SciELO - Scientific Electronic Library Online

vol.24 issue1Physiological quality of bean seeds produced in compacted soilVigor and field emergence in "high oleic" sunflower seeds, in southeast of Buenos Aires province author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista Brasileira de Sementes

Print version ISSN 0101-3122

Rev. bras. sementes vol.24 no.1 Londrina  2002 

Influência de sistemas de rotação de sementes de gramíneas forrageiras temperadas na composição do banco de sementes invasoras no solo1


Influence of temperate grass seed rotation systems on weed seed soil bank composition



Renato Borges MedeirosI; Jeffrey J. SteinerII

IProf. Adjunto, bolsista da CAPES, Depto. de Plantas Forrageiras e Agrometeorologia, Faculdade de Agronomia, UFRGS, Cx. Postal 776, 91501-970, Porto Alegre-RS; e-mail:
IIRes. Agronomist, PhD, National Forage Seed Production Research Center, USDA-ARS, 3450 SW Campus Way, Corvallis, OR, 97331; e-mail:




Em consequência da mudança nas práticas agronômicas nos cultivos de gramíneas perenes para sementes no oeste de Oregon, EUA, sistemas alternativos de rotação estão sendo considerados com a finalidade de reduzir a infestação de plantas invasoras. De um modo geral, inexistem informações dos efeitos de práticas agronômicas alternativas e aplicações de herbicidas na composição do banco de sementes no solo durante esta transição. Seis sistemas alternativos de rotação foram aplicados a partir de 1992 em uma área historicamente cultivada com azevém-perene (Lolium perenne L.), para produção de sementes em sistema de monocultura. Parcelas medindo 20x30m, para cada sistema, foram arranjadas em blocos completos, casualizados com quatro repetições. Vinte a trinta amostras de solo foram retiradas de cada parcela em junho de 1997. A composição das espécies de invasoras das amostras de solo foram determinadas em sucessivos ensaios de germinação conduzidos em casa de vegetação. Além da densidade de sementes, índices de heterogeneidade para equitabilidade, riqueza e diversidade de espécies foram determinados. As espécies mais abundantes foram Juncus buffonius L. e Poa anua L. Mudanças na composição do banco de sementes foram decorrentes dos diferentes herbicidas utilizados nas culturas componentes da rotação. Trigo (Triticum aestivum L.) e aveia (Avena sativa L.) plantados em cultivo mínimo, na primavera, reduziram a riqueza e a densidade de espécies em comparação com os outros sistemas de rotação de cultura, mas não alterou os ídices de equitabilidade e heterogeneidade de espécies. Quando "meadowfoam" (Limnanthes alba Hartweg ex Benth.) sucedeu trigo na rotação, a riqueza de espécies não foi afetada, mas equitabilidade e a diversidade diminuíram em comparação com outros sistemas de rotação. "Meadowfoam" sucedendo trevo-branco (Trifolium repens L.), o método de estabelecimento (preparo mínimo e convencional do solo) não afetou a densidade, equitabilidade e riqueza de espécies.

Termos para indexação: azevém-perene, espécies invasoras, práticas culturais, produção de sementes, sementes enterradas.


Due to changing cropping practices in perennial grass seed crops in western Oregon, USA, alternative rotation systems are being considered to reduce weed infestations. Information is generally lacking regarding the effects of alternative agronomic operations and herbicide inputs on soil weed seed bank composition during this transition. Six crop rotation systems were imposed in 1992 on a field that had historically produced monoculture perennial ryegrass (Lolium perenne L.) seeds. Each system plot was 20 x 30 m, arranged in a randomized complete block design, replicated four times. Twenty to thirty soil cores were sampled in June 1997 from each plot. The weed species composition of the cores was determined by successive greenhouse grow-out assays. In addition to seed density, heterogeneity indices for species evenness, richness, and diversity were determined. The most abundant species were Juncus bufonius L. and Poa annua L. Changes in seed bank composition were due to the different herbicides used for the rotation crop components. Compared to the other rotation systems, no-tillage, spring-planted wheat (Triticum aestivum L.) and oat (Avena sativa L.) reduced overall weed seed density and richness, but did not affect weed species evenness or diversity. When meadowfoam (Limnanthes alba Hartweg ex Benth.) succeeded wheat in rotation, weed species richness was unaffected, but evenness and diversity were reduced, compared to the other rotation systems. For meadowfoam in sequence after white clover (Trifolium repens L.), crop establishment method (no-tillage and conventional tillage) had no effect on weed seed species density, evenness, or diversity.

Index terms: buried seeds, cropping practices, perennial ryegrass, seed production, weed species.




The size and composition of the weed seed bank as well as the above-ground flora reflect the impact of past and present weed, crop, and soil management practices (Roberts, 1981; Cardina et al., 1991). In general, large seed banks are associated with arable sites (Fenner, 1995), and its composition is richer than the composition of plants that cover it (Symonides, 1986). The amount of species diversity within seed banks increases by disturbance amount from plowing to disking to no-tillage practices (Feldman et al., 1997).

It is normally assumed that seed banks have clustered (patchy) spatial distributions (Dessaint et al., 1991; Forcella et al., 1992; Boccanelli & Lewis, 1994). Methods including indices of variance-to-mean ratio, Lloyd's mean crowding and Lloyd's patchiness (Dessaint et al., 1991), and Morisita's index (Boccanelli & Lewis, 1994) are used to determine weed aggregation patterns.

The viable seed fraction in the seed bank is the main source of weed recruitment and infestation in crop fields (Cavers & Benoit, 1989). Reducing the size of weed seed banks has been a long-term goal of weed management strategies, especially for continuously cropped fields (Schweizer & Zimdahl, 1984). Weed seed bank changes occur temporally and by management practices including crop rotation sequences and tillage practices (Chauvel et al., 1989). Tillage practices influence the vertical distribution and density of weed seeds (Buhler, 1995). Limiting the amount and depth of tillage leaves seeds close to the soil surface where most seeds have the ability to germinate and become established (Buhler & Mester, 1991; Yenish et al., 1992; Buhler, 1995).

Crop rotation can disrupt the continuous dominance of specific weeds in a field, decrease the build-up of weed populations, and prevent major shifts in weed species composition (Ball, 1992). However, the lack of rotation crop complexity in a corn and soybean rotation was shown do not influence weed seed bank composition (Mulugeta & Stoltenber, 1997).

Herbicides to control weeds in different rotation systems are a key component of conservation tillage management (Buhler, 1995). However, single herbicide chemistries rarely control all weed species when applied at selective rates, and repeated crop/herbicide combinations may rapidly shift the composition of weed seed banks (Roberts, 1981). Selective foliar-applied post-emergent herbicides are preferable to soil-applied herbicides for weed management in conservation tillage systems (Lewis, 1985).

Weed seed contaminants are a major problem for maintaining grass seed end-product quality (Schweitzer, 1994). If product purity standards are not met, entire seed lots can be rejected or require recleaning (Churchill et al., 1995). Weed control practices in turf grass and forage seed systems have historically relied on herbicide control methods, but fewer chemicals are available because of withdrawal of registration labels (Mueller-Warrant, 1994). Detection of herbicides in ground-water (Halberg, 1989) and the appearance of herbicide resistant weeds in many rotation systems (Maxwell et al., 1990) necessitate approaches to optimize efficacy when using herbicides (Mulugeta & Stoltenberg, 1997). However, the reduced herbicides inputs and greater reliance on mechanical methods may result in increased weed management problems over time (Mulugeta & Stoltenberg, 1997).

There is a common view that the introduction of no-till crop establishment challenges conventional approaches used to manage weeds in crops. Few reports are available that describe the effects of short-term crop rotations combined with tillage practices on the composition of soil weed seed banks (Feldman et al., 1997). Specifically, there is a lack of information concerning the composition of the prevalent weed seed bank in Willamette Valley, Oregon turf and forage seed rotation systems.

The objective of this research was to determine the impact of six perennial ryegrass rotation system components that differ by rotation sequence and tillage method on soil weed seed species diversity and spatial aggregation.



The field experiment was conducted during 1997 to 1998 at the USDA-ARS long-term sustainable grass seed rotation and management systems evaluation area in Linn County (44º35'N latitude and 123º17'W longitude), Oregon, USA. The farm surrounding the research area has been under grass seed production cultivation for decades. A cropping systems project was established on the area in 1992. The soil is Woodburn (fine-silty, mixed, mesic Aquultic Agixeroll) that is poorly drained and normally saturated from late-autumn to early-spring. The climate is typical for humid temperate marine climatic conditions with 800mm of precipitation during the rainfall period, with little or no precipitation during July and August.

Each tillage-crop rotation sequence combination (Table 1) was 20x30m, with each treatment replicated four times and arranged in a randomized complete block design. There were a total of 24 plots at the research site. The CT (conventional-till) system consisted of two passes with a tractor-powered, rear-mounted roto-tiller. A roller mounted behind the roto-tiller compressed the soil with each pass. In the NT (No-till), crops were planted directly into the partially removed crop residue (approximately 10% of the total above-ground post-harvest phytomass produced by the crop). A conventional double-disk opener planter with modifications was used to plant the crop seeds.

Prescription-based management principles were used to determine when to conduct fertilizer and weed control practices. Standard weed control practices were used for each rotation system, using registered herbicides. Record of the specific herbicides used for the crops harvested in 1997 is given in Table 2. The plots were fertilized according to standard recommendations for each crop.

The soil seed bank was sampled by removing 20 to 30 cores of 5cm diameter by 7cm depth, along two V-shaped pattern transects in each plot, spaced 1m apart. A total of 640 soil cores were collected from all plots. The sampling was done during June 1997, immediately after crop harvest. The soil cores were air-dried in paper bags and stored at approximately 20ºC for five months until analyzed for seed content.

The soil cores were broken by hand and ground for two seconds in a grinder (Custom Laboratory Equipment, Orange City, Fl). The approximate 250ml of ground soil was added to 325ml of vermiculite (grade #4) and mixed (Burrell Scientific, Pittsburgh) for 30 seconds at 300 shakes min-1. The mixed samples were placed in plastic zip-lock-bags with 250ml of dionized water. The hydrated samples were stored in a dark growth chamber (Percival Model E-54, Boone, IA) for 10 days at 5ºC to break secondary seed dormancy. After stratification, each sample was spread in aluminum trays (20x20cm) to a depth of three cm ±1cm depth. The trays were placed at random on greenhouse benches and maintained at 20±2ºC and 18±2ºC (day and night, respectively). Natural light was supplemented with high intensity light (PPF of 170µmol.m-2.s-1) for 14 hours day-1. The samples in the trays were lightly watered daily or when necessary to maintain uniform wetness. Each soil sample was subjected to three cycles of drying for seven days, hydrating, seedling emergence and counting, and soil stirring. Preliminary experiments indicated that three cycles generally depleted the seed bank of readily non-dormant viable seeds, as has been reported by others (Lush, 1988).

The weed seedlings were identified, counted, and removed with forceps as they emerged until germination ceased. Species identifications were confirmed by growing seedling reference samples to a stage of development that was easily recognized. Reference seedling samples, representative of all identified and unidentified species, were maintained in a herbarium. The assays were conducted between 1 December 1997 and 30 April 1998.

The weed seed species data from the soil cores were logarithmically transformed [log10 (x+1)] to adjust for heterogeneity of variance and non-additivity (Mulugeta & Stoltenberg, 1997). Differences among transformed means were separated using Fisher's protected least significant difference test. The treatment means, calculated from log-transformed values, were inversely transformed using the antilogarithm [antilog10 (x-1)] for interpretation. The individual weed species differences among the six rotation systems were subjected to analysis of variance.

The density and spatial distribution patterns of the individual weed seeds in the soil cores collected along the sampling transects included: (i) number of seeds; (ii) percentage of cores with seeds; (iii) relative percentage of species (R); (iv) mean number of seeds per core; (v) coefficient of variation (CV) and (vi) Morisita's aggregation index (Johnson & Anderson, 1986; Boccanelli & Lewis, 1994). The PCR-COR software (McCune & Mefford, 1997) was used to calculate the heterogeneity indices: (i) richness (S), (ii) Shannon-Weaver diversity [H'=-∑ Pi * ln (Pi)] and (iii) evenness (E = H'/ln richness).

Multidimensional scaling and cluster analyses were used to test similarity among the six rotation systems based on the weed seed densities of the four individual replications and means. Multivariate common factor analysis (Norusis, 1994) was used to determine the soil seed bank density relationships based of the 15 most prevalent species among the aggregated samples. Four factor groupings were produced using factor rotation by the equimax method. Pearson's correlation (r) and Spearman's rank correlation (rs) coefficients were used to determine placement of significantly associated species within the factor groupings. Based on the placement of the 15 weed species within the four factor groups, the six rotation system treatments were placed into three groups. Analysis of variance was used to determine differences among three rotation system groups for the six measures of soil seed bank density and spatial distribution. Pearson correlation coefficients were determined among seed bank density and spatial distribution characteristics by individual species, individual species by rotation systems, and heterogeneity indices by the six rotation systems. All results were reported at P ≤ 0.05, unless otherwise stated.



Weed species characterization - A total of twenty weed species were recorded in the top 7cm of the soil seed bank among all rotation system treatments. The soil seed bank was almost exclusively composed of annual species (Table 3). The only exception was the perennial species was Lotus corniculatus L., which occurred at a low density and likely was an introduced contaminant with crop seeds previously sown in the experimental area. The weed flora was dominated by two species, Juncus bufonius L. and Poa annua L. that represented 59 and 22% of the total seed bank composition, respectively (Table 3). These were also the most abundant species recovered from all samples, being present in 80 and 88% of the soil cores, respectively. The remaining 16 species account only for 19% of the seed bank, of which Veronica arvensis L.and Cerastium viscosum L. represented 11% of the total. The five least abundant species, Lupinus sp., Solanum nigrum L., Oxalis sp., Echinochloa crusgalli (L.) Beauv. and Lotus corniculatus, represented only 0.02% of the seed bank and were excluded from the statistical analyses (Dessaint et al., 1991). The species present at this research site are indicative of soils that are saturated from early-winter to early-spring an that have an abundance of Juncus bufonius in the soil seed bank (Thompson & Grime, 1979). Normally, this kind of environment contains a high frequency of annual species (Levassor et al., 1990).

The coefficient of variation (CV) and Morisita's index values indicate that the seed bank samples were very different from each other (Table 3). Juncus bufonius and Poa annua were fairly randomly distributed compared to the very aggregated species such as Crepis setosa Hall f. and Kickxia elatine (L.) Dumort. Morisita's indices that equal 1 have random spatial distribution patterns (Johnson & Anderson, 1986). Aggregated patterns have been observed in old-field systems (Bigwood & Inouye, 1988; Boccanelli & Lewis, 1994), as well as in cultivated systems (Benoit et al., 1989; Dessaint et al., 1991; Forcella et al., 1992). Sonchus asper (L.) Hill and Senecio vulgaris L. were the most evenly distributed dicotyledonous species among all samples, and had relatively low Morisita's index values (Table 3).

Generally high CV values were observed for all weed species (Table 3), and the CV was correlated with Morisita's index (r = 0.94, ≤ = 0.0001). In addition, the percentage of soil cores that contained a weed species was seed density dependent (r = 0.82, ≤ = 0.0001). Other systems have reported that the CV was seed density dependent (Forcella et al., 1992) and that aggregation and density were inversely proportional (Dessaint et al., 1991). In this experiment, both CV and aggregation were not associated with weed seed density (r = -0.43 and -0.26, respectively). This may be the result of the greater number of species present in our systems compared to others (Forcella et al., 1992).

Rotation system effects - The soil seed bank size and species composition among rotation system treatments was differentially affected by crop sequence history and the associated herbicide applications (Table 4). Multidimensional scaling (MDS) and cluster analyses classified the six rotation systems into three groupings: (i) WCNT-MFNT and WCCT-MFCT (Group-A); (ii) SpWhtNT-MFNT and F-WCNTWC (Group-B); (iii) PRGNT-SpOatNT and PRGNT-SpWhtNT (Group-C), based on the density of the soil seed bank constituents (Figure 1). Between the MDS and cluster analysis methods, there was strong support for groupings of the respective rotation systems into Groups 1 and 3. The two rotation systems in Group-A only differed by establishment of the meadowfoam crop, and the two systems in Group-C by kind of spring-planted cereal (wheat and oats). The aggregation of the two systems into Group-B was not as assured as with the other two groups based on the MDS method (Figure 1A), probably because of the difference in kinds of crops that composed the rotation sequence.



Relationship of species density and heterogeneity with rotation system groups - The Group-A meadowfoam-after-white clover seed rotation systems had the same soil seed bank richness, evenness, and diversity indices (Table 5), and had generally similar species densities (Table 4). Crop sequence (Table 1), and thus the similarity of herbicide management (Table 2), appeared to be the cause for this similarity, with no effect resulting from the utilization of conventional and no-tillage establishment methods. This may have been due to the relatively few years that no-till management has been employed at the site. This is contrary to reports that even moderate disturbance increased species diversity (Rykiel, 1985). In other longer-term system comparisons, the number of species found in no-tillage management were either greater or the same as in conventional tillage (Cardina et al., 1991).



The Group-B rotation systems (SpWht-MFNT and F-WCNTWC) had different crop sequence histories, but shared nearly the same species composition (Table 4). The Group-B systems had similar richness, but differed in evenness and diversity, with the SpWht-MFNT system having lower E and H' than all of the six systems (Table 5). The low E value for SpWht-MFNT may be due to the dominance of Juncus bufonius in this system, compared to the other five systems. The F-WCNTWC system had more Digitaria saguinalis (L.) Soop. than the SpWht-MFNT system. Because diversity was directly associated with evenness (r = 0.88, ≤ = 0.05), SpWht-MFNT also had the lowest H' index value. The high weed seed densities in the meadowfoam systems (both systems in Group-A and SpWht-MFNT in Group-B) may be due to poor weed control of early germinating species in late-winter or early-spring because of a sole reliance on the herbicide Prism for post-emergence control.

The Group-C systems, which were comprised of the two spring-planted cereals that were planted without tillage and which followed winter fallow applications of glyphosate, had the lowest total soil seed bank densities of all systems (Table 4), in addition to the lowest species richness (Table 5). The PRG-SpOatNT system had a total seed bank size greater than the PRG-SpWhtNT system, but was similar to the F-WCNT-WC system. The PRG-SpWhtNT system had one post-emergence application of the selective herbicide Hoelon and PRG-SpOatNT did not (because of lack of crop tolerance). The Group-C systems also had equal heterogeneity indices, but the amount of species richness was less than that in WCNT-MFNT and WCCT-MFCT.

Relationship of individual weed species densities and their associations with rotation system groups - The individual weed species composition of the soil seed bank was differentially affected by the different rotation systems (Table 4). Multivariate common factor analyses classified the individual weed species based on the Pearson correlation among weed seed densities of the six rotation systems (Table 6). Those species that did not have significant factor loadings based on Pearson correlation coefficients were classified by their Spearman rank correlation coefficients. This approach provided insight into which species were the cause of the group classifications of the rotation systems (Figure 1). Factor-1 contained those weed species that were associated with Group-A rotations systems (WCNT-MFNT and WCCT-MFCT). Factor-2 contained weed species that were associated with Group-B (SpWht-MFNT and F-WCNTWC). Factor-3 weed species were associated with the rotation systems in Group-C (PRG-SpWhtNT and PRG-SpOatNT).

Juncus bufonius whichhad the greatest seed density of all species, was favored by the three meadowfoam systems, but was nearly eliminated in two winter-fallow systems (PRG-SpOatNT and PRG-SpWhtNT). The high potential densities of J. bufonius at this research site were due to the saturated soil conditions from winter until early-spring with rapid growth and high seed production in late-spring (Popai et al., 1995). Veronica arvensis responded similarly to the rotation systems as J. bufonius. (r = 0.99; ≤ = 0.0001), with great soil seed bank reduction by the two cereal rotation systems.

Poa annua had the greatest seed density of all treatment in the WCNT-MFNT and WCCT-MFCT systems. However, the introduction of winter chemical-fallow system in conjunction with white clover seed production (F-WCNTWC) or following perennial ryegrass seed production (PRGNT-SpOatNT and PRGNT-SpWhtNT), greatly reduced its density. Poa annua has become a problematic grassy weed in perennial ryegrass seed production systems since the exclusion of thermal post-harvest residue management (Mueller-Warrant, 1994). Rotation systems that include winter chemical-fallow components appeared to be effective weed management strategies to reduce the size of P. annua seed banks.

The soil seed bank density responses of Festuca myuros L., Trifolium dubium Sibth. and Cerastium viscosum were similar to Poa annua (r = 0.96, 0.81, and 0.83; ≤ = 0.002, 0.06, and 0.04, respectively). These three weeds may form an association of weeds with P. annua, but may respond independently in their response to different rotation systems because of theirelimination from the PRG-SpOatNT and PRG-SpWhtNT systems. Poa annua seed densities were greatly reduced in the cereal systems, so this may indicate that its response is still in transition. The response of Senecio vulgaris was inversely related to that of P. annua (r = -0.78; ≤ = 0.07), and may become a problematic weed in those systems that are able to have P. annua control.

The summer annual grass Panicum capillare L. had the greatest soil seed bank densities in the PRG-SpOatNT and PRG-SpWhtNT systems (Group-C). This was probably due to later germination time after the spring chemical-fallow application of glyphosate than the winter and spring annuals germinate, thus escaping the effect of the herbicide application. Digitaria sanguinalis (L.) Scop. had its greatest density in the PRG-SpOatNT (Group-C) and F-WCNTWC (Group-B) rotation systems, probably because hoelon herbicide is not used in the spring oat system and summer annual grasses germinate after the period of winter grass control in white clover seed rotation component. Gnaphalium palustre Nutt. and Kickxia elatine are summer annual dicotyledonous species with similar responses to the different rotation systems (r = 0.95; ≤ = 0.003) that had the greatest soil seed bank densities in the WCNT-MFNT, WCCT-MFCT (Group-A) and F-WCNTWC (Group-B) systems. The seedlings of both species probably also escaped the effects of winter- and spring-applied herbicides because of their late germination time, compared to other species (Medeiros & Steiner, unpublished data, 1998). The reduction in G. palustre and K. elatine in PRG-SpOatNT and PRG-SpWhtNT was possibly due to the combined competition effects from perennial ryegrass the prior year in the rotation sequence, as well as the competition from spring-planted cereals. These two summer annual grass species have independent niche responses to the six rotation systems, as indicated by the lack of correlation among their seed bank densities (r = -0.04; ≤ = 0.94). Similarly, Panicum capillare was not associated with G. palustre and K. elatine (r = -0.63 and -0.51; ≤ = 0.18 and 0.30, respectively). Crepis setosa responds to the rotation systems similarly as G. palustre and K. elatine, but was among the least abundant species in the total seed bank. This species does not appear to have the potential to become a problematic weed for grass seed rotation systems. The soil seed bank responses for Digitaria sanguinalis and K. elatine were similar (r = 0.87; ≤ = 0.03).

The species Sonchus asper, Senecio vulgaris and Daucus carota L. were unaffected by the six rotation system treatments and all had relatively low factor loading values (Table 6). Cardamine oligosperma Nutt. increased in PRG-SpWhtNT, but was present at relatively low seed densities in all treatments. There were also no correlation among C. oligosperma and any other weed species and appears that this is a non-problematic weed in these grass seed rotation systems.



♦ The weed flora recorded among all rotation system treatments was almost annual species, dominated by Juncus bufonius L. and Poa annua L.;
♦ high and low seed species densities were associated with random and aggregation seed distribution patterns, respectively;
♦ the charateristic of the soil seed bank among rotation system treatments was differentially affected by crop sequence history and the associated herbicide applications;
♦ rotation systems that included winter chemical-fallow system in conjunction with white clover seed production or, spring oat or wheat following perennial ryegrass seed production, greatly depressed the Poa annua L. density;
♦ there is a clear indication that rapid changes in weed seed bank size and composition can occur in different perennial ryegrass seed rotation systems.



The authors thank W. Gavin, B. Robinson and C. Reid for technical assistance. Thanks also to G. Mueller-Warrant M. Wilson for suggestions regarding field sampling strategies. This research was supported by a grant from the Oregon Department of Agriculture Alternatives to Field Burning research program, Grant No. 9302 012-1 and by CAPES, Brasil.



BALL, D.A. Weed seed bank response to tillage, herbicide and crop rotation sequence. Weed Science, Lawrence, v.40, n.4, p.654-659, 1992.         [ Links ]

BENOIT, D.L.; KENKEL, N.C. & CAVERS, P.B. Factors influencing the precision of seed bank estimates. Canadian Journal of Botany,Otawa, v.67, p.2833-2840, 1989.         [ Links ]

BIGWOOD, D.W. & INOUYE, D.W. Spatial pattern analysis of seed banks: an improved method and optimized sampling. Ecology, College Park, v.69, n.2, p.497-507, 1988.         [ Links ]

BOCCANELLI, S.I. & LEWIS, J.P. The seed bank of an old pampean prairie and its relation with the standing vegetation. Pesquisa Agropropecuária Brasileira, Brasília, v.29, n.12, p.1833-1840, 1994.         [ Links ]

BUHLER, D.D. Influence of tillage systems on weed population dynamics and management in corn and soybean in the Central USA. Crop Science, Madison, v.35, n.5, p.1247-1258, 1995.         [ Links ]

BUHLER, D.D. & MESTER,T.C. Effect of tillage system on the emergence depth of giant foxtail (Setaria faberi) and green foxtail (Setaria viridis). Weed Science, Lawrence, v.39, n.2, p.200-203, 1991.         [ Links ]

CARDINA, J.; REGNIER, E. & HARRISON, K. Long-term tillage effects on seed banks in three Ohio soils. Weed Science, Lawrence, v.39, n.2, p.186-194, 1991.         [ Links ]

CAVERS, P.B. & BENOIT, D.L. Seed bank in arable land. In: LECK, M.A.; PARKER, V.T. & SIMPSON, R.L. (eds.). Ecology of soil seed banks. New York:Academic Press, 1989. p.309-328.         [ Links ]

CHAUVEL, B.J.; GASQUEZ, J. & DARMENCY, H. Changes of weed seed bank parameters according to species, time and environment. Weed Research, Oxford, v.29, n.1, p.213-219, 1989.         [ Links ]

CHURCHILL, D.B.; ALDERMAN, S.C.; MUELLER-WARRANT, G.W. & BISLAND, D.M. Weed seeds survival in composed grass seed straw. In: YOUNG, W.C.III. (ed.). Seed poduction research. Corvallis: Oregon State University Extension and USDA-ARS, 1995. p.30-32.         [ Links ]

DESSAINT, F.; CHADOEUF, R. & BARRALIS, G. Spatial pattern analysis of weed seeds in the cultivated soil seed bank. Journal of Applied Ecology, Oxford, v.28, n.2, p.721-730, 1991.         [ Links ]

FELDMAN, S.R.; ALZUGARAY, C.; TORRES, P.S. & LEWIS, P. The effect of different tillage systems on the composition of the seed bank. Weed Research, Oxford, v.37, n.1, p.71-76, 1997.         [ Links ]

FENNER, M. Ecology of seed banks. In: KIEGEL, J. & GALILI, G. (eds.). Seed development and germination. New York: Marcel Dekker Inc., 1995. p.507-528.         [ Links ]

FORCELLA, F.; WILSON, R.G.; RENNER, K.A.; DAKKER, J.; HARVEY, R.G.; ALM, D.A.; BUHLER, D.D. & CARDINA, J. Weed seed banks of the U.S. corn belt: magnitude, variation, emergence, and application. Weed Science, Lawrence, v.40, n.4, p.636-644, 1992.         [ Links ]

HALBERG, G.R. Pesticide pollution of groundwater in the humid United States. Agricultural Ecosystem Environmental, Amsterdam, v.26, n.2, p.299-367, 1989.         [ Links ]

JOHNSON, R.G. & ANDERSON, R.C. The seed bank of tallgrass prairie in Illinois. American Midland Naturalist, Notre Dame,v.115, n.1, p.123-130, 1986.         [ Links ]

LEVASSOR, C.; ORTEGA, M. & PECO, B. Seed bank dynamics of Mediterranean pastures subject to mechanical disturbance. Journal of Vegetation Science, Uppsala, v.1, n.1, p.339-344, 1990.         [ Links ]

LEWIS, W.M. Weed control in reduced-tillage soybean production. In: WIESE, A.F. (ed.). Weed control in limited tillage systems. Champaign: Weed Science Society of America, 1985. p.41-50.         [ Links ]

LUSH, W.M. Biology of Poa annua in a temperate zone golf putting green (Agrostis stolonifera/Poa annua). II. The seed bank. Journal Appllied Ecology, Oxford, v.25, n.3, p.989-997, 1988.         [ Links ]

MAXWELL, B.D.; ROUSH, M.L. & RADOSEVICH, S.R. Predicting the evolution and dynamics of herbicide resistance in weed populations. Weed Technology, Lawrence, v.4, n.1, p.2-13, 1990.         [ Links ]

McCUNE, B. & MEFFORD, M.J. PC-ORD - Multivariate analysis of ecological data. Gleneden Beach: MjM Software, 1997. 256p.         [ Links ]

MUELLER-WARRANT, G.W.; YOUNG, W.C.III. & W.C. MELLBYE. Influence of residue removal method and herbicides on perennial ryegrass seed production: I. Weed control. Agronomy Journal, Madison, v.86, n.4, p.677-690, 1994.         [ Links ]

MULUGETA, D. & STOLTENBERG, D.E. Weed and seed bank management with integrated methods as influenced by tillage. Weed Science, Lawrence, v.45, n.5, p.706-715, 1997.         [ Links ]

NORUSIS, M.J. SPSS Professional Statistics 6.1. Chicago: Illinois, 1994. 385p.         [ Links ]

POPAI, A.I.; COX, T.I.; INGLE, A. & KERR, R. Seasonal emergence of weeds in cultivated soil in New Zealand. Weed Research, Oxford, v.35, n.2, p.429-436, 1995.         [ Links ]

ROBERTS, H.A. Seed banks in soils. Advances in Applied Biology, Toronto, v.6, n.1, p.1-55, 1981.         [ Links ]

RYKIEL, E.J. Towards a definition of ecological disturbance. Australian Journal of Ecology, Oxford, v.10, n.2, p.361-365, 1985.         [ Links ]

SCHWEITZER, L.R. Reducing grass seed field volunteer population with combine-mounted seed cleaners or recovery systems and by selective harvest timing. In: YOUNG, W.C. III (ed.). Seed production research. Corvallis: Oregon State University Extension & USDA-ARS, 1994. p.19-24.         [ Links ]

SCHWEIZER, E.E. & ZIMDAHL, R.L. Weed seed decline in irrigated soil after six years of continuous corn (Zea mays) and herbicides. Weed Science, Lawrence, v.32, v.1, p.76-83, 1984.         [ Links ]

SYMONIDES, E. Seed bank in old-field successional ecosystems. Ekologia Polska, Warszawa, v.34, n.1, p.3-29, 1986.         [ Links ]

THOMPSON, K. & GRIME, J.P. Seasonal variation in the seed banks of herbaceous species in ten contrasting habitats. Journal of Ecology, Cambridge, v.67, n.3, p.893-921, 1979.         [ Links ]

WEED SCIENCE SOCIETY OF AMERICA. Composite list of weeds. Weed Science, Lawrence, v.32, suppl.2, p.1-50,1984.         [ Links ]

WILKINSON, L. & HILL, M.A. Systat for the Macintosh, Version 5.2.1. Evanston: Illinois. 1992. 724p.         [ Links ]

YENISH, J.P.; DOLL, J.D. & BUHLER, D.D. Effect of tillage on vertical distribution and viability of weed seed in soil. Weed Science, Lawrence, v.40, n.3, p.429-433, 1992.         [ Links ]



Aceito para publicação em 30.12.2001.



1 Parte do Projeto de Pesquisa de Pós-Doutorado realizado pelo primeiro autor no "National Forage Seed Production Research Center", USDA-ARS, Corvallis, OR.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License