PHYSIOLOGICAL RESPONSES OF TWO TROPICAL WEEDS TO SHADE

Ipomoea asarifolia (Desr.) Roem. & Schultz (Convolvulaceae) and Stachytarpheta cayennensis (Rich) Vahl. (Verbenaceae), two weeds found in pastures and crop areas in Brazilian Amazonia, were grown in controlled environment cabinets under high (800-1000 μmol m2 s1) and low (200-350 μmol m2 s1) light regimes during a 40-day period. For both species leaf dry mass and leaf area per total plant dry mass, and leaf area per leaf dry mass were higher for low-light plants, whereas root mass per total plant dry mass was higher for high-light plants. High-light S. cayennensis allocated significantly more biomass to reproductive tissue than low-light plants, suggesting a probably lower ability of this species to maintain itself under shaded conditions. Relative growth rate (RGR) in I. asarifolia was initially higher for high-light grown plants and after 20 days started decreasing, becoming similar to low-light plants at the last two harvests (at 30 and 40 days). In S. cayennensis, RGR was also higher for high-light plants; however, this trend was not significant at the first and last harvest dates (10 and 40 days). These results are discussed in relation to their ecological and weed management implications.


INTRODUCTION
In agricultural ecosystems the light environment can be extremely variable and in many instances, may determine the composition of the weedy flora of the area.This is particularly true for tropical pastures, where selective grazing may create zones with relatively dense or sparse plant cover consequently creating contrasting light environments over relatively small areas.In these environments, weed seedlings may have to start their development under conditions of permanent shade or, during their development, be shaded by the fast growing tropical grasses or pasture weeds.Information on the physiological responses of weedy species to distinct light conditions can be critical in determining their occurrence potential and in helping to understand their competitive ability under different management situations so as to devise more effective control strategies.Managing tropical pasture weeds through an improved understanding of their ecophysiology is particularly important for areas like Brazilian Amazonia, where ecologically sound weed management strategies could decrease further deforestation by helping to increase the agricultural sustainability of already deforested areas.
Ipomoea asarifolia (Desr.)Roem.& Schultz and Stachytarpheta cayennensis (L.C.Rich) Vahl are important weedy species of pasture and agricultural areas in Brazilian Amazonia (Dias-Filho, 1990).Stachytarpheta cayennensis is also considered a noxious weed in other regions of Brazil (Lorenzi, 1991) and throughout the tropics (Nair et al., 1982;Haselwood & Motter, 1983;Kleinschmidt & Johnson, 1987), and many species of the genus Ipomoea are noxious invaders of agricultural areas (Elmore et al., 1990).The occurrence pattern of these species in the field suggests that light, among other resources, might be an important determinant of their success.
Because photosynthesis and stomatal behavior are associated with dry matter accumulation, and because light is known to limit carbon accumulation, and nitrogen content understanding these processes in weeds may provide insight as to their effects on crop growth, help predict their occurrence, and ultimately provide the needed information for their management.
A previous work determined the sun-shade acclimation characteristics of I. asarifolia and S. cayennensis by measuring their growth and bio-mass allocation patterns in response to high and low irradiance conditions (Dias-Filho, 1999).The objective of this study was to determine photosynthetic features, stomatal conductance and leaf nitrogen content of I. asarifolia and S. cayennensis developed at contrasting irradiance environments.

Plant material and growing conditions
Details on plant material and growing conditions were previously described (Dias-Filho, 1999).Briefly, Ipomoea asarifolia (Convolvulaceae) and Stachytarpheta cayennensis (Verbenaceae) were initially cultivated in 15 cm plastic pots in a greenhouse.Following expansion of the first two true leaves (about 21 days after germination), plants were transferred from the greenhouse to two Conviron growth chambers (Model PGW36.Controlled Environments, Inc. Pembina, ND) set at 28/24°C day/night air temperatures, 70% relative humidity, and 12-h photoperiod.
Within each growth chamber, a high-and a low-irradiance environment were simulated by placing multiple layers of neutral density shade cloth beneath the light source of half of each growth chamber.Light, monitored throughout the experiment with a light meter (Model LI-1905 light meter, Li-Cor, Inc. Lincoln, NE) at canopy height, ranged between 800 and 1000 µmol m -2 s -1 for high-irradiance and between 200 and 350 µmol m -2 s -1 for low-irradiance treatments.
All pots were watered daily to full capacity with deionized water and fertilized every five days with 50 mL per pot of a water soluble fertilizer solution (15:30:15; N:P 2 O 5 :K 2 O; 3.5 g L -1 ).

Gas exchange
Gas exchange parameters were measured on plants just before the fifth harvest, i.e., when the plants had been in the growth chamber for 40 days.A photosynthesis (A) versus irradiance (PFD) response curve (starting at ca. 1400 µmol m -2 s -1 and decreasing in five steps to 0), was measured in the laboratory on the most recent, fully expanded intact leaf of all plants using a portable photosynthesis system (Model LI-6200, LI-COR, Inc., Lincoln, NE).Gas exchange parameters were calculated on a leaf area basis.The amount of leaf area in the chamber (cuvette) was approximately 2.2 x 10 -3 m 2 for I. asarifolia and 1.5 x 10 -3 m 2 for S. cayennensis.The light levels were obtained by combining neutral-density filters between the leaf chamber and the light source, a high-pressure sodium vapor lamp (1000 W).Gas exchange measurements were made after an adjustment period of about 20 min to each new irradiance level.To absorb long-wave radiation and maintain constant leaf temperatures in the cuvette, a 10 cm deep Plexiglas circulating bath of cool water was suspended between the light source and the leaf cuvette.For S. cayennensis, leaf temperature during gas exchange measurements was maintained at 25.2 ± 0.9°C (low-irradiance plants, mean ± s.d.) and 25.5 ± 0.8°C (high-irradiance plants), except when it was necessary to measure gas exchange in the dark, which resulted in mean leaf temperature of 23.3 ± 0.8°C (low-irradiance) or 23.3 ± 0.7°C (high-irradiance).For I. asarifolia, leaf temperatures were 26.7 ± 0.7°C (low-irradiance) or 26.7 ± 0.8°C (high-irradiance); leaf temperatures in the dark were 23.3 ± 0.2°C (low-irradiance) and 23.3 ± 0.1°C (high-irradiance).After exposure to the lowest PFD, the plants were left for 30 min in the dark and measurements were made to obtain dark respiration (R d ) values.Five replicate plants of each species per experiment and one leaf per plant were sampled.The leaf-to-air vapor pressure gradient inside the cuvette during all measurements was 9-12 mmol mol -1 .
Photosynthesis vs. irradiance response data provided direct estimates of highest measured value of photosynthetic rate (A max ).Apparent quantum yield of CO 2 to incident irradiance (α) was calculated according to Thompson et al. (1992): the slope of a regression of five R d measurements and five low irradiance (ca.200 µmol m -2 s -1 ) measurements of CO 2 exchange, for each species and irradiance regime combination.The light compensation point was calculated as the ratio, R d /α.

Leaf nitrogen analysis
Leaf nitrogen was measured on each individual leaf used for the gas exchange measurements, and also on all leaves of each individual plant at each harvest date and light treatment combination.In all cases, leaf nitrogen content was measured with a nitrogen analyzer (Model NA 1500, series 2. Carlo Erba Strumentazione, Milan, Italy).The instrument was calibrated with both an NBS citrus leaf and organic standard.Precision was ± 0.17% of N.

Experimental design and statistical analysis
The experiment was arranged in a randomized complete block design with five replications.All pots were rotated within replications (blocks) every day to minimize variation in growth due to possible temperature and irradiance differences within the growth chamber.
Differences in net photosynthesis and stomatal conductance response curves for species and irradiance regime treatments were analyzed by analysis of variance with repeated measures (Von Ende, 1993).The betweensubject main effects were species and irradiance regimes, and the within-subjects or repeated measures effect was irradiance level.Homoscedastic residuals were obtained with logarithmically transformed values for photosynthesis and stomatal conductance.No severe departures from the normality assumption were detected by examining halfnormal probability plots of z-transformed within-cell correlations for both data sets.Compound symmetry of the covariance matrix of each analysis was tested by the Mauchlys sphericity test.The Huynh-Feldt (Von Ende, 1993) correction was considered for both analyses.Post hoc contrasts comparing species and irradiance regime differences in photosynthesis and stomatal conductance, within each irradiance level, were calculated.

RESULTS AND DISCUSSION
There were significant species and species × irradiance regime main effect differences in CO 2 assimilation rates; no significant difference could be detected for the irradiance regime main effect.Among the within-subjects main effects only irradiance level and its interactive effect with the irradiance regimes were statistically significant.For both species, low-irradiance-grown plants had a tendency to display higher assimilation values under the lowest irradiance levels (Fig. 1).Post hoc contrasts indicated that in I. asarifolia significant differences in the light response curve between irradiance environments were observed at the first (ca.1400 µmol m -2 s -1 , F 1.16 = 5.4,P = 0.03), second (ca.900 µmol m -2 s -1 , F 1.16 = 5.4,P = 0.03) and fifth (ca.200 µmol m -2 s -1 , F 1.16 = 19.5,P < 0.001) irradiance levels.High-irradiance grown I. asarifolia leaves had higher CO 2 assimilation rates at the two highest irradiance levels (i.e., ca.1400 and 900 µmol m -2 s -1 ), but lower assimilation at the lowest irradiance level (i.e., ca.200 µmol m -2 s -1 ).For S. cayennensis, significant differences in CO 2 assimilation rates between irradiance regimes could be detected only at the fifth (ca.200 µmol m -2 s -1 , F 1.16 = 23.03,P < 0.0001) irradiance level with higher CO 2 assimilation for low-irradiance grown plants.
Post hoc contrasts on differences between species, (ca.200 µmol m -2 s -1 , F 1.16 = 11.1,P= 0.004) irradiance levels, and in the dark (F 1.16 =4.6, P= 0.04); low-irradiance S. cayennensis leaves had higher g in these irradiance levels, but lower g in the dark (Fig. 2).No significant differences in g between irradiance regimes could be detected for I. asarifolia.
Light-saturated photosynthesis (A max , Table 1, Fig. 1) within each species was higher for plants subjected to the high-irradiance treatment.Between species, A max was higher for S. cayennensis, under both irradiance regimes.Within species differences in R d , assessed by post hoc contrasts of the light response curves, revealed that high-irradiance grown I. asarifolia leaves displayed higher (more negative) R d than low-irradiance leaves (F 1.16 = 8.9, P = 0.008).No other statistically significant difference in R d within and between species and treatments could be ) within each light regime, showed that under the high irradiance regime the two species differed in CO 2 assimilation only at the fourth (ca.450 µmol m -2 s -1 , F 1.16 = 4.5, P = 0.048) and the fifth (ca.200 µmol m -2 s -1 , F 1.16 = 7.6, P = 0.013) irradiance levels, with higher rates for S. cayennensis.Comparison of low-irradiance grown plants revealed higher assimilation rates for S. cayennensis at all irradiance levels (F 1.16 ≥ 7.09, P ≤ 0.017) (Fig. 1).
Leaf stomatal conductance, g, in S. cayennensis behaved differently from that of I. asarifolia; it was sensitive to growing conditions and was significantly reduced by growth at high photon flux density (Fig. 2).Comparisons between irradiance regimes assessed by post hoc contrasts showed significant differences in g for S. cayennensis at the first (ca.1400 µmol m -2 s -1 , F 1.16 = 7.9, P= 0.01) and fifth detected.However, a tendency for higher R d values for high-irradiance grown plants from both species and for S. cayennensis compared to I. asarifolia, can be observed in Table 1.
The apparent quantum efficiency, α, was sensitive to the light regime during growth (Table 1).Growth at the low-irradiance environment caused an average increase in α of 19% and 50%, respectively in I. asarifolia and S. cayennensis relative to high-irradiance plants; this difference, however, was statistically significant only for S. cayennensis (t=7.67,P < 0.0001).If we consider the reciprocal of α, the quantum requirement (quanta CO 2 -1 ), as a measure of efficiency of photosynthesis, high-irradiance S. cayennensis was significantly (t=32.9,P < 0.0001) more efficient than low-irradiance plants (mean ± s.d., 44.62 ± 2.5 vs. 31.27± 2.6), while no significant difference in photosynthetic efficiency could be detected between high-(54.92±11.6) and low-irradiance (43.58 ±7.3) I. asarifolia.Light compensation point, R d /α, was also influenced by the irradiance regimes, being at least 45% higher in high-irradiance grown I. asarifolia and S. cayennensis.As observed in Table 1, most of the difference in R d /α between species can be attributed to differences in α.As a result of the higher α observed in S. cayennensis, R d /α in this species was lower than that in I. asarifolia (Table 1).
In both species, nitrogen content per unit leaf weight was higher for low-irradiance plants (Fig. 3).However, when expressed on a leaf area basis, nitrogen content in high-irradiance S. cayennensis was higher than in low-irradiance plants.For I. asarifolia, no systematic tendency was evident, as nitrogen content, on a leaf area basis, varied throughout time in an inconsistent pattern between irradiance regimes (Fig. 3).A positive correlation between nitrogen content on a weight basis and CO 2 assimilation under the highest irradiance level (ca.1400 µmol m -2 s -1 ) was observed in high-irradiance S. cayennensis and I. asarifolia; in low-irradiance plants, a weaker correlation was found for S. cayennensis, but not for I. asarifolia, which displayed a stronger correlation between CO 2 assimilation and leaf nitrogen at low irradiance (Table 2).Stomatal conductance, g, in S. cayennensis was positively and significantly correlated with leaf nitrogen content (Table 2), the slope of this relationship being steeper for low-irradiance plants.Thus, there was a stronger response of g to leaf nitrogen on a mass basis in low-irradiance plants.I. asarifolia behaved differently from S. cayennensis, no relationship could be detected between g and leaf nitrogen in high-irradiance plants, yet, a weak, but nonsignificant correlation between these attributes was also apparent in low-irradiance I. asarifolia (Table 2).
In S. cayennensis, the amount of biomass allocated to reproductive tissue (i.e., flowering bearing spikes) was negatively correlated to CO 2 assimilation (r=-0.53 for high-irradiance and r=-0.33, for low-irradiance plants), stomatal conductance (r=-0.74 for high-irradiance and r=-0.33, for low-irradiance plants) and also to the nitrogen content (on a mass basis) from the leaf used for the gas exchange measurements (the youngest fully expanded leaf) (Table 2).Yet, none of these relationships were statistically significant (P > 0.05).However, when mean nitrogen content from all leaves of each individual plant was considered, as opposed to TABLE 1. Photosynthetic parameters of Stachytarpheta cayennensis and Ipomoea asarifolia under high-and low-irradiance regimes.Symbols and units; A max is light saturated photosynthesis rate (mmol m -2 s -1 ); a is apparent quantum efficiency (mol CO 2 photon -1 ); R d is dark respiration (mmol m -2 s -1 ) and R d /a is light compensation point (mmol m -2 s -1 ).Values are means (±s.d.), n=5.during those periods by low-(r= -0.98, P< 0.0001, n= 8) and high-irradiance (r= -0.79, P= 0.02, n= 8) S. cayennensis plants.When leaf nitrogen on an area basis was considered, a similar relationship was observed for low-(r= -0.86, P= 0.007) and high-irradiance (r= -0.67, P= 0.07) plants.Due to the nonlinear relation between mean leaf nitrogen from all leaves of each plant and the corresponding reproductive biomass of high-and low-irradiance plants, these correlations were measured by Spearman rank order correlation (Potvin & Roff, 1993).

Parameter High-irradiance Low-irradiance
The shape of the light response curve observed for I. asarifolia and, to some extent, for S. cayennensis, grown under the contrasting light environments (Fig. 1), was similar to those observed by Ramos & Grace (1990) and Riddoch et al. (1991b) for a group of tropical trees and by Regnier et al. (1988) for the weed Abutilon theophrasti, grown under high and low irradiance.Low-irradiance grown plants had higher assimilation rates under lower irradiance and lower assimilation under higher irradiance levels.However, in the present study, high-irradiance S. cayennensis leaves had a tendency to exhibit lower CO 2 assimilation under the highest irradiance level, when compared to lowirradiance plants.This unexpected behavior of high irradiance S. cayennensis leaves may be related to the amount of biomass allocated to reproductive tissue by these plants, as a negative relationship between the amount of biomass allocated to flower bearing spikes and CO 2 assimilation was suggested by our data.This, in turn, was probably associated with the developing reproductive structures acting as a competitive sink with leaves for nitrogen (and probably other resources), leading to the lower leaf  the leaf used for the gas exchange measurements, a strong and highly significant negative correlation was found between leaf nitrogen (on a mass basis) at the 30-and 40-day evaluation periods (when most plants have started reproduction) versus the corresponding biomass allocated to flower bearing spikes nitrogen content in plants with a higher proportion of biomass allocated to reproduction, as also suggested by our data.It is possible that the low leaf nitrogen content of high-irradiance plants may have rendered these plants more susceptible to photoinhibition as observed for other tropical species (Seemann et al., 1987).Developing reproductive structures acting as a sink for nitrogen has also been suggested for tropical legumes (Humphreys, 1991) and for crop species (Sadras et al., 1993;Sklensky & Davies, 1993).
The positive relationship between leaf CO 2 assimilation or stomatal conductance and leaf nitrogen content per unit area observed for S. cayennensis and, to some extent in I. asarifolia (Table 2), was expected and has also been reported in other studies involving tropical trees (Ramos & Grace, 1990;Riddoch et al., 1991b;Tuohy et al., 1991) and crops (Sinclair & Horie, 1989).Interestingly, no relationship between leaf nitrogen and stomatal conductance could be detected in low-irradiance I. asarifolia (Table 2).
Analysis of the parameters calculated from the light response curves data (Table 1) shows that within species, statistically significant differences between light treatments in the initial slope of the photosynthetic light response curves can be observed only for S. cayennensis.Low-irradiance treated plants of this species displayed a 50% higher quantum efficiency.Yet a tendency for higher quantum efficiency in low-irradiance I. asarifolia is also apparent (Table 1).Other studies have shown that light regime during growth does not normally alter apparent quantum efficiency in temperate (Björkman, 1981;Mahall & Schlesinger, 1982) or tropical (Sims & Pearcy, 1989;Ramos & Grace, 1990;Riddoch et al., 1991b) species.However, a tendency for higher α values in low-irradiance plants is usually observed in these studies.The larger α value in low-irradiance S. cayennensis relative to high-irradiance plants can explain the rapid raise in assimilation rate with increasing light observed in this species at the low-irradiance treatment (Fig. 1).When compared to low-irradiance I. asarifolia, low-irradiance S. cayennensis was particularly more efficient at the lower irradiance levels (i.e., ca.200 and 450 µmol m -2 s -1 ), displaying significantly higher CO 2 rates at these irradiance levels than high-irradiance plants (Fig. 1).
The observed tendency for lower R d values in low-irradiance plants (Table 1) is a common response also reported for other tropical species under similar conditions (Ramos & Grace, 1990;Riddoch et al., 1991b;Thompson et al., 1992), and may, in part, have contributed to the net photosynthetic advantage exhibited by shade grown plants at the lower irradiance levels (Fig. 1).Light compensation point, R d /α (Table 1), of both species followed the same trend observed in other studies involving tropical species, in which higher light during growth leads to higher values of R d /α (Ramos & Grace, 1990;Thompson et al., 1992).The values of R d /α calculated for I. asarifolia and S. cayennensis agree with values reported by Riddoch et al. (1991a) for two weeds of gaps from Nigeria, and do not suggest differences between these species in relation to their shade tolerance.However, it has been suggested from work with tropical tree seedlings that shade tolerance may not necessarily depend on leaf characteristics such as light compensation point (Thompson et al., 1992).
An important finding of this study was that, irrespective of light environment during growth, leaf nitrogen content and CO 2 assimilation were inversely related to the allocation of biomass to developing reproductive structures in S. cayennensis.Based on this, it could be speculated that this species might be more vulnerable to stresses like mowing or fire at the initial stages of flowering because of reduced vigor as expressed by lower carbon gains, which could, in turn, restrict resprouting ability.Under natural conditions this limitation could be further enhanced as plants in the field, unlike the ones in the present study, could experience more limiting edaphic conditions.Such information might also have important practical implications for planning the most appropriate time for the control of this species.

CONCLUSIONS
1.The overall effect of shade is more deleterious to S. cayennensis, which has its photosynthetic efficiency and reproductive capacity negatively affected by shade.
2. Developing reproductive structures act as a sink for leaf nitrogen in S. cayennensis at both light and shade environments.

FIG. 1 .
FIG. 1.Light (PFD) response curves of photosynthesis (A) of I. asarifolia and S. cayennensis grown under high-(open symbols and solid lines) and low-irradiance (closed symbols and dashed lines) conditions.

FIG. 2 .
FIG. 2. Light (PFD) response curves of stomatal conductance (g) of I. asarifolia and S. cayennensis grown under low-(closed symbols and dashed lines) and high-irradiance (open symbols and solid lines) conditions.Curves fitted by distance-weighted least squares.