|
|
||||||||
a Dep. of Agronomy, Univ. of Kentucky, Lexington, KY 40546-0091 USA
brue1{at}pop.uky.edu
| ABSTRACT |
|---|
|
|
|---|
Abbreviations: CER, carbon dioxide exchange rate NCI, nodal carbon input CGR, crop growth rate
| INTRODUCTION |
|---|
|
|
|---|
, including soybean, is a function of the number of seeds produced per unit area and the average weight of the individual seed (seed size) (Shibles et al., 1975). Yield variation can result from differences in seed size and/or seed number, but environmentally induced variation is usually associated with changes in seed number (Shibles et al., 1975; Egli, 1993).
Soybean seed number per unit area seems to be related to canopy photosynthesis during flowering and pod set. Treatments that alter canopy photosynthesis during this period, such as CO2 enrichment (Cooper and Brun, 1967; Jones et al., 1984), or increased irradiance (Schou et al., 1978) caused corresponding changes in seed number. Herbert and Litchfield (1984), Ramseur et al. (1985), Egli and Yu (1991), Egli (1993), Board and Harville (1994), and Board et al. (1995) reported linear relationships between crop growth rate (CGR), measured during flowering and podset, and seed number in soybean. The variation in CGR was created by differences in row spacing and plant population (Herbert and Litchfield, 1984; Board and Harville, 1994), irrigation and intra-row spacing (Ramseur et al., 1985), shade (Egli and Yu, 1991; Egli, 1993), and defoliation (Board et al., 1995). The relationship for a single cultivar was consistent across years, but there were large differences between cultivars (Egli and Yu, 1991; Egli, 1993). At a constant CGR, large seeded cultivars produced fewer seeds per unit area than small seeded cultivars. These observations clearly document the involvement of both source activity (i.e., canopy photosynthesis) and sink characteristics in determining seed number per unit area in soybean.
The studies linking seed number and photosynthesis were conducted with plant communities. However, it is difficult to evaluate mechanisms translating changes in canopy photosynthesis into changes in seeds per unit area at the community level. Vegetative growth and the production of new nodes continues throughout much of flowering and podset so that alterations in canopy photosynthesis could change the number of potential fruiting sites (Board and Tan, 1995). Continuous new node production also creates large differences in the timing of reproductive development among nodes (Egli and Crafts-Brandner, 1996), making any investigation of the relationship between photosynthesis and the fate of individual flowers and pods hopelessly complex.
Successful investigations of possible mechanisms relating photosynthesis to seed number seems to require a simpler system, one where changes in photosynthesis can be directly related to flower and pod development. A single phloem-isolated node with its subtending leaf provides such a system. At a phloem-isolated node, sucrose cannot be transported to sinks other than those at the isolated node and no sucrose can reach the node from other leaves. Manipulation of photosynthesis at such an isolated node will result in variation in the sucrose supply to the developing sink and, for example, the effect of this variation can be related to specific stages of flower and fruit development.
The objectives of this research were, first, to develop and evaluate an isolated node system that could be used to investigate the relationship between photosynthesis and fruit and seed number, and, second, to use the system to evaluate the relationship between photosynthesis and seed number in soybean cultivars with differences in sink characteristics (final seed size and individual seed growth rate).
| Materials and methods |
|---|
|
|
|---|
Reproductive growth stages (Fehr and Caviness, 1977) were determined at weekly intervals on control plants. At reproductive growth stage R1, plants were thinned to 6 plants per meter of row. On 6 July 1995 and 8 July 1996 (Elgin 87), 13 July 1995 and 15 July 1996 (Emerald), 1 Aug. 1995 and 25 July 1996 (Essex) an individual node (sixth node, where unifoliolate node is node one) on the main stem of each plant was isolated by girdling the stem below the node and removing the portion of the plant above the node. The stem was girdled when the first flowers opened at the sixth node. Girdling was accomplished by wrapping a bare constantan wire (approximately 0.7 mm diameter) around the stem and electrically heating it. The wire was connected to a power source and current flow was regulated by a power controller set at 5% of the maximum power level. The wire was heated for 5 s. The girdle was considered a success if, after 2 d, a complete band of dead tissue was observed around the stem.
Differences in assimilate production at the girdled nodes were created by removing a portion (0%, non-defoliated; approximately 34%, middle leaflet removed; 66%, two side leaflets removed; 83%, two side leaflets and half of middle leaflet removed; 100%, entire leaf removed) of the original leaf area when the plants were girdled. Leaf area of the center and side leaflets removed from the 100% defoliation treatment was measured with a Li-Cor LI-3100 area meter (LI-COR, Inc., Lincoln, NE) to estimate leaf area removed from each treatment. There was also a control treatment (not girdled or cut [upper portion of the plant was not removed]), and a control-cut treatment (non-girdled plant with the portion of the plant above the sixth node on the main stem removed). The control-cut treatment was included to determine the effect of removing the top of the plant on photosynthesis of the leaf at the sixth node. Each cultivar was a separate experiment and the experimental design was completely randomized, with 6 to 12 replications (plants) per treatment. The girdling treatment was not always successful resulting in a variable number of girdled plants per treatment.
After the treatment was applied, leaf carbon dioxide exchange rate (CER) and stomatal conductance were measured with a Li-Cor LI-6200 (1995) or a LI-6400 (1996) portable photosynthesis system. The CER of leaves (four per treatment) at the girdled nodes was measured at approximately weekly intervals on relatively sunny days (PAR generally > 800 µmols m-2 s-1) between 1100 and 1400 h EDST. The leaves at the girdled node were continuously exposed to full sunlight because the top of the plant was removed and the individual plants were widely spaced. Leaves on the control treatment were exposed to full sunlight only during the CER measurement.
Seed growth rate was measured on control plants to characterize cultivar differences in sink characteristics which are related to seed number (Egli and Yu, 1991; Egli, 1993). At growth stage R5, 100 pods of similar size, without consideration of their nodal position, on control plants were marked with acrylic paint and two replications of 20 pods were collected twice, 14 d apart during the linear phase of seed growth to determine the seed growth rate for each cultivar.
At maturity, the number of pods, seeds, and flower scars (Jiang and Egli, 1993) at the girdled node were determined. The combined flower and pod abortion was calculated from flower number and mature pod number.
| Results and discussion |
|---|
|
|
|---|
effects on leaf CER. Leaves of both treatments were exposed to full sunlight during the CER measurements, but leaves on the control plants were partially shaded at all other times by the top of the plant. This difference in radiation environment had little effect on CER.
|
Defoliation and CER
Defoliation of the leaf at the girdled node increased CER for the first 20 d following defoliation in 1995 (Fig. 2)
. The differences between the high and low levels of defoliation were significant
for Elgin 87 and Essex, but not for Emerald. Treatment differences in CER were associated with differences in stomatal conductance (data not shown). In 1996, defoliation at the girdled node had no effect on leaf CER of cultivars Emerald and Essex (Fig. 3)
, but the 83% defoliation treatment for Elgin 87 had a significantly higher (2555%) CER than the non-defoliated plants during the first 10 d following defoliation. These differences in CER corresponded with differences in stomatal conductance, which was higher (approximately 5070%) in the 83% defoliation treatment than the non-defoliated treatment (data not shown).
|
|
Despite defoliation effects on CER per unit leaf area in most experiments, the defoliation treatments provided differences in the total nodal carbon input (NCI), which was calculated by multiplying leaf CER by the leaf area at the girdled node (Fig. 4 and 5) . Throughout the sampling period, NCI was inversely related to the proportion of leaf area at the isolated node, with fourfold differences between the 0 and 83% defoliation treatments. Hence, variation in leaf CER did not compromise the defoliation treatments which produced large variation in photosynthetic input at a single node.
|
|
|
|
The relationship between average NCI and pod or seed number was not consistent across years for pod number (Essex) and seed number (Essex and Elgin 87) at high levels of NCI (Fig. 6) indicating that something besides assimilate supply was affecting pod and seed number. It is possible that differences in temperature could be responsible, as has been reported for wheat (Triticum aestivum L.) (Fischer, 1985) and sunflower (Helianthus annuus L.) (Cantagallo et al., 1997). The number of cells in soybean cotyledons may be sensitive to environmental conditions (Egli et al., 1989), so the variation between years could also be due to variation in sink characteristics (cell number), but there were no large differences in individual seed growth rate across years for Essex (Table 2) . However, the comparison across years could not be made for Elgin 87 because its individual seed growth rate was not measured in 1995. Additional research is needed to evaluate this variation between years.
|
At low levels of NCI, pod and seed number for each cultivar were closely associated with NCI across years. This relationship indicates that the primary effect of the environment on pod and seed number was expressed through its effect on NCI. These results agree with other reports of a close association between canopy photosynthesis and seed number (Herbert and Litchfield, 1984; Ramseur et al., 1985; Egli and Yu, 1991; Egli, 1993; Board and Harville, 1994; Board et al., 1995). In addition, treatments designed to alter canopy photosynthesis (CO2 enrichment, Cooper and Brun, 1967; Jones et al., 1984; light enrichment, Schou et al., 1978) and the availability of assimilate also produced corresponding changes in pod and seed number.
The ability of the girdled node technique to mimic, in part, canopy level investigations of the relationship between seed and pod number and photosynthesis suggests that this technique can be used to investigate further the mechanisms involved in the determination of seed number. However, pod and seed number always reached a maximum at relatively low levels of average NCI, representing a significant departure from the linear relationships commonly observed at the canopy level and predicted by models of pod and seed set proposed by Sheldrake (Sheldrake, 1979), Duncan (Egli, 1998) and Charles-Edwards (Charles-Edwards et al., 1986). This deviation does not negate the usefulness of the isolated node technique but does highlight the importance of canopy level factors such as fertile nodes per unit area (Board and Tan, 1995; Jiang and Egli, 1993).
The curvilinear response of pod and seed number to increasing NCI occurred in both years and for all three cultivars. This response also occurred in a greenhouse experiment with Elgin 87 (Bruening, 1997). Understanding the factors that influence pod and seed number and developing realistic algorithms for predicting pod and seed number in crop models requires an explanation of this curvilinear response. The response did not appear to be caused by a lack of flowers, as substantial flower and pod abortion occurred at high NCI levels. Sheldrake (1979), Duncan (Egli, 1998), and Charles-Edwards (Charles-Edwards et al., 1986) hypothesized that there was an assimilate threshold, or assimilate concentration (Wardlaw, 1990), below which sink development would not occur. Once seeds from the first pods to develop entered the linear phase of seed growth, the cumulative assimilate utilization by the seeds at the girdled node may have been great enough to permanently reduce the assimilate concentration in the phloem below the threshold level, preventing the development of additional pods. This hypothesis would explain the cultivar differences as cultivars (e.g., Essex) with slow growing seeds would require more seeds to reduce the assimilate concentration below the threshold than cultivars with high seed growth rates (e.g., Emerald). Alternatively, hormones produced by the early set fruit may trigger abortion in late developing fruit (Huff and Dybing, 1980) and produce the curvilinear response. Identification of the correct explanation will require additional research, which should be facilitated by use of the girdled node technique.
In conclusion, the girdled node technique described here removes much of the complexity of vegetative and reproductive growth at the canopy level and provides a simple system where the development of the reproductive sink is totally dependent upon sucrose from a single leaf. In many respects, the system mimics relationships reported in the literature (positive relationship between photosynthesis and pod and seed number, inverse relationship between pod and seed number, and individual seed growth rate) suggesting that it can be used further to investigate the factors determining pod and seed number in soybean. The results also demonstrate that the carbon supply at these single nodes was much greater than that needed to support the developing sink. This finding merits further investigation.
| ACKNOWLEDGMENTS |
|---|
| NOTES |
|---|
|
|
|---|
Received for publication July 26, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. B. Egli and W. P. Bruening Shade and Temporal Distribution of Pod Production and Pod Set in Soybean Crop Sci., August 1, 2005; 45(5): 1764 - 1769. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. B. Egli and W. P. Bruening Synchronous Flowering and Fruit Set at Phloem-Isolated Nodes in Soybean Crop Sci., September 1, 2002; 42(5): 1535 - 1540. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Bissuel-Belaygue, A. A. Cowan, A. H. Marshall, and J. Wery Reproductive Development of White Clover (Trifolium repens L.) is Not Impaired by a Moderate Water Deficit That Reduces Vegetative Growth: II. Fertilization Efficiency and Seed Set Crop Sci., March 1, 2002; 42(2): 414 - 422. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Kumudini, D. J. Hume, and G. Chu Genetic Improvement in Short Season Soybeans: I. Dry Matter Accumulation, Partitioning, and Leaf Area Duration Crop Sci., March 1, 2001; 41(2): 391 - 398. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. B. Egli and W. P. Bruening Potential of Early-Maturing Soybean Cultivars in Late Plantings Agron. J., May 1, 2000; 92(3): 532 - 537. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||