 (Harvest 2) when

S. palmata

had completed itsannual leaf production.

The effects of shade and rhizome severance on sea-sonal changes in the quantity of leaves were investigatedby counting the number of leaves present and those alreadyfallen for 10–20 randomly selected culms in each of the10 plots in each treatment, but sample size was half in thesecond year. The census was repeated at 1-month intervalsuntil the end of the experiment except during Decemberto March when the ground was covered by snow.

All plant materials harvested from each plot (1 m

2

)were divided into leaves, above-ground culms, below-ground culms, rhizomes and roots. The number ofabove-ground culms and the leaf area (cm

2

) were alsorecorded. Fresh weight of each component was meas-ured and a sample dried at 75

°

C to constant weightbefore re-weighing. The bulk density (g cm

–3

) of eachcomponent except for leaves was measured and cal-culated as dry weight per volume. The volume of eachcomponent was measured by a volume meter using airpressure (Daiki Corp.). Dead parts of each componentwere excluded from the measurement.

Above-ground biomass was defined as comprising leavesand above-ground culms, while below-ground biomass

included below-ground culms, rhizomes and roots.Specific leaf area (SLA, leaf area per leaf mass) and leafmass ratio (LMR, ratio of leaf mass to ramet mass)were calculated.

Statistical tests were performed with the

statis-tical program (SAS Institute 1995). Effects of lightconditions, rhizome connection and harvest (1997,1998) on biomass (above- and below-ground biomassand ramet mass), bulk density (above- and below-ground culms, rhizomes and roots), culm density, SLAand LMR were analysed by three-way

. TheTukey–Kramer multiple comparison test was used toidentify differences in the biomass, culm density, SLAand LMR among the four different treatments foreach harvest. Bulk density was also compared betweenharvests using one-way

for each treatment.Effects of treatments on the number of leaves per culmwere analysed by three-way

and the Tukey–Kramer test. Data were log- or root-transformed tomeet the assumptions of the

s (Bartlett test)where necessary.

Results

Total ramet mass (i.e. including several clonal fragments)was significantly affected by light, rhizome connectionand time of harvest (Table 1). At harvests 1 and 2, rametmass was lower in shaded than in open conditions(Fig. 2). By harvest 2, shading decreased ramet mass byabout 40% in the intact and by about 90% in the severedtreatment. The reduction in ramet mass due to shadingwas significantly greater in the severed than the intactrhizome treatment only at harvest 2, resulting in a sig-nificant interaction among the three factors (Table 1).

Culm density was significantly affected by lightconditions and rhizome connection, but not affectedby time of harvest (Table 1). At both harvests 1 and 2,culm density was lower when shaded than in the open(Fig. 2b). By harvest 2 in shaded conditions, culmdensity was decreased by just over 45% in the intactbut by about 95% in the severed treatment, resulting ina significant interaction between light and rhizomeconnection (Table 1).

Both above- and below-ground biomass were signific-antly affected by light condition, rhizome connectionand time of harvest (Table 1). In the first year (Harvest 1),shading reduced biomass of above-ground but not below-ground parts in both the rhizome connection treatments(Fig. 3). In the second year (Harvest 2), below-groundbiomass parts also responded to shade but only in thesevered rhizome treatment (90% reduction Fig. 3b).

In both the above- and below-ground culms and roots,bulk densities were reduced by shading but were notaffected by rhizome connection (Table 1); In below-ground

Fig. 1 Schematic representation of Sasa palmata plants grow-ing under four different treatments: (a) intact rhizome underopen conditions; (b) severed rhizome under open conditions;(c) intact rhizome under shaded conditions; and (d) severedrhizome under shaded conditions.


81



Journal of Ecology

,

90

, 78–85

culms and roots, bulk densities decreased with time(Table 1, Fig. 4).

The numbers of newly emerged leaves which remainedattached per shoot in each treatment are shown inFig. 5. In 1997, almost all the annual leaf productionwas completed by early August in all treatments.Subsequently, leaves in the shaded conditions turnedyellow and the number of leaves attached to culmsdecreased abruptly compared with a gradual decreaseover the growing season in open conditions. The decrease

in the number of leaves caused by shading was statis-tically significant throughout the growing season(two-way

;

F

= 28.9,

P

< 0.0001), whereas littledifference was observed between intact and severedtreatments across light conditions (

F

= 0.3,

P

= 0.5815).By the following spring, all leaves had been shed inthe shaded conditions, whereas 2–4 leaves per shootremained in the open.

In July 1998, two-way

showed that thenumber of leaves was significantly reduced by shading(

F

= 27.4,

P

< 0.0001) but not by rhizome connection(

F

= 2.2,

P

= 0.1382). Leaves that unfolded in thesecond year under shaded conditions remained greenuntil the end of the experiment.

Fig. 2 Effect of light heterogeneity and rhizome connectionon changes in (a) ramet mass and (b) culm density (m–2) of Sasapalmata. s, intact /open; h, severed/open; d, intact /shaded;j, severed/shaded. Values within columns that do not share acommon letter are significantly different from each other atP < 0.05; Tukey–Kramer multiple comparison test after .

Fig. 3 Effect of light heterogeneity and rhizome connectionon changes in dry mass of (a) above-ground and (b) below-ground parts of Sasa palmata. s, intact /open; h, severed/open;d, intact /shaded; j, severed/shaded. Statistics as in Fig. 2.

Table 1

Results of significance tests for biomass, culm density and bulk density of

Sasa palmata

. The effects of light conditions (homogeneous: open-open,heterogeneous: open-shaded), rhizome connection status (intact, severed), harvest (1997, 1998) and their interaction tested in three-way ANOVAs. Valuesare

F

ratio and its significance for effect

Treatment

Biomass

Culm density

Bulk density

Ramet mass

Above-ground parts

Below-ground parts

Above-ground culms

Below-ground culms Rhizomes Roots

Light 153.37*** 366.11*** 51.60*** 197.95*** 71.08*** 7.48* 2.50 5.11*Connection 27.44*** 31.74*** 22.66*** 8.92** 2.21 0.94 3.63 1.28Harvest 20.31*** 12.12** 63.61*** 0.05 1.58 10.71** 2.50 6.83*Light

×

connection 26.85*** 21.54*** 30.07*** 10.67** 0.77 2.21 0.05 0.02Connection

×

harvest 5.29* 0.00 7.08* 0.04 0.25 0.24 2.11 0.12Light

×

harvest 18.37*** 0.40 18.35*** 0.42 0.04 0.44 0.23 0.60Light

×

connection

×

harvest 22.59*** 3.13 28.00*** 0.53 0.64 0.41 3.39 8.38**

*

P

< 0.05, **

P

< 0.01, ***

P

< 0.001.


82

T. Saitoh, K. Seiwa & A. Nishiwaki


Journal of Ecology

,

90

, 78–85

Both LMR and SLA were significantly affected byrhizome connection and light condition, and LMR wasalso affected by the time of harvest (Table 2). LMR andSLA were lower in the shade than in the open in bothrhizome connection treatments at all harvests (Fig. 6).

Discussion

S A S A

P A L M A T A

Both ramet mass and culm density of

S. palmata

weregreatly reduced under shaded compared with openconditions, but the extent of the reduction was lowerwhen the rhizome connection was intact than when

it was severed. This suggests that growth reductionin shaded ramets was compensated for by their con-nection to clone parts which remained in the open.Persistence of the dwarf bamboo,

S. palmata

, in light-limited environments is therefore strongly enhancedby physiological integration within a clonal fragmentwhich enables translocation of photosynthates fromramets in environments where light is abundant tothose under shaded conditions. A substantial num-ber of culms of

S. palmata

persist in the understoreyeven though their population density and biomass aresmaller than in gaps or at the forest edge (Kawahara& Tadaki 1978; Kawahara 1984; Saitoh

et al

. 2000).Saitoh

et al

. (2000) also showed that, although bothrelative PPFD and the coverage of

S. palmata

weremaximum in the gaps and decreased with increasingdistance from the gap edge into the understorey, the

Fig. 4 Bulk density of Sasa palmata: (a) above-ground culms; (b) below-ground culms; (c) rhizome; and (d) roots. Solid and cross-hatched bars show 1997 and 1998 values, respectively.

Fig. 5 Leaf-survival patterns of Sasa palmata. s, intact /open; h, severed/open; d, intact /shaded; j, severed/shaded. Broken lineis the cumulative number of leaves which emerged on a shoot and solid line is the number of newly emerged leaves actuallyattached to a shoot for leaves in (a) 1997 and (b) 1998.


83



Journal of Ecology

,

90

, 78–85

rate of decrease in

Sasa

coverage was significantly lowerthan that in relative PPFD. We also observed thatsingle clonal fragments of

S. palmata

frequently grewunder heterogeneous light conditions, i.e. extending fromgap to forest understorey, with long rhizomes (length5–10 m) connecting the various above-ground culms(Saitoh & Seiwa unpublished data); however we wereonly concerned with physiological integration at dis-tances up to 2 m.

Sasa palmata

, a macrophytic clonalplant, may therefore depend on physiological integra-tion to persist in the heterogeneous light environmentscharacteristic of the understorey.

In temperate deciduous forests, light conditionsare favourable for understorey plants, including dwarfbamboos, in early spring and autumn, i.e. before expan-sion and after shedding of canopy leaves (e.g. Seiwa1998). Lei & Koike (1998) also showed that evergreenleaves enable

Sasa

species to maintain high rates ofphotosynthesis during the period between the meltingof snow and the emergence of canopy leaves andbetween leaf fall and snow fall. In Japanese beechforests, however, light in the understorey is poorthroughout the growing season, because

Fagus crenata

completes its annual leaf production prior to snowthaw and does not shed its leaves until just before snowfalls in autumn. Nevertheless

S. palmata

persists in theunderstorey of

F. crenata

forests, further suggesting thatphotosynthate is translocated from high light environ-ments such as gaps to the shaded understorey.

S A S A

P A L M A T A

Both above- and below-ground biomass were muchmore greatly decreased by shading when rhizome con-nections were severed than when they were intact, whereabove-ground biomass, particularly that of leaves, fellearlier in the season. Relative PPFD in the shaded con-ditions was approximately 1% of that of the open, andwas below the light compensation point of

S. senanensis

(Lei & Koike 1998), so that if S. palmata is similar, thecost of respiration in its leaves should exceed the gainby photosynthesis. It is also well known that, for woodyplants, respiration is usually slower in below- thanabove-ground organs (Kramer & Kozlowski 1979) andunder such severely shaded conditions, clonal fragmentsof S. palmata would therefore benefit if the above-ground biomass, particularly of leaves, is reduced com-pared with that below ground.

LMR was found to be greatly decreased under theshaded conditions irrespective of ramet connection,probably as a consequence of the functional divisionof labour. Several authors have reported that whenclonal fragments are exposed to spatially heterogeneousenvironments there was a relatively low proportion ofbiomass allocated to leaves in ramets in limited lightconditions (Friedman & Alpert 1991; Stuefer et al. 1994;Stuefer et al. 1996; Alpert 1999). As well as allowingeach ramet to maximize resource acquisition from itsown microenvironment (Friedman & Alpert 1991; Stueferet al. 1994) this may also minimize the resource loss.Although we do not know how the open ramets thatare connected the shaded ramets respond, tracingthe movement of 15N within fragments showed moretranslocation under heterogeneous light conditions(from ramets of shaded to those of open conditions)than between ramets in open conditions (Saitoh andSeiwa unpublished data). Carbon may therefore betranslocated from the open ramets to the shadedramets and the reverse may be true for the nitrogenwithin S. palmata fragments, suggesting a functionaldivision of labour.

Table 2 Results of significance tests for LMR and SLA ofSasa palmata. The effects of light conditions (homogeneous:open-open, heterogeneous: open-shaded), rhizome connectionstatus (intact, severed), harvest (1997, 1998) and their inter-action tested in three-way ANOVAs. Values are F ratio and itssignificance for effect

Treatment LMR SLA

Light 419.49*** 74.82***Connection 9.12** 4.73*Harvest 40.98*** 0.27Light × connection 2.78 3.90Connection × harvest 8.59** 9.71**Light × harvest 32.63*** 9.91**Light × connection × harvest 6.27* 7.37*

*P < 0.05, **P < 0.01, ***P < 0.001

Fig. 6 (a) Leaf mass ratio (LMR) and (b) specific leaf area(SLA) of Sasa palmata. s, intact /open; h, severed/open;d, intact /shaded; j, severed/shaded.


84T. Saitoh, K. Seiwa & A. Nishiwaki

© 2002 British Ecological Society, Journal of Ecology, 90, 78–85

In S. palmata, SLA was greater in the shaded condi-tions than in the open, irrespective of rhizome connec-tion. This suggests adaptive plasticity, with individualleaves changing their morphology to maximize lightacquisition at a particular microsite even though theclonal fragment is exposed to a heterogeneous lightenvironment (Alpert 1999). In S. palmata, physiolog-ical integration among intraclonal ramets would notmodify the plastic response of individual leaves.

Under shaded conditions, below-ground biomassdecreased more slowly than that above-ground, evenwhen rhizomes were severed. Longer preservation ofbelow-ground parts would enhance recovery when lightconditions improve because above-ground productiondepends largely on the number of dormant buds onrhizomes and the amount of stored reserves (Oshima1961b,c). The longer persistence of rhizomes underlimited light conditions may be due to effective hori-zontal carbon translocation. Furthermore, bulk densityof rhizomes was not reduced even in the second year,although the value for of above-ground culms fellrapidly in the first year (Fig. 4). These rhizome traits mayalso be important for the persistence of Sasa palmatain the understorey.

Shaded clone parts were substantially supported bythe translocation of photosynthates from connectedramets in the open up to 2 m away, although it needsfurther study to establish whether translocation occursover longer distances in the field. We conclude thatphysiological integration within a clonal fragmentand longer preservation of below-ground parts wouldenhance the persistence of this dwarf bamboo in aheterogeneous resource environment such as the gap–understorey continuum in temperate forests.

Acknowledgements

The author is very grateful to Shigetoshi Akasaka andHiroshi Kanno for assistance with fieldwork. I thankP. Alpert and two anonymous referees for their valuablecomments and suggestions on the manuscript.

References

Alpert, P. (1991) Nitrogen sharing among ramets increasesclonal growth in Fragaria chiloensis. Ecology, 72, 69–80.

Alpert, P. (1999) Effects of clonal integration on plant plasti-city in Fragaria chiloensis. Plant Ecology, 141, 99–106.

Alpert, P. & Mooney, H.A. (1986) Resource sharing amongramets in the clonal herb, Fragaria chiloensis. Oecologia, 70,227–233.

Ashmun, J.W., Thomas, R.J. & Pitelka, L.F. (1982) Trans-location of photoassimilates between sister ramets intwo rhizomatous forest herbs. Annals of Botany, 49, 403–415.

Bloom, A.J., Chapin, III, F.S. & Mooney, H.A. (1985) Resourcelimitation in plants – an economic analogy. Annual Reviewof Ecology and Systematics, 16, 363–392.

Chazdon, R.L. (1988) Sunflecks and their importance toforest understorey plants. Advances in Ecological Research,18, 1–63.

DeByle, N.V. (1964) Detection of functional intraclonal aspen

root connections by tracers and excavation. Forest Science,10, 386–396.

Denslow, J.S. (1987) Tropical rainforest gaps and tree speciesdiversity. Annual Review of Ecology and Systematics, 18,431–451.

Denslow, J.S., Ellison, A.M. & Sanford, R.E. (1998) Treefallgap size effects on above- and below-ground processes in atropical wet forest. Journal of Ecology, 86, 597–609.

Evans, J.P. (1992) The effects of local resource availabilityand clonal integration on ramet functional morphology inHydrocotyle bonariensis. Oecologia, 89, 265–276.

FAO-UNESCO (1998) World reference base for soil resources.World soil resource reports, vol. 88. UNESCO, Rome.

Friedman, D. & Alpert, P. (1991) Reciprocal transport betweenramets increases growth of Fragaria chiloensis when lightand nitrogen occur in separate patches but only if patchesare rich. Oecologia, 86, 76–80.

Grime, J.P. (1979) Plant Strategies and Vegetation Processes.Wiley, Chichester.

Hartnett, D.C. & Bazzaz, F.A. (1983) Physiological integrationamong intraclonal ramets in Solidago canadensis. Ecology,64, 779–788.

Hester, M.W., McKee, K.L., Burdick, D.M., Koch, M.S.,Flynn, K.M. & Patterson, S. (1994) Clonal integration inSpartina patens across a nitrogen and salinity gradient.Canadian Journal of Botany, 72, 767–770.

Kawahara, T. (1984) Geographical distribution and biomass inSasa communities. Bamboo Journal, 2, 10–15. (In Japanese.)

Kawahara, T. & Tadaki, Y. (1978) Studies on Sasa com-munities (III) Relationship between light intensity and biomassof Sasa nipponica. Journal of Japanese Forestry Society, 60,244–248. (In Japanese.)

Kramer, P.J. & Kozlowski, T.T. (1979) Physiology of WoodyPlants. Academic Press, San Diego.

Lechowicz, M.J. & Bell, G. (1991) The ecology and genetics offitness in forest plants. II. Microspatial heterogeneity of theedaphic environment. Journal of Ecology, 79, 687–696.

Lei, T.T. & Koike, T. (1998) Functional leaf phenotypes forshaded and open environments of a dominant dwarf bamboo(Sasa senanensis) in northern Japan. International Journalof Plant Sciences, 159, 812–820.

Lötscher, M. & Hay, M.J.M. (1997) Genotypic differences inphysiological integration, morphological plasticity and util-ization of phosphorus induced by variation in phosphatesupply in Trifolium repens. Journal of Ecology, 85, 341–350.

Makita, A. (1998) Population dynamics in the regenerationprocess of monocarpic dwarf bamboos, Sasa species. Popu-lation Biology of Grasses (ed. G.P. Cheplick), pp. 313–332.Cambridge University Press, New York.

Nakashizuka, T. (1984) Regeneration process of climax beech(Fagus crenata Blume) forest IV. Gap formation. JapaneseJournal of Ecology, 34, 75–85.

Oshima, Y. (1961a) Ecological studies of Sasa communities II.Seasonal variations of productive structure and annual netproduction in Sasa communities. Botanical Magazine ofTokyo, 74, 280–290.

Oshima, Y. (1961b) Ecological studies of Sasa communities III.Dry matter production and distribution of products amongvarious organs in Sasa kurilensis community. BotanicalMagazine of Tokyo, 74, 473–479.

Oshima, Y. (1961c) Ecological studies of Sasa communities IV.Photosynthesis and respiration of Sasa kurilensis. BotanicalMagazine of Tokyo, 74, 349–356.

Pitelka, L.F. & Ashmun, J.W. (1985) Physiology and integra-tion of ramets in clonal plants. Population Biology andEvolution of Clonal Organisms (eds J.B.C. Jackson, L.W. Buss& R.E. Cook), pp. 399–437. Yale University Press, NewHaven.

Saitoh, T., Seiwa, K., Nishiwaki, A., Kanno, H. & Akasaka, S.(2000) Spatial distribution patterns of Sasa palmata in rela-tion to light conditions across gap-understory continuum


85Physiological integration in dwarf bamboo

© 2002 British Ecological Society, Journal of Ecology, 90, 78–85

in a beech (Fagus crenata) forest. Journal of Japanese ForestrySociety, 82, 342–348. (In Japanese.)

Seiwa, K. (1998) Advantages of early germination for growthand survival of seedlings of Acer mono under different over-storey phenologies in deciduous broad-leaved forests. Journalof Ecology, 86, 219–228.

Slade, A.J. & Hutchings, M.J. (1987) An analysis of the costsand benefits of physiological integration between ramets inthe clonal perennial herb Glechoma hederacea. Oecologia,73, 425–431.

Stuefer, J.F., de Kroon, H. & During, H.J. (1996) Exploitationof environmental heterogeneity by spatial division of labourin a clonal plant. Functional Ecology, 10, 328–334.

Stuefer, J.F., During, H.J. & de Kroon, H. (1994) High benefitsof clonal integration in two stoloniferous species, in responseto heterogeneous light environments. Journal of Ecology,82, 511–518.

SAS (1995) JMP User’s Guide. Version 3 of JMP. SAS Insti-tute Inc., Cory, NC.

Suzuki, S. (1978) Index to Japanese Bambusaceae, pp. 1–384.Gakken, Tokyo. (In Japanese.)

Tadokoro, K. & Yajima, T. (1990) Growth and activity ofrhizomes of Sasa nipponica. Journal of Japanese ForestrySociety, 72, 345–348. (In Japanese.)

Usui, H. (1961) Phytosociological revision of the dominantspecies of Sasa-type undergrowth – Silvicultural applica-tion of the researches on Japanese forest vegetation. SpecialBulletin of the College of Agriculture, Utsunomiya University,11, 1–35. (In Japanese.)

Wijesinghe, D.K. & Handel, S.T. (1994) Advantages of clonalgrowth in heterogeneous habitats: an experiment with Potentillasimplex. Journal of Ecology, 82, 495–502.

Yamamoto, S. (1989) Gap dynamics in climax Fagus crenataforest. Botanical Magazine of Tokyo, 102, 93–114.

Received 13 February 2001 revision accepted 3 July 2001


Importance of physiological integration of dwarf bamboo to persistence in forest understorey: a field experiment

Documents