Crop Science
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (27)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Andrade, F. H.
Right arrow Articles by Casanovas, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Andrade, F. H.
Right arrow Articles by Casanovas, M.
Agricola
Right arrow Articles by Andrade, F. H.
Right arrow Articles by Casanovas, M.
Related Collections
Right arrow Maize
Crop Science 42:1173-1179 (2002)
© 2002 Crop Science Society of America

CROP PHYSIOLOGY & METABOLISM

Kernel Number Prediction in Maize under Nitrogen or Water Stress

F. H. Andrade*, L. Echarte, R. Rizzalli, A. Della Maggiora and M. Casanovas

Unidad Integrada INTA Balcarce, Facultad de Ciencias Agrarias UNMP. CC276, 7620 Balcarce, Buenos Aires, Argentina

* Corresponding author (fandrade{at}balcarce.inta.gov.ar)


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Final kernel number in maize (Zea mays L.) is closely associated with the physiological condition of the crop during the critical period bracketing silking. The objective of this study was to determine whether there is a common underlying relationship between kernel number per plant (KNP) and plant growth rate (PGR) during that critical period when plant growth varies because of different abiotic stresses. A relationship between KNP and PGR obtained from a previous study of variation in plant density and incident radiation was used as reference. KNP and PGR were measured in experiments in which incident radiation per plant, nitrogen (N), and water availabilities were the experimental sources of variation. The equation fitted to the data obtained at different radiation levels explained 72% of the variation in the data obtained at different levels of N or water availability. Moreover, different models for each set of data did not provide a significantly better fit than a single model for the two sets of data combined. A common relationship between KNP and PGR was also obtained when N supply and water availability were variable. The relationship between KNP and PGR obtained for treatments in which PGR was varied through plant density and shading also could predict KNP for conditions in which PGR was affected by water and/or N deficiencies. The PGR during the critical period of kernel set is a good predictor of the capacity of the maize plant to set kernels under a wide range of environmental and management practices.


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
MAIZE GRAIN YIELD is closely associated with kernel number at harvest and this yield component is a function of the physiological condition of the crop during the period bracketing flowering (Tollenaar, 1977; Edmeades and Daynard, 1979; Kiniry and Ritchie, 1985; Jacobs and Pearson, 1991; Otegui and Andrade, 2000).

Early kernel growth and development in maize is highly dependent on assimilate supply from concurrent photosynthesis (Boyle et al., 1991; Schussler and Westgate, 1991, Zinselmeier et al., 2000). This is probably the reason for the close relationship between kernel number per plant (KNP) and plant growth rate during the period bracketing flowering (PGR) reported by several authors (Tollenaar et al., 1992; Kiniry et al., 1997; Andrade et al., 1999).

In previous work, linear or curvilinear relationships between KNP and PGR were reported for maize grown without water or nutrient deficiencies (Tollenaar et al., 1992; Kiniry and Knievel, 1995; Andrade et al., 1999). Andrade et al. (1999) reported that PGR at flowering was a good predictor of KNP when plant growth rate was affected by plant density, plant-to-plant variability, incident radiation or year effects. When PGR was expressed per unit thermal time, it also predicted the effects of variable night temperature on KNP.

Much of the effect of water and nitrogen deficiencies on crops can be explained through radiation interception and radiation use efficiency that relate directly to carbon assimilation and plant growth (Boyer, 1970; Gifford et al., 1984; Uhart and Andrade, 1995a). Moreover, much of the seed loss at low water potential can be accounted for by the lack of assimilate supply at flowering (Boyle et al., 1991; Schussler and Westgate, 1991). Thus, drought or nitrogen deficiency-induced effects on kernel number could be related to photosynthesis or plant growth at flowering.

Other mechanisms, however, appear to regulate the abortion process in maize, such as (i) direct effects of water and nitrogen deficiencies (Schussler and Westgate, 1991; Zinselmeier et al., 1995b; Below et al., 2000) and (ii) stress induced changes in rates of hormone synthesis, flux, and/or turnover (Morris, 1996; Jones and Setter, 2000). These alternative effects may work in parallel or in series with the assimilate supply in determining kernel number under nitrogen or water stress conditions. The hypotheses of this work were that (i) carbon assimilation is an acceptable predictor of kernel number fixation when nitrogen or water availability is variable and (ii) a curvilinear relationship between KNP and PGR obtained by varying PGR at flowering through plant density and incident radiation is valid when PGR at flowering varies because of changes in nitrogen or water availability.

The objective of this study was to examine the relationship between KNP and PGR during a period bracketing silking when plant growth was modified by nitrogen or water availability, and to compare it with that obtained when plant growth was modified through variations in incident radiation per plant without nutrient or water deficiencies. The goal was to explore whether there was a common relationship between KNP and PGR when PGR was varied by nitrogen availability, water deficits, shading, or plant density. This research would benefit modelers to predict KNP under conditions of abiotic stress.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
The data were obtained from 11 experiments conducted at the Instituto Nacional de Tecnología Agropecuaria Balcarce Experimental Station (37° 45' S, 58° 18' W, 130 m alt.) (Tables 1 and 2). Data on a per plant basis in five of these experiments (1, 2, 3, 6a, and 6b) were derived from studies already published (Otegui et al., 1995; Uhart and Andrade, 1995b). The soil was a Typic Argiudoll with a depth of 1.5 m and with an organic matter content of approximately 56 g kg-1 in the first 25 cm of depth. The area is characterized by low average temperatures during the growing season (17.8°C) and a frost-free period of about 150 d. More details about the climate of the Balcarce region were presented in Andrade (1995). Maize was sown in mid-October. Plant density at harvest was 7.5 to 9.5 plants m-2. The experiments were each conducted with three to four replications. The size of the plots was at least four rows 0.70 m apart and 12 to 15 m long. Plots were fertilized with 30 kg P ha-1, with some variations among experiments, to provide adequate P nutrition. Weeds and insects were adequately controlled.


View this table:
[in this window]
[in a new window]
  		Table 1. Experiments 1 to 4 in which incident radiation and N or water availabilities were variable. Hybrid Dk 636 was sown in October at optimal plant densities. Relative values of N and water use are shown. R = incident radiation; N = nitrogen availabilities; W = water availabilities. Radiation levels changed in plots with optimum N and water availabilities; N availability varied in nonshaded plots without water deficiencies and water availability varied in nonshaded plots without N deficiencies.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Experiments 5 to 8 in which N and/or water availabilities were variable. Crops were sown in October at optimal plant densities. Relative values of N uptake and water use are shown. N = nitrogen availability; W = water availability.

 
In Exp. 1, 2, 3, and 4, incident radiation and N or water availability were variable (Table 1) and only one hybrid (Dekalb 636) was used. For the data used herein, radiation was varied in plots supplied with optimum N and water; and N and water availability was varied in nonshaded plots. Incident radiation was modified during a period bracketing flowering (from approximately 1–2 wk before to 2–3 wk after flowering). Plots were shaded with black synthetic cloth of differing mesh, stretched above the crop on frame and wire structures. Relative values of incident radiation ranged from 1 (nonshaded control) to 0.45 or 0.55 according to the experiment (Table 1). In Exp. 1, 2 and 3, N availability varied in nonshaded plots and soil water was kept at more than 50% of maximum soil available water during the entire growing season by irrigation (Table 1). The treatments without N deficiency received 180 kg N ha-1 (Exp. 1 and 2) or 6 g N plant-1 (Exp. 3) and the relative N uptake until 3 wk after flowering for these treatments is indicated as 1 in Table 1. Within each experiment, N uptake for the treatments with less N availability because of reduced N fertilization rate or N immobilization was expressed as relative to that shown by the treatment without N deficiencies. More details about these experiments can be found in Uhart and Andrade (1995a)(b). In Exp. 4, two water availability levels were imposed on nonshade plots without N deficiencies: a control and a severe water stress treatment. Drip irrigation was applied to the control treatment to keep plant available soil water at more than 50% of its maximum in the first meter of depth during the entire growing season. The severe water stress treatment during the period bracketing flowering was imposed by limiting irrigation. The growing season was particularly dry during December-January. The treatments applied in this experiment resulted in the relative values of water availability shown in Table 1.

Data from previous work with the same hybrid, in which PGR varied because of changes in plant density or radiation levels (Andrade et al., 1999), were used as a reference and are means of three or four replications. Plant density varied from 2.1 to 16.9 plants m-2 and shading level from 0 to 55%.

In Exp. 5 (a, b, and c), 6 (a and b), 7, and 8, only N or water availability was varied and shading treatments were not applied. Hybrids used in these experiments are indicated in Table 2. These hybrids are similar in cycle length and are all well adapted to the Balcarce area. Exp. 5a, b, and c were conducted without irrigation in different years and consisted of a combination of tillage (no till vs conventional tillage) and two levels of N fertilization (0 and 120–200 kg N ha-1) that resulted in the relative N and water use values indicated in Table 2. In these experiments, plots were limited by N availability, by water availability, or by both. Within each experiment, relative N uptake until 3 wk after flowering for the treatment corresponding to conventional tillage with fertilization was maximum and is indicated as 1 in Table 2. Nitrogen uptake for other treatments was expressed relative to this value. Average evapotranspiration for the period bracketing flowering was greatest in Exp. 5b (4.4 mm d-1), intermediate in Exp. 5a (4.0 mm d-1), and least in Exp. 5c (2.0 mm d-1). Water use values during the period bracketing flowering were expressed relative to the maximum water use observed in Exp. 5b. This last experiment received the greatest amount of precipitation during that period. In Exp. 6a, 6b, 7, and 8, different water regimes were imposed during the period bracketing flowering and N was not limiting growth. In control treatments, soil water was kept over 50% of maximum plant available water in the first meter of depth during the entire growing season by drip (Exp. 7 and 8) or furrow (Exp. 6a and b) irrigation. Drought treatments were generated by placing black plastic covers on the ground to prevent moisture from entering the soil. A period of limited water supply was generated from approximately 20 d before to 20 d after silking. In Exp. 7 and 8, intermediate levels of water deficiencies were achieved by irrigating a fraction of the amount provided to the control plots. Water use values during the period bracketing flowering were expressed relative to the treatment with the greatest water use within each experiment (Table 2). More details about Exp. 6 were presented in Otegui et al. (1995).

Experiments 3 and 7 were conducted with plants growing in pots (soil volume = 0.020 m3) that were distributed in the field simulating a normal crop stand. Pots were filled with a mixture of sand and soil and they were partially buried in the soil. Adequate mineral nutrition (except N in Exp. 3) was provided.

Aboveground dry matter was measured approximately 2 wk before and 2 to 3 wk after silking by taking samples of 8 to 10 plants from the central rows (approximately 1 m2), leaving borders between adjacent harvests. The samples were separated in stems+sheaths+tassels, leaves and ear (when visible), oven-dried (with air circulating at 60°C) to constant weight, weighed, and ground to pass a 1-mm mesh screen. These data were used to calculate mean PGR during the period bracketing flowering. PGR values were also expressed per unit thermal time (growth rate divided by t - tb, in which t is mean day temperature and tb is the base temperature for maize, 8°C) to account for possible differences in the duration of the critical period for grain number determination. Mean temperature for the period bracketing flowering was 20.5 ± 0.8°C.

Grain and total dry matter yields (dry weight basis) were determined at physiological maturity by hand harvesting all the ears from 7.15-m length of the two center rows of the plot (10 m2). Individual kernel weight was determined for each experimental unit by weighing two representative samples of 500 kernels each. The number of kernels per plant was calculated on the basis of grain yield, individual kernel weight, and plant density and related to PGR.

When nitrogen availability was variable, N concentration in plant tissues was determined following Method A (without salicylic acid) reported by Nelson and Sommers (1973). Nitrogen content in each plant part was determined as the product of N concentration (on a dry weight basis) and dry weight. Total N uptake was calculated as the sum of the N contents of the different plant parts. N uptake until 3 wk after flowering was calculated for all treatments.

When water availability was a treatment variable, soil water content was measured weekly, gravimetrically in the upper soil layer (0–0.1 or 0.3-m depth) and with a calibrated neutron probe (Troxler 103A or Troxler 4300, Troxler Electronic Lab., Research Triangle Park, NC) between 0.1 or 0.3 m and maximum soil depth explored by the roots (except in Exp. 7). Actual crop water use was calculated with a hydrological balance model (Dardanelli et al., 1991). In Exp. 7, water use values were calculated as the amount of irrigation provided to each pot.

Relative values of KNP were used in Exp. 5a, 5b, 5c, 6a, 6b, 7, and 8 to eliminate differences in potential kernel number among the different hybrids. In the experiments conducted with hybrid Ax 777 (5a, b, and c), a value of 1 for KNP was given to the N fertilized treatment under conventional tillage in the wetter year (control treatment). Other values were expressed relative to this control. In the experiments conducted with hybrids SPS 240 (6a and b), Dekalb 636 (7) and Dekalb 639 (8), a value of 1 for KNP was given to the irrigated control treatments. Within each experiment, the value of KNP for the rest of the treatments was expressed relative to their respective control treatments. The reference data in which PGR varied because of plant density, shading, and year effects (Andrade et al., 1999) were also expressed as relative values, and a value of 1 for KNP was given to the average PGR value shown by the control treatment (8.5 plants m-2, no shading, October planting).

Data from each experiment were analyzed by ANOVA procedures. Appropriate standard errors of the means were calculated. Inverse (KNP = a + b/PGR) and rectangular hyperbola {KNP = a(PGR - b)/[1 + c(PGR - b)]} equations were fitted (Jandel Tablecurve, 1992) to the data from Exp. 1 to 4, to the data from Exp. 5a, b, and c, to the data from Exp. 5 to 8 (with KNP expressed in relative values), and to the reference data (with KNP expressed both in relative and absolute values). The PGR values were expressed per unit time and per unit thermal time. In Exp. 1 to 4, different equations were fitted to data obtained with variable N/water supply and to data obtained with variable incident radiation. In Exp. 5a to c, different equations were fitted to data obtained with variable N, to data obtained with variable water supply, and to data obtained with variable N and water supply. Appropriate standard errors of the estimates and prediction intervals were calculated. Parameters of the equations were compared by t tests. The variation produced by the N/water treatments that was explained by the equation fitted to data obtained with variable incident radiation was calculated for Exp. 1 to 4. A similar analysis was done for the sources of variation in Exp. 5 a, b, and c. Similarly, the variation produced by the N and/or water treatments that was explained by the reference equation was calculated for Exp. 1 to 4 and for Exp. 5 to 8. In the reference equation, predicted KNP was taken as 0 for PGR lower than the threshold PGR value for kernel set.

A separate analysis was performed on data from Exp. 1 to 4. First, an equation was fitted to all KNP-PGR data from these experiments (reduced model or single-equation fit) and the residual sum of squares was calculated. Second, two equations were fitted, one to data obtained with variable N/water and the other to data obtained with variable incident radiation (full model or two-equation fit) and their respective residual sums of squares were calculated and added. Finally, a F-test was performed to determine the adequacy of the single-equation fit to explain variation in KNP compared to the two-equation fit (Gallant, 1987). The test was

where q is the difference in number of parameters between the two models, p is the number of parameters in the full model, and n the total number of observations. The numerator indicates how much the error was reduced for each parameter added to the more complex model. The denominator is an estimator of the error variance.

A similar analysis (reduced model vs. full model) was performed to compare the effects of nitrogen and water deficiencies in Exp. 5a, b, and c.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
The relationship between KNP (uppermost ear) and PGR obtained in previous studies with hybrid Dekalb 636 was used as a reference data set (Fig. 1) . Parameters of the equations fitted to these data were a = 594 ± 17 and b = -753 ± 48 (inverse equation) and a = 412 ± 96, b = 1.17 ± 0.2 and c = 0.67 ± 0.2 (rectangular hyperbola equation). This curvilinear relationship explained kernel number per uppermost ear when plant growth rate varied because of plant density, incident radiation, or year effects (Andrade et al., 1999). The general features of this relationship are (i) kernel number per plant responded to increases in plant growth, (ii) at PGR greater than 4 g d-1, increases in growth rates produced only small increments in kernels per uppermost ear, and (iii) a threshold value of PGR of approximately 1 g d-1 for kernel set. Similar relationships between KNP and PGR were reported by Edmeades and Daynard (1979) and Tollenaar et al. (1992).



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 1. Relationship between kernel number per uppermost ear and plant growth rate during a period bracketing silking for maize plants (Dekalb hybrid 636) growing without nutrient nor water limitations, in which plant growth rate was modified by plant density, year and shading level. Values are means of 3 or 4 replications. The inverse equation (solid line) was KNP = 594-753/PGR (R2 = 0.88, n = 34). See Andrade et al. (1999) for more details.

 
The KNP and PGR values obtained in Exp. 1 to 4 (variable incident radiation and variable N or water availability) are shown in Fig. 2 . Shading and N or water deficiencies significantly reduced both KNP and PGR. An inverse or rectangular hyperbola model accurately described the relationship between KNP and PGR when N and water availabilities were variable or when incident radiation was variable (Fig. 2). For these experiments, the inverse or the rectangular hyperbola equations fitted to the data obtained at a range of radiation levels explained 72% of the variation in the data obtained at different levels of N or water availability. Moreover, two different equations, one for data with variable N or water and the other for data with variable incident radiation (full model or two-equation fit) did not provide a significantly better fit than a single equation for all the data combined (reduced model or single-equation fit) (F = 0.51; P > 0.5). Accordingly, the parameters of the equations that fit the data obtained at different levels of N or water availability did not significantly differ (P > 0.05) from those that fit the data obtained at different radiation levels. For example, parameters a and b of the inverse equation were 539 ± 37 and -517 ± 79 for data obtained at different levels of N or water availability and 598 ± 48 and -757 ± 130 for the data obtained at different radiation levels. When data from these experiments were expressed on a per unit ground area basis, variable radiation and variable N also produced similar effects on the relationship between kernel number and growth (Uhart and Andrade, 1995b).



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 2. Relationship between kernel number per plant and plant growth rate during a 30-d period bracketing silking for maize hybrid Dekalb 636 growing under (i) variable incident radiation at optimum N and water availabilities and (ii) variable N or water availabilities without shading, in Exp. 1 to 4. Plants did not show prolificacy, so kernel number per plant was equal to kernel number per uppermost ear. Each point is the average of three replications. When N or water were variable, best fit was given by the rectangular hyperbola equation: KNP = 375(PGR-0.82)/[(1 + 0.58(PGR-0.82)] (R2 = 0.83, n = 12). When radiation was variable, best fit was given by the inverse equation: KNP = 598-757/PGR (R2 = 0.82, n = 10). Reference equation from Fig. 1 is also shown. Standard error of the means were 0.22, 0.16, 0.11, and 0.30 g d-1 for PGR and 17, 19, 20, and 28 kernels for KNP, for Exp. 1, 2, 3, and 4, respectively.

 
The equations fitted to the reference data explained 74% (P > 0.01) of the variation observed when N and water were variable in Exp. 1 to 4. With only one exception, the parameters of the inverse and rectangular hyperbola equations fitted to the data of these experiments did not differ (P > 0.05) from those obtained for the reference data, even though the range of the x variable was quite different. Similar results were obtained when only the data from Exp. 1, 2, and 3 (variable N and variable radiation) were considered.

In Exp. 5a, 5b, and 5c, both N and water availability were variable. Inverse or rectangular hyperbola models described accurately the relationship between KNP and PGR, and the effect of reducing N availability was similar to the effect of reducing water availability or both N and water availability (Fig. 3) . The inverse equation fitted to the data obtained at different levels of water availability explained 93% of the variation in the data obtained at different levels of N supply. Similarly, the inverse equation fitted to the data obtained at different levels of N supply explained 86% of the variation in the data obtained at different levels of water availability. Moreover, two different equations, one for data with variable N and the other for data with variable water availability (full model or two equation fit) did not provide a significantly better fit than a single equation for the data combined (reduced model or single equation fit) (F = 0,43; P > 0.5). The parameters of the equation that fit the data for variable N did not statistically differ (P > 0.05) from those of the equation that fit the data for variable water or for variable N and water. Parameters a and b of the inverse equation were 667 ± 19 and -705 ± 53 for data obtained with variable N supply; 699 ± 41 and -837 ± 124 for data obtained with variable water availability; and 699 ± 42 and -862 ± 101 for data obtained with variable N and water availability.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 3. Relationship between kernel number per plant and plant growth rate during a 30-d period bracketing silking for maize hybrid Asgrow 777 growing with and without fertilization in three growing seasons with different water availability during the period bracketing silking (Exp. 5a, 5b, and 5c). Plants did not show prolificacy, so kernel number per plant was equal to kernel number per uppermost ear. Each point is the average of four replications. The control corresponds to fertilized treatments in the wettest year (Exp. 5b). Reduced N corresponds to non fertilized treatments in the wettest year (Exp. 5b). Reduced W corresponds to fertilized treatments in the years with lowest and intermediate available water (Exp. 5a and 5c). Reduced N and W corresponds to non fertilized treatments in the years with lowest and intermediate available water (Exp. 5a and 5c). Inverse equations were KNP = 667-705/PGR (R2 = 0.99, n = 4) for variable N at high water availability; KNP = 699-837/PGR (R2 = 0.92, n = 6) for variable available water in fertilized treatments and KNP = 699-862/PGR (R2 = 0.95, n = 6) for variable water and N supply. Standard error of the means varied from 0.11 to 0.26 g d-1 for PGR and from 15 to 16 kernels for KNP.

 
The KNP-PGR data from Exp. 5 to 8, in which N and water availabilities were variable, are shown in Fig. 4a . The KNP data were expressed as relative values within a hybrid to exclude differences in potential kernel number among the hybrids. These data were accurately described by the inverse or rectangular hyperbola equations fitted to the reference data also expressed as relative values. The reference equations (in relative values) explained 88% of the variation observed when N and water were variable in Exp. 5 through 8. Moreover, the values obtained from these experiments fell within the 95% prediction interval of the equation that fit the reference data. The PGR, expressed per unit thermal time, has been shown to be a good predictor of KNP (Andrade et al., 2000). However, the relationship was not improved when PGR data were expressed on a thermal time basis (Fig. 4b). This is probably because average temperature during the period bracketing silking varied less than 2°C among the different experiments presented in Fig. 4. When KNP was expressed as relative values and PGR in thermal time units, the reference equations fitted to the reference data explained 73% of the variation observed when N and water were variable in Exp. 5 through 8 (Fig. 4b). Moreover, the values obtained from these experiments fell within the 95% prediction interval of the equation that fit the reference data (Fig. 4b).



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 4. Relationship between kernel number per plant (expressed as relative values) and plant growth rate during a 30-d period bracketing silking for maize hybrids Asgrow 777 (Exp. 5), SPS 240 (Exp. 6), Dekalb 636 (Exp. 7), and Dekalb 639 (Exp. 8) growing under variable levels of nitrogen or water availabilities. KNP values for each hybrid are expressed relative to those obtained for the control without deficiencies. PGR is expressed per unit time (a) or per unit thermal time (b). Each point is the average of three replications. Plants did not show prolificacy, so kernel number per plant was equal to kernel number per uppermost ear. The reference data from Fig. 1 (with KNP data also expressed as relative values), the equation fitted to these data (solid line) and its 95% prediction interval (dashed lines) are also shown. KNP values corresponding to a relative value of 1 were 520, 385, 465, and 600 for Exp. 5 (a, b, and c), 6 (a and b), 7, and 8, respectively. SE of the means in absolute values ranged from 0.11 to 0.40 g d-1 for PGR and from 4 to 16 kernels for KNP. Equations fitted to reference data (n = 34) were KNP = 1.0(PGR-1.17)/[1+0.67*(PGR-1.17)] (R2 = 0.88) for Fig. 4a; and KNP = 8.97(PGR-0.06)/[1+5.36(PGR-0.06)] (R2 = 0.83) for Fig. 4b.

 
These results show that (i) PGR was a good predictor of KNP when nitrogen and water supply were variable; (ii) the relationship between KNP and PGR found when plant growth was modified by nitrogen and/or water supply was similar to that obtained when plant growth was changed by variations in plant density and incident radiation, and (iii) the effect of reducing N availability was similar to the effect of reducing water availability. Thus, the objective of this study was accomplished since a common relationship between KNP and PGR could be used to predict the effect of water deficits, nitrogen stress, plant density, and incident radiation on kernel set.

The PGR during flowering is a good predictor of KNP because it is correlated to growth of reproductive structures (Andrade et al., 1999) and because early seed development and kernel set in maize appears to be highly dependent on a continued supply of assimilates from concurrent photosynthesis (Zinselmeier et al., 2000). Stress induced abortion is at least partially mediated by assimilate flux to reproductive structures at flowering (Boyle et al., 1991; Schussler and Westgate, 1991), which is correlated with total plant assimilate production. Supporting this statement, studies of stress-induced kernel abortion in maize show that exogenous carbohydrate supply and short-term reserves in young ovules are crucial to kernel set in conditions of low water availabilities (Reed and Singletary, 1989; Boyle et al., 1991; Zinselmeier et al., 1995a).

Direct effects of N or water deficiencies on spikelet growth and kernel set have been reported. Below et al. (2000) suggested that N has a direct role in reproductive development by controlling the ability of the kernel to utilize carbon. Similarly, Schussler and Westgate (1991) and Zinselmeier et al. (1995b) found evidence of direct effects of drought on kernel set since low ovary water potential affected dry matter partitioning to the ear by reducing ovary sink strength. These direct effects of stress on kernel set are related to reductions in dry matter partitioning to the ear and/or in the number of grains set per unit of dry matter allocated to the ear during the critical period for grain number determination. The response of kernel number to drought or nutritional-induced changes in PGR, however, was similar to the response of kernel number to radiation or plant density-induced changes in PGR. The errors in the method used to estimate PGR may have made it difficult to detect differences among relationships established from different abiotic stresses. Nevertheless, these results suggest that direct effects of water and nitrogen deficiencies on kernel number are relatively small and/or correlated with the effect of stress on PGR.

Stress induced changes in rates of hormone synthesis, flux, and/or turnover appear to regulate kernel set (Jones and Setter, 2000). These processes, however, appear to link a decrease in assimilate supply with the developmental response leading to kernel abortion. Moreover, such a decrease affects sugar sensing systems that initiate changes in gene expression leading to enhanced C conservation when sugar supply is limited or C utilization when sugars are abundant (Koch, 1996). Thus, the signaling system would be related to carbon assimilation and assimilate flux to the reproductive structures, supporting the value of PGR as a predictor of KNP.

The relationship presented in Fig. 4 corresponds to different modern hybrids that are similar in cycle length and are well adapted to the Balcarce area. Relative values of KNP were used to eliminate differences in potential kernel number among the different hybrids. However, this may not be the only difference among hybrids in the underlying relationship. Studies that compared modern and old hybrids showed different relationships between KNP and PGR (Tollenaar et al., 1992). Modern hybrids showed lower PGR threshold for kernel set, higher potential kernel number, and a lower curvature than older hybrids (Echarte et al., 2001). Moreover, stress-tolerant hybrids may set more grains per unit PGR under stress conditions than less tolerant hybrids. These aspects must be taken into account for accurate predictions of kernel number.

In conclusion, PGR during a period bracketing silking, taken as an indicator of the amount of carbon available to the plant, is a good predictor of the capacity of the maize plant to set kernels under a wide range of environmental and management practices.


   ACKNOWLEDGMENTS
 
Thanks go to A. Cirilo, S. Uhart and M. Otegui who provided part of the data used in this work.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
This work was supported by Instituto Nacional de Tecnología Agropecuaria (INTA); Facultad de Ciencias Agrarias Univ. Nac. de Mar del Plata; CONICET; and Monsanto Argentina.

Received for publication December 20, 2000.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 Crop Sci.Home page
M. Coque and A. Gallais
Genetic Variation for Nitrogen Remobilization and Postsilking Nitrogen Uptake in Maize Recombinant Inbred Lines: Heritabilities and Correlations among Traits
Crop Sci., September 1, 2007; 47(5): 1787 - 1796.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
T. Sarlangue, F. H. Andrade, P. A. Calvino, and L. C. Purcell
Why Do Maize Hybrids Respond Differently to Variations in Plant Density?
Agron. J., June 5, 2007; 99(4): 984 - 991.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
F. E. Miguez and G. A. Bollero
Winter Cover Crops in Illinois: Evaluation of Ecophysiological Characteristics of Corn
Crop Sci., May 18, 2006; 46(4): 1536 - 1545.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
K. E. D'Andrea, M. E. Otegui, A. G. Cirilo, and G. Eyherabide
Genotypic Variability in Morphological and Physiological Traits among Maize Inbred Lines--Nitrogen Responses
Crop Sci., April 25, 2006; 46(3): 1266 - 1276.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
L. Echarte and M. Tollenaar
Kernel Set in Maize Hybrids and Their Inbred Lines Exposed to Stress
Crop Sci., February 24, 2006; 46(2): 870 - 878.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
P. M. O'Neill, J. F. Shanahan, and J. S. Schepers
Use of Chlorophyll Fluorescence Assessments to Differentiate Corn Hybrid Response To Variable Water Conditions
Crop Sci., February 1, 2006; 46(2): 681 - 687.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
P. Monneveux, C. Sanchez, D. Beck, and G. O. Edmeades
Drought Tolerance Improvement in Tropical Maize Source Populations: Evidence of Progress
Crop Sci., December 2, 2005; 46(1): 180 - 191.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
P. M. O'Neill, J. F. Shanahan, J. S. Schepers, and B. Caldwell
Agronomic Responses of Corn Hybrids from Different Eras to Deficit and Adequate Levels of Water and Nitrogen
Agron. J., November 1, 2004; 96(6): 1660 - 1667.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
L. Echarte, F. H. Andrade, C. R. C. Vega, and M. Tollenaar
Kernel Number Determination in Argentinean Maize Hybrids Released between 1965 and 1993
Crop Sci., September 1, 2004; 44(5): 1654 - 1661.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (27)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Andrade, F. H.
Right arrow Articles by Casanovas, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Andrade, F. H.
Right arrow Articles by Casanovas, M.
Agricola
Right arrow Articles by Andrade, F. H.
Right arrow Articles by Casanovas, M.
Related Collections
Right arrow Maize

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
Environmental Quality		 The Plant Genome