| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
CSIRO Div. of Plant Industry, Australian Cotton Co-operative Research Center, Locked Bag 59, Narrabri, NSW 2390, Australia
* Corresponding author (Steve.Milroy{at}csiro.au)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSSCALING FROM LEAF PHOTOSYNTHESIS...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: CA, leaf CO2 assimilation rate (µmol CO2 m-2 s-1) • CAdark, leaf CO2 assimilation rate at PAR 0 µmol m-2 s-1 (i.e. the y-intercept of the light response curve) • CAmax, leaf CO2 assimilation rate measured at PAR 2000 µmol m-2 s-1 • DAS, days after sowing • k, canopy light extinction coefficient • LAI, leaf area index • LAIcum, cumulative leaf area from the top of the canopy to the level of interest • LI, proportion of radiation intercepted by the crop canopy measured at midday • LID, proportion of light intercepted by the crop canopy over the day • PAR, photosynthetically active radiation (400–700 nm) • RUE, radiation use efficiency (g MJ-1, glucose equivalents) • SLN, specific leaf nitrogen (g N m-2 of leaf) • SLNgrad, vertical SLN gradient within the canopy
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSSCALING FROM LEAF PHOTOSYNTHESIS...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
To explore the significance of N gradients on the relationship between RUE and average canopy SLN, we used a simple method presented by Hammer and Wright (1994) which scales from photosynthesis at the leaf level to canopy RUE. The approach allows for the vertical N distribution in the canopy, the average canopy SLN, and the light extinction coefficient (k) to be manipulated when calculating the net carbon gain of the canopy on the basis of the photosynthetic rates of leaves in various strata within the canopy. Hammer and Wright (1994) successfully used this approach to explore the effects of radiation environment and canopy nitrogen dynamics on RUE of peanut (Arachis hypogaea L.). In the work we present here, we used this method to explore the sensitivity of cotton RUE to changes in canopy SLN, SLN gradient and k and to test whether these traits can explain the observed ontogenetic changes in RUE (e.g., Sadras, 1996). The purpose is to assess whether there is significant benefit from accommodating the developmental changes in N gradient and k in the simulation of biomass production by cotton. The first step was to assess the capacity of the scaling approach to capture variation in RUE in cotton.
To calculate the whole canopy photosynthetic rate using this approach requires the relationship between photosynthesis and light and the impact of nitrogen on this response. The response of leaf level photosynthesis to leaf nitrogen status has been reported for cotton by Reddy et al. (1997). However, the experiments were conducted under controlled environment conditions and have not been validated against plants grown in the field. Thus an additional aim of our work was to quantify the influence of leaf N status on the response of photosynthesis to light for field grown cotton plants over a wide range of leaf N.
An important consideration in exploring the impacts of leaf nitrogen status on photosynthesis is to determine the relative influence of leaf age. Constable and Rawson (1980) and Wullschleger and Oosterhuis (1990) have studied the influence of leaf age on photosynthesis. However, photosynthesis and nitrogen concentration decline in unison as the leaf ages (Constable and Oosterhuis, In Press). The interaction of nitrogen and age on photosynthesis of cotton leaves has not been assessed. A final aim of this work was therefore to assess whether or not the impact of leaf age on photosynthesis can be detected independently of the effect of leaf nitrogen status.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSSCALING FROM LEAF PHOTOSYNTHESIS...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
There were three nitrogen treatments: 0, 60, 120 kg ha-1 in both experiments. Nitrogen was applied as anhydrous ammonia, 54 d before sowing in P1 and 85 d before sowing in P2. The experimental designs were RCBD with four replicates (plots 6 by 11 m).
Measurements
Starting 36 and 44 d after sowing (DAS) for Exp. P1 and P2, respectively, the number of main stem nodes of 10 plants in each plot was counted on a weekly basis. Plotting these numbers against DAS allowed the age of sample leaves to be estimated from the node of insertion on the main stem.
In Exp. P1, photosynthesis was measured on 12 occasions at 1- to 2-wk intervals between 30 November (48 DAS) and 23 March (152 DAS). In the second Exp. P2, photosynthesis was again measured on 12 occasions at similar intervals between 12 December (45 DAS) and 23 March (146 DAS). On the first four occasions, in both experiments, the youngest fully expanded main stem leaf was measured from two plants in each plot. As the plants produced more nodes, one plant per plot was sampled with the youngest fully expanded leaf and a leaf from the lower half of the canopy being measured. In Exp. P2, the youngest fully expanded leaf and a leaf from the lower and from the middle third of the canopy were measured.
On the day of sampling, each leaf was first exposed to full sunlight for a minimum of 15 min and on most occasions for about 1 to 2 h. The CO2 assimilation rate was then measured with an open chamber infrared gas analysis system (LI-COR 6400, Lincoln, NE) with a 6-cm2 chamber. Measurements were recorded when both leaf CO2 assimilation rate and stomatal conductance were stable. Ambient temperature and humidity were used. An average was taken of five readings spaced evenly over 120 s. Readings were taken with an LED light source which supplied 2000 µmol m-2 s-1 of photosynthetically active radiation (PAR). The light source emitted in the wavelength range 660 to 675 nm. Although a light source was used, heavily overcast days were avoided to minimize the time required for the leaves to adjust to the light intensity used. Readings were taken within 3 h of solar noon, for the same reason and to avoid any possible effects of more rapidly changing temperature in the early morning. Stomatal conductance and leaf temperature were also recorded.
After the photosynthesis reading had been taken, the node of insertion of the leaf was recorded and the leaf removed. The area of each leaf was measured with a planimeter (LI-COR 3100) and it was then dried and ground. Nitrogen content was measured by the micro Kjeldahl technique.
On six occasions in Exp. P2, a series of light response measurements were taken on the sample leaves before removal. Photosynthesis was measured at 2000 µmol m-2 s-1 PAR and then at 1500, 1000, 500, 250, 125 and 0 µmol m-2 s-1 PAR. CO2 assimilation measured at 0 µmol m-2 s-1 PAR was taken as dark respiration (CAdark).
Data Analysis
The relationship of photosynthesis measured at 2000 µmol m-2 s-1 PAR (CAmax, µmol CO2 m-2 s-1) to specific leaf nitrogen (SLN, g N m-2 of leaf) of individual leaves was fitted to the response reported by Sinclair and Horie (1989):
 | [1] |
The response of photosynthetic rate (CA) to light intensity (PAR) was fitted by means of an exponential rise to a maximum function of similar form to that of Constable and Rawson (1980):
 | [2] |
The regression analyses and analysis of parallelism were conducted by means of the procedures in Genstat Version 5 (Lawes Agricultural Trust, IACR, Rothamsted, UK).
Crop Growth Experiments
Cultural Details
Three irrigated crop growth experiments were conducted at the same location as the photosynthesis experiments. The first two, referred to as G1 and G2, have been described previously (Milroy et al., 2001). Briefly, Exp. G1 was sown on 11 Oct. 1995 and comprised cultivar Siokra L22 grown with high nitrogen application (150 kg ha-1) and full irrigation. Eight plants per square meter were established. There were three replicates. Experiment G2 was sown on 16 Oct. 1996 and comprised cultivar Sicala V-2i grown with high (150 kg ha-1) or nil nitrogen application. The experimental design was a RCB with four replicates. A density of 13 plants m-2 was established.
Experiment G3 was sown on the 16 Oct. 1998 and comprised cultivar Sicala V-2i grown with high (100 kg ha-1), low (50 kg ha-1), or nil nitrogen application. Plots were 25 by 8 m (8 rows) and the experimental design was an RCB with four replicates. Nitrogen was applied as anhydrous ammonia, 35 DAS. A density of 11 plants m-2 was established. In all growth experiments commercial equipment and management practices were used.
Measurements
Plant samples were taken approximately every 2 wk. Before first flower, normal samplings involved taking all plants from within 1 m of row. Fresh biomass per square meter was measured and a subsample of four plants was taken for further processing. Beginning around the time of first flower, destructive stratified samplings were made. For these samples, a 1-m2 area of the crop canopy in each plot was cut in four successive layers of equal vertical thickness. Details of the method have been published previously (Milroy et al., 2001). All samples were selected randomly from the inner two or four (from eight row plots) rows of each plot.
All samples were partitioned into leaves (laminae), stem (including petioles), squares, and bolls. A square was defined as being present when the leaf that subtended the square had unfolded. Fruit were defined as bolls from the day of anthesis. The leaf area (and hence leaf area index, LAI) in the subsample of normal samplings and in each layer of stratified samplings was then measured with the planimeter. All samples were dried in a forced draft oven at 70°C for at least 48 h and weighed. Normal biomass samplings were converted back to unit area by means of a drying ratio. All masses are presented on an oven-dry basis.
The leaf material was ground and analyzed for N concentration on a mass basis (g N g-1 dry weight) with a near infrared refractometer (Perten Inframatic 8100, Germany) calibrated against the Kjeldahl method or, if the sample was too small, with complete combustion and thermal conductivity analysis (LECO FP-228, St Joseph MI, USA) also calibrated against the Kjeldahl method. Stratified samplings continued until the crops were approaching maturity. A total of seven stratified harvests were made in Exp. G1, six in G2, and five in G3.
Measurements of PAR were taken in each plot above and below the canopy at ground level at approximately weekly intervals. At each date, three measurements were made on each plot with a ceptometer (Delta-T Devices Ltd., Cambridge, UK). Measurements were taken between 1100 and 1300 h.
Data Analysis
For each plot, we calculated the proportion of radiation intercepted by the crop (LI) at each day of measurement and then related it to DAS using an exponential rise to a maximum function to allow interpolation between measurement dates. Using the interpolated values of LI for the day of the stratified biomass samplings, we calculated a canopy light extinction coefficient (k) for each plot from a nonlinear regression between LI and the measured leaf area index (LAI):
 | [3] |
Vertical gradients of specific leaf nitrogen (g N m-2 of leaf) within the canopy (SLNgrad) and average canopy SLN were calculated for each harvest date from the data collected on leaf weight, leaf area, and N concentration for each stratified layer. SLNgrad was calculated from the linear regression of SLN in each layer against cumulative LAI (LAIcum) from the top of the canopy to the midpoint of each layer (Shiraiwa and Sinclair, 1993).
The RUE was calculated for each date on which a stratified harvest was taken with data from three consecutive sampling dates, centered on the date of interest. For each set, RUE was calculated from the linear regression of accumulated biomass on cumulative intercepted PAR over the three dates. In calculating biomass, the high synthesis cost of cotton fruit relative to vegetative growth was taken into account by adjusting the reproductive biomass for biosynthetic production costs (g glucose per g dry matter) of reproductive and vegetative tissues by means of the conversion factors of Wall et al. (1994). The conversion includes the growth respiration component.
The proportion of light intercepted by the crop canopy over the day (LID) was estimated from the proportion of PAR intercepted around noon (LI) by means of the equation of Charles-Edwards and Lawn (1984):
 | [4] |
The validity of this conversion has been tested for cotton by Sadras and Wilson (1997). Daily values of LID were derived by interpolation as for LI. The cumulative intercepted radiation was then derived from the daily values of LID and the daily radiation (PAR) recorded by a fully serviced weather station at the research institute.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSSCALING FROM LEAF PHOTOSYNTHESIS...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
 | [5] |
 |
 |
|
To attempt to identify the source of unexplained variability in the response of CAmax to SLN a number of variables were considered. First, to test for the effect of leaf age on the relationship between CAmax and SLN, a multiple regression analysis was conducted for each season with leaf age and SLN as independent variables. In Exp. P1, there was no improvement in the response; however, in Exp. P2, there was a significant improvement (R2 = 0.58; P < 0.001). When the data from both experiments were combined, the response of CAmax to SLN and leaf age (d) was:
 | [6] |
 |
 |
 |
(R2 = 0.50; n = 533; P < 0.0001).
Thus while including age as a variable gave a statistically significant improvement in the fit, there was still a large amount of residual variation.
A number of other variables were also considered. The possibility of water stress was excluded by omitting all measurements taken less than 4 d before an irrigation or, alternatively, by omitting all measurements with less than the average stomatal conductivity. In neither case was an improvement in the regression of CAmax against age and SLN achieved. Similarly, no improvement was found when only leaves from the upper portions of the canopy were considered in an attempt to exclude leaves which had undergone some degree of acclimation to low light. A weak correlation was found between leaf temperature and photosynthesis, but again, adding temperature to the response after the age and SLN terms did not improve the response.
The influence of SLN on the response of CA to PAR was explored in the second season (Exp. P2) on a subset of the leaves. The leaves were categorized into seven groups, each spanning 0.5 units of SLN. An analysis of parallelism was conducted with the light response function of Sinclair and Horie (1989) (Eq. [2]) to compare the response of CA to PAR across the seven different categories of SLN. As expected, the shape of the response varied to reflect the reduction in CA with declining SLN (Fig. 2)
. The analysis showed that this difference in the response to SLN was due to changes in CAmax and dark respiration (CAdark) (P < 0.001) but not in the parameter Q, implying no change in the initial light use efficiency. Calculating the initial slope arithmetically also showed it to remain constant at around 0.047 mol CO2 mol-1 quanta. The responses of CAmax and CAdark, the latter calculated as the mean of the measurement taken at PAR = 0, to the average SLN (Fig. 3)
were:
 | [7] |
 |
 |
|
|
 | [8] |
 |
(r2 = 0.94; P < 0.01) (Fig. 3b).
So, in the form of Eq. [2], the change in the response of photosynthesis (CA) to light intensity (PAR) across a range of SLN levels can therefore be represented by the function:
 | [9] |
 |
Crop Growth
The canopy characteristics needed for scaling from leaf photosynthesis to RUE were measured on the growth experiments and presented in Table 1
along with the measured RUE in glucose equivalents. It can be seen from this table that both the light extinction coefficient (k) and the nitrogen gradient varied during the development of the crop. The change in SLN with cumulative LAI was linear in the majority of cases. There was no simple correlation in the data between RUE and average canopy SLN, k, or SLNgrad alone (r2 < 0.12). Our aim was to use the approach of Hammer and Wright (1994) to investigate the impact of the variation in the nitrogen gradients and k on the observed RUE.
|
|
SCALING FROM LEAF PHOTOSYNTHESIS TO CANOPY RADIATION USE EFFICIENCY |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSSCALING FROM LEAF PHOTOSYNTHESIS...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
1. The proportion of light interception by sunlit and shaded leaf area in each stratum in the canopy is estimated from the observed k and the leaf area above the midpoint of the stratum (Table 1).
2. The N content of the leaves in the layer is estimated from the observed average canopy SLN and SLNgrad (Table 1).
3. Potential photosynthetic rate of leaves is calculated from the response of CAmax to SLN (Eq. [1]).
4. Photosynthetic rate of the sunlit and shaded leaf area is calculated from the light intensity received at that level (Eq. [2]).
5. The photosynthetic contribution of the layer is calculated by multiplying the photosynthetic rate by the area of the shaded and sunlit leaves.
6. Daily gross photosynthesis is estimated by summing the contribution of each layer and integrating the rates over the day.
7. Net photosynthetic gain is estimated by deducting maintenance respiration for the various organ types from the gross photosynthesis. Maintenance respiration was calculated from values for cotton taken from Hesketh et al. (1971) and Mutsaers (1976).
8. RUE is estimated by dividing the net photosynthetic gain by the integrated PAR interception for the day.
Further process refinements to the procedure explored in this study were to use SLN to estimate the other parameters (which included respiration and the initial slope of the light response curve) defining the light response curve (Eq. [7] to [9]).
No allowance was necessary within the framework for the glucose requirement for the synthesis of different tissue types or growth respiration. This is because RUE calculated from the experimental data has been converted to a glucose equivalent basis by the coefficients of Wall et al. (1994). This circumvents the need to apply an average synthetic efficiency or alter the coefficients within the framework for plants with different partitioning of biomass due to ontogeny.
Scaling Results
Three modifications of the existing method (Hammer and Wright, 1994) to scale from leaf photosynthesis to canopy RUE were assessed. The results of each were compared with the observed RUE (Fig. 4) . The first approach (Fig. 4a) used the functions published in the framework that estimate carbon assimilation (CA) from light intensity, and uses the response for CAmax versus SLN taken from this study (Eq. [5]). The second approach (Fig. 4b) used Eq. [9] for estimating CA from light intensity with the relationships of CAmax and CAdark to SLN being derived from the same data (Eq. [7] and [8]). The third approach (Fig. 4c) is the same as the second approach but uses Eq. [5] instead of Eq. [7]. The strength of Eq. [5] is that it is based on a much larger sample size, and includes data from both years. In each case actual day length, light intensity, k, average canopy SLN, SLNgrad, and LAI as measured in the growth experiments (Table 1) were used.
|
Substituting a single overall SLNgrad into the calculation instead of the one measured on each sampling date had little impact on the capacity of the method to estimate RUE accurately within the observed range of canopy SLN. Use of a uniform vertical nitrogen distribution, that is SLNgrad equals zero, for all samplings resulted in little change in the regression equation but slightly increased variability in predicted RUE (Fig. 5) . Use of a pooled light extinction coefficient (k) caused no significant change in the reliability of the estimation of RUE.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSSCALING FROM LEAF PHOTOSYNTHESIS...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The asymptote of the CAmax curve is comparable to other field-based measurements for cotton (Wells, 1988; Pettigrew and Turley, 1998; Stiller, 2000) and the glasshouse trial of Reddy et al. (1997). However, it was approximately 30% greater than the highest rates measured by Constable and Rawson (1980) and Wullshleger and Oosterhuis (1990).
The impact of age on leaf photosynthesis has been studied by Constable and Rawson (1980) and Wullschleger and Oosterhuis (1990). Although the latter measured leaves in situ within field canopies, thus confounding the influence of leaf age with increasing shading, in both studies, CAmax had a strong correlation with leaf age. In this study, we also found that SLN and leaf age were strongly correlated (data not shown); however, there was still a 9% improvement in the ability to account for the variation in CAmax when age was added as a variable after SLN.
The high variation in the relationship between CAmax, age and SLN is surprising. The effects of temperature (El-Sharkawy and Hesketh, 1964; Bednarz and Van Iersel, 2001) and moisture stress (Ackerson et al., 1977; McMichael and Hesketh, 1982; Turner et al., 1986) on photosynthesis of cotton are well documented. However, incorporating these terms provided no improvement in the variability, nor did making allowance for shade acclimation, which has been shown to affect CAmax in other species (Evans 1989). It is possible that there were residual effects of age which were not taken into account. Photosynthesis per unit leaf area in cotton usually peaks just before full expansion (Constable and Rawson, 1980). However, some of the upper leaves sampled may have been still in the increasing phase of photosynthesis, thus fitting poorly with the rest of the data which were in the declining phase with leaf age.
The estimates of CAmax derived from the light response functions for the leaves of different SLN classes were consistent with the overall SLN function, showing a similar rise to a maximum (Fig. 2). The concurrent increase of CAmax and CAdark to increasing SLN (Fig. 3) is consistent with the general relationship of CAdark to the preceding rate of CA (McCree, 1974). More specifically, the response of CAdark to SLN (Fig. 3) was consistent with the hypothesis of Barnes and Hole (1978) that respiration rate can be standardized by expressing it on a per unit N basis. The rate of respiration was greater than rates reported by Constable and Rawson (1980), -0.2 to -0.4 µmol CO2 m-2 s-1, as theirs were averaged over the whole night period. Ludwig et al. (1965) reported values of -2 to -2.5 µmol CO2 m-2 s-1, which are closer to those reported here. However, they do not report the N status of the leaves, so direct comparison is not possible. The lack of response in initial light use efficiency to SLN was consistent with the results of Pons et al. (1989) who showed that apparent quantum use efficiency of Lysimachia sp. was far less responsive to SLN than CAmax or respiration and those of Connor et al. (1993) who found no response in apparent quantum use efficiency for sunflower.
Thus, overall, the response functions for photosynthesis to SLN are of a type consistent with those for other crops and the CAmax values obtained are similar to those previously published for cotton in field and glasshouse studies. The results should provide robust relationships for field grown cotton under favorable conditions, suitable for inclusion in simulation models intended to reflect growth of commercial crops.
Scaling from Leaf to Canopy
The results of this study further demonstrate the robustness of the approach of Hammer and Wright (1994) for scaling from leaf photosynthesis to canopy RUE. The pattern of variation in the calculated RUE was good when compared with the observed data, although there was a significant bias at low RUE. Using more refined responses to account for the effects of nitrogen and light on the photosynthesis of cotton leaves significantly improved our ability to account for the variation in RUE (Fig. 4). We then used the scaling process to explore the relative importance of the dynamics of canopy nitrogen and light on the observed variation in RUE.
The variation in the RUE observations was due to both ontogenetic variation and differences in soil N supply. The values correspond well to those obtained by Sadras (1996). The ability of the method to account for these changes is indicated in Fig. 4. Given this ability, the sensitivity analysis indicates that the calculated RUE is responsive to SLNgrad and to a lesser extent to k (Fig. 6). When assessed for their impact on simulating the actual RUEs, however, these parameters had minimal impact. While the importance of these characteristics has been demonstrated in other crops, it has also been shown that they are of most significance at high LAI (Hirose and Werger, 1987). The LAIs measured in our crops were relatively low. When the impact of SLNgrad was tested at different LAI, no impact was evident at the lowest LAIs in this experiment but a large impact was evident at an LAI of 5.0 (Fig. 6). The LAI values in the growth experiments are typical of those observed in temperate (Constable and Hearn, 1981; Constable et al., 1990) and well managed tropical crops in Australia (Basinski et al., 1975; Ockerby et al., 1993), although high N rates can generate much larger canopies (Basinski et al., 1975; Ockerby et al., 1993). Thus, there is no evidence that ontogenetic changes in SLNgrad or k generally contribute to the dynamics of RUE for Australian cotton crops.
The approach of Hammer and Wright (1994) was developed from one previously published (Sinclair and Horie, 1989) by including the provision for SLNgrad. It appears that there is some value in including the gradient in that it gave a small improvement in the r2 of the regression against observed RUE values (Fig. 5). However, for simulating cotton growth, a single value for the gradient appears adequate and there is no benefit in varying the gradient for crop development.
The reasons for the bias in the calculated RUE at low values needs to be explored. Possible reasons include the estimation of maintenance respiration rate, the degree of partitioning to the roots, and the effects of reproductive organs on canopy light interception and distribution. Limited data are available on these aspects of cotton growth. In support of the idea that root partitioning is the cause, most of the points which lie furthest from the 1:1 line are the earliest sampling in each treatment. At these samplings, partitioning to the roots would have been highest and this would have increased the observed RUEs if it had been taken into account, but no adequate data are available. Excluding these six points from the regression reduced the bias, increased the slope to 0.75, and increased the r2 to 0.77.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSSCALING FROM LEAF PHOTOSYNTHESIS...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The overall shape of the relationship between photosynthesis and SLN for cotton was consistent with that for other species. Incorporating the influence of age improved the relationship significantly but by less than 10%, in spite of the correlation of SLN and age. The photosynthesis relationship developed should be appropriate for use in simulating cotton crops.
|
ACKNOWLEDGMENTS |
|---|
Received for publication April 12, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSSCALING FROM LEAF PHOTOSYNTHESIS...DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. A. Reddall, L. J. Wilson, P. C. Gregg, and V. O. Sadras Photosynthetic Response of Cotton to Spider Mite Damage: Interaction with Light and Compensatory Mechanisms Crop Sci., September 1, 2007; 47(5): 2047 - 2057. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Env | ||||
