| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Coop. Ext. Serv., Univ. of Arkansas, P.O. Box 391, Little Rock, AR 72203
b Dep. of Agron., 210 Waters Hall, Univ. of Missouri, Columbia, MO 65211
* Corresponding author (jjennings{at}uaex.edu)
Received for publication December 11, 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
Abbreviations: +I+F, insecticide and fungicide added • -I-F, no insecticide or fungicide added • L1, L2, and L3 are Locations 1, 2, and 3, respectively
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
Interseeding more alfalfa to extend the productive life of a declining alfalfa stand would be a desirable alternative to destroying the old plants and complete re-establishment, but interseeding is rarely successful. In New Hampshire, establishment of sod-seeded alfalfa was significantly inhibited by old stands at densities of only 10 plants m-2 (Mueller-Warrant and Koch, 1981). Soil-borne pathogens may be a problem (Godfrey et al., 1986), but Shroyer et al. (1994) reported establishment failures in Kansas when no-till planting into existing old, thin alfalfa stands, even when fungicide was applied to alfalfa seed at above labeled rates. In Iowa, fungicides applied to soil or seed at different rates and combinations did not improve alfalfa seedling survival when planted after alfalfa or orchardgrass (Dactylis glomerata L.) (Hurd et al., 1994). Alfalfa seedling mortality was higher when planted after alfalfa, even with fungicide treatment, than when planted after orchardgrass, and this result was attributed to autotoxicity.
Observations in Missouri indicate that alfalfa stand densities commonly decline to approximately 50 to 60 plants m-2 by the third or fourth year after establishment regardless of plant density the seeding year (Nelson et al., 1998). Competition and/or an allelopathic chemical or chemicals produced by the seedlings and young plants may contribute to the decline because surviving plants tend to be evenly spaced. Rice (1984) suggests that most, if not all, spatial patterns of plants are due to a combination of allelopathy and competition and not to either factor alone.
Allelochemicals from both foliage and root exudates contribute to allelopathy in certain ecosystems (Tukey, 1969; Putnam, 1985). Extracts of alfalfa top growth were reported to be more autotoxic to young alfalfa seedlings than were extracts from roots or crowns (Miller, 1996). Production and deposition of autotoxic chemicals from an established plant may contribute to development of a zone of influence around it within which establishment and growth of new alfalfa plants would be reduced. In Ontario, Canada, growth of peach (Prunus persica L.) seedlings was strongly inhibited within 0.9 m of old peach tree stumps or the sites where old stumps had been removed, but seedling growth was normal beyond that distance (Patrick, 1955). Fumigation with methyl bromide, ethylene dibromide, or a propane–propene mixture did not alleviate the problem, suggesting that the effect was not due to seedling disease or soil insects.
Observations from our previous field studies suggested that established alfalfa plants may also develop such a zone of autotoxic influence. Attempts to thicken two old alfalfa stands, with densities of 23 and 26 plants m-2, by no-till interseeding in April resulted in weak, stunted seedlings that died during the first summer (Jennings and Nelson, 1991). Well-defined rows of new alfalfa seedlings were very apparent just outside the plot area where drill rows extended into tall fescue (Festuca arundinacea Schreb.). However, plant density decreased sharply in rows near established alfalfa plants, leaving a bare zone of several centimeters around the established plants.
It would be advantageous for producers to be able to interseed alfalfa into a declining alfalfa stand to maintain productivity and extend stand persistence. Defining the zone of autotoxic influence around established alfalfa plants would be important to determine a threshold plant density below which interseeding into a declining stand might be possible. Our objective was to characterize the lateral zone of autotoxic influence around established alfalfa plants as influenced by pesticides, whether the established plant remained alive, and the effect of clipping the established alfalfa plant.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
|
Cody alfalfa was seeded around each test plant about 1 May 1992. To achieve similar interplant competition within the plot, alfalfa was planted in a pattern of eight concentric hexagons. The test plant was located in the center, with the six corners of the center hexagon being positioned 12.5 cm from the test plant. Each succeeding hexagon was 12.5 cm larger in radius than the one previous so that the plants in the corners of the eighth and outermost hexagon were positioned 1 m from the test plant. Inoculated alfalfa seed were planted at the six corners and at a 12.5-cm spacing along the perimeter of each hexagon. In this arrangement, alfalfa seed were planted at 216 positions around the test plant at an equidistant spacing of 12.5 cm within and between each hexagon. A plywood template with holes at the proper spacing was used to facilitate planting. Three seeds were planted through each hole in the template. After emergence, plants were thinned to one seedling at each position. Before planting, benefin [N-butyl-N-ethyl-2, 6-dinitro-4-(trifluoromethyl) benzenamine] herbicide at 1.25 kg a.i. ha-1 was incorporated with hand tools into the surface 2.5 cm of soil in each plot to reduce infestation by annual weeds.
Sixteen plot areas were prepared, eight at both L1 and L2, to compare the zone of influence as affected by frequent and infrequent clipping of test plants and combinations of control or experimental rates of insecticide and fungicide. The treatments where insecticide and fungicide were added (+I+F) received metalaxyl {(R)-[(2,6-dimethylphenyl)-methoxyacetyl-amino]-propionic acid methyl ester} fungicide seed treatment at 0.2 g a.i. kg-1 seed plus 15% chlorpyrifos [O,O-diethyl-O-(3,5,6-trichloro-2-pyridinyl) phosphorothioate] insecticide granules that were broadcast by hand at 1.12 kg a.i. ha-1 just before seeding.
Unfortunately, about half the test plants were killed by a late-spring frost shortly after pesticides were applied and plots were seeded but before the seedlings emerged. This left insufficient live test plants for the clipping treatments. Although the number of replications was not even, the objective was changed to compare the effects of live vs. dead test plants. Responses would be due to competition and autotoxicity (live plants) compared with autotoxicity alone (dead plants). Location 1 had two dead and six live test-plant replications with +I+F and three dead and five live test-plant replications when no insecticide or fungicide was added (-I-F). Location 2 had four dead and four live test-plant replications with +I+F and five dead and three live test-plant replications with -I-F.
Top growth of live test plants was supported by wire mesh cylinders, approximately 15-cm diam., to reduce lodging of stems and subsequent shading of seedlings. All test plants were clipped to a 10-cm height in early June at midbloom as the seedlings became established. Malathion (O,O-dimethyl phosphorodithioate of diethyl mercaptosuccinate) insecticide was applied to the test plants to control larvae of alfalfa weevil (Hypera postica Gyll.).
For descriptive purposes, young alfalfa plants grown around the test plant are referred to as seedlings. Alfalfa seedlings emerging at the corners of each of the eight concentric hexagons were used for data collection. They were marked soon after emergence by a wooden peg placed 2 cm to the side of each seedling. In the few cases where no seedling emerged at the corner position, the adjacent seedling in that hexagon was used. The 87.5-cm position was designated as the control for distance effects because it extended beyond the suspected zone of influence observed in a previous experiment. The outermost hexagon at the 100-cm position was not used as a control for distance effects because, due to the hexagon plot design, alfalfa seedlings were not bordered by other seedlings on all six sides.
Height of seedling plants was measured 43 d after planting. Plots were harvested one time when the seedling plants reached late bloom. At harvest, data were recorded for number of surviving seedlings and total dry matter yield of the six seedlings in each hexagon. Data were analyzed by the General Linear Models procedure because of the uneven number of replications involving dead and live test plants. Significant differences were reported at the 0.05 level of probability.
Location 3
In the spring of 1992, test plants were selected from a 5-yr-old stand of ‘Cimarron’ alfalfa. The soil at L3 was a Waben very cherty silt loam (loamy-skeletal, siliceous, mesic Ultic Hapludalf). Soil test levels for L3 were pH of 7.1, Bray-1 P of 31 mg kg-1, exchangeable K of 203 mg kg-1, and organic matter of 1.7%. Plots were fertilized annually with P, K, and B according to Missouri soil test recommendations for a yield of 11 Mg ha-1. As in L1 and L2, alfalfa plants within a 2-m radius of the test plants were killed with herbicide, and the area was left fallow. Test plants were held upright by a wire cylinder and cut each time they reached bloom stage during 1992. On 5 May 1993, Cody alfalfa was seeded in four 1.2-m-long rows spaced 90° apart in a spoke-like arrangement with the test plant at the center. This planting design facilitated easier and faster harvest than the hexagon design used in L1 and L2. All plots received granular chlorpyrifos insecticide in the seeded row, and seed were treated with metalaxyl fungicide at the same rates used in L1 and L2. Plots were hand-weeded or sprayed as needed with imazethapyr {2-[4,5-dihydro-4-methyl-4-(1-methylethyl)-5-oxo-1H-imidazol-2-yl]-5-ethyl-3-pyridinecarboxylic acid} at labeled rates.
Seed were inoculated and planted, spaced at 2.5-cm intervals in the 1.2-m-long rows extending from the test plant. Approximately four seed were planted at each 2.5-cm interval. After emergence, seedlings in each row were thinned to one per 2.5 cm. For data collection, each row was divided into five 20-cm sections, starting at the test plant. The last 20 cm of row (from 1–1.2 m) was planted to provide competition on the outer end of the 80- to 100-cm row section, allowing the 80- to 100-cm row section to be the control.
During the seeding year, test plants were either clipped at 2-wk intervals to a stubble height of 25 cm to reduce competition on surrounding seedlings or clipped to a 5-cm stubble when seedlings reached early bloom. Each clipping treatment was replicated four times. Top growth of the test plants was supported by wire mesh cylinders (approximately 15-cm diam.) to reduce stem lodging. Seedlings and test plants in both treatments were cut to a 5-cm stubble each time the seedlings reached early bloom.
Alfalfa seedlings were harvested three times in 1993, the seeding year, and five times in 1994. For consistency, young alfalfa plants grown around the test plant are referred to as seedlings for both years. At each harvest in 1993, the number of seedlings and dry matter yield were recorded for each 20-cm row section. During 1994, the number of stems per row section was also recorded at each harvest. Frequent clipping of test plants was discontinued in 1994, and test plants in both treatments were harvested when seedling plants in the rows were harvested at early bloom. However, data were still recorded separately for both clipping treatments.
In June 1995, about 26 mo after seeding, two replications of the clipped treatment were selected to determine if alfalfa root growth followed response patterns similar to top growth with increasing distance from the test plants. Parallel trenches were dug with a backhoe approximately 1.5 m deep and about 0.2 m from either side of two selected rows for each of the two plots. Alfalfa seedlings in the rows were held in place during root excavation by clamping the stems at ground level between two 2.4-m long boards. This allowed us to carefully remove the soil from around and below the intact and suspended alfalfa roots to a 1-m depth. Once the plants were excavated, the top growth was removed 5 cm above the cotyledonary node. Number of plants, number of root branches, root weight, and crown weight of the alfalfa seedlings were determined in each row section. Root volume was measured by water displacement. Data for the two rows within each replication were combined for analysis.
We tested the directional orientation of the radial rows of seedling plants in L3 and could not determine a row directional effect on the yield or plant density response. Data were analyzed by analysis-of-variance procedures. Significant differences were reported at the 0.05 level of probability.
Theory: Separating Autotoxicity from Competition
Separating plant interference with another plant from competition is difficult because the two processes occur simultaneously. In this case, the test plant competes with the seedlings for resources and sustains an autotoxic environment. Competition models have been established to describe the relationship between plant density and yield for many self-thinning plant populations. Interplant competition results in an inverse relationship between plant density and plant size and is described by the "-3/2 thinning law" (Silvertown, 1987). According to the -3/2 thinning law, the density of plants in a dense population decreases due to mortality as individual plants become larger with age, leading to a population change from one with many small individuals to one with a few large individuals. Every change of three log units in mean plant weight corresponds to a change of only two log units in plant density. When the size and number of plants in a population reaches the carrying capacity of the environment, the relationship between log plant density and log mean plant weight becomes -1. At this point, any increase in mean plant weight causes a proportionate decrease in plant survival and is referred to as the "law of constant yield" (Watkinson, 1986).
Weidenhamer (1996) proposed a model (Fig. 1) to demonstrate the presence of allelopathy based on deviation of plant density and yield from these established competition models. This model shows that in the presence of a phytotoxin, as plant density decreases, the slope of the line relating log mean plant weight and log plant density will deviate from the negative slope. The slope of the line will approach zero and then reverse to a positive slope. The position of the change depends on the phytotoxin concentration. At high toxin concentrations, the maximum individual plant weight will occur at an intermediate plant density. In soils containing low toxin levels or when toxins are not replenished in the soil, the slope of the line may only decrease and not reverse. Data for L3 were fitted to this model to help explain observed responses.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
Distance Effects
Patterns were similar, but locations differed for the response of seedling plant height (Fig. 2A)
and number of plants (Fig. 2B) to distance from the test plants, so data are presented separately. Data for live–dead test plants and pesticide treatments were combined for distance effects. Seedling height at 12.5 cm from the test plant was significantly lower than at the 87.5-cm control distance for both L1 and L2 (Fig. 2A). At L1, there was no difference in seedling height between 37.5 and 87.5 cm from the test plant, but height at the 25-cm distance was lower than at distances of 37.5 through 62.5 cm from the test plant. At L2, height was not different between 25 and 87.5 cm, except at 50 cm where seedlings were shorter than at 62.5 cm from the test plant.
|
The location x distance interaction was not significant for total dry matter yield, and variances were similar, so data for locations were combined for presentation. Dry matter yield of seedlings planted at 12.5 cm from the test plant was significantly lower than at 87.5 cm from the test plant (Fig. 2C). Yield was not different between 37.5 and 87.5 cm from the test plant, but yield at the 37.5- and 62.5-cm distances was significantly higher than at 25 cm.
Location 3
Clipping Effects
Clipping test plants frequently at L3 to reduce competition did not improve establishment or yield of seedling alfalfa. Seedling plant density was not significantly different between the clipped or unclipped test-plant treatment. But, yield of seedlings was significantly lower in the clipped test-plant treatment than in the unclipped test-plant treatment, suggesting that clipping established plants did not reduce the overall effect. Seedling yield of the clipped treatment ranked significantly lower (p < 0.05) than in the unclipped treatment for all three harvests in 1993 (seeding year) and for three of five harvests in 1994 (Table 2). There was no interaction of clipping treatment with distance for yield or plant density.
|
|
Seedling yield per row section averaged over test-plant treatments was significantly affected by distance from the test plant. Total yield in 1993 (seeding year) was lowest within 20 cm of the test plant (Fig. 4) . Yield of the 40- to 60-cm row section was significantly higher than for both the control distance and the 20- to 40-cm row section. In 1994, yield per row section continued to rank significantly lower for the 0- to 20-cm row section than for the control. Yields for the 40- to 60- and 60- to 80-cm row sections were significantly higher (p < 0.05) than those of the control (Fig. 4). In both years, seedling yield in the control row section (80–100 cm) was not significantly different from the 20- to 40-cm row section.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
Response to the live or dead test-plant treatment was significant only at L2 for the interaction of treatment with distance for weight per seedling and the treatment main effect for seedling yield. The lack of significant differences for all factors at L1 and for seedling height and seedling number at L2 indicates that both autotoxicity and competition from the test plants were involved. Although there was no competition from the dead test plants, seedling density and total yield followed similar patterns with respect to distance from both live and dead test plants, resulting in a nonsignificant interaction of treatment with distance. The higher yield of seedlings at L2 in the dead test-plant treatment compared with the live test-plant treatment suggests that the negative response of autotoxicity is lessened without competitive effects of the live plant. The inconsistent response between locations may have been due to differences in soil texture, especially with high summer rainfall (Table 1), which can influence the rate of autotoxin movement (Jennings and Nelson, 1998).
Size of the Zone of Influence
Negative effects of the test plants on seedling growth were generally consistent with respect to distance from the test plant across experiments and years. For both plot designs, the size of the zone of autotoxic influence ranged from 20 to 25 cm in radius for most parameters measured. This zone remained consistent even though rainfall was above normal through the summer months of 1992, when L1 and L2 were established, and was below normal in May and July in 1993 when L3 was established (Table 1).
Effects of autotoxicity should increase near the test plant. The same would be true for competition. Autotoxic effects are greater from top-growth extracts than from root extracts (Miller, 1996). However, the mechanisms and rates of transfer from the leaves to the soil are unknown. Transfer could be from washing of the leaves by rain or by fall of older or diseased leaves at the base of the canopy. Removing the existing alfalfa plants surrounding the test plant 1 yr before planting the experiments should have allowed dissipation of the autotoxic chemical from the soil (Miller, 1996), except for the zone-of-influence area where it was being recharged by the test plant. The different response between L1 and L2 for the live–dead test-plant treatment suggests that both competition and autotoxicity affect growth of new alfalfa planted near the test plants, but the relative effect of each will vary depending on the conditions.
At each location, plant density was proportionately less affected than dry matter yield in positions closest to the test plant. Many laboratory tests show percentage germination of alfalfa is affected less by extracts from alfalfa tops than is root growth (Miller, 1996; Jennings and Nelson, 1998; Chon et al., 2000). Thus, alfalfa stands affected by sublethal autotoxicity may have good plant density but low yields compared with a control. In Missouri, alfalfa planted 6 mo after killing an old alfalfa stand, which allowed for some of the autotoxin to dissipate, had equal stands but significantly lower yield than the 18-mo control (Jennings and Nelson, 2002). Similar results were reported in Wisconsin (Cosgrove, 1996) where plant density of alfalfa no-till–planted into an old alfalfa field after a 36-wk interval was 100 to 117% that after 1 yr of corn (control), but yield was only 58 to 66% of the control.
Response to Clipping
Results from L3 suggest that trying to reduce competition from an old stand by frequent cutting or grazing may not be beneficial for alleviating autotoxicity during reseeding. Reduced yield of alfalfa in positions nearest the test plants may have been partially due to intense competition from the test plants, but the lower yield of the clipped treatment compared with the nonclipped treatment in L3 (Table 2) suggests that the yield reductions were not due to competition alone. The higher yielding and more vigorous test plants in the unclipped treatment were expected to be more competitive with seedlings than test plants that were frequently clipped during the season.
Plant stress is known to increase production and effects of allelochemicals (Einhellig, 1989, 1996), which could be a factor when alfalfa is repeatedly defoliated. Seedlings closest to the test plants would likely be most affected. If frequent clipping caused the test plants to produce more autotoxins, then root growth of seedlings could be inhibited, likely causing a concomitant reduction in top growth. In L3, root mass was reduced in a manner that closely matched that of top growth at each position (Fig. 5C). Data from the clipping treatments also suggest that seedlings become conditioned during establishment to have a lower productive capacity than unaffected plants as the early response remained through the second year. This autoconditioning effect was also noted in our field experiments (Jennings and Nelson, 2002).
Separating Autotoxicity from Competition
At L3, the inhibition of seedlings within 20 cm of test plants was due to a combination of autotoxicity and competition, whereas seedlings at the midpoint of the row appeared to have stimulated growth compared with the control. Plant density was significantly higher at 40 to 60 cm and dry matter yield at 40 to 80 cm from the test plant than for the control (Fig. 3A and 4). Other studies have shown stimulated growth of alfalfa seedlings that were exposed to low concentrations of alfalfa extracts (Chon et al., 2000) and in other species when phytotoxin concentrations were diluted to low levels per seedling (Winkle et al., 1981; Thijs et al., 1994). In these cases, the presence of a phytotoxin may result in maximum plant size at an intermediate seedling density (Fig. 1) and reduced size at both low density (due to high phytotoxicity) and high density (due to intense resource competition) (Weidenhamer, 1996). Essentially, growth may be stimulated by low concentrations of phytotoxins at some intermediate seedling density at the periphery of the zone of influence. Alfalfa roots can have a lateral spread of >0.5 m, especially when depth of penetration is limited (Weaver, 1926). Lateral spread of roots of excavated test plants at L3 was observed to be 15 to 20 cm, which corresponds with the 0- to 20-cm row section where lowest plant density and yield occurred. The 40- to 60-cm row section, where greatest plant density and yield occurred, was beyond the observed lateral spread of test-plant roots.
In L3, seedling density decreased between Year 1 and Year 2, but annual dry matter yield increased. Such a shift of decreasing plant density due to interplant competition followed by increasing size of remaining plants has been noted for alfalfa populations (Nelson et al., 1998). However, in L3, highest seedling yield and highest yield per seedling (data not shown) occurred at the highest seedling densities (Fig. 3A and 4). This effect is not adequately explained by competition. Further, the maximum density and yield of seedlings were higher than that of the control, suggesting a stimulatory effect.
Regression analysis of the relationship between seedling density and seedling yield per section (Fig. 6A) and of the relationship between log seedling density and log mean yield per seedling (Fig. 6B) showed positive slopes in contrast to a negative slope of -1 suggested by the law of constant yield. Data points associated with each line represent specific row sections for each harvest, but they are not necessarily sequential with distance from the test plant because yield and seedling density responses peaked between 40 and 60 cm. The positive slopes of the observed effect remained relatively consistent among all harvests as seedling density decreased from harvest to harvest. This result supports the conclusion that the zone of influence surrounding alfalfa test plants was not simply due to competition and that a form of interference, probably autotoxicity, was involved. This is supported by the live–dead test-plant treatment where the zone of influence around the dead plants was similar in size to that of the live plants, which also competed for resources.
|
|
SUMMARY |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
In Missouri, alfalfa stands for hay production are generally considered marginal at 32 plants m-2 because yield is reduced and weed encroachment increases. That density is more than four times greater than the maximum density that would allow some seedlings to be established between autotoxic zones and be fully productive. Still, those seedlings that establish within the 20-cm zones of influence would contribute little to yield.
Consistent effects observed for the zone of influence of established alfalfa plants on seedling establishment and growth suggest that attempts at interseeding alfalfa to increase density of declining stands should be discouraged at densities of 
8 plants m-2. This quantitative assessment is relatively consistent with the empirical density of 10 plants m-2 in New Hampshire (Mueller-Warrant and Koch, 1981). Relationship of plant density and dry matter yield at L3 and the effects of live and dead test plants suggest that observed effects were due to both autotoxicity and competition although the relative impact of either factor could not be partitioned. Further study is needed to determine if climatic regions and soil types affect the zone of influence and whether management practices such as increased seeding rates or use of autotoxicity-tolerant cultivars may alter the density relationship to improve success of thickening old, thin alfalfa stands.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S.-U. Chon, C. J. Nelson, and J. H. Coutts Osmotic and Autotoxic Effects of Leaf Extracts on Germination and Seedling Growth of Alfalfa Agron. J., November 1, 2004; 96(6): 1673 - 1679. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
