Heteroblastic seedling supplement of Setaria seed: Ⅲ. Seedling supplement behavior
- (0) Download
https://www.eduzhai.net International Journal of Plant Research 2012, 2(6): 165-180 DOI: 10.5923/j.plant.20120206.01 Setaria faberi Seed HeteroblastyBlueprints Seedling Recruitment: III. Seedling Recruitment Behavior Kari Jovaag1, Jack Dekker2,*, Brad Atchison2 1Weed Biology Laboratory, Department of Agronomy, Department of Statistics, Iowa State University, Ames, Iowa, 50011, USA 2Weed Biology Laboratory, Department of Agronomy, Iowa State University, Ames, Iowa, 50011, USA Abstract Seedling recruit ment of heterogeneous Setaria faberi seed entering the soil post-abscission is elucidated herein. This is the third in a series of three articles providing evidence that weedy Setaria seedling recruit ment behavior is predicated on dormancy state heterogeneity at abscission (seed heteroblasty), as modulated by environmental signals. Co mplex oscillating patterns of seedling emergence were observed during the first half of the growing season in all 503 soil burial cores of the 39 populations studied. These patterns were attributed to six distinct dormancy phenotype cohorts arising fro m inherent somatic poly morphis m in seed dormancy states (canalized phenotypes). Early season cohorts were formalized using a mixtu re model consisting of four normal distributions. Two, numerically low, late season cohorts were also observed. Variation in emergence patterns among Setaria populations revealed a fine scale adaptation to local conditions. Seedling recruit ment patterns were influenced by both parental-genotypic (time of embryogenesis) and environmental (year, co mmon nursery location, seed age in the soil) parameters. The influence of seed heteroblasty on recruitment behavior was apparent in that S. faberi populations with higher dormancy at the time of dispersal had lower emergence numbers the follo wing spring, and in many instances occurred later, co mpared to those less dormant. Heteroblasty was thus the first determinant of behavior, most apparent in recruit ment nu mber, less so in pattern. Environ ment modulated seedling numbers, but mo re strongly influenced pattern. The resulting pattern of emergence revealed the actual “hedge-bet” structure for Setaria seedling recruit ment investment, its realized niche, an adaptation to the predictable mortality risks caused by agricultural production and interactions with neighbors. These comp lex patterns in seedling recru it ment behavior support the conjecture that the inherent dormancy capacities of S. faberi seeds provides a germinability ‘memory’ of successful historical exp loitation of local opportunity, the inherent starting condition that interacts in both a determin istic and plastic manner with environ mental signals to define the consequential heterogeneous life history trajectories. Keywords Seed Heteroblasty, Seed Do rmancy, Setaria, Co mmunity Assembly, Seedling Emergence, Canalized Phenotypes 1. Introduction Setaria faberi is a very successful invasive agricu ltural weed because of its ability to form long-lived soil seed pools of heterogeneous seed (see first and second articles in this series) that cycle annually between dormancy and germination candidacy until environmental conditions allow germination and emergence to occur (Figure 5, second article in this series). This paper examines seedling emergence fro m the soil of those same populations and is the third in a series on the seed-seedling life h istory of this weedy species. Seedling recru it ment is an irreversib le, co mmitted step in the life history of a plant. It is the time when the annual S. faberi plant resumes autotrophic growth, assembles in local agro-commun ities, and begins interactions with neighbours. * Corresponding author: jdekker@iast ate.edu (Jack Dekker) Published online at https://www.eduzhai.net Copyright © 2012 Scientific & Academic Publishing. All Rights Reserved As such, its timing is crucial to subsequent behaviours, including survival and reproductive success. Seedling emergence timing is a trade-off between the risks of mortality fro m disturbance and competition with neighbours and explo itation of opportunity space-time to accu mulate biomass and produce seed. S. faberi has evolved in agro-ecosystems characterized by annual d isturbance and selection by predictable farming practices (e.g., p lanting, tillage, herbicides, harvesting). The variable risks in these agricultural habitats have not led to a single, best time to germinate. Is there a pattern to the period icity of S. faberi seedling recruit ment? Past research only reveals a qualitative pattern: an early flush of seedling emergence in the spring (April-June) wh ich decreases as temperatures warm, followed by low numbers during the late season (e.g., Forcella, et al., 1992, 1997). Is it possible to discover a mo re precise, quantitative pattern to seedling emergence? If so, does it bear a relationship to seed heteroblastic patterns among the same seeds at abscission? A successful survival strategy for a p rolific seed 166 Kari Jovaag et al.: Setaria faberi Seed HeteroblastyBlueprints Seedling Recruitment: III. Seedling Recruitment Behavior producing species like Setaria may be to disperse heteroblastic seeds able to take advantage of as many recruit ment opportunities as are available, in both seasonal time and field space. The resulting pattern of seedling emergence timing may reveal the ‘hedge-bet’ for individual Setaria fitness. We hypothesize that there exists a pattern to Setaria seedling recruit ment timing which is predicated in the first instance on the dormancy state heterogeneity (seed heteroblasty) at abscission. This pattern in the second instance is modulated by the environ ment (year, co mmon nursery location, seed age in the soil). Thus, patterns in Setaria emergence will reflect the h istorically advantageous recruit ment times (opportunity) enabling survival and reproductive success. Table 1. Location (name, latitude, longitude) and nomenclature (lot no., name) of S. faberi populations grouped by year of abscission and collection. 1Population names consist of 3 parts: a prefix letter designating the population location, two digits designating the collection year, andtwo digits designating the Julian week of abscission and collection. All locations were in Ames, IA, except S.E. Research Station, near Crawfordsville, IA; R was 100m west of Hinds (H) Seed Collection Location Name Lat it ude 1997 Curtiss Farm 42o00'23"N Lon git ude 93o38'90"W Oakwood Drive 42°01'83"N 93°36'54"W 1998 Curtiss Farm 42o00'23"N 93o38'90"W Hinds Farm 42°03'65"N 93°36' 96"W Johnson Farm 41°58'88"N 93°38'49"W Mort en sen Road Wessex Road 42°00'57"N 41°59'67"N 93°40' 59"W 93°38'30"W Whiteoak Drive 41°59'75"N 93°38'90"W 1999 Curtiss Farm 42o00'23"N 93o38'90"W Hinds Farm 42°03'65"N 93°36' 96"W Johnson Farm 41°58'88"N 93°38'49"W S.E. Research St at ion 41°12'36"N S.E. Research St at ion 41°12'26"N Whiteoak Drive 41°59'75"N 91°29' 73"W 91°29'68"W 93°38'90"W Nomen clat ure Lot # Name1 3728 C97-36 3732 O97-34 3733 O97-36 3734 O97-37 3740 C98-32 3746 C98-36 3754 C98-40 3739 H98-32 3744 R98-34 3747 H98-36 3752 H98-40 3738 J98-32 3749 J98-36 3751 J98-40 3745 M98-34 3750 M98-40 3743 X98-34 3758 X98-39 3741 W98-32 3748 W98-36 3753 W98-40 3769 C99-35 3770 C99-38 3771 C99-42 3773 H99-32 3774 H99-36 3775 H99-41 3776 J99-35 3777 J99-38 3778 J99-42 3766 B99-32 3767 B99-36 3768 B99-40 3779 K99-32 3780 K99-36 3781 K99-40 3782 W99-32 3783 W99-36 3784 W99-42 Burial Date Johnson Crawfordsville 09.29.97 09.11.98 09.11.98 09.11.98 09.29.97 09.11.98 10.04.97 09.10.98 09.10.98 09.10.98 10.04.97 09.10.98 08.22.98 09.11.98 10.14.98 08.22.98 08.31.98 09.19.98 10.14.98 08.22.98 09.19.98 10.14.98 08.31.98 10.14.98 08.31.98 10.01.98 08.22.98 09.19.98 10.14.98 08.21.98 09.10.98 10.13.98 08.21.98 09.01.98 09.22.98 10.13.98 08.21.98 09.22.98 10.13.98 09.01.98 10.13.98 09.01.98 09.29.98 08.21.98 09.22.98 10.13.98 09.01.99 09.30.99 10.27.99 08.09.99 09.14.99 10.20.99 08.22.98 09.19.98 10.14.98 08.24.99 09.14.99 10.13.99 08.24.99 09.14.99 10.13.99 08.09.99 09.14.99 10.20.99 09.01.99 09.29.99 10.26.99 08.10.99 09.14.99 10.19.99 08.21.98 09.22.98 10.13.98 08.24.99 09.14.99 10.12.99 08.24.99 09.14.99 10.12.99 08.10.99 09.14.99 10.19.99 International Journal of Plant Research 2012, 2(6): 165-180 167 2. Materials and Methods 2.1. Experi mental 2.1.1. Setaria Populations For this study, a Setaria population was defined as a specific co mb ination of Setaria location and seasonal time of abscission (table 1). Seeds collected fro m the sameccc location but at different seasonal times were treated as samples fro m d ifferent populations rather than repeated measures of the same population. The reason for this was that seed dormancy phenotypes developing early in the season differ fro m those developing later in the season due to the strong influence of photoperiod on dormancy induction (Dekker, 2004; Dekker et al, 1996). Locations. S. faberi seeds were collected fro m a local area around Ames, IA (table 1). Most were less than 3km fro m each other. The geographic range was restricted to allow characterizat ion of localized population phenotypic (heteroblastic) dormancy structure (see second article in this series). In 1999, seeds were also collected fro m two distant locations in Iowa (near Crawfordsville, about 194km fro m Ames) to gain some initial perspective on regional heteroblastic variation. All seeds were gathered from Setaria monocultures, ensuring no contamination fro m non-target Setaria species occurred. When seed was collected fro m the same site in different years the site was protected fro m human interference. Seasonal time of abscission. Setaria seed development occurs continuously fro m July through November (Julian week (JW) 26-48) depending on environmental conditions. To evaluate the effects of time of abscission during this seed rain period, seed was collected fro m most of the locations at discrete intervals roughly corresponding to August (early; JW 31-35), September (middle; JW 36-39) and October (late; JW 40-42) (table 1). Our Julian week calendar assumes 29 days for every February. 2.1.2. Seed Collection and Preparation Ripe, recently abscised (e.g., 0-5 days; Dekker et al., 1996) seed were collected fro m each population and dried at room temperature (20℃) and ambient hu midity. Twelve hundred hard, dark, mature seeds fro m each lot were counted manually to provide six rep lications of 200 seeds per population (3 rep licat ions at each of the two nurseries). Further procedural details are given in Jovaag, Dekker, and Atchison 2009a. 2.1.3. Seed Burial Seeds were buried in t wo co mmon location, agricu ltural nurseries. Nurseries were on the Iowa State University Agricultural Experiment Research Station, at Johnson Farm, Ames, and the Southeast Research Station near Crawfordsville, Iowa (table 1). Field soil seedbank site preparation. The areas selected for this study had been previously maintained as turf grass (primarily bro megrass, Bromus sp.) for more than 12 years. Well in advance of burial, these areas were treated with a 3% solution of glyphosate (isopropylamine salt of N-(phosphomethyl) glycine; Roundup) to kill the grass sod. After allowing time for the herbicide to work, the dead sod was tilled to a depth of about 20cm using a rototiller. Time was allo wed for continued decay of the sod and the area was tilled again. Finally, the soil was raked s mooth and even for seedbed preparation. Soil cores were removed periodically in the experimental areas to confirm that no previous, contaminating S. faberi seeds were present. Crop (soybean) rows were planted 96cm apart. Individual S. faberi seed burial sites (soil cores) were located in the inter-ro w area. Burial sites were arranged in a co mp letely randomized design using three replications. Burial procedure. At each burial site, three equidistant metal-staffed marker flags were used to define an 11cm diameter round “core” area. A 7cm diameter, 10cm deep (volu me: 390cm³), soil core was extracted fro m the centre of the three flags. The soil was placed in a dishpan and thoroughly mixed with 200 seeds from a population (seed density: 0.5seeds/cm³). The soil was then carefully returned to the hole fro m wh ich it was extracted. Seed burial occurred one to three weeks after the seed was harvested (table 1). Field management. No-till Roundup Ready soybeans were p lanted each spring between the soil core rows, simu lating normal stand density and row spacing and ensuring the seed-soil cores were undisturbed. Both chemical and manual weed control methods were used to keep the plots weed free. Soybeans were harvested by hand and threshed, then the residue was redistributed over the entire plot area. 2.1.4. Seedling Emergence Data Collection Each week in 1997-2000, fro m spring through autumn when the soil was not frozen, e merged seedlings in each core were counted, then killed by severing them at the soil surface to prevent them fro m influencing subsequent germination and emergence. 2.2. Anal ysis 2.2.1. Statistical Model Seedling emergence in the first half of the season (Julian weeks 14-31) showed a co mplex oscillat ing pattern (figure 1). Mixture models, that is, models wh ich are a co mbination of two or more simp le distributions, are often flexib le enough to fit such complex patterns, yet still provide parameters for interpretation and comparison. Determination of the number of distributions in the mixture is a difficu lt problem which has not been completely resolved (McLach lan & Peel 2001). The goal is to have the smallest number o f co mponents which are different, have nonzero mixing proportions, and are compatible with the data. For our data, a min imu m of four distributions were generally necessary for the mixture model to be co mpatible with the 168 Kari Jovaag et al.: Setaria faberi Seed HeteroblastyBlueprints Seedling Recruitment: III. Seedling Recruitment Behavior data. Mixtures of five distributions were unnecessary since they generally contained one distribution with a mixing proportion close or equal to zero. Letting g(x) represent the expected proportion of seeds emerg ing at t ime x ( x ∈ [14,31]) the four d istribution mixture model is: 4 g(x) = ∑ πi fi (x) i=1 where: 0 ≤ π i ≤ 1, π1+…+π4=1 and: fi (x) ~ N (µi ,σi ) The πi are the mixing proportions, and the fi ( x) are the component distributions. Further, the standard deviations of the norma l distributions were constrained to be equal. That is, fi ( x) ~ N (µi ,σ ) . The constraint disallowed models containing two distributions with nearly identical means and very different standard deviations. Because of the critical importance of the timing of emergence, it seems reasonable that if there are distinct emergence phenotypes, that they would have distinct mean times, rather than identical mean times and varying standard deviations. This constraint also resulted in models with eight parameters (4 means, 1 standard deviation, 3 independent proportions) rather than eleven. Distribution Names. Herein, the four mixture model distributions are called early spring, mid-spring, late spring and early su mmer. Two addit ional late season emergence cohorts are called su mmer and autumn. Co mparison of emergence and agricultural d isturbances. Table 2 shows historical times (ca. 1980-1996) of agricultural disturbances typical of management practices for central and southeast Iowa corn-soybean rotation fields fro m wh ich S. faberi populations were taken (personal communicat ion, R. Hart zler, M.D.K. Owen, Iowa State Univ.; Farnham, 2001; Whigham, et al., 2000). To allow comparisons between disturbance timing and seedling recruit ment observations, emergence times for the six cohorts reported herein are also provided. 2.2.2. Co mparison of Emergence and Heterogeneity at A bs cis s io n To test the hypothesis that Setaria seedling recru it ment timing is predicated on the dormancy state heterogeneity at abscission, seed germinability of S. faberi populations at abscission was compared to their cumu lative seedling emergence the following spring. Direct co mparison of seed germinability nu mbers and emergence numbers was not possible because of the differing experimental methods needed in each case. Thus, the comparison was made on the relative rankings of the populations. Setaria populations were first ranked fro m least to most germinable at the time of abscission (rankings given in Jovaag, 2006). The populations were then ranked fro m least to most cumulative emergence at five different times during the first year after burial (Julian weeks 18, 20, 24, 28, and 48). The first three t ime points used (JW18, 20, 24) were intermediate to the four distribution mean times used to model emergence (table 2). Julian week 28 was shortly after the fourth mean. The last date was chosen to contain all first year emergence. The germinability ranking at abscission was then compared to the emergence ranking at each chosen time point using Spearman’s rank correlation, a nonparametric test of association that does not require specific distributional assumptions. To further co mpare emergence and heterogeneity at abscission, the emergence patterns of the seed dormancy (after-ripening, A R) groups developed in the first article of this series were also analyzed using four distribution mixtu re mo d els . International Journal of Plant Research 2012, 2(6): 165-180 1 International Journal of Plant Research 2012, 2(6): 165-180 Table 2. Calendar of hist orical, seasonal t imes (Julian week, month) of agricult ural field dist urbances (seedbed preparat ion; plant ing; weed cont rol, including t illage and herbicides; t ime aft er which all cropping operat ions cease, layby; harvest and autumn tillage), and seedling emergence timing for central and southeastern Iowa, US, Setaria faberi population cohorts (all S. fabericombined: time, +/- S.E.; mean proportion; see table 3) Maize Month Julian Week Seedbed Prep AP RIL | MAY | JUNE | JULY | AUGUST | SEPTEMBER | OCT OBER | NOVEMBER | DEC 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 P lant in g Weed Control Layby Harvest Autumn Tillage S. faberi Early Spring 12 Recruit men t Mid-Spring 38 Cohorts Late Spring 30 Early Summer 20 Soybeans Summer Aut umn Seedbed Prep 0.2 0.1 P lant in g Weed Control Layby Harvest Autumn Tillage 169 170 Kari Jovaag et al.: Setaria faberi Seed HeteroblastyBlueprints Seedling Recruitment: III. Seedling Recruitment Behavior 3. Results 3.1. Spring Emergence A consistent pattern was observed in the number of Setaria seedlings emerging during the spring (JW 14-31): periods of high emergence (local maxima) followed by intervening min ima (figure 1): Spring and Earl y Summer Emergence equal standard deviations (see model discussion in materials and methods: analysis). The model had four emergence periods (local maxima) separated by intervening minima. However, when mean times were close, or the proportion in one of the distributions was very small, two or more o f the distributions combined to form one peak, resulting in a model with two or three peak emergence periods. This resulted in the close coincidence of the models (solid curved lines in figures 2-3) with the frequency histograms of the actual data (bars in figures 2-3). In the following sections “peak” is used when referring to a local maxima in the model. The “distribution”, “mean t ime” or “proportion” refers to one of the four underlying normal distributions (dotted lines in figures 2-3). Separate models were obtained for each year and common nursery. Additionally, separate models were obtained for seeds of different ages in the soil, for seed collected from co mmon locations during the early, middle and late periods of the seed rain, and for the AR groups developed in the first article in this series. Mi d Summer through Autumn Emergence Figure 1. S. faberi seedling emergence (no.) with time (Julian week (JW)) during the first year (spring and early summer: JW14-31, left; mid summer and autumn: JW32-50, right) after seed burial. Each line represents seedling emergence from one of the 503 cores used in the st udy This oscillating pattern occurred in all 503 cores of the 39 populations studied over three years in two common burial n u rseries , alt hou gh t he mag n it ude an d t imin g o f t he os cillat ions varied amon g th e co res . To mo del th ese oscillat ions, spring emergence data was analy zed using a mixtu re model consisting of four normal d istributions with 3.1.1. General Emergence Pattern Overall, S. faberi seedling emergence during the first half of the first year after burial (Julian weeks 14-31) had an oscillating pattern with three sequentially decreasing peaks (figure 2, top left). Fifty percent of the emergence during this period occurred within the first peak, wh ich consisted of the small early spring distribution and the large mid spring distribution (table 3). The late spring distribution formed the second, somewhat smaller (30%) peak. The early summer distribution formed the third and smallest peak. Th is overall response of S. faberi can be deconstructed to reveal the separate influences of years (1997-1999) and co mmon burial nurseries (Crawfordsville, Johnson) on the spring emergence pattern. Years. Seedling emergence patterns differed between years primarily in the proportions of emergence occurring within each distribution (table 3, figure 2). The proportion of emergence during mid spring decreased fro m the 1998 season to the 1999 season, and again from the 1999 seas on to the 2000 season. The proportion of emergence during early and late spring increased from the 1998 season to the 1999 season, and again fro m the 1999 season to the 2000 season. The proportion of emergence during early su mmer was greater during the 2000 season than during the 1998 or 1999 seasons. Further, in the 1998 season (1997 populations), the mid spring distribution had the largest proportion. In the 1999 season (1998 populations), the mid and late spring proportions were about equal, and larger than the other two. In the 2000 season (1999 populations), the late spring distribution had the largest proportion. International Journal of Plant Research 2012, 2(6): 165-180 171 Table 3. Seedling emergence (mean times (Julian week, JW), std. deviations (JW), top portion of table; proportions, cumulative no. (per core of 200, JW14-31), bottom portion of table) of S. faberi populations within each of the four normal mixture model components (early, mid and late spring and early summer) during the first year after burial, grouped by burial year, nursery, and burial year at each nursery (Crawfordsville (Craw); Johnson (John)). In all cases, numbers in the same column, portion of the table and group (year, nursery, year by individual nursery) with the same uppercase letter are not statistically different (comparison of 95% confidence intervals, probability (p)>.05). Numbers in the same row with the same lowercase letter are not st at ist ically different (hypot hesis t est , p>.05) Table 3. Top Portion: Populations Distribution mean Time (JW) Early Spring Mid Spring Late Sp ring Early Su mmer Co mmon Standard Deviation (JW) All S. faberi 1997 1998 1999 17.0 17.2 A 16.9 A 16.9 A 19.1 18.8 A 19.5 B 20.1 C 22.8 21.7 A 22.9 B 23.0 B 26.5 26.6 A 26.6 A 26.6 A 1.05 ____ 0.90 B 0.78 A 1.18 C Cra wfords ville Jo h ns o n 16.9 A 16.9 A 20.0 B 18.9 A 23.7 B 22.5 A 27.2 B 26.4 A 0.96 B 0.88 A 1997 Craw 1998 Craw 1999 Craw 15.6 A 16.7 B 17.0 C 19.4 A 19.8 B 20.2 C 21.2 A 24.0 C 23.3 B 26.8 A 27.1 B 27.3 B 0.80 A 0.72 A 1.16 B 1997 John 1998 John 1999 John 18.5 C 17.0 B 16.5 A 21.0 B 19.4 A ---- 23.5 C 22.4 A 22.6 B 26.5 B 26.6 B 26.2 A 0.92 B 0.54 A 1.26 C Table 3. Bottom Portion: Proportion of seeds within each distribution Early Spring Mid Spring Late Sp ring Early Su mmer Cu mulat ive Nu mber Emerged (JW14-31) All S. faberi 1997 1998 1999 0.12 0.06 A a 0.17 B b 0.20 C b 0.38 0.71 C c 0.36 B c 0.14 A a 0.3 0.08 A a 0.34 B c 0.43 C d 0.19 0.14 A b 0.13 A a 0.23 B c 37 44 B 35 A 34 A Cra wfords ville Jo h ns o n 0.20 B b 0.37 A d 0.32 B c 0.09 A a 0.41 B d 0.27 A c 0.11 A a 0.22 B b 33 A 40 B 1997 Craw 1998 Craw 1999 Craw 0.01 A a 0.19 B b 0.26 C b 0.54 C d 0.39 B d 0.21 A b 0.18 A c 0.35 B c 0.42 C c 0.03 A b 0.07 A a 0.11 B a 28 A 26 A 42 B 1997 John 1998 John 1999 John 0.79 C d 0.16 B b 0.08 A a 0.08 A b 0.33 B c ----- 0.02 A a 0.37 B d 0.37 B b 0.12 A c 0.14 A a 0.55 B c 48 B 44 B 25 A 172 Kari Jovaag et al.: Setaria faberi Seed HeteroblastyBlueprints Seedling Recruitment: III. Seedling Recruitment Behavior aft er burial for populat ions buried in 1997 (t ier 4), 1998 (t iers 5; including second year after burial), and 1999 populations buried in 2000 (bottom tier). Bars provide a relative frequency histogram of seedling emergence. The solid line is the seedling emergence est imate (proport ion) from the mixt ure model (4 normal components with equal variance, see materials and methods). Dashed lines show the model’s four normal component s (weighted by the mixing proportions) Mid and late spring t iming means also differed between years, though timing had a lesser effect than proportion on the emergence pattern. The mid spring mean t ime for the 1998 season occurred earlier than for the 1999 season which in turn occurred earlier than the mean time fo r the 2000 season. The late spring mean time for the 1998 season occurred earlier than for either the 1999 or 2000 season. Mean times for early spring and early summer were similar between years. The variability of the distributions also differed between years. The 1999 season was less variable than the 1998 season which was less variable than the 2000 season. On average, the total number of seeds emerging during the spring (JW 14-31) was greater during the 1998 season than during the 1999 or 2000 seasons. Co mmon nurseries. The mid spring through early summer mean t imes were about a week earlier at Johnson than at Crawfordsville, the mean time of the early spring distribution was similar at both nurseries (table 3, figure 2). The proportions of emergence during mid spring and early summer we re 7-8% greater at Johnson than at Crawfordsville, while the proportions during early and late spring were 7-9% lower. Variability was greater at Crawfordsville than at Johnson. On average, the total number of seeds emerging during the spring (JW 14-31) was greater at Johnson than at Cra wfords ville. Year by co mmon nursery interaction. Overall, the influence of year was greater than that of co mmon nursery on the spring emergence patterns during the 1998 and 1999 seasons. However, the mean time for each distribution was earlier during the 1998 season than during the 1999 season at Crawfordsville, but later or similar at Johnson (table 3). A lso, the greatest proportion of emergence in 1998 occurred during early spring at Johnson, but during mid spring at Crawfordsville. Differences between the common nurseries were mo re apparent during the 2000 season. At Crawfordsville in 2000, emergence occurred primarily fro m early spring through late spring, with the greatest proportion (42%) during late spring (table 3). At Johnson in 2000, emergence primarily occurred during late spring and early summer, with la rgest proportion (55%) during early summer. Figure 2. S. faberi seedling emergence (proportion relative to total spring no.) with time (Julian week, JW) during the spring and early summer of the first year after burial for all S. faberi populations (top tier), each nursery (second tier), 1997 populations (tiers 3 and 4), 1997 second and third years 3.1.2. Seed Age in the So il Emergence patterns among seeds of different ages in the soil were co mpared within a season (1999 or 2000; figure 3). Both years, the total nu mber of seeds emerg ing during the spring (JW 14-31) declined rapidly as seed age in the soil increased (table 4). Patterns in emergence timing and proportion between the first and second year after burial differed depending on the season. Data for the third year after burial was only availab le for the 2000 season. International Journal of Plant Research 2012, 2(6): 165-180 173 Table 4. Seedling emergence (mean times (Julian week, JW), std. deviations (JW), top portion of table; proportions, cumulative no. (per core of 200, JW14-31), bottom portion of table) of S. faberi populations within each of the four normal mixture model components (early, mid and late spring and early summer) during the first-third years after burial, grouped by season (1999, 2000). Data for the third year after burial was available only for the 2000 season. In all cases, numbers in the same column, portion of the table and group (season) with the same uppercase letter are not statistically different (comparison of 95% confidence intervals, probability (p)>.05). Numbers in the same row with the same lowercase letter are not statistically different (hypothesis test, p>.05) Table 4. Top Portion: Distribution mean Time (JW) Early Spring Mid Spring Late Sp ring Early Su mmer Co mmon Standard Deviation (JW) 1999 Season 1st Year 2nd Year 16.9 A 17.1 B 19.5 A 19.5 A 22.9 B 22.6 A 26.6 A 26.7 A 0.78 B 0.39 A 2000 Season 1st Year 2nd Year 3rd Year 16.9 A 17.1 A 17.3 A 20.1 A 20.8 B 20.6 AB 23.0 A 23.7 B 23.8 B 26.6 A 27.5 B 27.3 B 1.18 C 1.00 B 0.77 A Table 4. Bottom Portion: 1999 Season 1st Year 2nd Year Proportion of seeds within each distribution Early Spring Mid Spring Late Sp ring Early Su mmer 0.17 A b 0.18 A b 0.36 A c 0.54 B d 0.34 B c 0.24 A c 0.13 B a 0.04 A a 2000 Season 1st Year 2nd Year 3rd Year 0.20 B b 0.06 A a 0.10 A a 0.14 A a 0.20 AB b 0.26 B b 0.43 A d 0.48 A d 0.57 B c 0.23 B c 0.25 B c 0.08 A a Cu mulat ive Nu mber Emerged (JW14-31) 35 B 29 A 34 C 15 B 4 A 174 Kari Jovaag et al.: Setaria faberi Seed HeteroblastyBlueprints Seedling Recruitment: III. Seedling Recruitment Behavior Figure 3. S. faberi seedling emergence (proportion relative to total spring no.) from populations collected during the early (Julian week (JW) 32, top tier, left), middle (JW36, top tier, right) and late (JW40, tier 2) periods of the seed rain (1999 and 2000 seasons combined), and the 1998 (tiers 3 and 4) and 1999 (tiers 5 and 6) inter-AR groups (Group 1, most germinable; group 2 middle germination; group 3 (in 1998) or group 4 (in 1999) least germinable; developed in first article in this series) with time (Julian week, JW) during the spring and early summer of the first year after burial. Bars provide a relative frequency histogram of seedling emergence. The solid line is the seedling emergence estimate (proportion) from the mixture model (4 normal components with equal variance, see materials and methods). Dashed lines show the model’s four normal components (weighted by the mixing proportions). Emergence timing for the third year after burial was similar to that for the second year after burial. Emergence proportion for the third year after burial was greater during late spring and less during early su mmer than for the first or second years. Emergence proportion for the third year after burial was similar to that of the second year during early and mid spring. 3.1.3. Seasonal Time of Abscission Populations of seed that matured during JW32, JW 36 and JW40 were collected fro m eight locations; four in each of 1998 and 1999. These twenty-four populations provided a basis for the co mparison of emergence patterns among seed with varying seasonal times of abscission. The 1998 populations (during the 1999 season) and 1999 populations (during the 2000 season) were modelled separately because of the strong influence of annual climat ic conditions. Models with the years combined were also obtained to exa mine general patterns. Co mparison of early-middle-late. During the 1999 season, the total number of seeds emerg ing during the spring (JW14-31) was lower for the early (JW32, August 1998) maturing seed than for seed maturing during the middle (JW36, September 1998) seasonal period, which in turn was lower than the total emergence fro m late (JW40, October 1998) maturing seed (table 5). During the 2000 season, there was no statistical diffe rence in the number of seeds emerging fro m the early, middle and late maturing 1999 populations. Emergence timing was similar among early, middle and late maturing populations, with a few exceptions. The mean time of the early spring 1999 distribution was earlier fo r the middle 1998 populations than for the 1998 early or late populations. The 2000 early spring distribution was also earlier fo r the 1999 early and middle populations than for the 1999 late populations. The mean time of the 2000 early summer distribution was earlier for the 1999 midd le populations than for the 1999 early and late populations. Patterns in the emergence proportions among early, midd le and late maturing populations were equivocal, and varied with year.
... pages left unread,continue reading
Free reading is over, click to pay to read the rest ... pages
0 dollars，0 people have bought.
Reading is over. You can download the document and read it offline
0people have downloaded it