| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Cátedra de Cereales, Dep. de Producción Vegetal, Facultad de Agronomía, Univ. de Buenos Aires, Av. San Martín 4453, Buenos Aires (C1417DSE), Argentina
* Corresponding author (jcarcova{at}agro.uba.ar)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
15% increase in KN per plant. Pollination gaps of 2 and 4 d reduced KN per plant drastically (up to 51%), but the reduction was smaller for the 6-d gap. This study (i) gives evidence of the negative impact of delayed pollination timing among silks on kernel set, which was not related to reduced silk receptivity, and (ii) defines the time gap for maximum interference of early- on late-pollinated ovaries, a period shorter than 4 d.
Abbreviations: ACC, 1-aminocyclopropane-1-carboxylic acid • DAS, days after silking • En, ear n • H, hybrid • HD, high plant population density • KN, kernel number • LD, low plant population density • NP, natural pollination • PP, plant population • Gn, n-day gap between early- and late-pollinated silks
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Kernel abortion can be partially overcome by increasing assimilate supply of plants under stress conditions (Boyle et al., 1991; Schussler and Westgate, 1991, 1995). Nevertheless, evidence from recent studies indicates that assimilate availability per fertile spikelet seems not to be the only factor controlling kernel set when water and nutrients are not limiting growth (Cárcova et al., 2000). It is well known that kernel set in the subapical ear depends on synchronous silking and pollination of both ear shoots (Harris et al., 1976; Motto and Moll, 1983; Sarquís et al., 1998). Considering pollination synchrony within the ear, a significant reduction in kernel set has been observed when the pollination interval between early- and late-appearing silks in the apical ear was artificially increased (Freier et al., 1984). On the other hand, maize kernel set can be improved significantly (8–31%) through synchronous pollination, both between ears at low plant population and within the apical ear at high stand densities (Sarquís et al., 1998; Cárcova et al., 2000). Apparently, delayed fertilization of early-silking ovaries allowed some of the late-silking ones to compensate for their ontogenic delay and set kernels (Struik and Makonnen, 1992). All these studies suggest that the rate of silk emergence and pollination timing may explain part of the genotypic differences observed in final KN.
There is little information about the consequences of altering the natural synchrony of silk emergence and pollination within the apical ear (Freier et al., 1984). Floret development along the ear follows an acropetal pattern of differentiation. Synchrony in floret development along the ear may be related to the number of spikelets per ear row, which depends on the duration of spikelet production and the rate of spikelet initiation. Duration and rate of spikelet production are under temperature control (Otegui and Melón, 1997), and genotypic differences exist for all these traits (Edmeades et al., 1993; Bassetti and Westgate, 1993a, b). It is still unknown to what extent growth synchrony among spikelets within and between ears is defined by the order of ontogenic differentiation, and if this order modifies subsequent sink activity. Increasing inflorescence temperature can increase the metabolic rate and hence its "sink strength" for assimilate partitioning (Ou-Lee and Setter, 1985). Thus, a selective temperature treatment might be an effective method for altering partitioning between competing sinks on the ear. To tests these possibilities, we imposed selective ear heating treatments in an attempt to modify the rate of silk emergence and/or sink activity of competing ovaries. We also increased the time gap between early- and late-pollinated silks. Tip ear heating was expected to minimize the advantage of early silking ovaries. Lateral heating and pollination gaps were expected to exaggerate this advantage.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Experiment 2
The experiment was carried out in 1999–2000 to analyze in more detail the effect of pollination timing on kernel set. Hybrids DK752 and DK664 (semi-prolific, 16 kernel rows ear-1) were sown on 26 October at Salto (34°33'S, 60°33'W), Argentina, on a silty clay loam soil (Typic Argiudol). Two plant populations (PP) were used, 3 and 9 plants m-2. Each hybrid- plant population combination (H by PP) covered two 20-row by 15-m plots, with rows spaced at 0.70 m. As in Exp. 1, plots were kept free of weeds and pests, and there was no water or nutrient stress.
Ear Heating Treatments
In Exp. 1, three different ear temperature treatments were applied to the apical ears of at least 15 plants per treatment in an attempt to modify synchrony of silk emergence among spikelets: (i) heating the ear tip (tip ear heating), (ii) heating one side of the ear (lateral ear heating), and (iii) control (no ear heating). For each temperature treatment, three or four groups of five plants each were tagged at random before silking. The date when silks first appeared (Day 1) was registered for both ears (E1 = apical ear, and E2 = sub-apical ear) of each tagged plant. Heating treatments (+5°C with respect to air temperature, but never above 35°C) were performed with a 120-
resistance fixed to an aluminum foil attached to the ear. The limit of 35°C was set on the basis of results from Dupuis and Dumas (1990), who determined that fertilization rate is highly reduced in maize when pollinated spikelets are exposed to temperatures above 36°C. Each heating foil also had a temperature sensor (LM35Z, National Semiconductors, CA) fixed to it to monitor the output temperature. For tip ear heating, two pieces of aluminum foil were bound to the upper third of the outermost husks of the ear (about 10 cm) with elastic bands. For lateral ear heating, the foil was placed within the outer-most husks along one side of the ear, and also fixed with elastic bands. The abaxial side or adaxial side of the ear was chosen at random among plants for lateral heating. In both heating treatments, aluminum foils never had direct contact with the kernels (i.e., at least two or three husks were always between them), to avoid possible confounded effects of temperature and the foils themselves. The foils were moved and/or replaced by larger foils accordingly to ear growth (Otegui and Bonhomme, 1998), in order to keep the same proportion of heated area along the treatment period. Individual heating systems were connected in groups of five (i.e., five plants) to a central control unit, which displayed the output temperature. Air temperature was also measured at each central unit by means of an LM35A, and used as reference to control the output temperature generated by the 120- resistor of each individual heater. This control was performed with an individual operational amplifier (one per heater), which added 5°C to the reference air temperature. A second operational amplifier (also one per heater) detected the moment when reference air temperature +5°C 
35°C, and controlled current to the heating resistor. All operational amplifiers were located at the central unit (i.e., 10 per unit). Heating treatments were performed all along a 
14-d period, ranged in average between 2 to 3 d before silking and 12 d post-silking.
The dynamics of silk emergence were determined on E1 of each tagged plant. Exposed silks were cut each 2 d between silking and 9 d after silking (DAS), and all new silks (i.e., those with a bisected hairy end) were counted (Cárcova et al. 2000). Silks were identified, preserved in ethanol (700 g kg-1), and counted by hand. The ears of all tagged plants were harvested at physiological maturity, and KN was determined in both E1 and E2. All ears with at least 10 kernels ear-1 were considered fertile (Tollenaar et al., 1992) and included in the analysis. Consequently, silk number and KN per ear could be matched for each tagged plant in each heating treatment. Treatments were compared by ANOVA, and a t-test analysis was used to determine significant (P < 0.05) differences between means for both the number of silks and KN.
Pollination Treatments
In Exp. 1, three pollination treatments were imposed to modify the number of simultaneously pollinated silks: (i) natural pollination, (ii) split pollination, and (iii) synchronous pollination. These treatments were done on both inner-plot plants (i.e., at 9 plants m-2) and border plants (i.e., at 4.5 plants m-2). At least 20 per pollination treatment were tagged at random for each plant population condition. The date of silking of both ears was registered for each tagged plant. In plants destined to be synchronously pollinated, both E1 and E2 were bagged before silking, and simultaneously hand pollinated on 5 DAS. This treatment granted synchronous pollination of a large proportion of ovaries per ear (Cárcova et al., 2000). Silks were pollinated between 1000 and 1200 h, with fresh pollen collected from donor plants. For pollen collection, tassels that had just started anthesis were bagged late in the afternoon, and pollen obtained from them was recovered the following morning. After being hand pollinated, ears were left unbagged, allowing natural pollination of late-appearing silks. More details on the development of synchronous pollination are given in Cárcova et al. (2000). Exposed silks from all bagged ears were cut just before hand pollination, and counted as previously described for heating treatments. Ears of plants destined to split pollination were left unbagged and allowed to be pollinated naturally during the first 2 DAS. Then, all ears were covered up to 8 DAS, when they were unbagged again and hand-pollinated as described above. Subapical ears (E2) were bagged before silking and remained covered until 8 DAS, when they were also pollinated. In this way, a 6-d gap (G6) was established between pollination of early- and late-pollinated silks. The objective of this treatment was to enhance the pollination asynchrony within the apical ear.
In Exp. 2, four pollination treatments were performed at each H x PP plot. Within each plot, at least 20 plants were tagged before silking for each pollination treatment. Silking date was recorded individually for each ear. Pollination treatments were (i) control or natural pollinated (NP) plants, whose ears were always unbagged and received pollen naturally; (ii) 2-d gap (G2), (iii) 4-d gap (G4), and (iv) 6-d gap (G6). For all gap treatments, silks from E1 received pollen naturally during the first 2 DAS, then ears were bagged to avoid pollination for 2, 4, or 6 d depending on the treatment. After each gap treatment, ears were left unbagged and all exerted silks were allowed to receive fresh pollen naturally, as described above. In all gap treatments, ears were hand-supplemented with fresh pollen on 8 DAS. This hand pollination was performed to assure pollen availability at the end of the pollen production period, and was not aimed to synchronize pollination, as in bagged ears of Exp. 1 on 5 DAS. Sub-apical ears (E2) were bagged until apical ears were unbagged and both ears started receiving pollen naturally. The extension of the gaps without pollination was planned to avoid the confounded effects of (i) reduced kernel set related to interference between early- and late-pollinated ovaries (the one we wanted to test), and (ii) reduced kernel set related to lack of ovary fertilization due to reduced silk receptivity (Bassetti and Westgate, 1993a,b), which was probably the reason of decreased kernel number in Freier et al. (1984) work. For this reason, the longest gap tested in our experiment did not exceed 6 d (G6).
At maturity, KN was determined for E1 and E2 in each treatment. Data were analyzed by ANOVA, and a t-test was used to determine significant (P < 0.05) mean differences between pollination treatments at each H x PP combination.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
4.5°C by the heat treatment. Ear temperature profiles over a 48-h period are shown in Fig. 1
. The same trend was observed during the rest of the heating period, which spanned 14 d.
|
|
 | [1] |
Both heating and pollination treatments affected KN per ear to a greater extent (73% variation between the lowest and the greatest KN per ear values) than did the number of silks exposed on 5 DAS (6% variation between the lowest and the greatest silk number E-11 values) (Table 1).
|
|
|
The artificial 6-d pollination gap (G6), created between early and late appearing silks in Exp. 1, produced a significant (P < 0.05) reduction in KN per plant at both plant populations (9 and 4.5 plants m-2) (Table 1). In Exp. 2, both hybrids reduced KN per E1 under the G2 and G4 treatments at 9 plants m-2 (Table 3, Fig. 3b), with no kernel set in E2. Kernel number did not vary between natural-pollinated plants and G6 at this plant population. At the low plant population (3 plants m-2), all gap treatments reduced KN per plant, but this reduction was smaller for G6 (Table 3). Split pollination of silks exposed from E1 had a negative effect on kernel set of both ears at this plant population, but differences between treatments were larger in E2 than in E1.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
227°C days (8°C base temperature) before silking (Otegui and Bonhomme, 1998), while the heating treatment in this work started at 35°C days before silking. On the other hand, precise tip ear heating was almost impossible to perform at an earlier stage without injuring the plant, because of the uncertain identification within the husks and the corresponding leaf sheath of the exact position of the apical third of E1 (Otegui, 1997).
Lateral ear heating resulted in a reduction in KN per ear, which was not related to the rate of silk emergence. On the basis of silking dynamics, ovary fertilization took place at similar rates on both heated and nonheated ear sides. Thus, increased kernel abortion in the latter cannot be attributed to the negative effects of asynchronous pollination like those promoted through split pollinations (i.e., Gn). For example, the G2 treatment promoted an ontogenic gap between early- and late-pollinated ovaries equivalent to 30°C day. In the lateral heating treatment, the ontogenic difference on 2 DAS between the heated (39°C day) and the nonheated (30°C day) sides was only 9°C day. Thus, reduced KN induced by lateral heating can be attributed to a differential metabolic activity between heated and nonheated ovaries during this early post-fertilization period, which may have promoted an uneven distribution of assimilates among ovaries of similar position along the ear. Results from Jones et al. (1984) support this hypothesis. They found that both high (35°C) and low (15°C) temperatures during early kernel growth (i.e., lag phase) had a similar detrimental effect on kernel development and final kernel mass. They attributed the adverse effect of extreme temperatures to a reduction in the number and/or size of endosperm cells formed during this stage, therefore reducing kernel sink capacity. In their experiment, the best condition for kernel development was obtained when a 30°C thermal environment was provided to the ears. In our study, the temperature regime on the heated side (between 27.2 and 30.9°C and never above 35°C) was close to the optimum quoted by Jones et al. (1984), and may have promoted enhanced metabolic activity in that region of the ear, with a concomitant increase in partitioning of assimilates (Ou-Lee and Setter, 1985). If so, reduced KN in the nonheated side was the result of kernel abortion promoted by reduced assimilate availability (Charles-Edwards, 1984; Zinselmeier et al., 1995; Egli, 1998) induced by temperature effects on assimilate partitioning (Ou-Lee and Setter, 1985) rather than by dominance effects due to a significant pollination gap.
Synchronous pollination increased kernel set at high plant population, and final KN of this treatment did not differ from values obtained with border-row plants, grown under less interplant competition. This result is in agreement with previous studies demonstrating the beneficial effects of synchronous pollination on kernel set (Motto and Moll, 1983; Sarquís et al., 1998; Cárcova et al., 2000). However, there is little information about the impact of an increased interval between early and late appearing silks on KN. Freier et al. (1984) discussed limitations to kernel set imposed by within-ear fertilization synchrony, but their split pollination results did not clearly distinguish between kernel abortion and lack of ovary fertilization due to loss in silk receptivity. In their study, the 7-d period between early and late pollinated ovaries may have reduced fertilization among the latter due to silk aging (Bassetti and Westgate, 1993a). In the present work, almost all split pollinations (i.e., gap between early and late pollinated silks) promoted a reduction in kernel set when compared to the natural pollinated control. This reduction was entirely due to kernel abortion, as indicated by the presence of embryos that arrested their development soon after fertilization. This result gives evidence of an interference effect among ovaries (i.e., dominance of early-formed ovaries from the base of the ear on the late-formed ones from the tip). At a given plant population, this interference depended upon the pollination/fertilization timing linked to ontogeny under natural conditions (Bassetti and Westgate, 1993a; Otegui, 1997), and was independent of assimilate availability, because a similar effect was observed at two contrasting source-sink ratios (3 and 9 plants m-2).
Kernel set in Exp. 2 was consistently greater for G6 than for the shorter periods examined, G2 and G4, suggesting that the dominance of early-pollinated ovaries on the late-pollinated ones diminished when the delay between pollinations was greater than 4 d. There are two possible interpretations for the split pollination results of Exp. 2, one based on direct assimilate partitioning effects and the other on the transport of a plant growth regulator that promotes abortion in late-pollinated ovaries. The first interpretation considers a "double effect" on assimilate partitioning among ovaries, which results from the balance between timing and intensity of pollination across treatments. As the gap between early and late pollinations increased the asynchrony between ovary fertilizations, it also increased the number of ovaries that were pollinated in the second group of late-appearing silks. It may be expected the first process (i.e., increased time gap between pollinations) being increasingly detrimental for final kernel set (Freier et al., 1984) and the second (i.e., increased number of silks that are pollinated simultaneously at the end of the longest gap) being increasingly beneficial (Cárcova et al., 2000). The first process sets an ontogenic gap among ovaries, which results in different sink activity related to sink size (i.e., early pollinated larger activity than late pollinated), with the concomitant direct effect on assimilate partitioning. In the second process, a large number of simultaneously late-pollinated ovaries (like in G6) is assumed to represent a high-activity sink (i.e., large assimilate demand established abruptly on 8 DAS), which may be able to counterbalance the high sink activity of the early-pollinated ones and set kernels. There is evidence in literature (Hall et al., 1981; Freier et al., 1984; Boyle et al., 1991) to support the assumptions related to the first process, but no research has been carried out to verify the second.
The other interpretation for our split pollination results is based on the activation and/or transport of a plant growth regulator in early-set kernels, that may be responsible of growth arrest in the late-pollinated ones (Bangerth, 1989). This approach is founded in two aspects of the response pattern observed in our work that deserve consideration: (i) the transient nature of the interference process (e.g., kernel set decreased in G2 and G4 and then increased in G6), and (ii) the degree of interference. The former aspect suggests the process of interference is mediated by a diffusive and ephemeral substance, which does not affect the source organs (i.e., early-fertilized ovaries) but is sensed by neighbor structures that are delayed in development. The degree of interference (i.e., reduction in kernel set) may be related to the amount of substance, which may vary with the source size (i.e., number of early-fertilized ovaries) and the production dynamics of the substance, and to the sensitivity of the younger florets.
The above-mentioned characteristics could be associated to a compound like ethylene, a plant growth regulator that could be involved in the transmission of the signal among ovaries within the ear. Its production was found to increase following pollination in cut flowers, suggesting movement of a stimulus for ethylene production from the stigma and style to the ovary, receptacle, and petals (Reid and Wu, 1991). In maize, Cheng and Lur (1996) found that dry weight accumulation in apical kernels ceased within 96 h after pollination in response to a shading treatment, but ethylene and 1-aminocyclopropane-1-carboxylic acid (ACC) increased rapidly at about 24 h after pollination in all kernels from shaded plants. On the other hand, dry weight accumulation in apical kernels was reduced only when ACC was applied to ears at 32 h after pollination. An application of ACC on 5 d after pollination did not affect these kernels. Cheng and Lur (1996) concluded that the stage most sensitive to ethylene was the initial period of endosperm nuclei division, prior to the stage of maximum cell division that takes place 6 to 9 d after pollination. The response pattern they observed is consistent with data obtained from split pollination treatments in our work.
Recent studies have also focused on the effects of cytokinins on kernel set. Endogenous levels of cytokinins increased in kernels from silking onwards, reaching their maximum at 9 to 12 d after pollination (Cheikh and Jones, 1994; Dietrich et al., 1995) coincident with the peak of mitotic activity within the endosperm. Stem infusion of a synthetic cytokinin (benzylaminopurine) at pollination increased KN per ear at maturity (Dietrich et al., 1995) and partially recovered apical kernel growth from high temperature stress (Cheikh and Jones, 1994). Nevertheless, studies with exogenous applied cytokinins do not help understand the nature of its endogenous production in relation to pollination timing among ovaries.
Finally, data presented in this work give evidence to support the existence of a "correlative signal" in the determination of kernel set, proposed in theory by Bangerth (1989). Results from the lateral heating experiment are more likely attributable to differences in assimilate partitioning due to alter sink activity mediated by temperature than to a "primigenic dominance." The latter, which suggests an inhibition promoted by the early developed sinks, involves the transfer of a "dominance signal." Results from split pollinations (gaps) indicate this process took place to some extent in ears of those treatments. Its transient nature suggests the transfer of the dominance signal would be mediated by a "second messenger" like ethylene (Bangerth, 1989). Next studies on maize kernel set should focus on these aspects, but keeping in mind the high correlation between assimilate portioning due to kernel growth and a primigenic dominance which involves a signal transfer.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Received for publication December 11, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Carcova and M. E. Otegui Ovary Growth and Maize Kernel Set Crop Sci., May 31, 2007; 47(3): 1104 - 1110. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
|
|
S. R. Anderson, M. J. Lauer, J. B. Schoper, and R. M. Shibles Pollination Timing Effects on Kernel Set and Silk Receptivity in Four Maize Hybrids Crop Sci., March 1, 2004; 44(2): 464 - 473. [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 Environmental Quality |
The Plant Genome | |||
