Responses to Light and Water Availability of Four Invasive Melastomataceae in the Hawaiian Islands Author(s): Zdravko Baruch, RobertR.

Pattison, Guillermo Goldstein Reviewed work(s): Source: International Journal of Plant Sciences, Vol. 161, No. 1 (January 2000), pp. 107-118 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/10.1086/314233 . Accessed: 03/04/2012 12:00
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to International Journal of Plant Sciences.

http://www.jstor.org

Int. J. Plant Sci. 161(1):107118. 2000. 2000 by The University of Chicago. All rights reserved. 1058-5893/2000/16101-0011$03.00

RESPONSES TO LIGHT AND WATER AVAILABILITY OF FOUR INVASIVE MELASTOMATACEAE IN THE HAWAIIAN ISLANDS
Zdravko Baruch,1,* Robert R. Pattison,2, and Guillermo Goldstein
*Deptartamento de Estudios Ambientales, Universidad Simo n Bol var, Apartado 89000, Caracas 1080, Venezuela; and Department of Botany, University of Hawaii, 3190 Maile Way, Honolulu, Hawaii 96822, U.S.A.

Plant invasion by Neotropical Melastomataceae is prominent in Hawaii. To understand life history traits of four successful invasive Melastomataceae, two shade-intolerant herbs (Arthrostema ciliatum and Tibouchina herbacea) and two shade-tolerant woody species (Clidemia hirta, a shrub, and Miconia calvescens, a tree) were subjected to three light levels and two watering regimes in a greenhouse. Plant height, leaf number and area, biomass allocation, relative growth rate (RGR), carbon assimilation (A), leaf nutrient content, leaf construction costs (CC), specic leaf mass (SLM), and leaf spectral properties were determined at the end of the experimental period. Plant size, total biomass, RGR, A, CC, and SLM decreased, whereas leaf light transmittance and leaf N increased under low light in all species. The effects of water stress were weaker than light-stress effects. Relative growth rate of herbs grown in sun and partial shade (0.046 and 0.033 g g1 d1, respectively) was higher than in the woody species (0.027 and 0.020 g g1 d1). Woody species allocated more biomass to leaf production than herbs, which allocated more biomass to stem production. Shade increased allocation of biomass to leaves, and water stress increased the root-shoot ratio in all species. Partial shade increased leaf area ratios more in the herbs (140%) than in woody species (68%). Miconia calvescens and C. hirta had higher leaf absorbance (92%) than both herbs (79%). Maximum A under all light treatments was similar in all species, and there was substantial acclimation to the different light levels. Leaf construction cost was higher in the apparently long-lived leaves of the woody species. Relative growth rate, carbon allocation, and SLM showed larger changes to light and water stress than A and related photosynthetic parameters. All species showed responses qualitatively similar to those of other tropical species including the high acclimation potential to light, but the herbs exhibited the largest quantitative responses. When compared with a large group of native species, the four melastomes appear to be better suited to capture and use light, which is consistent with their rapid spread in mesic and disturbed Hawaiian environments. Keywords: carbon assimilation, construction costs, leaf nutrients, melastomataceae, nitrogen use efciency, plant growth, plant invasions, specic leaf mass, ecophysiology.

Introduction
Several species of the Neotropical family Melastomataceae are among the most aggressive invaders of the Hawaiian and other Pacic islands that are particularly susceptible to plant invasions (Vitousek et al. 1987; Loope and Mueller-Dombois 1989; Smith 1992; Meyer and Florence 1996; Medeiros et al. 1997). Introduced accidentally or as ornamentals, these species have displaced native plants throughout the islands following disturbances such as deforestation, re, and overgrazing. In addition to reproductive traits such as small seed size and high dispersion capability (Rejmanek and Richardson 1996), high growth rate, large carbon assimilation, and acclimation capacity (Baruch et al. 1985; Bazzaz 1986; Williams and Black 1994; Pattison et al. 1998; Baruch and Goldstein 1999) also contribute to the success of an invader plant. Apart from al1 Author for correspondence; fax 58-2-9063064; e-mail zbaruc@ usb.ve. 2 Current address: Department of Botany, Washington State University, Pullman, Washington 99164-4238, U.S.A.

Manuscript received April 1999; revised manuscript received September 1999.

tering species composition and the structure and diversity of island communities, these invasions may also alter ecosystem function (Vitousek 1994). Light and water are the most limiting resources for plant life, as they are essential for the acquisition of carbon and mineral nutrients. Plant species usually differ in their adaptations to the low availability of these two limiting resources. For example, shade-tolerant species benet from adaptations that maximize light capture at low light levels and minimize respiratory costs (Pearcy and Sims 1994). Low light availability generally increases the proportion of biomass allocated to leaves in order to compensate for reduced carbon xation (Schulze and Chapin 1987). Shade-intolerant plants face water, nutrient, and photoinhibition stresses that can adversely interact with the maintenance of high photosynthetic rates (Pearcy and Sims 1994). In contrast, water stress increases the root-shoot ratio and would tend to increase root surface area for nutrient uptake (Chapin 1991). It is also possible that conicting situations occur where both light and water shortages take place simultaneously. Growth and biomass allocation responses to resource shortage are specic for each life-form and act as balancing mechanisms to maintain equilibrium in the
107

108

INTERNATIONAL JOURNAL OF PLANT SCIENCES pacity to invade open habitats and persist under early stages of forest succession, we consider it as intermediate in terms of light requirements. This shrub shows leaf senescence, cessation of growth, and death of shoot tips during rainless periods (Smith 1992), indicating that it is also a drought-sensitive species. Arthrostema ciliatum is an annual or biannual herb, and Tibouchina herbacea is a perennial herb. Both are up to 1.5 m tall and found in open, disturbed, or ruderal sites (Wester 1992; Medeiros et al. 1997). On the basis of their habitat, both herbs are probably shade-intolerant but relatively drought-tolerant species. For convenience, these species will hereafter be cited only by genus.

internal budget of carbon and nutrients (Orians and Solbrig 1977; Schulze 1981; Schulze and Chapin 1987). Growth, carbon assimilation, and allocation and nutrient dynamics responses of tropical plants to light and to drought stress have been studied in some detail in relation to habitat (open or forest gap vs. understory) and to succesional dynamics. Some ndings support that shade results in an increase of plant height and specic leaf mass and decrease of growth rate (Bjorkman 1981; Chazdon and Field 1987; Fetcher et al. 1987; Denslow et al. 1990; Thompson et al. 1992; Ellison et al. 1993; Kitajima 1996), leaf number and leaf construction costs decrease under shaded conditions (Williams et al. 1989; Denslow et al. 1990; Poorter and Villar 1997), and assimilation rate on a leaf area basis decreases and large acclimation capacity characterizes plants subjected to light limitation (Denslow et al. 1990; Newell et al. 1993; Ellsworth and Reich 1996; StraussDebenedetti and Bazzaz 1996). A compilation of these studies has been recently published (Mulkey et al. 1996). Although several melastomes have been included, specially by Ellison et al. (1993), none of the species studied here has been discussed. From the 15 invading melastomes described for Hawaii (Almeda 1990), we selected four species from different habitats and with different growth forms as the most aggressive invaders: the forest tree Miconia calvescens DC, the shrub Clidemia hirta L. D. Don, and the herbs from open sites, Arthrostema ciliatum Pav. Ex D. Don. and Tibouchina herbacea (DC) Cogn. Here we attempt to answer the question How do light and water availability, representative of the contrasting environments where these invading species colonize and persist, affect growth, carbon assimilation, and biomass allocation? On the basis of their distribution in the Hawaiian Islands, we predict that the herbs from open sites will be most sensitive to shade, whereas the forest understory seedlings of the tree will be better able to acclimate to different light regimes. The shrub, which grows in both open and relatively shaded habitats, should exhibit intermediate responses to light. In addition, we will compare our results with those of Pattison et al. (1998) and Baruch and Goldstein (1999) for native species that may provide new insights to understand the success of the invasive melastomes in Hawaii.

Seed Source and Experimental Treatments

Miconia and Tibouchina seedlings were collected on the island of Maui, the former from a lowland secondary rain forest near Hana and the latter from an open site along the road from Kahului to Hana. Clidemia seedlings were collected at the Lyon Arboretum in Honolulu (Oahu Island), and Arthrostema seedlings were grown from seeds collected from many plants in the same area. Care was taken to collect seedlings with intact roots, and they were promptly transplanted to small pots. All species were grown for 1 mo under one layer of shade cloth (25% of full sunlight) in the greenhouse. Before the treatments, all plants were transplanted into 4-L pots on a 2 : 1 soil (Supersoil Potting Mix, San Francisco) to perlite mix, except those of Arthrostema, which were planted in 2.5-L pots because of their initial smaller size. All plants were fertilized with a N, P, and K controlled-release mixture (14 : 14 : 14) at a dose of 5.5 g (4 L) and 2.2 g (2.5 L). After 1 mo, ve individuals of each species were randomly selected and harvested to determine plant height and leaf number and area and the dry weight of leaves, stems, and roots in order to obtain the initial data for relative growth rate calculation. Leaf area was measured with an LI-3000A leaf area meter (LiCor, Lincoln, Nebr.). Twenty individuals of each species were randomly assigned to each of the following treatments: high light (full sunlight in the greenhouse; HL), medium light (25% of full sunlight; ML), and low light (4%5% of full sunlight; LL). Light levels were obtained by covering the greenhouse benches with none, one, or two layers of neutral density lters (shade cloths). Each light treatment was subdivided into two water treatments, each with 10 individuals per species. Watered plants received water every 23 d. Water was withheld for the drought-stress plants until the rst symptoms of wilting were evident. Then they were watered to eld capacity. All stressed plants had at least two drying cycles. Every 2 wk, leaf numbers were counted, plant height was measured up to the top leaf, and phenology were determined in all plants. At midexperiment, the plants were randomly rotated under all treatments. The experimental period was 76 d for Arthrostema, 79 d for Tibouchina, 81 d for Clidemia, and 85 d for Miconia in order to allow for time for gas exchange measurements and to process the samples.

Material and Methods Species and Habitat

Miconia calvescens is a tree (up to 15 m tall) that was introduced as an ornamental in the Pacic islands. In Hawaii, it became one of the most serious threats in sites of up to 500 m elevation with more than 1800 mm of rainfall (Medeiros et al. 1997; Meyer and Malet 1997). This tree has large deep green leaves with a purple underside and produces large numbers of seedlings in the understory of invaded rain forests (Meyer and Malet 1997). Its growth is enhanced when light becomes more available after the removal of parent trees from the canopy (Meyer and Malet 1997), indicating that M. calvescens seedlings may be shade tolerant and relatively drought sensitive. Clidemia hirta is a small to medium-height (0.53 m tall) shrub that in Hawaii occurs up to 1250 m in disturbed mesic sites and behaves as a typical secondary succession pioneer (Smith 1992; Medeiros et al. 1997). Because of its ca-

Environmental and Physiological Measurements

Air temperature and relative humidity in the greenhouse were monitored with Vaisala sensors (Helsinki). Leaf temper-

BARUCH ET AL.LIFE HISTORY OF INVASIVE MELASTOMES ature was measured with thermocouples attached to their abaxial surface. Photosynthetic photon ux density (PPFD) was measured with Li-Cor quantum sensors. All data were recorded and averaged at 10-min intervals with a Campbell CR10 data logger (Campbell Scientic, Logan, Utah). During the last third of the experiment, carbon assimilation (A) and stomatal conductance (gs) were measured with a Li-Cor 6200 gas exchange system on young fully expanded leaves of four randomly selected individuals per species and per treatment during the mornings of sunny days under the different growing light conditions. Before the nal harvest, at the end of the last drying cycle, predawn leaf water potential (W) was measured on leaf segments of three randomly selected individuals per species and per treatment with a thermocouple psychrometer (Model SC-10, Decagon Devices, Pullman, Wash.). The percentage radiation reected (r) and transmitted (t) by young fully expanded leaves of three randomly selected individuals per species of the watered plants was measured on their adaxial surface with a Li-Cor 1800-12S integrating sphere connected to a quantum sensor. The percentage of radiation absorbed by the leaf (a) was calculated as a = 100 (r t). Carbon assimilation (A) and gs were measured on the youngest fully expanded leaf of three watered individuals per treatment per species. After an initial dark respiration measurement, PPFD was increased stepwise to 1500 mmol m2 s1. Under each PPFD level, the leaf was allowed to equilibrate for 30 min before measurement. Light was generated with a QBEAM 2001 lamp (Quantum Devices, Barneweld, Wis.). Quantum yield, light saturation, and compensation points were calculated from the relationship between A and PPFD, corrected by the leaf absorbance of each species. On completion of light response curves, leaf disks were taken for specic leaf mass (SLM) determination.

109

the main factors. When applicable, a one-way ANOVA was applied for differences among species. A Tukey test was used to analyze the differences between the means of the treatments (SYSTAT 1997). The relationships between RGR and other variables were analyzed by multiple regression.

Results Growth Conditions, Mortality, and Phenology

Mean maximum and minimum air temperature and relative humidity for the treatments were 34.4 and 19.8C and 85.1% and 39.1%, respectively. Mean daylight maximum leaf temperatures were 32.8 and 29.8C for HL and LL conditions, respectively. Mean maximum SE PPFDs were 783 163, 208 48, and 34 6 mmol m2 s1 in the HL, ML, and LL treatments, respectively. The lowest predawn Ws attained during water stress ranged from 0.67 MPa in Tibouchina to 0.82 MPa in Arthrostema. The predawn Ws of Clidemia and Miconia minima were intermediate to those of the two herbs. The watered control plants ranged between 0.25 and 0.29 MPa for all species. All species except Clidemia exhibited some mortality under the LL treatment. The highest mortality was 60% and 70% for Tibouchina under the watered and waterstressed treatments, respectively. Water-stressed Miconia and Arthrostema had 20% and 10% mortality, respectively. Arthrostema owered after 60 d from the beginning of the experiment but only under the HL and ML watered treatments. The biomass of reproductive parts was !1% of the total plant biomass.

Plant Size, Leaf Number, Leaf Area, and Leaf Specic Mass
Except under LL, watered plants were the tallest in all species (table 1). Water stress affected the height of Arthrostema the most, reducing it to 25% of the watered controls. Except in watered Arthrostema and Miconia, ML plants were the tallest, and LL plants were the smallest. Under both watering treatments, plant height of the herbs Arthrostema and Tibouchina showed the largest responses to light (table 1). Watering increased leaf area and number in all species, and LL induced the production of fewer leaves with lower total leaf area per plant. The largest leaf area under drought stress occurred under ML (table 1). The largest leaf area in watered Arthrostema and Clidemia was also observed under ML. The effects of water and light regimes on leaf number and total leaf area were much more pronounced in Arthrostema and Tibouchina, whose leaves were three to 11 times smaller than those of Clidemia and Miconia (table 1). Specic leaf mass was lowest in Arthrostema, which was unaffected by water stress. In the other species, drought induced the production of thicker leaves (table 1). Under all watering and light conditions, the woody Clidemia and Miconia had the highest SLM (table 1). All species produced leaves with signicantly lower SLM under LL (table 1). Leaf senescence, although not measured specically, was more pronounced in the herbs than in the woody species. Also, leaf production, measured as the increase of leaf number between censuses, was higher in the herbs (data not shown).

Growth Analysis, Nutrient and Energy Concentration, and Statistics

At harvest, plant height, leaf number and area, and biomass of all plants were measured as indicated above. Relative growth rate was calculated per plant and per leaf number and area, plant height, and total biomass were calculated as RGR = ln Wf ln Wi /tf ti, where W is both the initial and nal values of the variable measured, and the denominator is the time elapsed between the initial and nal measurements. Leaf area ratio (LAR) was calculated as the relationship between leaf area and total plant mass. Leaf N, P, and K concentrations were determined by micro-Kjeldahl (Nelson and Sommers 1972) and inductive plasma spectrometry methods (Isaac and Johnson 1983). An aliquot of dried leaves was employed for the determination of energy concentration of the tissues on an ash-free dry weight basis with a bomb calorimeter (Parr Instruments, Moline, Ill.). Leaf construction costs (CC) were calculated from N (considering ammonium as the source of N), energy, and ash concentrations (Williams et al. 1987). Instantaneous photosynthetic nitrogen use efciency (PNUE) was calculated as A/[N], and the rate of photosynthesis by unit leaf CC was calculated as A/CC. In both cases, the A corresponds to that measured under the maximum PPFD. Statistical analysis was performed separately for each species with a two-way ANOVA with light and water treatments as

110

INTERNATIONAL JOURNAL OF PLANT SCIENCES Table 1

Mean and Standard Error (in Parentheses) of Architecture-Related Traits under All Treatments Treatments Watered Traits and species High light Medium light Low light High light Water stressed Medium light Low light Water ANOVA Light Interaction

Plant height (cm): Herbs: Arthrostema ....... Tibouchina ......... Woody: Clidemia ............ Miconia ............ Leaf area (cm2): Herbs: Arthrostema ....... Tibouchina ......... Woody: Clidemia ............ Miconia ............ Leaf number: Herbs: Arthrostema ....... Tibouchina ......... Woody: Clidemia ............ Miconia ............ Specic leaf mass (g m2): Herbs: Arthrostema ....... Tibouchina ......... Woody: Clidemia ............ Miconia ............ Note.

29.6 (3.6) 44.8 (5.6) 30.4 (1.4) 12.7 (1.9)

53.3 (4.2) 77.1 (2.0) 40.1 (2.3) 13.7 (1.2)

26.2 (1.6) 32.0 (5.5) 31.1 (2.0) 11.7 (1.5)

115.6 (4.3) 81.6 (3.8) 37.3 (1.7) 18.7 (1.3)

89.3 (2.1) 87.3 (1.3) 49.1 (2.0) 16.2 (0.7)

23.0 (1.5) 37.7 (3.3) 30.3 (1.8) 11.1 (0.5)

* *

* *

* *

* *

* *

* *

111.4 (17.0) 455.2 (44.6) 724.0 (53.5) 382.1 (40.7)

411.2 (35.4) 525.0 (25.6) 765.6 (70.2) 562.7 (75.7)

184.6 (25.7) 45.7 (19.9) 411.9 (33.4) 153.7 (37.0)

986.9 (60.2) 1810.7 (126.6) 1139.0 (81.8) 1019.8 (160.6)

1144.0 (71.0) 1339.7 (214.8) 1287.8 (112.8) 822.6 (105.7)

127.1 (28.7) 69.7 (23.8) 404.1 (29.0) 219.5 (37.9)

* *

* *

* *

* *

* *

* *

20.8 (2.1) 62.2 (5.9) 35.7 (3.4) 7.2 (0.2)

37.6 (1.9) 52.5 (5.3) 23.4 (1.8) 7.2 (0.3)

20.5 (2.2) 11.6 (1.3) 14.6 (0.8) 5.2 (0.2)

122.2 (10.6) 204.4 (16.1) 54.2 (4.0) 10.5 (1.2)

90.8 (6.8) 98.1 (16.0) 32.5 (1.7) 7.1 (0.4)

17.4 (2.2) 11.7 (1.9) 14.6 (0.9) 5.8 (0.2)

* *

* *

* *

* *

* *

* *

29 (1) 50 (3) 59 (1) 65 (4)

15 (0.6) 26 (1) 36 (1) 42 (1)

9 (0.1) 16 (3) 25 (1) 30 (3)

27 (1) 37 (2) 49 (2) 60 (1)

16 (1) 21 (1) 33 (1) 37 (2)

12 (3) 18 (5) 20 (1) 24 (2)

ns *

* *

* *

* *

* *

ns ns

Values of P ! 0.05 (*) are statistically signicant; ns = not signicant.

Biomass Production, Allocation, and Relative Growth Rate

In all species and treatments, plant biomass decreased from HL to LL. This was more pronounced in both herb species. Water-stressed Arthrostema and Miconia in HL and ML treatments had similar total plant biomass. Watered plants produced more biomass than stressed plants. This difference was more evident under HL (table 2). With the exception of Miconia, the root-shoot ratio de-

creased with decreasing light intensity in watered plants. In drought-stressed plants, however, it decreased in Arthrostema and Clidemia but increased in Miconia and Tibouchina as light intensity decreased (table 2). In both cases, the changes in rootshoot ratio in the woody Clidemia and Miconia were the smallest. In all species, the root-shoot ratio increased under water stress. Proportionally, Clidemia and Miconia allocated more biomass to leaves than either of the two herbs, whereas Arthrostema and Tibouchina allocated a higher proportion to

BARUCH ET AL.LIFE HISTORY OF INVASIVE MELASTOMES Table 2

Mean and Standard Error (in Parentheses) of Total Biomass, Relative Growth Rate (RGR), and Leaf Area Ratio (LAR) Treatments Water stressed Traits and species Total biomass (g): Herbs: Arthrostema ...... Tibouchina ....... Woody: Clidemia .......... Miconia ........... Root-shoot ratio: Herbs: Arthrostema ...... Tibouchina ....... Woody: Clidemia .......... Miconia ........... RGR (g g1 d1): Herbs: Arthrostema ...... Tibouchina ....... Woody: Clidemia .......... Miconia ........... LAR (cm2 g1): Herbs: Arthrostema ...... Tibouchina ....... Woody: Clidemia .......... Miconia ........... Note. High light Medium light Low light High light Watered Medium light Low light Water ANOVA Light

111

Interference

1.39 (0.20) 5.44 (0.88) 8.26 (0.61) 4.32 (0.58)

1.88 (0.18) 3.49 (0.21) 5.30 (0.67) 3.71 (0.43)

.38 (0.06) 0.28 (0.11) 1.90 (0.18) 0.96 (0.12)

12.84 (0.97) 20.63 (1.56) 10.52 (1.04) 9.99 (1.59)

5.76 (0.38) 6.79 (1.11) 7.50 (0.64) 4.21 (0.47)

0.24 (0.04) 0.31 (0.06) 1.45 (0.16) 0.87 (0.16)

* *

* *

* *

* *

* *

ns ns

0.72 (0.04) 0.41 (0.03) 0.47 (0.03) 0.46 (0.03)

0.29 (0.01) 0.21 (0.01) 0.35 (0.03) 0.33 (0.02)

0.25 (0.02) 0.67 (0.18) 0.37 (0.02) 0.60 (0.10)

0.33 (0.01) 0.54 (0.03) 0.33 (0.02) 0.34 (0.01)

0.18 (0.01) 0.14 (0.01) 0.24 (0.01) 0.21 (0.02)

0.17 (0.02) 0.28 (0.03) 0.28 (0.02) 0.34 (0.04)

* *

* *

* *

* *

* *

ns ns

0.0152 (0.002) 0.0300 (0.002) 0.0284 (0.001) 0.0146 (0.001)

0.0195 (0.001) 0.0253 (0.001) 0.0224 (0.001) 0.0129 (0.001)

0.0012 (0.002) 0.0053 (0.004) 0.0105 (0.001) 0.0023 (0.002)

0.0440 (0.001) 0.0481 (0.002) 0.0315 (0.001) 0.0244 (0.002)

0.0345 (0.001) 0.0322 (0.002) 0.0276 (0.001) 0.0149 (0.001)

0.0060 (0.002) 0.0033 (0.002) 0.0076 (0.001) 0.0043 (0.002)

* ns

* *

* ns

ns *

* *

* *

86.3 (11.3) 87.0 (5.8) 88.3 (3.1) 93.7 (8.5)

223.0 (8.8) 152.7 (8.9) 151.7 (7.9) 149.1 (4.3)

500.4 (21.0) 171.9 (51.5) 222.7 (11.7) 156.1 (9.0)

77.7 (2.5) 88.5 (3.9) 111.6 (5.0) 110.1 (10.2)

201.1 (8.3) 198.3 (4.0) 172.3 (6.7) 193.7 (17.2)

541.5 (72.6) 203.8 (45.7) 293.1 (21.4) 258.4 (26.6)

* *

ns ns

ns ns

* *

* *

ns ns

Symbols as in table 1.

stems (data not shown). The young stems of both herbs are green and presumably capable of carbon assimilation. With the exception of Tibouchina, root biomass was highest in the drought-stressed and HL treatment. Herbs had higher RGRs in the watered high- and mediumlight treatments. Watering increased RGR in all species but was only statistically signicant in Arthrostema and Miconia.

Light treatment had a signicant affect on all species: plants under LL had the lowest and even negative RGR as determined by changes in total biomass. Clidemia was the least affected by LL (table 2). The negative RGRs under LL were caused mainly by decreases in stem and root biomass (data not shown). In all species, LAR increased signicantly as light levels decreased. Watered plants had higher LARs than stressed

112

INTERNATIONAL JOURNAL OF PLANT SCIENCES CO2 mol photons1 and increased signicantly as the light intensity of the treatment decreased. Dark respiration decreased signicantly from HL to lower light intensities only in Clidemia but did not show any apparent trend in Miconia or Tibouchina (table 5). Under HL, gs of both herbs was the highest, but this difference disappeared under LL. In all species, gs decreased as light regime decreased (table 4).

plants (table 2). Interspecic differences in LAR were smaller under HL and ML than under LL. Arthrostema had the highest LAR under LL (table 2).

Leaf Nutrients, Construction Costs, and Spectral Properties

Under any treatment, Arthrostema leaves had the highest leaf N concentration on a mass basis (table 3). Water-stressed plants of all species had signicantly lower leaf N concentration, whereas shade increased N content in all species except Clidemia. On an area basis, leaf N was higher in the woody plants because of their higher SLM. Leaf K responded similarly to N, whereas leaf P concentration did not show any consistent pattern except that it decreased signicantly as light decreased in Clidemia (table 3). Leaf CC was lowest in Tibouchina and showed a tendency to decrease as the intensity of the light level decreased (table 3). The two woody species and Arthrostema had the highest leaf CC (table 3). Lack of replicates prevented further statistical analysis. Except in Tibouchina and Miconia, leaf reectance decreased under shade (table 4). Leaf transmittance increased signicantly in all species, possibly because of the thinner leaves developed under LL. Leaf absorbance decreased under LL in all species except Arthrostema (table 4). The lightdemanding herbs Arthrostema and Tibouchina had the highest leaf reectance and transmittance, whereas the relatively shade-tolerant Miconia and Clidemia showed the highest leaf absorbances. The highest leaf absorbance was that of Miconia, with their red-pigmented abaxial surface and high SLM.

Discussion
Light and water availability and their interaction affected most of the measured variables, but the effects of water stress were weaker than light-stress effects. The magnitude of the responses was related to growth form. With few exceptions, the studied melastomes showed similar qualitative results to other tropical plants. Our hypothesis was conrmed, as the herbs from open habitats were most sensitive to light deprivation, whereas the seedlings of Miconia were better able to acclimate to low light. Light intensity of the treatments was similar to that found in natural habitats such as large forest gaps, medium-sized gaps, and the understory of forests (Denslow et al. 1990; Chazdon et al. 1996). The range of temperatures was similar to most mesic sites in the Hawaiian Islands. The high mortality of Tibouchina under LL indicates that this species may not be able to meet its metabolic requirements under deep shade, thereby restricting its habitat to open sites. The early owering of Arthrostemas watered HL and ML plants indicates plasticity in phenology that may constitute a valuable trait for an invader species (Bazzaz 1986). This owering event may have triggered senescence-related processes that deteriorated the photosynthetic machinery, as evidenced by the low photosynthetic rates measured after owering.

Photosynthetic Gas Exchange

During gas exchange measurements, leaf temperature and relative humidity in the cuvette ranged between 25 and 29C and between 50% and 60%, respectively. Flowering of Arthrostema under ML and HL probably induced changes in carbon metabolism that resulted in extremely low and unreliable values of A (g. 1). However, it was possible to extrapolate A from previous measurements under HL in the greenhouse. Therefore, Arthrostema data are shown only for comparison. Maximum photosynthesis on an area basis decreased signicantly in all species with decreasing PPFD growing conditions (table 5). In contrast, because of the decrease in SLM, A on a mass basis showed no signicant differences among treatments in any species (table 5). Arthrostema and Tibouchina were the most efcient photosynthetic N users (table 5), and under HL, they were also the most efcient in xing carbon per unit leaf CC (table 4). Assimilation rate strongly affected RGR of all species but more in herbs than in the woody plants, as evidenced by a steeper slope of the regression between A and RGR in the former (herbs, RGR = 0.0053 0.0056 A; R2 = 0.90, P = 0.0001; woody species RGR = 0.0015 0.0030 A ; R2 = 0.69; P = 0.0007). In all species except Arthrostema, HL and ML plants showed similar light response curves, and the maximum A measured decreased with decreasing PPFD growing conditions (g. 1). In all species, light saturation took place in the range between 400 and 1000 mmol m2 s1 for HL and ML treatments, whereas it occurred at 100200 mmol m2 s1 for LL plants (g. 1). Quantum yield ranged between 0.053 and 0.114 mol

Plant Size, Leaf Number and Area, and Specic Leaf Mass
All ML plants, except watered Arthrostema, were as tall or even taller than HL plants. This response was not accompanied by a parallel increase in plant mass, although the proportion of plant biomass allocated to stems did increase (data not shown), indicating that the stems developed under ML were more slender and less compact. Plants under LL also showed a tendency to grow taller, particularly in the shade-tolerant Clidemia and Miconia species. This behavior has been interpreted as a response to enhance the capture of additional light, mediated by the phytochrome system and activated by a decrease in the red : far red ratio under shade (Lee at al. 1996). This was most likely not the case in this experiment since it was performed under the shade provided by neutral density lters that do not alter the spectral composition of light. Increased plant height under shade was also found in seedlings of Heliocarpus appendiculatus, an early successional shadeintolerant tropical tree (Fetcher et al. 1987). Plant leaf area of all species also increased under ML, which resulted in the highest individual average leaf size in ML plants (data not shown). This response to light was more pronounced in the herbs, which, as fast-growing species, could reallocate carbon faster than the woody Clidemia and Miconia plants. For all species, SLM decreased as light availability decreased.

BARUCH ET AL.LIFE HISTORY OF INVASIVE MELASTOMES Table 3

Mean and Standard Error (in Parentheses) of Leaf Nutrients and Leaf Construction Costs Treatments Water stressed Traits and species Leaf N (%): Herbs: Arthrostema ................. Tibouchina ................... Woody: Clidemia ..................... Miconia ...................... Leaf P (%): Herbs: Arthrostema ................. Tibouchina ................... Woody: Clidemia ..................... Miconia ...................... Leaf K (%): Herbs: Arthrostema ................. Tibouchina ................... Woody: Clidemia ..................... Miconia ...................... Construction costs (g glucose g1): Herbs: Arthrostema ................. Tibouchina ................... Woody: Clidemia ..................... Miconia ...................... High light Medium light Low light High light Watered Medium light Low light Water ANOVA Light

113

Interaction

1.92 (0.26) 1.69 (0.08) 1.34 (0.05) 1.51 (0.19)

3.00 (0.07) 2.52 (0.01) 1.73 (0.08) 1.76 (0.01)

2.98 (0.08) NA 1.87 (0.11) 2.47 (0.13)

2.45 (0.16) 2.04 (0.10) 2.11 (0.08) 1.85 (0.19)

3.02 (0.14) 2.85 (0.11) 2.10 (0.07) 2.88 (0.10)

3.91 ) 2.91 ) 1.94 (0.03) 2.53 (0.11)

* )

* )

ns )

* *

* *

* *

0.24 (0.01) 0.14 (0.01) 0.50 (0.02) 0.34 (0.06)

0.28 (0.02) 0.20 (0.01) 0.24 (0.01) 0.24 (0.01)

0.29 (0.01) NA 0.27 (0.01) 0.29 (0.03)

0.32 (0.02) 0.18 (0.01) 0.43 (0.02) 0.32 (0.06)

0.29 (0.02) 0.21 (0.01) 0.28 (0.02) 0.30 (0.03)

.35 ) 0.24 ) 0.27 (0.02) 0.42 (0.03)

ns )

ns )

ns )

ns ns

* ns

ns ns

3.43 (0.18) 0.51 (0.03) 0.85 (0.09) 0.84 (0.11)

2.90 (0.22) 0.93 (0.07) 1.18 (0.11) 1.06 (0.05)

5.44 (0.27) NA 1.10 (0.12) 1.54 (0.13)

3.79 (0.31) 0.69 (0.10) 1.26 (0.10) 0.82 (0.09)

2.94 (0.42) 0.87 (0.06) 1.41 (0.09) 1.39 (0.02)

5.22 ) 0.75 ) 1.29 (0.10) 1.79 (0.05)

ns )

* )

ns )

* *

ns *

ns ns

1.27 1.18 1.27 1.26

1.21 1.18 1.25 1.23

NA NA 1.22 NA

1.31 1.17 1.29 1.24

1.24 1.20 1.24 1.22

NA NA 1.22 1.20

) ) ) )

) ) ) )

) ) ) )

Note. Symbols as in table 1; NA = data not available. For the treatments in which there was not enough material for analysis, the replicates were pooled and do not show standard error or ANOVA.

This response to low light is common to melastomes (Denslow et al. 1990; Ellison et al. 1993) and other tropical plants (Bjo rkman 1981; Chazdon and Field 1987; Fetcher et al. 1987; Thompson et al. 1992; Ellsworth and Reich 1996; Pattison et al. 1998). Less dense and thinner leaves may permit light to reach deeper chloroplasts in the leaf mesophyll, which is an important adaptive trait in shaded habitats (Bjo rkman 1981). The simultaneous increases in leaf area and decreases of SLM under low light have been explained as a response to increase carbon capture and to recover internal carbon balances (Bjo rkman 1981; Schulze and Chapin 1987). Drought stress signicantly increased SLM in all species un-

der all light treatments. The production of thicker leaves under water stress is a common response of plants to drought (Waring 1991). In addition, thick leaves may increase the resistance to water ow, thus reducing water loss and improving water use efciency (Nobel 1991). The plasticity in SLM may allow a plant to regulate and to balance various internal processes and biomass allocation in response to environmental demands, acting as a homeostatic mechanism (Lambers and Poorter 1992). Although the qualitative responses to water and light stresses were similar in all species, the quantitative changes in plant height, leaf area, and SLM were largest in the herbs, indicating that these plants can respond faster to changes in

114

INTERNATIONAL JOURNAL OF PLANT SCIENCES Table 4

Mean and Standard Error (in Parentheses) of Maximum Photosynthesis Rate per Unit of Leaf Construction Costs (Amax CC1) (in Equivalents of Glucose per Gram Tissue Dry Weight), Stomatal Conductance (gs), Leaf Transmittance, and Absorbance of Watered Plants Treatments Traits and species Amax CC (mmol CO2 g glucose s ): Herbs: Arthrostema ...................... Tibouchina ........................ Woody: Clidemia .......................... Miconia ........................... gs (mmol H20 m2 s1): Herbs: Arthrostema ...................... Tibouchina ........................ Woody: Clidemia .......................... Miconia ........................... Leaf transmittance (%): Herbs: Arthrostema ...................... Tibouchina ........................ Woody: Clidemia .......................... Miconia ........................... Leaf absorbance (%): Herbs: Arthrostema ...................... Tibouchina ........................ Woody: Clidemia .......................... Miconia ...........................
1 1 1

High light

Medium light

Low light

0.360 (0.031) 0.192 (0.011) 0.131 (0.023) 0.120 (0.022)

NA 0.335 (0.047) 0.165 (0.028) 0.111 (0.022)

NA NA

) *

0.017 (0.015) 0.125 (0.014)

ns ns

0.213 (0.092) 0.203 (0.020) 0.163 (0.012) 0.190 (0.020)

NA 0.275 (0.035) 0.166 (0.019) 0.106 (0.021)

0.113 (0.023) 0.146 (0.007) 0.142 (0.029) 0.105 (0.021)

* *

ns ns

10.11 (1.22) 9.01 (1.04) 3.27 (0.52) 0.75 (0.08)

11.08 (0.46) 11.83 (0.20) 6.55 (0.28) 0.96 (0.13)

15.43 (0.27) 16.17 (1.45) 11.26 (0.83) 3.37 (0.15)

* *

* *

77.69 (1.85) 82.36 (1.49) 88.24 (0.70) 96.71 (0.05)

79.68 (0.80) 78.85 (0.14) 85.78 (0.41) 97.11 (0.09)

77.30 (0.44) 73.79 (1.68) 81.16 (0.88) 94.11 (0.13)

ns *

* *

Note. P indicates the signicance of the one-way ANOVA on the difference between the light treatments. Symbols as in table 1. Amax CC1 and gs data for high-light Arthrostema are from greenhouse measurements.

environmental demands. Under similar full-light conditions, the melastomes had lower SLM than a large group of native Hawaiian plants (melastomes: 39.5 15.0; natives: 81.9 41.7 g m2) (Pattison et al. 1998; Baruch and Goldstein 1999). Low SLM appears to be a common trait of invader plants in Hawaii as compared with native species and may contribute to their success as larger assimilatory surfaces are produced for a given amount of carbon xed.

Biomass Production, Allocation, and Relative Growth Rate

Shade and water stress decreased total biomass in all species. However, this decrease was proportionally lesser in the relatively slow-growing woody Miconia and Clidemia than in both herbs. Under optimal growing conditions, herbs allocated proportionally more biomass to stems, whereas the woody species

BARUCH ET AL.LIFE HISTORY OF INVASIVE MELASTOMES

115

Fig. 1 Light response curves of the plants grown under high light (open circles), medium light (triangles), and low light (lled circles). Bars indicate one standard error of the mean. Different letters for each light level indicate statistical signicant means at P ! 0.05 . High- and mediumlight Arthrostemaplants displayed leaf senescence.

allocated more to leaves. In an open habitat, tall herbs have competitive advantage. In contrast, woody and relatively slowgrowing plants would nd no advantage by promoting stem growth, as they could not reach the canopy in the short term. For these species at the seedling stage, carbon allocation to leaves would be an available optimal avenue to increase resource capture. In response to water shortage, all species except Tibouchina increased the root-shoot ratio under HL and ML. This may help to increase nutrient and water absorption when carbon incorporation is not limited (Schulze and Chapin 1987). The combined stresses of light and water deprivation lead to mortality, particularly in Tibouchina. Under light limitation, LAR increased in all species, with Arthrostema showing the largest response. This was probably a consequence of a larger decrease in total plant biomass than in leaf area, resulting in an increased carbon incorporation that tended to balance the internal carbon reserves (Schulze and Chapin 1987). A similar response to light limitation was found for a group of invasive species in Hawaii (Pattison et al. 1998). Carbon allocation changes in response to water and light stresses were the largest in Arthrostema and the smallest in Clidemia. In this regard, the former showed the largest acclimation responses. The RGR of both herbs was in the range of pioneer species (Kitajima 1996) and other invasive plants in Hawaii (Pattison

et al. 1998). Relative growth rates of the woody and shadetolerant Miconia and Clidemia were substantially lower than those of the herbs but similar to those of other melastomes in Central America (Denslow et al. 1990) and shade-tolerant plants in general (Kitajima 1996). The RGR of all species responded strongly to both water and/or light stresses as expected. In all species, RGR decreased signicantly as the intensity of the growing light conditions decreased. The main cause of this decrease was the strongly correlated decrease of A. A parallel decrease in SLM and the increase in LAR was not enough to counter the strong reduction in A that might have helped to maintain RGR levels. Clidemia was unique in maintaining positive RGR even under a level of LL that, considering only RGR responses, would categorize it as the most shade-tolerant species. Under similar conditions, the RGR of the four melastomes was 250% larger than that for a group of native trees and herbs (melastomes, 0.036 g g1d1; natives, 0.010 g g1d1) (Pattison et al. 1998). High RGR is one of the main traits that characterize invader species (Bazazz 1986), and it apparently contributes to the high invasive potential of the melastomes.

Leaf Nutrients, Spectral Properties, and Construction Cost

The signicantly higher leaf Nmass in the plants under LL observed in this experiment contrasts with observations on

116

INTERNATIONAL JOURNAL OF PLANT SCIENCES other melastomes (Denslow et al. 1990) and tropical trees (Thompson et al. 1992). This reduction of leaf Nmass was probably caused by a large decrease in SLM that accompanied growth under LL, as discussed in Plant Size, Leaf Number and Area, and Specic Leaf Mass. The herbs were more efcient in absorbing and using N than the woody species, consistent with their higher RGR (Lambers and Poorter 1992). Under similar conditions of full sunlight and watering, the melastomes exhibited higher leaf N and P concentration than a large group of Hawaiian native species (melastomes, 2.11 0.30; natives, 1.36% 0.63% N, and melastomes 0.31 0.11; natives, 0.08% 0.05% P) (Baruch and Goldstein 1999). Considered in isolation, this would make invasive species less efcient in nutrient use. However, a large foliar N content promotes carbon assimilation and enhances the growth potential of invasive species (Field and Mooney 1986; Baruch and Goldstein 1999). Leaf reectance did not show any particular trend in response to light levels in any species. The signicant increase in leaf transmittance in all species under LL was most likely a consequence of the parallel decrease in SLM already discussed. This increase resulted in a decrease in leaf absorbance under LL except in Arthrostema. This was unexpected, as increased absorbance should be a valuable response to low light availability. The two herbs from open habitats had lower leaf absorbances than the woody shade-tolerant plants. These results are different than results obtained by Lee and Graham (1986), who found no signicant difference in leaf absorbance between a large group of tropical sun and shade plants. The deep green leaves of Miconia with purple underside were the most absorptive of all, as would be expected from this shadetolerant species (Lee and Graham 1986). The higher leaf reectance and lower leaf absorbance of the herbs probably help in reducing the radiation load and transpirational water loss of these open habitat species (Nobel 1991). Leaf CC was similar to other herbs and woody plants (Poorter and Villar 1997). Both woody species had higher leaf CC than Tibouchina, as expected from their denser and possibly longer-lived leaves. Although the differences in CC were small, they can lead to moderate differences in RGR (Poorter and Villar 1997). The relatively high leaf CC in Arthrostema could be attributed to the presumably high energy content of the sticky secretions produced by its leaf glandular hairs that may act as herbivory deterrents. Leaf CC decreased from HL to LL in all species. This could be the result of the thinner and less dense leaves produced under LL with the concurrent probable decrease of high-energy lignin content (Poorter 1994). Higher CC of sun leaves was also found for a group of tropical Piper species and other plants (Williams et al. 1989; Poorter and Villar 1997). Under similar light conditions, leaf CC of the melastomes was lower than that for a large group of native Hawaiian species (melastomes, 1.25 0.06; natives, 1.39 0.01 equivalents of glucose g1) (Baruch and Goldstein 1999). Low leaf CC has been associated with plants with high RGR (Lambers and Poorter 1992; Poorter and Villar 1997). The lower leaf CC of the melastomes supports that these species use carbon more efciently than native species by investing less energy per unit of leaf produced.

Table 5
Mean and Standard Error (in Parentheses) of Maximum Photosynthesis Amax on Leaf Area and Weight Basis, Photosynthetic Nitrogen Use Efciency (PNUE), and Dark Respiration of Watered Plants Treatments Traits and species Amax (mmol m s ): Herbs: Arthrostema ...... Tibouchina ....... Woody: Clidemia .......... Miconia ........... Amax (mmol g1 s1): Herbs: Arthrostema ...... Tibouchina ....... Woody: Clidemia .......... Miconia ........... PNUE (mmol gN1 s1): Herbs: Arthrostema ...... Tibouchina ....... Woody: Clidemia .......... Miconia ........... Dark respiration (mmol m2 s1): Herbs: Arthrostema ...... Tibouchina ....... Woody: Clidemia .......... Miconia ...........
2 1

High light

Medium light

Low light

10.63 (.77) 7.26 (.81) 8.29 (.29) 7.72 (1.40)

NA 8.33 (1.21) 6.14 (.76) 4.54 (1.23)

2.90 (.38) 2.85 (.94) 3.59 (.27) 2.89 (.42)

* *

* *

0.482 (.051) 0.225 (.013) 0.169 (.027) 0.150 (.031)

NA 0.404 (.057) 0.206 (.035) 0.136 (.027)

0.242 (.055) 0.268 (.076) 0.208 (.018) 0.151 (.017)

* ns

ns ns

19.59 (1.01) 11.15 (1.14) 8.15 (1.60) 8.19 (1.61)

NA 14.17 (2.04) 9.95 (2.06) 4.82 (1.11)

6.19 (1.40) 6.57 ) 10.69 (.81) 5.99 (.69)

* ns

ns ns

NA 0.833 (.310) 0.914 (.183) 0.927 (.067)

NA 0.821 (.230) 0.405 (.065) 0.441 (.020)

0.601 (.107) 0.875 (.392) 0.502 (.064) 0.946 (.030)

) ns

* *

Note. P indicates the signicance of the one-way ANOVA on the difference between the light treatments. Symbols as in table 1. Data for high-light Arthrostema are from greenhouse measurements.

BARUCH ET AL.LIFE HISTORY OF INVASIVE MELASTOMES

117

Gas Exchange
The maximum A values obtained were in the range of those reported for a large group of early and late successional tropical plants grown under contrasting light regimes (StraussDebenedetti and Bazzaz 1996). However, A of Miconia calvescens was higher than for other Miconias in the eld (Denslow et al. 1990; Newell et al. 1993; Ellsworth and Reich 1996). On an area and mass basis, A was highest in the herbs. On an area basis, A decreased monotonically as light intensity decreased, whereas it was maintained or even increased when considered on a mass basis. This is likely to be a consequence of the decrease in SLM from HL to LL. No substantial differences were observed in the shape of the light curves among the shade-tolerant Miconia, the shade-intolerant Tibouchina, and the intermediate Clidemia. The term acclimation as used here is similar to that of plasticity in Stauss-Debenedetti and Bazzaz (1996), in that it refers to the range of phenotypic expression under constant contrasting environments. All species showed light acclimation, i.e., light responses were proportional to the light regime in which they were grown (Bjo rkman 1981). This was expected for Miconia and Clidemia, as they grow under shade and in openings in the eld. This plasticity is even more valuable for Miconia, as it switches from a seedling and sapling in the forest understory to the full-light conditions in the canopy. This acclimation potential of Miconia was similar to that observed in other melastomes, tropical plants, and invaders in Hawaii (Denslow et al. 1990; Newell et al. 1993; Ellsworth and Reich 1996; Pattison et al. 1998). Under high-light conditions, the A of the melastomes was higher than that of a large group of Hawaiian native herbs, shrubs and trees (melastomes, 8.47 1.49 mmol m2 s1; natives, 6.9 4.62 mmol m2 s1) (Baruch and Goldstein 1999).

The higher A of the melastomes was strongly related to their high RGR. Both herbs had a signicantly higher PNUE than the woody species, which is the likely result of the higher A and lower SLM of the former. Under similar high-light conditions, PNUE of the melastomes was up to 400% higher than for a large group of native Hawaiian species (Baruch and Goldstein 1999).

Conclusion
In all species, all variables responded strongly to light limitation, whereas the responses to water stress were weaker but interacted with light responses. Quantitatively, biomass allocation, RGR, SLM, and CC of herbs responded more to the different treatments than the woody plants. All species showed acclimation to light. With few exceptions, the responses of the four melastomes to light stress were similar to those of other tropical plants. When compared with a large number of Hawaiian native species (Pattison et al. 1998; Baruch and Goldstein 1999), the four melastomes had higher A and RGR and lower SLM and CC. These traits probably contribute signicantly to the success of the invader melastomes in the disturbed mesic habitats of Hawaii. We hope that the results of this study will provide insights that might help land managers seeking to control these and other similar invasive species

Acknowledgments
Z. Baruch thanks Universidad Simo n Bol var for support during sabbatical leave at the University of Hawaii. Research funds were provided by the University of Hawaii Cooperative National Park Resources Studies Unit. We thank Dr. Clifford Smith for support during the study.

Literature Cited
Almeda F 1990 Melastomataceae. Pages 903917 in DL Wagner, DR Herbst, SH Sohmer, eds. Manual of the owering plants of Hawaii. University of Hawaii Press and Bishop Museum Press, Honolulu. Baruch Z, G Goldstein 1999 Leaf construction cost, nutrient concentration, and net CO2 assimilation of native and invasive species in Hawaii. Oecologia 121:183192. Baruch Z, MM Ludlow, R Davis 1985 Photosynthetic responses of native and introduced C4 grasses from Venezuelan savannas. Oecologia 67:288293. Bazzaz FA 1986 Life history of colonizing plants: some demographic, genetic and physiological features. Pages 96110 in HA Mooney, JA Drake, eds. Ecology of biological invasions of North America and Hawaii. Ecological Studies 58. Springer, New York. Bjorkman O 1981 Responses to different quantum ux densities. Pages 57107 in OL Lange, PS Nobel, CB Osmond, H Ziegler, eds. Physiological plant ecology. Vol 1. Responses to the physical environment. Springer, Berlin. Chapin FS 1991 Effects of multiple environmental stresses on nutrient availability and use. Pages 6788 in HA Mooney, WE Winner, EJ Pell, eds. Response of plants to multiple stresses. Academic Press, San Diego, Calif. Chazdon RL, CB Field 1987 Determinants of photosynthetic capacity in six rainforest Piper species. Oecologia 73:222230. Chazdon R, R Pearcy, D Lee, N Fetcher 1996 Photosynthetic responses of tropical forest plants to contrasting light environments. Pages 555 in SS Mulkey, RL Chazdon, AP Smith, eds. Tropical forest plant ecophysiology. Chapman & Hall, New York. Denslow JS, JC Schulz, PM Vitousek, BR Strain 1990 Growth responses of tropical shrubs to treefall gap environments. Ecology 71: 165179. Ellison AM, JS Denslow, BA Loiselle, D Brenes 1993 Seed and seedling ecology of neotropical Melastomataceae. Ecology 74: 17331749. Ellsworth DS, PB Reich 1996 Photosynthesis and leaf nitrogen in ve Amazonian tree species during early secondary succession. Ecology 77:581594. Fetcher N, SF Oberbauer, G Rojas, BR Strain 1987 Efectos del re gimen de luz sobre la fotos ntesis y crecimiento en plantulas de a rboles de un bosque lluvioso tropical de Costa Rica. Rev Biol Trop 35(suppl):97110. Field CB, HA Mooney 1986 The photosynthesis-nitrogen relationships in wild plants. Pages 2555 in TJ Givnish, ed. On the economy of plant form and function. Cambridge University Press, Cambridge. Isaac RA, WC Johnson 1983 High speed analysis of agriculture samples using inductively coupled plasma-atomic emission spectroscopy. Spectrochim Acta 38B:277282. Kitajima K 1996 Ecophysiology of tropical tree seedlings. Pages 559596 in SS Mulkey, RL Chazdon, AP Smith, eds. Tropical forest plant ecophysiology. Chapman & Hall, New York. Lambers H, H Poorter 1992 Inherent variation in growth rate be-

118

INTERNATIONAL JOURNAL OF PLANT SCIENCES

Poorter H, R Villar 1997 The fate of acquired carbon in plants: chemical composition and construction costs. Pages 3972 in FA Bazzaz, J Grace, eds. Plant resource allocation. Academic Press, San Diego, Calif. Rejmanek M, Richardson D 1996 What attributes make some plant species more invasive? Ecology 77:16551661. Schulze ED 1981 Plant life forms and their carbon, water and nutrient relations Pages 615676 in LO Lange, PC Nobel, CB Osmond, H Ziegler, eds. Encyclopedia of plant physiology. Vol 12B. Springer, Berlin. Schulze ED, FS Chapin 1987 Plant specialization to environments of different resource availability. Pages 120148 in DE Schulze, H Zwolfer, eds. Potentials and limitations of ecosystem analysis. Ecological Studies 61. Springer, Berlin. Smith CW 1992 Distribution, status, phenology, rate of spread, and management of Clidemia in Hawaii. Pages 241253 in CP Stone, CW Smith, JT Tunison, eds. Alien plant invasions in native ecosystems of Hawaii. University of Hawaii Press, Honolulu. Strauss-Debenedetti S, FA Bazzaz 1996 Photosynthetic characteristics of tropical trees along successional gradients. Pages 162186 in SS Mulkey, RL Chazdon, AP Smith, eds. Tropical forest plant ecophysiology. Chapman & Hall, New York. SYSTAT 1997 SYSTAT version 7.0 for Windows. SPSS, Chicago. Thompson WA, PE Kriedemann, IE Craig 1992 Photosynthetic response to light and nutrients in sun-tolerant and shade-tolerant rainforest trees. I. Growth, leaf anatomy and nutrient content. Aust J Plant Physiol 19:118. Vitousek PM 1994 Beyond global warming: ecology and global change. Ecology 75:18611876. Vitousek PM, LL Loope, CP Stone 1987 Introduced species in Hawaii: biological effects and opportunities for ecological research. Trends Ecol Evol 2:224227. Waring RH 1991 Responses of evergreen trees to multiple stresses. Pages 371390 in HA Mooney, WE Winner, EL Pell, eds. Response of plants to multiple stresses. Academic Press, San Diego, Calif. Wester L 1992 Origin and distribution of adventive alien owering plants in Hawaii. Pages 99154 in CP Stone, CW Smith, JT Tunison, eds. Alien plant invasions in native ecosystems of Hawaii. University of Hawaii Press, Honolulu. Williams DG, RA Black 1994 Drought response of a native and introduced Hawaiian grass. Oecologia 97:512519. Williams KF, CB Field, HA Mooney 1989 Relationships among leaf construction cost, leaf longevity, and light environment in rain-forest plants of the genus Piper. Am Nat 133:198211. Williams KF, F Percival, J Merino, HA Mooney 1987 Estimation of tissue construction cost from heat of combustion and organic nitrogen content. Plant Cell Environ 10:725734.

tween higher plants: a search for physiological causes and ecological consequences. Adv Ecol Res 23:188261. Lee DW, K Baskaran, M Mansor, H Mohamad, SK Yap 1996 Irradiance and spectral quality affect Asian tropical rain forest tree seedling development. Ecology 77:568580. Lee DW, R Graham 1986 Leaf optical properties of rainforest sun and extreme shade plants. Am J Bot 73:11001108. Loope LL, D Mueller-Dombois 1989 Characteristics of invaded islands, with special reference to Hawaii. Pages 257280 in JA Drake, HA Mooney, F Di Castri, RH Groves, FJ Kruger, M Rejmanek, M Williamson, eds. Biological invasions: a global perspective. Scientic Committee on Problems of the Environment (SCOPE). Vol 37. Wiley, Chichester. Medeiros AC, LL Loope, P Conant, S McElvaney 1997 Status, ecology and management of the invasive plant Miconia calvescens DC (Melastomataceae) in the Hawaiian Islands. Bishop Mus Occas Pap 48:2336. Meyer JY, J Florence 1996 Tahitis native ora endangered by the invasion of Miconia calvescens. J Biogeogr 23:775781. Meyer JY, JP Malet 1997 Study and management of the alien invasive tree Miconia calvescens DC (Melastomataceae) in the islands of Raiatea and Tahaa (Society Islands, French Polynesia), 19921996. Technical Report 111. Hawaii Cooperative National Park Resources Studies Unit, Honolulu. 56 pp. Mulkey SS, RL Chazdon, AP Smith, eds 1996 Tropical forest plant ecophysiology. Chapman & Hall, New York. Nelson DW, LE Sommers 1972 Determination of total nitrogen in plant material. Agron J 65:109112. Newell EA, EP McDonald, BR Strain, JS Denslow 1993 Photosynthetic responses of Miconia species to canopy openings in a tropical lowland rainforest. Oecologia 94:4956. Nobel PS 1991 Physicochemical and environmental plant physiology. Academic Press, San Diego, Calif. 635 pp. Orians GH, OT Solbrig 1977 A cost-income model of leaves and roots with special reference to arid and semi-arid areas. Am Nat 111:677690. Pattison RR, G Goldstein, A Ares 1998 Growth, biomass allocation and photosynthesis of invasive and native Hawaiian rainforest species. Oecologia 117:449459. Pearcy RW, DA Sims 1994 Photosynthetic acclimation to changing light environments: scaling from the leaf to the whole plant. Pages 145174 in MM Caldwell, RW Pearcy, eds. Exploitation of environmental heterogeneity by plants. Academic Press, San Diego, Calif. Poorter H 1994 Construction costs and payback time of biomass: a whole plant perspective. Pages 111127 in J Roy, E Garnier, eds. A whole plant perspective on carbon-nitrogen interactions. SPB Academic, The Hague.