Agronomy Journal Grow Your Career With ASA
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 ISI 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 ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cárcova, J.
Right arrow Articles by Ghersa, C. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Cárcova, J.
Right arrow Articles by Ghersa, C. M.
Agricola
Right arrow Articles by Cárcova, J.
Right arrow Articles by Ghersa, C. M.
Related Collections
Right arrow Crop Growth and Development
Right arrow Maize
Right arrow Root Development
Right arrow Maize Management
Agronomy Journal 92:1256-1265 (2000)
© 2000 American Society of Agronomy

CORN

Long-Term Cropping Effects on Maize

Crop Evapotranspiration and Grain Yield

Jorgelina Cárcovaa, Gustavo A. Maddonnia and Claudio M. Ghersab

a Cátedra de Cerealicultura, Universidad de Buenos Aires, Av. San Martín 4453 (1417), Buenos Aires, Argentina
b Cátedra de Ecología Vegetal, Facultad de Agronomía, Universidad de Buenos Aires, Av. San Martín 4453 (1417), Buenos Aires, Argentina

jcarcova{at}agro.uba.ar


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 
Long-term cropping in the Rolling Pampa of Argentina may reduce available soil water for maize (Zea mays L.) production. Grain yield, however, has not decreased possibly because of genetic improvement. Our objectives were to: (i) evaluate root and shoot growth and its relation to crop evapotranspiration (ETa), (ii) analyze interactions among hybrids and length of the cropping period on grain yield and its components, and (iii) establish functional relations between grain yield components and their determinant processes. Three hybrids with contrasting production stability coefficients and representing different eras of maize breeding were grown at three farms in 1994 to 1996 in fields with short (S) and long (L) cropping periods on silty clay loam soils (fine, illitic, thermic Typic Argiudolls). Silking took place 5 to 6 d later in L than in S and a significant (P < 0.05) reduction (21–36%) of ETa was observed in L vs. S around this stage. Lower root abundance (reduction of 60%) and canopy size (reduction of 6–13%) in L have probably contributed to less ETa. In addition, the ontogenic delay exposed crops grown in L to more stressful meteorological conditions around silking. Kernel number (KN), closely related to grain yield (r2 > 0.65), was significantly associated with daily ETa around silking. The modern hybrid established a greater KN mm-1 d of ETa (853 kernels m-2 mm-1 d) compared with the older ones (386 kernels m-2 mm-1 d). This characteristic may mask the deleterious effects of long-term cropping on maize grain yield.

Abbreviations: ASW, available soil water • CO, organic carbon • ETa, actual crop evapotranspiration • ETo, reference evapotranspiration • fIPAR, fraction of incident PAR intercepted by the canopy • KN, kernel number • KW, kernel weight • L, field with a long cropping period • NT, total nitrogen • PAR, photosynthetically active radiation • PWG k-1, plant weight gain per kernel • S, field with a short cropping period • SS, index of relative soil aggregate stability • SWC, soil water content


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 
THE Rolling Pampa of Argentina (Hall et al., 1992) is the most fertile area within the largest area of Mollisols of the Southern Hemisphere (Soriano et al., 1991). This region is one of the most important temperate crop production areas of South America. Crop rotation is a common practice and before 1970 crop sequences differed with farm size. On small farms (50–200 ha), wheat (Triticum aestivum)–wheat–wheat/sunflower (Helianthus annus L.) was the dominant crop sequence. Beef cattle (Bos sp.) frequently grazed the crop stubble. On large farms (>200 ha), mixed pasture–maize–maize–maize–wheat/sunflower was a frequently used crop sequence and 20% of land was in leys for 2 to 10 yr. In the last 20 yr, a dramatic expansion of soybean [Glycine max (L.) Merr.] production took place and the most frequent crop sequence now is maize–wheat/soybean–maize. Together with the expansion of soybean, there was a steady increase in the intensity of land use, and cattle grazing has nearly disappeared on crop farms (Ghersa and Martínez Ghersa, 1991). This intensified cropping has led to a 60% reduction in the organic matter content of most arable soils, with the concomitant decline in favorable chemical and physical properties (Michelena et al., 1989).

Maize production in the Rolling Pampa is conducted predominantly under dryland conditions. Crop evapotranspiration during a significant proportion of the growing cycle depends on stored soil water and on the capacity of the root system to absorb it (Hall et al., 1992). Severe degradation of soil aggregation reduces infiltration rates (Musto, 1979) and can result in decreased available soil water. In soil profiles without physical constraints, root systems penetrate rapidly. Their ability to grow into inhospitable layers, however, could limit water absorption (Passioura, 1982). Topsoil compaction imposed experimentally affected root growth and distribution resulting in reduced ETa (de Willigen and Van Noordwijk, 1987; Tardieu et al., 1992). Our hypothesis is that a long period of continuous cropping in soils with similar genesis results in changes in water availability for maize crops.

Maize grain yield is mainly related to KN (Bolaños and Edmeades, 1993, 1996; Cirilo and Andrade, 1994; Otegui, 1995; Otegui et al., 1995b). Kernel weight (KW), however, also contributes to grain yield variability, depending on the hybrid analyzed (Otegui, 1995). A reduction of 4.7 kernels m-2 for each mm reduction in ETa around silking, was reported by Otegui et al. (1995a) for one maize hybrid in the Southern Pampa of Argentina (37° to 39° S, 57° to 63° W). Despite the hypothesized reduction in ETa by long-term cropping and its expected impact on maize grain yield, Presello et al. (1997) reported a steady increase of maize grain yield in the last four decades for the region under study. Mella et al. (1984) attributed 79% of grain yield increase to genetic improvement of Argentine maize hybrids. Moreover, comparisons of grain yield of hybrids released in the last four decades in Argentina, revealed that modern hybrids have a greater response to favorable environments (high production stability coefficient; Eberhart and Russell, 1966) and had the highest mean grain yield (average of the tested environments) (Mella et al., 1984). Thus, genetic improvement may partially compensate for deleterious effects of long-term cropping on soil chemical and physical conditions. Our second hypothesis is that the effect of long-term cropping on maize grain yield depends on the specific hybrid considered.

The objectives of this paper were to (i) evaluate root and shoot growth and its relation to ETa, (ii) analyze interactions among hybrids and length of the cropping period on grain yield and its components, and (iii) establish functional relations between grain yield components and their determinant processes (e.g., ETa and dry matter accumulation).


   Materials and methods
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
Results and discussion
 Conclusions
 REFERENCES
 
Experiments were conducted in the main maize production area of Argentina (32° to 35°S, 58° to 62°W) on silty clay loam soils (fine, illitic, thermic Typic Argiudolls) from 1994 through 1996. Three farms located within 30 km of each other were selected. Cropping history (last 30 yr) and soil properties for more than 100 fields from the three farms were compared. Soil properties considered were: organic carbon (Co); total nitrogen (NT); an index of relative soil aggregate stability (SS; aggregate size distribution of a soil expressed as percentage of a pristine one); mineral nitrogen (N–NO-3); labile carbon, and depth to the B2t horizon.

Based on the analysis of available information, fields with similar soil type but different length of cropping were classified at each farm in two groups. Cropping intensity and soil aggregation were the criteria used to constitute the groups (Fig. 1) . Thresholds for each criterion were fixed based on its representation within each farm. Soil aggregate stability was included as a criterion because it was the only soil property significantly related to the duration of the current cropping period . Thus, one group had fields with a short cropping period (S; <6 yr since the last pasture, <20 yr of harvested crops during the past 30 yr) and a Ss >30%. The other group had fields with a longer cropping period (L; >13 yr since the last pasture, >24 yr of harvested crops during the last 30 yr) and a Ss of <25 %.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 1 Criteria used to classify the length of the cropping period of a field as short (S) or long (L) based on its agricultural history and soil aggregate stability (Ss). Cumulative probabilities of (a) amount of fields with n years of harvested crops in the last 30 yr, and (b) amount of fields with n years since the last pasture. Fig. 1c represents Ss as a function of time. Zero indicates the beginning of a cropping period. Fields above thresholds indicated by the dotted line and included in the box defined by dotted lines in Fig. 1c were classified as L fields, and fields below thresholds indicated by the solid line and contained within the box defined by solid lines represent S fields. Data set corresponds to Farm 3

 
Cropping history was considered a classification factor, i.e., an intrinsic property of the fields and not something assigned by the experimenters (Hurlbert, 1984). Thus, among fields of each group (S and L), one of them was randomly assigned at each farm. A split-plot (randomized complete block with split subtreatments) design with three replicates (farms) was used with the length of cropping (S vs. L) as the main factor and hybrids (DK4F31, DK4F37, and DK752) as subfactors. In each field, hybrids were planted in a randomized complete block design experiment with three replicates (subplots). Hybrids differed in production stability coefficients (responsiveness to different environments, Eberhart and Russell, 1966) and were released in different eras (DK4F31: double-cross, stability coefficient 0.8, 1975 release; DK4F37: double-cross, stability coefficient 1.0, 1987 release; DK752: single-cross, stability coefficient 1.1, 1993 release). A higher value of the stability coefficient indicates more adaptability and less stability as a response to a higher quality environment. During the 3- to 5-yr period after release, >80% of the land used for maize production in Argentina was planted with the selected hybrids. In late October, hybrids were sown in a representative area (350 m2) of each field. Each subplot consisted of five rows, 0.7-m apart and 5 m long. Experiments were over-sown and thinned to 6 to 7 plants m-2 at the 3-leaf stage. The experiments were conducted under rainfed conditions with no fertilizer application. Fertilizer usage for maize production has been relatively restricted during the decades under analyses (Hall et al., 1992).

Soil Measurements
Physical and chemical soil properties were determined each year at the three farms in the two fields selected (S and L). A nondisturbed site with the same soil type on each farm was also sampled to characterize the pristine condition. This site was generally placed below the wire netting between fields. Samples (20-cm increments) were taken with a core (15-mm diam.) at sowing to a depth of 60 cm. Organic carbon (CO) (Richter and Von Wistinghausen, 1981); total nitrogen (NT) (Kjeldal procedure; Horwitz et al., 1975); extractable phosphorus P (Bray and Kurtz technique, Olsen and Sommers, 1982); bulk density (Blake and Hartage, 1986); pH (1:2.5 soil/solution ratio); and SS (Hénin et al., 1972) were determined in the upper 20-cm depth. Mineral nitrogen (N–NO-3) (Daniel and Marbán, 1989) was determined in the upper 60-cm depth (20-cm increments). Soil water content (SWC) was measured at sowing gravimetrically from 0 to the 60-cm depth. After sowing, gravimetric soil water measurements were made at 1- to 2-wk intervals, between the 0- and 20-cm depth, and by the neutron scatter technique (Model Troxler 3320, Troxler International, Triangle Park, NC) between the 20- and 140-cm depth at 20-cm intervals. One aluminum access tube was installed to a depth of 150 cm in the central row of each subplot. Independent calibration curves for volumetric SWC were developed for the soils on each farm and regression coefficients compared. As the coefficients of the equations did not differ significantly among farms, a single equation (Eq. [1]) was used.

(1)

where SWC is soil water content (cm3 cm-3).

To determine the limits of water extraction, three soil samples from each depth (20-cm intervals from the soil surface to 140-cm depth) were taken in each field at the beginning of the experiments. The upper and the lower limits for water extraction were determined in the laboratory with a pressure-membrane apparatus from disturbed soil samples. Water-holding capacity for the entire profile was then estimated as the sum of the differences between water content at the upper (-30 kPa) and the lower (-1500 kPa) limits in all soil layers. For silty clay loam soils, estimation of both limits with laboratory methods using undisturbed samples differs slightly (overestimation of upper limit <2%) from those obtained in field conditions (Ratliff et al., 1983). Disturbed samples, however, are likely to have higher water contents than undisturbed samples at -30kPa. This could result in higher laboratory estimates of water holding capacity than those obtained in the field.

Available soil water (ASW) was calculated as the ratio of measured SWC minus the lower limit to the water-holding capacity in the rooting depth. A new soil layer was included in the rooting depth when its water balance between two consecutive determinations indicated that a significant water extraction had taken place (P < 0.01).

Actual ETa was calculated as rainfall plus the change in soil water storage between two observation dates. Daily ETa was calculated for the presilking period, the period around silking (±15 d), the postsilking period, and the whole growing season (sowing-physiological maturity). When SWC exceeded the upper limit of a certain soil layer, the difference was considered to drain to the next layer.

Crop Measurements
Observations of crop phenology were made weekly on seven plants in each subplot and dates of specific stages were registered when 50% of the plants reached the stages of: (i) silking (at least one silk visible on the apical ear) and (ii) physiological maturity (the occurrence of black layer in grains of the midsection of the ear; Daynard and Duncan, 1969).

Canopy size was characterized indirectly by the fraction of incident photosynthetically active radiation intercepted by the canopy (fIPAR) (Jones and Kiniry, 1986). The fIPAR was calculated from PAR measured above the canopy and PAR measured below the green leaves but above the senesced leaves at the bottom of the canopy. The fIPAR was determined about every 14 d from the three-leaf stage to physiological maturity. Five measurements were taken within each subplot, between 1100 and 1400 h on clear days, with 0.70 m of a line-quantum sensor (LI-191 SA, LI-COR, Lincoln, NE) placed perpendicular to the rows (Gallo and Daughtry, 1986). In this manuscript only maximum fIPAR values are presented.

Two complementary methodologies were used to characterize the root system. Root sampling and root observations were made at silking in each field. One of the methodologies was semiquantitative, but allowed a complete visualization of the root profile. The other one was quantitative and permitted the determination of root length density. Root abundance (semiquantitative methodology) was estimated for hybrid DK752 in each field following the methodology proposed by Massé (1982) on a vertical wall of a trench perpendicular to the row direction. A 4 by 4-cm grid of 0.7-m width by 0.8-m depth, was placed against the wall, centered on a plant. Root abundance was defined in each square (4 by 4 cm) with a five category semiquantitative scale described below.


A single observer made all observations. Root abundance was summed for each layer and for the whole profile to an 80-cm depth.

To measure root length density, soil samples were taken from each subplot between two plants in the row and at 17.5 cm from the rows, with a hand auger of 20 cm height having an inside diameter of 1.5 cm. Samples were taken at 20-cm intervals up to a depth of 80 cm, preserved from drying in plastic bags, and stored at -18°C. After thawing for 24 h they were hand-washed. Organic matter and debris were eliminated and roots separated from the suspension with two mesh sieves (the finest <1 mm). Root length density was determined using the line-intercept method described by Tennant (1975) and expressed in units of cm cm-3. Values of root length density from the row and the interrow were averaged.

The number of spikelets ear-1 (apical and subapical) and its components (rows ear-1 and spikelets row-1) were determined at silking in six plants per subplot to characterize potential KN (Cirilo and Andrade, 1994).

Total dry matter production during the postsilking period was estimated as the difference between total dry matter at physiological maturity and total dry matter at silking. Samples were constituted by 7 plants per subplot (1 m2).

Grain yield, KN, KW, and prolificacy (ears plant-1) were determined from samples taken at physiological maturity. Kernel set of apical ear was estimated as the quotient between KN at physiological maturity and potential KN determined at silking.

Plant weight gain per kernel (PWG k-1) was estimated as the quotient between total dry matter production during the postsilking period and KN (Maddonni et al., 1998).

Meteorological Conditions
Temperature and rainfall were recorded daily at each field. Reference evapotranspiration (ETo) was estimated with the modified CERES-Maize model (Jones and Kiniry, 1986) from maximum and minimum temperature and total solar radiation, which was recorded at the nearest meteorological station (<30 km).

Statistical Analysis
Results were subjected to analysis of variance to evaluate the effects of treatments and their interactions on soil properties, ASW, ETa, root variables, and dry matter production. For presilking determinations and in 1995 to 1996, a split-plot design with three replicates (farms) was used with the length of cropping (S vs. L) as the main factor and hybrids (DK4F31, DK4F37, and DK752) as subfactor. In 1994 to 1995, the field with a short length of cropping of Farm 3 was lost 1 wk after silking. As a consequence, the variables studied from silking onward were analyzed within each farm (in Farms 1 and 2). This was possible because in each field (S and L), hybrids were randomized in completed block design with three replicates.

Correlations were evaluated and relationships among variables fitted using TBLCURVE (Jandel Scientific, 1992). Differences among hybrids in the relationship between KN and ETa around silking were tested by comparison of the linear regression coefficients and with an analysis of variance of the residual sum of squares (Mead and Curnow, 1983).

A rectangular hyperbola, such as that described in Cirilo and Andrade (1996), was fitted between KW and PWG k-1 during the postsilking period for the whole data set.


   Results and discussion
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
Conclusions
 REFERENCES
 
Soil chemical properties differed slightly (P < 0.05) between S and L (Table 1) . Only in 1994 was the NO-3–N greater in S than in L. Critical reductions were observed, however, in comparison with values of a pristine soil. For example, a 21 to 28% reduction of NT, 25 to 29% reduction of Co, and 75 to 85% reduction of P were detected. When long-term cropping effects on nutrient supply was studied, NO-3–N at sowing was not reflected in maize grain yields (Urricariet and Lavado, 1999).


View this table:
[in this window]
[in a new window]
  		Table 1 Physical and chemical soil properties of fields with long and short cropping periods and of pristine soils. Total nitrogen (NT, 0–20 cm), organic carbon (Co, 0–20 cm), mineral nitrogen content (NO-3–N, 0–60 cm), extractable phosphorus (P, 0–20 cm), pH (0–20 cm), soil aggregate stability (Ss, 0–20 cm), soil water content (SWC, 0–140 cm), and bulk density (0–20 cm) at maize sowing in 1994 and 1995. Soil water content (SWC) at the upper and lower limits (0–140 cm) is included. Values are the mean of three farms. Standard deviations of the means are presented. Different letters within a column and year indicate significant differences (P < 0.05) between short and long cropping periods

 
In the region under study, rainfall during winter–spring fallow generally results in SWC at the upper limits for soil water extraction in the 0- to 140-cm depth (Cárcova and Otegui, 1997). In our experiments, however, SWC at sowing (Table 1) was less than expected based on values estimated by Cárcova and Otegui (1997) from the 30-yr climatic record of the region. Nevertheless, ASW during the presilking period was always above threshold (20–30%) to reduce ETa in maize (Grant et al., 1989; Fig. 2) . Generally, ASW decreased throughout the growing season and was similar or less in S than in L (Fig. 2). No difference among hybrids and no interaction between cropping history and hybrids were found.



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 2 Relative available soil water (ASW) of maize in fields with short (S) and long (L) cropping periods in relation to days after sowing. Values are the mean of three hybrids. Asterisks indicate the dates at which significant (P < 0.05) differences in ASW were detected. Horizontal lines indicate silking period

 
Differences in ASW between S and L could be a result in differences in rooting depth, limits of water extraction, and/or rooting colonization of each soil layer. The depth of initial soil water depletion increased at a constant rate relative to time after sowing with no differences between lengths of the cropping period or among hybrids (Fig. 3) . In both S and L, water extraction below 100 cm was <10% (data not shown). Consequently, rooting depth where water uptake took place was not modified by the intensity of land use. The lower limit for water extraction determined in the laboratory did not differ between S and L (Fig. 4) . Moreover, minimum values registered in our experiments were similar in both S and L and close to the lower limit below the 80-cm depth. Between the 40 and 80-cm depth, the B2t horizon, values observed were always larger than the lower limit determined in the laboratory. Despite similar minimum values of ASW between S and L, they were reached earlier in the former (e.g., 30 d in Farm 2). These results suggest a constraint to water uptake other than rooting depth or limits for water extraction. Root abundance, measured at silking stage, was significantly (P < 0.05) greater in S than in L (620 and 387 in 1994–1995 and 638 and 402 in 1995–1996 in S and L, respectively). When analyzed by stratum, differences were statistically significant for most layers (Fig. 5) . These results are consistent with those reported by Tardieu (1984). He demonstrated that topsoil compaction imposed a restriction to root growth that was transmitted to the deeper layers. In our experiments, reduction of SS promoted by long-term cropping could have produced a similar negative effect on root growth as that of topsoil compaction. During both growing seasons, root length density decreased with depth and no difference was found between S and L in the first year. However, during 1995 to 1996 there was a significant interaction between the length of cropping and depth because of the more uniform root distribution in S than in L (Table 2) . This result is consistent with root abundance observations. Consequently, less root colonization of the soil profile in L seems to be partially explain the greater values of ASW. de Willigen and Van Noordwijk (1987) and Tardieu and Katerji (1991) observed an appreciable effect of the root spatial arrangement on water uptake.



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 3 Rooting depth determined by water extraction as a function of days after sowing for maize grown in fields with short (S) and long (L) cropping periods during 2 yr. Values for both S and L are the average of three hybrids. Dotted line indicates the linear regression fitted between rooting depth and days after sowing

 


View larger version (44K):
[in this window]
[in a new window]
  		Fig. 4 Limits for water extraction from the 0- to 140-cm soil depth. Values are the average of fields with short (S) and long (L) cropping periods at three farms. Crossbars indicate standard errors of the mean. Minimum (min) and maximum (max) soil water content measured in situ in S and L at Farm 2 is presented. Shaded rectangle indicates B2t horizon

 


View larger version (21K):
[in this window]
[in a new window]
  		Fig. 5 Root abundance for hybrid DK752 at silking in fields with short (S) and long (L) cropping periods. Values represent a summation of observations on a 0–4 scale with 0 representing no roots and 4 representing very abundant roots. Observations are summed over 18, 4-cm increments for each 4-cm depth increase. Values are the average of the three farms considered. Asterisks indicate significant differences (P < 0.05) in root abundance between S and L at each 4-cm depth

 

View this table:
[in this window]
[in a new window]
  		Table 2 Root length density interactions between hybrid and soil depth in 1994–1995 and length of cropping and soil depth in 1995–1996. Different letters indicate significant differences (P < 0.05) within a column

 
Root growth was not the only crop attribute modified by the length of cropping. Canopy size was smaller in L than in S, as estimated by maximum fIPAR (Table 3) . Less root colonization of the soil profile, which results in less soil water and nutrient uptake, apparently decreased canopy size in L when compared with S. The constraint in shoot growth was also reflected in crop phenology. Maize grown in L silked 5 to 6 d later than those grown in S. Contrary to the presilking period, ASW values during silking were below the threshold when ETa is reduced in maize (Grant et al., 1989; Fig. 2). In these experiments Cárcova et al. (1998) characterized plant water stress using infrared thermometry. During the period around silking crops had a water stress index ca. 24 and 45% for 1994 to 1995 and 1995 to 1996, respectively. In accordance with these results, in 1994 to 1995, mean daily ETa around the silking period was 67 to 96% of daily ET0 (5.7 mm d-1) (Table 3). In the drier 1995 to 1996 year, ETa was only 41 to 56% of daily ET0 (5.9 mm d-1). In both years, ETa around the silking period was always greater in S than in L. No interaction between the length of cropping and hybrid was observed in ETa. In addition to root growth restriction, canopy size (Table 3) could have contributed to differences found between S and L in ETa around silking (Al-Kaisi et al., 1989). The fIPAR values obtained in our experiments (>0.7) are related to a leaf area index of about 2 (Maddonni and Otegui, 1996), which is associated with a negligible evaporative flux (Al-Kaisi et al., 1989). In addition, crops grown in L were exposed to less available soil water around the silking period (Table 4) because of the delay in silking. For example, at Farm 2 in 1994 to 1995, SWC was greater in S than in L 15 d before silking. Rainfall during the period around silking, was greater in S than in L in 1995 to 1996 and at Farm 2 in 1994 to 1995.


View this table:
[in this window]
[in a new window]
  		Table 3 Mean daily evapotranspiration (ETa) for different maize growth stages and during the growing season (total) and maximum flPAR in three maize hybrids grown in fields with different lengths of cropping. During 1994–1995, the field with a short length of cropping of Farm 3 was lost 1 wk after silking; thus, only values of Farm 1 and 2 are presented. In 1995–1996 values presented are the average of the three farms. Different letters indicate significant differences (capital letters P < 0.10, lowercase letters P < 0.05) within each column

 

View this table:
[in this window]
[in a new window]
  		Table 4 Soil water content (SWC), rainfall, reference evapotranspiration (ETo), and actual crop evapotranspiration (ETa) before or around silking for maize grown in fields with different lengths of cropping. During 1994–1995, the field with a short length of cropping of Farm 3 was lost 1 wk after silking; thus, only values of Farm 1 and 2 are presented. In 1995–1996 values presented are the average of the three farms. In all the cases, values are the average of three hybrids. Different letters indicate significant differences (capital letters P < 0.10, lowercase letters P < 0.05) within each column

 
When comparing hybrids, DK752 in 1994 to 1995 had the greatest root length density below 20 cm (Table 2) and a more uniform root distribution within the profile with values higher than 1 cm cm-3 up to a depth of 80 cm. A more uniform root distribution suggests a potential for greater water extraction for DK752 (Boot, 1990). Nevertheless, at Farm 1, DK4F37 had a greater ETa than DK752 during the period around silking and in the postsilking period in 1994 to 1995 (Table 3). Greater ETa for DK4F37 could be associated with its high osmotic adjustment (Lemcoff et al., 1998). Differences among hybrids in water extraction may also be related to: the value of SWC below which transpiration rate starts to decline and/or the rate with which transpiration rate declines below a certain SWC (Ray and Sinclair, 1997).

Despite constraints for maize growth, the length of cropping did not modify potential KN (Table 5) . Likewise, kernel set was not limited by potential KN (Table 6) . Similar results were obtained in maize crops grown under different nutrient regimes (Uhart and Andrade, 1995), sowing dates (Cirilo and Andrade, 1994; Otegui and Melón, 1997), water availability (Otegui et al., 1995a), and plant populations (Otegui, 1997), suggesting that potential KN is a very stable genotype characteristic. Among hybrids, DK752 had the most spikelets ear-1, rows ear-1 and spikelets row-1 (Table 5). Kernel set of this hybrid, however, was the least because of the greater difference among hybrids in potential kernel number than in KN (Tables 5 and 6).


View this table:
[in this window]
[in a new window]
  		Table 5 Effects of the length of cropping and hybrids on potential kernel number components at silking. Values are the average of three farms. Different letters within a column and year indicate significant differences (P < 0.05)

 

View this table:
[in this window]
[in a new window]
  		Table 6 Effects of the length of cropping and hybrids on kernel set, grain yield, and grain yield components (kernel number, kernel weight, and prolificacy). During 1994–1995, the field with a short length of cropping of Farm 3 was lost 1 wk after silking; thus, only values of Farm 1 and 2 are presented. In 1995–1996 values presented are the averages of three farms. Different letters indicate significant differences (capital letters P < 0.10, lowercase letters P < 0.05) within each column

 
When averaged across cropping history and hybrids, grain yield had a positive linear relationship with KN . During 1994 to 1995, grain yield and its components differed significantly between S and L and among hybrids (Table 6). Maize grown in S had the greatest grain yield, KN, and KW. Hybrid DK752, in both S and L, had the greatest KN and the least KW. During 1995 to 1996, no differences in grain yield and its components were detected among treatments. During both years, DK4F37 was the most prolific hybrid (ears plant-1 > 1).

Kernel number was associated with mean daily ETa around silking (Fig. 6) . When a relationship between KN and ETa was fitted for each hybrid, parameters of the linear regression did not differ for DK4F31 and DK4F37. Thus, a single regression was fitted for both hybrids. Contrary, the slope of the linear regression for DK752 was significantly (P < 0.05) greater than that of the other hybrids. The nonzero origin of both fitted functions between KN and ETa, indicates a curvilinear response of KN to ETa as was found by Otegui et al. (1995a). Fitted functions also indicate an increase of 386 kernels m-2 (for DK4F31 and DK4F37) and 853 kernels m-2 (for DK752) for each mm d-1 of ETa, suggesting a different sensitivity among hybrids to water uptake. Probably, the greater spikelets ear-1 of DK752 (Table 5) explained the greater KN of this hybrid. Moreover, this hybrid had the fastest appearance rate of silks among hybrids (Maddonni et al., 1999).



View larger version (32K):
[in this window]
[in a new window]
  		Fig. 6 Kernel number (KN) of hybrids (DK4F31, DK4F37, and DK752) grown in fields with short (S) and long (L) cropping periods as a function of mean daily crop evapotranspiration (ETa) around silking in 1994 to 1995 (solid symbols) and 1995 to 1996 (open symbols). Dotted line represents linear regression between KN and ETa for DK752. Solid line represents linear regression between KN and ETa for both DK4F31 and DK4F37

 
A threshold of ETa above which differences in KN between length of cropping and among hybrids were reflected in KN apparently existed. For example, maize crops grown in S had a greater KN than those grown in L during 1994 to 1995, only when ETa was above 3 mm d-1. Likewise, DK752 had the greatest KN when ETa was above the same value.

A reduction of KN can affect KW values (Maddonni et al., 1998). Long-term cropping modified KW only during 1994 to 1995, when KW values were below the maximum for these hybrids grown in the region without water or nutrient limitation (Table 6). When averaged across cropping history and hybrids, KW had a positive relationship with PWG k-1 (Fig. 7) . Although PWG k-1 overestimates assimilate availability during the effective grain filling period, this methodology was useful to compare the effect of contrasting postsilking environmental conditions on KW (Maddonni et al., 1998). Plant weight gain per kernel differed between years (P < 0.05), with a mean value of 218 mg kernel-1 (min. 131 mg kernel-1, max. 330 mg kernel-1) and 282 mg kernel -1 (min. 176 mg kernel-1, max. 404 mg kernel-1) during 1994 to 1995 and 1995 to 1996, respectively. Crops grown in L during 1994 to 1995, had a lower PWG k-1 than those grown in S, resulting in a reduced KW (Fig. 7). In contrast, during 1995 to 1996, the low KN of crops grown in both S and L resulted in a nonlimiting PWG k-1, that was reflected in KW values close to maximum.



View larger version (39K):
[in this window]
[in a new window]
  		Fig. 7 Kernel weight (KW) of hybrids (DK4F31, DK4F37, and DK752) grown in fields with short (S) and long (L) cropping periods as a function of plant weight gain per kernel (PWG k-1) during the 1994 to 1995 (solid symbols) and 1995 to 1996 (open symbols) postsilking periods. Solid line represents the rectangular hyperbola fitted between KW and PWG k-1

 

   Conclusions
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
REFERENCES
 
In conclusion, long-term cropping in the Rolling Pampa apparently imposed some constraints for maize growth. Crops grown in L had less root abundance and a smaller canopy size than those grown in S. A 5- to 6-d delay of silking was observed in L, exposing crops to less ASW around this critical period. Less ASW and less water uptake capacity reduced ETa around silking. Kernel number (the main component of grain yield) was highly dependent on ETa around silking. The modern hybrid with the highest production stability coefficient had the greatest KN mm-1 of ETa, a characteristic that may have masked the deleterious effect of intensive land use. Breeders should consider the KN/ETa relationship as a selection criteria for hybrids to be used in fields with long-term cropping.


   ACKNOWLEDGMENTS
 
We thank A.J. Hall, M.E. Otegui, and G.A. Slafer for the critical review of this paper. We also extend our acknowledgment to P. Dodds, M. Amaya, A. Ochoa, and E. Hidalgo for their cooperation in field experiments. This work was partially supported by Dekalb Argentina S.A. and the University of Buenos Aires (UBACyT AG 108-TG40).

Received for publication September 24, 1999.
   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 




This article has been cited by other articles:


Home page
 Crop Sci.Home page
L. Borras, J. A. Cura, and M. E. Otegui
Maize Kernel Composition and Post-Flowering Source-Sink Ratio
Crop Sci., May 1, 2002; 42(3): 781 - 790.
[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 ISI 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 ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cárcova, J.
Right arrow Articles by Ghersa, C. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Cárcova, J.
Right arrow Articles by Ghersa, C. M.
Agricola
Right arrow Articles by Cárcova, J.
Right arrow Articles by Ghersa, C. M.
Related Collections
Right arrow Crop Growth and Development
Right arrow Maize
Right arrow Root Development
Right arrow Maize Management

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Crop Science 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