Crop Science Journal of Natural Resources and Life Sciences Education
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 Related articles in Crop Science
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 ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Agricola
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Related Collections
Right arrow Crop Physiology & Metabolism
Right arrow Soybean
Crop Science 42:1535-1540 (2002)
© 2002 Crop Science Society of America

CROP PHYSIOLOGY & METABOLISM

Synchronous Flowering and Fruit Set at Phloem-Isolated Nodes in Soybean

D. B. Egli* and W. P. Bruening

Dep. of Agronomy, Univ. of Kentucky, Lexington, KY 40546-0091

* Corresponding author (degli{at}uky.edu)


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Much of the variation in yield of soybean [Glycine max (L.) Merr.] results from variation in seeds per unit area, but the physiological regulation of seed number is not well understood. The relationship between the timing of flower development and pod and seed number was evaluated in three greenhouse experiments with a girdled-node system. Plants (‘Elgin 87’) were grown in 3.0-L pots filled with a soil-vermiculite mixture (2:1 v/v) and the main stem below Node 7 (one-node treatment) or below Node 5 (three-node treatment) was girdled when the first flower opened at Node 7. The leaves at Nodes 5 and 6 on the three-node treatment were removed and the main stem above Node 7 was removed on all plants. All nodes in the three-node treatment flowered at the same time and the total flower production was 68% greater than on the one-node treatment. The three-node treatment produced more pods (36–94%) and more seeds (36–78%) than the one-node treatment. Girdling-induced reductions in light saturated C exchange rate (CER) during flowering and pod set were consistently greater on the one-node than on the three-node treatment. However, the three-node treatment produced more pods per unit assimilate, suggesting that the increased CER did not account for the increase in pods and seeds. The number of pods and seeds at phloem-isolated nodes responded to synchronous flowering, suggesting that the timing of flower development may play a role in determining reproductive success in soybean.

Abbreviations: CER, carbon exchange rate • DAG, days after girdling


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
VARIATION IN THE NUMBER of seeds per unit area usually accounts for most of the environmental variation in yield of soybean and other grain crops (Shibles et al., 1975; Evans, 1993; Egli, 1998). Associations between seed number and canopy photosynthesis, crop growth rate, or plant growth rate during flowering and pod set (Christy and Porter, 1982; Egli and Zhen-wen, 1991; Vega et al., 2001) suggest that photosynthesis and the supply of assimilate mediates the relationship between environment and seed number. Seed number in soybean is also affected by treatments that affect photosynthesis, such as shade (Andrade and Ferrerio, 1996), increases in atmospheric CO2 concentration (Hardman and Brun, 1971), water stress (Shaw and Laing, 1966), and defoliation (Board and Tan, 1995), thus supporting the contention that seed number is related to the supply of assimilate.

Pod and seed number, however, were not closely associated with assimilate supply at single phloem-isolated nodes in soybean. Pod and seed number reached maximum levels at photosynthesis rates that were as low as 40% of the highest levels (Bruening and Egli, 1999; 2000). The availability of flowers did not seem to limit pod set as flower and pod abortion were always >50%. When photosynthesis increased with little increase in pod or seed number, starch accumulated in the leaves (Bruening and Egli, 2000). The failure of pod and seed number to increase in this C-rich environment with excess reproductive potential suggests that other factors besides assimilate availability may be partially regulating pod and seed number.

The timing of flower development and pollination can be an important determinant of seed number. For example, simultaneous pollination of silks within the apical ear or between the apical ear and the second ear of maize (Zea mays L.) increased seed set (Freier et al., 1984; Carcova et al., 2000). At low plant populations, most of the response to synchronous pollination occurred on second ears (39–535%), while at high populations the apical ear accounted for much of the response (8–31%; Carcova et al., 2000). Stress induced asynchronous pollination in maize reduces seed number (Jacobs and Pearson, 1991). Simultaneous development of reproductive structures in maize seems to improve seed set without a corresponding increase in assimilate supply.

Soybean is noted for its asynchronous flowering and pod set with 18 to 49 d between the first and last flower (Hansen and Shibles, 1978; Gai et al., 1984; Dybing, 1994). Synchrony has not been investigated extensively in soybean, but we speculated that the timing of flower development could provide an explanation for the curvilinear response of pod and seed number to assimilate supply at phloem-isolated nodes (Bruening and Egli, 1999). Although the first flowers to develop at a node are less likely to abort than later flowers (Huff and Dybing, 1980; Heitholt et al., 1986), detailed studies of flowering and pod set in soybean are complicated by its growth habit. Flowering occurs progressively toward the top of the main stem so that flowering, pod development, and rapid seed growth may occur simultaneously at different locations on the plant (Spaeth and Sinclair, 1984). The girdled-node system (Bruening and Egli, 1999; 2000), however, makes it possible to investigate the characteristics of flowering and pod set at a single node without interference from assimilate supplies or reproductive activity at other nodes.

Our objective was to evaluate the hypothesis that increasing the number of synchronous flowers while maintaining a constant assimilate supply would increase pod set in soybean. We used the girdled-node system to isolate target nodes and we increased the number of synchronous flowers by isolating three nodes instead of the single node used previously (Bruening and Egli, 1999, 2000).


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Soybean plants (Elgin 87, maturity group II) were grown in 3.0-L pots filled with a 2:1 (v/v) mixture of silt loam surface soil and vermiculite in a greenhouse at the University of Kentucky. Pots were overseeded and thinned to one plant per pot after emergence. Experiment 1 was planted on 14 Apr. 1999, Exp. 2 on 10 Aug. 1999, and Exp. 3 on 12 Apr. 2000. The plants were fertilized with 0.3 g per pot of Peters Professional fertilizer (20-20-20, N-P-K) at approximately weekly intervals. The seeds were not inoculated with Bradyrhizobium japonicum and the roots were not nodulated. Temperature in the greenhouse was maintained between 20 and 30°C and supplemental light ({approx}120 µmol m-2 s-1) was provided by high pressure sodium lamps (430 W, P.L. Light Systems, Grimsby, ON, Canada). The photoperiod was maintained at 14 h.

On 19 May 1999 (Exp. 1), 16 Sept. 1999 (Exp. 2), and 22 May 2000 (Exp. 3), when the first flower opened at the target node, the main stem was girdled by electrically heating a bare 0.7-mm diam. constantan wire wrapped around the stem as described previously (Bruening and Egli, 1999). The main stem above the target node was removed. The stem was girdled below Node 7 (Node 6 in Exp.1) for the one-node treatment (Node 1 was the node of the unifoliolate leaves). The main stem in the three-node treatment was girdled below Node 5 (Node 4 in Exp. 1), and the leaf and petiole were removed from Nodes 5 and 6 (Nodes 4 and 5 in Exp. 1), but not from Node 7 (Node 6 in Exp. 1). The leaf-bearing node was at the same nodal position in both girdled treatments in Exp. 2 and 3, but the girdle was two nodes lower on the main stem in the three-node treatment. In Exp. 1, there was a one-node difference in the position of the leaf-bearing node between treatments. Assimilate from the same leaf position, restricted to one node in the one-node treatment, was available to three nodes in the three-node treatment. A nongirdled control with the main stem above Node 7 removed was included in Exp. 2 and 3. A three-node treatment in which all of the flowers were removed daily from the middle and bottom nodes was also included in Exp. 3 to evaluate the relative effect of the one- and three-node treatments on pod set when there were no differences in flower production. New leaves developing on branches at the target nodes were removed daily in all treatments to eliminate additional sources of assimilate. There were six replications of each treatment in Exp. 1, 15 in Exp. 2, and 8 in Exp. 3, all arranged in a completely randomized design.

Newly opened flowers on the main raceme and subbranches were counted daily until flowering stopped. Newly opened flowers were identified by the appearance of the petals, but this technique was not entirely successful as total flower number by this technique was usually greater than the sum of flower scars and pods at maturity. This bias should be constant during the flowering period and probably did not affect treatment comparisons.

Light-saturated CER (PAR 2000 µmol m-2 sec-1, 4 or 5 leaves per treatment) was measured at approximately weekly intervals with a LI-6400 portable photosynthesis system fitted with a Li-Cor 6400-02 Red/Blue light source (Li-Cor Inc., Lincoln, NE). A model based on a PAR response curve and greenhouse environmental conditions was used to estimate total daily photosynthesis as described by Bruening and Egli (2000). The response curves were created by measuring CER at 13 PAR levels from 50 to 2000 µmol m-2 sec-1 on two replications of leaves from each treatment in Exp. 3 at 5 d after girdling (DAG) when there were large treatment effects on light saturated CER, and 30 DAG when treatment effects were minimal. The appropriate response curve was used to estimate CER for each 0.5-hr period during the day from PAR levels in the greenhouse (0.5-h means from two replications recorded with a Li-Cor 1000 data logger, Li-Cor, Inc.). These estimates of CER were adjusted for variation in air temperature (measured with two replications of shielded thermocouples at the top of the plants) in the greenhouse (Norman and Arkebauer, 1991), and treatment and temporal effects using the data from the weekly measurements of light-saturated CER with daily values estimated by linear interpolation.

The number of pods, seeds, and flower scars (Exp. 2 and 3) (Jiang and Egli, 1993) at the isolated nodes (top, middle, and bottom nodes on the three-node treatment) was determined at maturity. Combined flower and pod abortion in Exp. 2 and 3 was calculated from the number of flowers (flower scars + pods) and mature pods containing developed seeds.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Girdling the stem and removing the top of the plant photosynthetically isolates the targeted node(s), making it possible to study pod and seed set as a function of the assimilate supplied only by the leaf at the isolated node(s). This method also eliminates any possible interference from reproductive development on the rest of the plant so that pod set at the target nodes was not affected by reproductive activity at other nodes. This isolation by itself had minimal effects on reproductive development, as shown by a lack of significant differences in flower number between the control and one-node treatments (Table 1) and significant differences in pod and seed number only in Exp. 3 (Table 2) . Similar results were found in greenhouse and field experiments with Elgin 87 and two other cultivars (Bruening and Egli, 1999, 2000). As expected, the three-node treatment, with one leaf supporting reproductive development at three nodes, produced nearly 70% more flowers than the one-node treatment (Table 1). The main racemes flowered first (Gai et al., 1984) and all three nodes in the three-node treatment flowered at nearly the same time. Thus, the three-node treatment had a significant advantage over the one-node treatment in the number of early flowers with large differences occurring in the first 5 DAG in Exp. 2 and 3 and continuing into the second and third 5-day periods in Exp. 3 (Table 3) . More flowers were produced in Exp. 3 than in Exp. 2 as a result of more flowers on the subbranches (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. The distribution of flowers on main racemes and subbranches at phloem-isolated nodes on soybean plants.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Reproductive development at phloem-isolated nodes on soybean plants.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Flower production during flowering and pod set on girdled and nongirdled soybean plants.

 
The greater reproductive potential of the three-node treatment was associated with a significant (P = 0.05) increase in mature pods of 36 to 94% (Table 2). Seed number increased by similar amounts, so there were no significant treatment effects on seeds per pod, and the number of seeds was determined by the proportion of flowers that produced mature pods. There were more pods and seeds in all treatments in Exp. 3 than in Exp. 1 and 2, but the magnitude of the response to the three-node treatment was similar to the response in Exp. 2.

The leaf-bearing node (top node) of the three-node treatment produced more than twice as many pods and seeds as either the middle or bottom nodes, which produced equal numbers (Table 2). These results are consistent with results of 14C distribution studies showing preferential accumulation of radioactivity at the node of the labeled leaf (Stephenson and Wilson, 1977); however, the distribution of pods among the three nodes provides little support for the phyllotactic pattern of preferential distribution to alternate nodes described by Bloomquist and Kust (1971). Stephenson and Wilson (1977) also reported little evidence for a distribution pattern favoring alternate nodes.

Girdling the main stem of soybean frequently causes a substantial decrease in light saturated CER of the leaf at the girdled node (Bruening and Egli, 1999) and this response occurred in the experiments reported here (Fig. 1) . Girdling reduced the CER of the one-node treatment more than the three-node treatment, but, as in previous experiments (Bruening and Egli, 1999), the CER of the girdled plants recovered to control levels by {approx}25 DAG. The three-node treatment averaged 33% (Exp. 2) and 51% (Exp. 3) higher CER than the one-node treatment during the first 25 DAG. Although the mechanisms responsible for this reduction in CER are not known (Egli and Bruening, 1999), the effect was clearly less when the girdle was farther from the leaf-bearing node. The differences in daily total C exchange during the flowering period (Table 4) , modeled from PAR and temperatures in the greenhouse (Bruening and Egli, 2000), were smaller than the differences in light saturated CER, probably reflecting smaller treatment effects at the lower PAR levels commonly encountered in the greenhouse.



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 1. The effect of girdling the main stem on light-saturated C exchange rate (CER) during flowering and pod set, Exp. 2 and 3. Error bars represent ± SE of the mean.

 

View this table:
[in this window]
[in a new window]
  		Table 4. Total daytime C exchange during flowering and pod set of leaves at phloem-isolated nodes on soybean plants, Exp. 2 and 3.

 
The CER differences between treatments matched the differences in pods and seeds between the one- and three-node treatments, but further considerations argue against a cause and effect interpretation. First, the relative advantage in total C exchange for the three-node treatment was smaller than the increase in pods and seeds, resulting in an increase in pods per unit CER of {approx}12 and 32% (Table 4). Secondly, total pods and seeds on the three-node treatment were significantly (P = 0.05) reduced to 15.1 pods and 37.1 seeds per node (reductions of 42 and 45%, respectively) when flowering was restricted to the top node. There were more than enough flowers on this node (Table 1) to support the 26 pods produced on the three-node treatment (Table 2); thus, if the advantage for this treatment was simply a result of the higher CER levels (Fig. 1 and Table 4), removing the flowers would have had no effect on pod number. However, removing the synchronous flowers reduced pod and seed number, suggesting that it was the presence of these flowers that contributed to the advantage for the three-node treatment. And finally, in previous experiments with single girdled nodes (Bruening and Egli, 1999, 2000) pod and seed number did not respond much to CER rates that were >40 to 60% of the maximum rate. Maximum CER rates in those experiments were similar (between 5 and 20 µmol m-2 s-1 during flowering and pod set; Bruening, 1997) to those reported here; consequently, our CER rates should have been well above the level where pod and seed number are responsive to assimilate supply. The increase in pods and seeds in the three-node treatment seems to be a response to the presence of more flowers on the three nodes and was not simply a result of the enhanced assimilate supply.

Soybean has at least three axillary buds at each node (Gai et al., 1984). The middle bud can develop into a primary branch or a main raceme while the other two buds develop into subracemes or subbranches on reproductive nodes (Gai et al., 1984). The isolated nodes on most plants in our experiments developed subbranches that produced leaves and flowers. Most of the subbranches were extremely long and bore most of the flowers (65–88%, Table 1), in contrast with field plantings of 12 determinate and indeterminate cultivars, where 75% of the flowers were on the main raceme (Gai et al., 1984). The total number of flowers per individual node (12–56) was greater than the 8 to 10 reported by others (Brevedan et al., 1978; Hansen and Shibles, 1978; Gai et al., 1984; Jiang and Egli, 1993). Apparently, subracemes or subbranches do not make a major contribution to flower production on normal plants (Spaeth and Sinclair, 1984), although Gai et al. (1984) suggested that they may be more common in highly productive environments. The loss of apical dominance when the top of the plant was removed may have also contributed to subbranch development in our experiments as they were present on control plants and assimilate-rich girdled plants. Regardless of the cause, the large flower potential of the subbranches would contribute to the reproductive plasticity of the soybean plant and help insure that the production of pods would not be limited by the availability of flowers. The long subbranches and the large number of flowers provided an excellent system to study flower dynamics and pod and seed set.

More than 60% of the total pod and seed load was borne on the subbranches in Exp. 3 when the total pod load per node was large (control and one-node treatments, Table 5) . At very low pod loads in Exp. 2 (middle and bottom nodes on the three-node treatment), the majority of the pods and seeds were found on the main raceme and most of the flowers and pods on the subbranches aborted (Table 6) . These changing proportions probably reflect the relatively constant number of flowers on the main raceme (between 6 and 8, Table 1), which flowers first (Gai et al., 1984), and apparently provided adequate reproductive sites when pod numbers were low. But as pod number increased, the additional pods were set on the subbranches and, for extremely large pod numbers, the subbranches dominated. The subbranches were an important source of reproductive plasticity in the isolated-node system.


View this table:
[in this window]
[in a new window]
  		Table 5. Distribution of pods and seeds on main racemes and subbranches at phloem-isolated nodes on soybean plants.

 

View this table:
[in this window]
[in a new window]
  		Table 6. Flower and pod abortion on main racemes and subbranches at phloem-isolated nodes on soybean plants.

 
Pod or seed number are frequently modeled as a function of the assimilate supply from photosynthesis (Sheldrake, 1979; Charles-Edwards, 1984; W.G. Duncan quoted in Egli, 1998) based on the well-documented association between canopy photosynthesis, crop growth rate, or factors that influence canopy photosynthesis (e.g., shade, CO2 levels, water stress) and pod and seed number (Shaw and Liang, 1966; Hardman and Brun, 1971; Christy and Porter, 1982; Egli and Zhen-wen, 1991; Andrade and Ferrerio, 1996; Vega et al., 2001). Implicit in these models is the assumption that the number of flowers does not limit pod or seed number. Consistent high levels of flower and pod abortion in soybean support this assumption (Van Schaik and Probst, 1958; Hansen and Shibles, 1978). However, in the experiments reported here using the isolated-node system, increasing the number of flowers present at the same time with the three-node treatment increased pods and seeds at maturity. In a like manner, synchronous pollination within an ear and between the apical and subapical ears increased seed number in maize (Freier et al., 1984; Carcova et al., 2000). Apparently, initiating development of many pods or seeds simultaneously increases the number set per unit assimilate above that obtained when pods or seeds start development sequentially so that late developers initiate growth when the early ones are more advanced. The advantages of synchronous development, well documented in maize, also seem to be realized at phloem-isolated nodes in soybean.

Soybean is noted for its extreme reproductive plasticity. The number of nodes per plant and, ultimately, the number of flowers per unit area are sensitive to environmental conditions. The plant can also respond to a favorable environment by increasing the number of flowers per node. Consequently, it is not surprising that the number of flowers per unit area or per plant is always much greater than the number of pods. Within all of this reproductive plasticity that responds to the availability of assimilate, there also seems to be some role for the timing of flower development, as synchronous flowering increased pod set in the three experiments reported here. Failure of pod and seed number to respond to increasing assimilate supplies above a minimum level when excess flowers were available (Bruening and Egli, 1999, 2000) also suggests that assimilate supply may not be the only factor regulating pod and seed number in soybean.

It is not yet clear how our results from phloem-isolated nodes relate to an intact plant where asynchronous flowering is the normal pattern of development (Spaeth and Sinclair, 1984) with up to 3 or 4 wk between the pollination of the first and last flowers. Heitholt et al. (1986) found that flower and pod abortion at a specific node responded primarily to the physiological conditions at that node; thus, the variation in flowering on a plant may not be a detriment, as each nodal unit would operate independently. Recent reports suggesting that seed number is positively associated with the length of the flowering and pod or seed set period in soybean (Egli and Bruening, 2000) and other crops (Fischer, 1985) seems to contradict the potential value of synchronous flowering. Clearly, much remains to be learned about dynamics of flowering and seed set in soybean and other crops.

Investigators of flower and pod abortion in soybean frequently espouse one of two philosophies: (i) the flowers that abort and do not produce mature pods containing seed represent lost yield (Peterson et al., 1986), or (ii) that flower and pod abortion is an indication that the availability of flowers is not limiting pod and seed number, which are then solely limited by the assimilate supply (Egli, 1998). Dybing (1994) pointed out that interpreting the overproduction of flowers as indicating that events during flowering are not important may be an oversimplification. Our results support Dybing (1994). In spite of the reproductive plasticity at individual nodes and high levels of reproductive failure, the time of flower availability was important in determining pod and seed number at phloem-isolated nodes. A closer examination of the dynamics of flower and pod development at individual nodes is needed to determine the mechanisms responsible for the response to synchronous flowering reported here.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Published with the approval of the Director of the Kentucky Agric. Exp. Stn. as Paper 01-06-90.

Received for publication July 16, 2001.


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


Related articles in Crop Science:

This issue in Crop science

Crop Science 2002 42: 1393-1395. [Full Text]  




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 Related articles in Crop Science
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 ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Agricola
Right arrow Articles by Egli, D. B.
Right arrow Articles by Bruening, W. P.
Related Collections
Right arrow Crop Physiology & Metabolism
Right arrow Soybean

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Plant Registrations Soil Science Society of America Journal
Journal of Natural Resources
and Life Sciences Education		 Journal of
Environmental Quality