| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Agronomy, Iowa State Univ., Ames, IA 50011
b Dep. of Statistics, Iowa State Univ., Ames, IA 50011
* Corresponding author (Bryce.M.Lemke{at}monsanto.com)
Received for publication May 30, 2002.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Three primary obstacles that have limited widespread use of eastern gamagrass have been its ability to tolerate close grazing (Rechenthin, 1951), is establishment difficulty (Ahring and Frank, 1968), and its low seed production (Polk and Adcock, 1964). Several grazing strategies that incorporate deferment and residual height recommendations have been developed to improve stand persistence (Brejda et al., 1996; Aiken, 1997; Aiken and Springer, 1998). Recent work has led to techniques for improved establishment of eastern gamagrass (Anderson, 1985; Kindiger, 1994; Mueller et al., 2000; Aberle et al., 2000). However, little has been done in improving seed production.
One reason for low seed yields in eastern gamagrass is the indeterminate nature of its reproductive growth, with appearance of spikes on the same plant occurring over a considerable time period (Jackson and Dewald, 1994). Seed harvest is further complicated because mature seed easily shatters from the plant, and seed ripens over a period of several weeks (Polk and Adcock, 1964; Wright et al., 1983).
Harvestable seed yield of eastern gamagrass may be increased by use of N application and defoliation treatments. Application of N can increase seed yield for grass species by increasing inflorescence density per unit area (Hill and Loch, 1993). Masters et al. (1993) found that application of 67 kg ha-1 of N significantly increased the density of reproductive tillers and seed produced for big bluestem (Andropogan Gerardi Vitman). Defoliation also has been shown to create conditions that increase reproductive tiller development for warm-season grass crops (Hulbert, 1988). For example, Vogel and Bjugstad (1968) found that clipping little bluestem [Schizachyrium scoparium (Michx.) Nash], big bluestem, and indiangrass [Sorghastrum nutans (L.) Nash] for 3 successive years at the seed-ripened stage or later increased spring-time tillering of plants. Hulbert (1988) found that removing plant debris in the spring substantially increased flowering stem density in big bluestem and indiangrass by improving the light environment of emerging shoots.
There is limited information regarding cultural practices for seed production of eastern gamagrass. Dewald and Kindiger (2000) recommended planting in wide rows (90–120 cm). For soil fertility, they recommended that soil P should be maintained "on the high side" and that annual application of N is needed. They state "timing of harvest is judgmental and should be based on frequent inspections to determine caryopses content or grain fill of the cupulate fruitcase enclosure." Precise information regarding N rates, optimum harvest time, and methods of overcoming problems of reproductive indeterminacy and seed shattering are needed to increase harvestable seed yields of eastern gamagrass. The objectives of this research were to evaluate the effects of N application and defoliation on inflorescence appearance, seed loss, and viability of harvested seed in two cultivars of eastern gamagrass.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Precipitation and daily high and low temperatures were measured during the growing season (1 April–31 October) on the agronomy farm at Boone, IA, for the 2000 and 2001 seasons. The 50-yr average precipitation and high and low temperature totals were obtained from a weather station in Ames, IA, approximately 5 km northeast of the experimental site. Daily growing degree units (GDU), with 10°C as the base temperature, were figured using the equation:
 |
For the current study, eastern gamagrass stands were burned on 14 Mar. 2000 and 6 Apr. 2001, before green-up. In 2000, each block of gamagrass was divided into four whole plots composed of 12 rows each. Each block was randomly assigned to receive 0, 56, 112, or 224 kg N ha-1. The N treatments were applied with a Gandy Model 6505 drop spreader (Wikco Industries, Lincoln, NE) on 26 Apr. 2000 and 1 May 2001, with NH4NO3 as the N source. In addition, whole plots were divided into four-row split plots; each split-plot was randomly assigned to receive one of three defoliation treatments consisting of a control (no defoliation), spring defoliation, or fall defoliation. The spring defoliations were done on 22 May 2000 and 29 May 2001, after the plants reached the 4 to 5 collared leaf stage. Plant height at this stage was approximately 60 cm. A fall defoliation treatment, taken on 17 Aug. 2000, was included as a treatment for the second year of the study to assess its influence on seed harvest during the 2001 season. Defoliation was done with a Bush Hog RDTH 72 finishing mower (Bush Hog LLC., Selma, AL) at a 45-cm height.
Data Collection
Anthesis counts, cupule counts, and seed harvests were used to characterize flowering and seed production. Anthesis counts were taken weekly from June through August on an interior row of each plot to minimize border effects. Seed tillers were considered to be at anthesis when anthers began shedding pollen from the staminate portion of the inflorescence. Once tillers reached this stage, biodegradable flagging tape was tied below the inflorescence to signify that the tiller had flowered and was counted. Terminal and lateral inflorescences were counted independently (terminal inflorescences emerge from the apical meristem of a reproductive tiller while lateral inflorescences emerge from axillary buds located where leaf blades attach to the reproductive tiller.). Cupule counts were taken weekly from 17 July to 5 Sept. 2000 and 25 July to 28 Aug. 2001 on four randomly selected plants from the middle two rows of each plot to characterize shattering. Only cupules on inflorescences that had reached anthesis were counted, and data was collected on both terminal and lateral inflorescences.
Seed harvests were taken weekly for 6 wk from 27 July through 30 Aug. in 2001. Seed was harvested weekly from randomly selected plants located in the interior row of each plot not used for anthesis counts. Intact inflorescences were dried for 4 wk at room temperature in mesh harvest bags. After drying, inflorescences were broken at joints, and germination percentage and viability of the cupules were determined. Germination tests were conducted by placing the cupules for each treatment in 13 by 13 by 3.5 cm covered containers containing two layers of Anchor Steel blue seed germination paper (Anchor Paper Co., St. Paul, MN) moistened with distilled water. To accommodate all treatments in a limited amount of germination space, each germination box contained two treatments randomly placed in separate halves of the box. One half of the box, containing 25 randomly sampled cupules for a treatment, was considered as an experimental unit.
All germination tests were performed at 20/30°C alternating temperature (Ahring and Frank, 1968), with light (four 40-W cool-white fluorescent lights vertically oriented on both the left and right sides of the germinator) and 30°C for 8 h daily. Germination counts were made every 7 d for 28 d. Seeds were considered germinated if the coleoptile exceeded the seed in length and the seedling was normal according to the seedling evaluation criteria of the Association of Official Seed Analysts for comparable grasses (AOSA, 1992). Normal seedlings were removed as they were counted. Water was added to each germination box as needed to maintain optimum moisture levels. After 28 d of incubation, ungerminated caryopses were examined by tetrazolium tests and classified as dormant or dead.
Plant basal area was determined from crown circumference measurements made at the soil surface. All inflorescence and seed data measurements were converted to a per crown area basis to account for the wide range of plant size variability that accompanies a gamagrass stand and to allow comparisons to be made from a common unit size.
Statistical Design and Analysis
The experimental layout was a randomized complete block in a split-split-split plot treatment arrangement with four replications. Each cultivar represented a whole plot. The first split represented the N treatments. The second split represented the defoliation treatments, and the final split represented the dates on which counts were taken. Statistical analysis was performed using the Mixed Models procedure of the Statistical Analysis System (Littell et al., 1996) using the compound symmetry covariance structure. Main effects and the interaction of cultivar, N, defoliation, and date were tested for significance using analysis of variance. Mean comparisons were made by using an F-protected LSD (Steele and Torrie, 1980). The significance level for all comparisons was P 
0.05.
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Inflorescence Appearance
In both seasons inflorescence production was influenced by an interaction between cultivar and sampling date. Pete and Iuka eastern gamagrass had similar amounts of production during the first 2 wk of appearance of either the lateral or terminal inflorescences (Fig. 1). Thereafter, Pete produced more inflorescences than Iuka. For the 2000 season, peak anthesis for terminal inflorescences occurred between 15 and 25 June, while peak anthesis for lateral inflorescences occurred approximately 3 wk later, between 12 and 18 July. Flowering occurred later in 2001, with peak anthesis of terminal inflorescences occurring between 26 June and 3 July, and peak anthesis for lateral inflorescences near 12 July. Anthesis timing of the two cultivars was similar. Season total inflorescence production is presented in Table 1. Pete produced 53 and 36% more terminal inflorescences and 81 and 63% more lateral inflorescences than Iuka in 2000 and 2001, respectively.
|
|
There was a Defoliation x N x Sampling date interaction for lateral inflorescences in 2001. Each of the defoliation treatments was analyzed separately to determine its causes. The N x Date interaction was significant in the control (no defoliation) treatment (P = 0.01). This interaction appears to have been caused by the 224 kg ha-1 fertilization rate, affecting the peak lateral inflorescence time to be delayed by 2 wk as compared with the other N rates. The spring defoliation treatment had significant differences for N level (P = 0.03) and a N x Date interaction (P = 0.02). The mean number of lateral inflorescences for the season was 326, 390, 468, and 343 at 0, 56, 112, and 224 kg ha-1, respectively (LSD0.05 = 88). The N x Date interaction may have resulted from a drop in lateral inflorescence production after the peak period, followed by a second flush at the end of July in the 0, 56, and 112 kg ha-1 N, while inflorescence production remained steady for several weeks after the initial peak in lateral inflorescence appearance at the 224 kg ha-1 rate.
Seed Loss
Timing of eastern gamagrass seed harvest is quite difficult due to the indeterminate reproductive growth pattern of this species. Inflorescence and cupule appearance on the same plant occurs over a considerable time period, and seeds easily shatter from the inflorescence on maturity. In addition, there is considerable variability in maturity among plants within the synthetic varieties of eastern gamagrass. Over time, there was a net seed loss because shattering occurred faster than new cupules were formed.
Seed number on terminal and lateral inflorescences varied with sampling date in both seasons. Pete produced more terminal cupules than Iuka in 2001 and more lateral cupules in both years (Fig. 2). There was a significant interaction between cultivar and sampling date for terminal cupule amounts in 2001 and for lateral cupules in both seasons. This interaction occurred because seed shattered more rapidly from Pete than from Iuka. Trends across all treatments in 2000 showed that the greatest number of cupules on terminal inflorescences in both Pete and Iuka occurred between 17 and 26 July, while the greatest numbers of lateral cupules occurred between 26 July and 1 August (Fig. 2). Combined lateral and terminal cupule load was greatest for the period 16 July to 1 August, when Pete and Iuka contained >2800 and 1700 cupules m-2 basal area, respectively. For the 2001 season, the greatest number of terminal cupules was counted on 25 July, while the greatest amount of lateral cupules occurred between 25 July and 1 August. The period of greatest total cupule load was 25 July to 1 August with >6000 and 4000 combined terminal and lateral cupules per m2 basal area for Pete and Iuka, respectively. Pete produced 58 and 94% more total terminal and lateral cupules, respectively, than Iuka for the 2000 season (Table 1). In 2001, Pete produced 28 and 76% more total terminal and lateral cupules than Iuka, respectively. Since the number of cupules per inflorescence was similar for the two cultivars, differences in cupule numbers resulted from more inflorescences being produced in Pete than Iuka.
|
Nitrogen application affected cupule production in 2001 only. Cupule production on both terminal and lateral inflorescences responded to N application (P < 0.01). A greater number of cupules were produced when N was applied as compared with no added N (Table 2). Orthogonal contrasts between the combined N addition levels and no added N demonstrated that N increased both the number of terminal and lateral inflorescences (P = 0.05 and 0.03, respectively) and the number of cupules on these inflorescences (P = 0.01 and 0.02, respectively).
|
|
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Heat units, expressed as growing degree units (GDU) after anthesis, have been correlated to seed development and optimal harvest time for crested wheatgrass [Agropyron desertorum (Fisch. ex Link) Schult.], Russian wildrye [Psathyrostachys juncea (Fisch.) Nevski], intermediate wheatgrass [Thinopyrum intermedium (Host) Barkw. & D.R. Dewey subsp. intermedium], and western wheatgrass [Pascopyrum smithii (Rydb.) A. Löve] (Berdahl and Frank, 1998). Similar to eastern gamagrass, these four species suffer from excessive seed shattering, causing economic losses to seed producers. If seed harvests were taken at the optimum time suggested by our study (8 August), the harvest timing would have occurred at 550 GDU after peak terminal anthesis and 200 GDU after peak lateral anthesis in 2000, and 500 GDU after peak terminal anthesis and 325 GDU after peak lateral anthesis in 2001. It appears that using GDUs to determine harvest timing may not be the best method for eastern gamagrass because GDUs from anthesis to peak seed load varied between years and determining peak anthesis would be difficult for a seed producer.
A visual indicator may be the simplest way for a seed manager to decide seed harvest timing in eastern gamagrass. Optimal harvest time in this study was approximately 2 wk after terminal cupules began shattering. Seed harvested 1 wk before this period had 15 and 5 percentage points more immature seed on terminal and lateral inflorescences, respectively. Beyond this period, seed shattered at a rate of 50 and 32 cupules d-1 in 2000 for Pete and Iuka, respectively. In 2001, seed shattered at 178 and 109 cupules d-1 for the two cultivars. At these rates, approximately 13 and 20% of the seed was lost in the first week after peak cupule load in 2000 and 2001, respectively.
Cultivar had the greatest influence on seed yield of any treatment in this study. Pete produced greater total numbers of terminal and lateral inflorescences and cupules than did Iuka in both years. These differences between cultivars may be attributed to Iuka being selected for forage production. Iuka was developed from two accessions selected from material originally collected in Texas, Oklahoma, Kansas, and Arkansas, based on visual forage value (T.L. Springer, personal communication, 2002). Visual forage value took into consideration such things as crown diameter, leafiness, potential yield, and growth habit. In contrast to Iuka, Pete was developed as a composite of 70 accessions originating from native populations in Kansas and Oklahoma, presumably with little selection for forage quality (Fine et al., 1990). The techniques used to develop Iuka may have resulted in selecting toward vegetative tillers at the expense of reproductive tillers. A greater amount of vegetative tillers relative to reproductive tillers would contribute to greater leafiness and visual forage quality. It also would reduce the amount of seed produced by this cultivar.
Total seed production was much greater in the second year of this study than in the first, which could be attributed to the age of the stand (3 vs. 4 yr), the production environment, or a combination of the two. Terminal cupule production increased 166 and 133% and lateral cupules increased 98 and 80% for Iuka and Pete, respectively, from 2000 to 2001. In 2000, growing season rainfall amounts were nearly 20 cm less than in 2001, with the largest differences occurring in the spring months of April, May, and June.
Management of eastern gamagrass for both forage and seed would improve the diversity of options that livestock producers have for stands. However, early season defoliation has decreased seed yields in other warm-season grass, for example, switchgrass (Brejda et al., 1994). Our defoliation treatments were applied to simulate early season grazing (spring defoliation) and a combine harvest with removal of vegetative tissue late in the season (fall defoliation) as compared with plants that were not defoliated.
Reductions in seed production occurred with spring defoliation in the cultivar Iuka for the first year of the study and in Pete during both years, although reductions in both terminal and lateral cupules did not occur in either year. Lateral cupules were decreased by 20% during the peak seed load in 2000, but cupules shattered less rapidly from the defoliated plants than from the control plants. After 14 August, more cupules were retained on the defoliated plants. In 2001, cupule loads for spring-defoliated plants were not significantly different from that of nondefoliated plants. Combined, these responses suggest that spring defoliation to a height of 45 cm may have affected the seed available during the period of greatest seed yield, but seed yields in the spring-defoliated plants were not much different from the nondefoliated plants after mid-August. If seed and livestock producers with eastern gamagrass stands are willing to accept moderate seed yield reductions, they have the option of managing stands for both forage and seed production. Defoliated fields should be the last fields harvested for seed.
The spring defoliation treatment may have decreased lateral inflorescence production due to a decrease in carbohydrate reserves caused by the defoliation treatment. Dewald and Louthan (1979) found that tillers remain vegetative in the season in which they are initiated and become reproductive in later seasons, after achieving a minimum size for reproductive tiller formation. Jackson and Dewald (1994) discovered that carbohydrate reserve concentrations reach minimum levels when vegetative tillers "bolt" to create reproductive tillers, which may control the number of reproductive tillers formed in eastern gamagrass. The defoliation treatment in this study may have decreased carbohydrate reserves by removing vegetative tissue early in the season, which caused the decrease in lateral inflorescence production in the 2000 season.
Fall defoliation increased seed load above that of the spring defoliation treatment in Pete during the second year, suggesting that defoliation during seed harvest may be beneficial. This is similar to results of Vogel and Bjugstad (1968), who reported increased tillering and seed yields of big bluestem and indiangrass plants following a clipping at the seed ripened stage.
Studies with switchgrass and big bluestem (Masters et al., 1993; George and Reigh, 1987; Harlan and Kneebone, 1953) found that N application increased reproductive tiller and seed formation in warm-season grasses. Addition of 56 kg ha-1 N increased seed load in the second year of this study. However, seed yield did not increase at N rates above this level. In fact, an increase in N from 112 to 224 kg ha-1 decreased cupule numbers. It may be possible that increased vegetative growth from excessive N addition decreased light penetration into the crown area of the plants, which then decreased lateral inflorescence formation. Light penetration into the canopy has been shown to increase reproductive stem density in other warm-season grasses. Knapp (1984) found that removing plant debris substantially increased flowering stem density in big bluestem by improving the light environment of emerging shoots.
The 0 kg ha-1 N rate produced the greatest percentage of viable seeds. This may be explained by the limited production of cupules later in the season for the 0 kg ha-1 N rate. There were greater amounts of mature seed at the 0 kg ha-1 rate during the later harvests. The differences in cupule production response to N between years may be explained by rainfall patterns. Drought conditions of 2000, especially the below normal rainfall received in April, May, and June, may have hindered availability and uptake of N. Higher rainfall amounts over the same period in 2001, resulted in greater N uptake by plants. Climatic factors influence both the transformation of N fertilizer in the soil and the amount of N available to the plant (Craswell and Godwin, 1984). Under typical rainfall patterns 56 kg ha-1 was sufficient to produce the greatest amount of harvestable seed. Rates above this level may decrease seed load, and combine harvest would be slowed because N promotes greater amounts of vegetative tissue to be processed through the combine.
While application of 56 kg ha-1 N increased cupule load by approximately 50% above no N application in 1 yr of the study, it appears that other practices could be developed to further increase the seed yield of eastern gamagrass. It was visually observed during the course of this study that only about one-tenth of the tillers on a plant advance to reproductive growth, while the rest remain vegetative. Future research should focus on understanding the mechanism that regulates the conversion of vegetative tillers into reproductive tillers. This unknown regulator may be carbohydrate reserves, light interception, soil moisture, growth hormones, or a combination of these and other factors. A better understanding of this trigger could lead to greater improvement in eastern gamagrass seed yields.
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Herrera-C., W. R. Ocumpaugh, J. A. Ortega-S., J. Lloyd-Reilley, G. A. Rasmussen, and S. Maher Environmental Influences on Seed Quality of Windmillgrass Ecotypes in South Texas Agron. J., June 23, 2008; 100(4): 1205 - 1210. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Wang, C. Mu, Y. Hou, and X. Li Optimum Harvest Time of Vicia cracca in Relation to High Seed Quality during Pod Development Crop Sci., March 19, 2008; 48(2): 709 - 715. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R. Gibson, E. Z. Aberle, A. D. Knapp, K. J. Moore, and R. Hintz Release of Seed Dormancy in Field Plantings of Eastern Gamagrass Crop Sci., January 31, 2005; 45(2): 494 - 502. [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 | |||
