Crop Science Grow Your Career with CSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 26 August 2005
Published in Crop Sci 45:2023-2029 (2005)
© 2005 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in Crop Science
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Willenborg, C. J.
Right arrow Articles by Shirtliffe, S. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Willenborg, C. J.
Right arrow Articles by Shirtliffe, S. J.
Agricola
Right arrow Articles by Willenborg, C. J.
Right arrow Articles by Shirtliffe, S. J.
Related Collections
Right arrow Seed Establishment
Right arrow Other Grain Crops
Right arrow Crop Ecology

CROP ECOLOGY, MANAGEMENT & QUALITY

Oat Germination Characteristics Differ among Genotypes, Seed Sizes, and Osmotic Potentials

Christian J. Willenborga,*, Jeffery C. Wildemanc, Aaron K. Millerb, Brian G. Rossnageld and Steven J. Shirtliffeb

a Department of Plant Science, University of Manitoba, 222 Agriculture Building, Winnipeg, MB R3T 2N2
b Dep. of Plant Sciences, Univ. of Saskatchewan, 51 Campus Drive, Saskatoon, SK, Canada S7N 5A8
c Cargill AgHorizons, P.O. Box 4200, Regina, SK, Canada S4P 3W5
d Crop Development Centre, Univ. of Saskatchewan, 51 Campus Drive, Saskatoon, SK, Canada S7N 5A8

* Corresponding author (christian_willenborg{at}umanitoba.ca)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Oat (Avena sativa L.) yield and quality on the northern Great Plains are consistently reduced by frequent drought and wild oat (Avena fatua L.) competition. Wild oat cannot be selectively removed from oat with herbicides. Identifying genotypes or seed size(s) with high germination potential under moisture stress may facilitate improved seedling vigor, stand establishment, and crop competitiveness. Therefore, a germination study was conducted to determine the effects of genotype and seed size on the germination of oat seed subjected to moisture stress. Large, medium, and small seeds of six common western Canadian oat genotypes were germinated in polyethylene glycol (PEG 8000) solutions with initial osmotic potentials ranging from 0 to –0.4 MPa at 5°C. Generalized linear mixed models fit to the data provided a statistically valid, appropriate, and convenient method to analyze germination data. In all genotypes examined, decreasing seed size and osmotic potential increased median germination time (MGT) and lowered final germination percentage (FGP). Among genotypes, CDC Bell had the fastest MGT while AC Mustang had the highest FGP. Delays in MGT and reductions in FGP resulting from increased moisture stress were similar to those observed in other cereals, suggesting that oat may be as capable of germinating under low spring soil moisture conditions as wheat and barley. The results of this study also indicate that large seed of genotypes such as AC Mustang and CDC Bell appear better suited to germinate under the range of osmotic potentials in this study. Although differences in MGT and FGP between seed sizes in this study were statistically significant, they were generally small and thus, oat germination characteristics may not be substantially improved by screening out small seed from farm-saved seed.

Abbreviations: FGP, final germination percentage • GLIM, generalized linear model • GLIMMIX, generalized linear mixed model • GLM, general linear model • MGT, median germination time • PEG, polyethylene glycol • REML, restricted maximum likelihood • TKW, 1000-kernel weight


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
SEED GERMINATION, beginning with imbibition and culminating in radicle emergence, is an essential process in plant development. Successful stand establishment, in terms of the rapidity and extent of establishment, are contingent on adequate seed germination and vigor. Moreover, the rapidity and extent of stand establishment are important factors determining yield and maturity of spring crops (Briggs and Aytenfisu, 1979). Nevertheless, inadequate stand establishment frequently occurs on the semiarid northern Great Plains of the USA and Canada, an area characterized by short growing seasons, frequent droughts, and low temperatures (Ashraf and Abu-Shakra, 1978; Hao and de Jong, 1988). Several studies have indicated that germination is delayed and reduced by low water potential (Rao and Dao, 1987; Hao and de Jong, 1988), low temperature (Livingston and de Jong, 1990; Willenborg et al., 2004), high salinity (Romo and Haferkamp, 1987), and combinations of these factors (Willenborg et al., 2004; Livingston and de Jong, 1990; Hao and de Jong, 1988). The resulting late and inadequate germination and subsequent seedling establishment critically affect crop production this region (Livingston and de Jong, 1990).

Oat is one of the major economic grain crops grown in the northern Great Plains region, planted annually on 1.7 million hectares of land in western Canada alone (Statistics Canada, 2004). Oat is a crop that normally produces seed of varying size as a result of the multifloret habit of the oat spikelet (Doehlert et al., 2002, 2004). Oat seed size is inherently nonuniform because oat may produce one, two, or three seeds per spikelet. The innermost seed, called the primary seed, is the largest and seed size and mass decrease with increasing seed order (Doehlert et al., 2002). This relationship results from reduced sink strength and photoassimilate acquisition of higher order seeds (Palagyi, 1983; Doehlert et al., 2002).

The influence of seed size on germination and subsequent seedling emergence of various crop species has been the subject of numerous published studies (Mathur et al., 1982; Lafond and Baker, 1986; Kawade et al., 1987; Roy et al., 1996; Guberac et al., 1998; Larsen and Andreasen, 2004). However, results from these studies have been mixed, varying widely among species. Large seed of pearl millet (Pennisetum glaucum L.) demonstrated increased germination and emergence compared to small seed (Kawade et al., 1987). Larsen and Andreasen (2004) also observed increased germination percentage and decreased median germination time with increasing seed mass in slender creeping red fescue (Festuca rubra L. subsp. litoralis Vasey), perennial ryegrass (Lolium perenne L.), and Kentucky bluegrass (Poa pratensis L.). In contrast, small wheat (Triticum aestivum L.) seed germinated faster than large under various temperature and moisture stress combinations (Lafond and Baker, 1986). Similarly, small seed of a modern rice (Oriza sativa L.) cultivar (cv. BR1) began germination earlier than large seed (Roy et al., 1996). In wheat and rice, it was concluded that small seed completed water imbibition more rapidly, and thus began germination earlier (Roy et al., 1996) and germinated more rapidly (Lafond and Baker, 1986). While Lafond and Baker (1986) did not observe a seed size effect on final germination percentage, Roy et al. (1996) indicated that large rice seed, although slower to initiate germination, had higher final germination than small seed. In oat, Guberac et al. (1998) observed increased germination in large seeds (99%) versus small (95%), but the study included only one genotype. Likewise, Mathur et al. (1982) noted that Indian oat varieties with large seed had higher overall germination percentages and germination energy indices than varieties composed of small seeds. However, the oat seeds were germinated in deionized water and thus were not subjected to moisture stress. Germination of oat seed on the northern Great Plains frequently occurs under low soil moisture conditions.

Little is known about the effect of seed size on the germination of various oat genotypes, particularly under moisture-limited conditions. Both rapid and complete germination are critical for the establishment of a competitive crop. The competitive advantage gained by earlier emerging plants frequently results from the size bias and resulting asymmetric competition that follows in which one plant obtains a disproportionately large amount of resources (O'Donovan et al., 1985; Gonzalez Ponce, 1987; Freckleton and Watkinson, 2001). This is especially important in oat, where wild oat cannot be selectively removed from the crop using herbicides because of genetic and morphological similarities. Early planting coupled with rapid germination and subsequent earlier emergence may result in oat gaining a competitive advantage over wild oat. Identifying either genotypes or seed sizes with high germination potential under moisture stress may facilitate improved seedling vigor, stand establishment, and crop competitiveness. The objective of this study was to determine the effects of genotype and seed size on the germination of oat seed subjected to moisture stress. We postulate that the response of oat germination to seed size will vary among genotypes and moisture levels.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Seed Material and Grading
Six common western Canadian oat genotypes were examined: AC Mustang, CDC Bell, CDC Pacer, OT 288, Riel, and Triple Crown. Genotypes were chosen on the basis of diversity of pedigrees and relatively high seed lot germination percentage, which exceeded 97% for all genotypes. All seed lots originated from one field harvested in 2001 at the Kernen Crop Research Farm, Saskatoon, SK, Canada (lat 52°09', long 106°33'). Within each seed lot, oat seed was separated into three size classes consisting of small, medium, and large seed. Small seeds were those that passed through a 1.95- x 8.33-mm slotted sieve (Canada Seed Equipment Limited, Saskatoon, Canada), medium seeds were those retained on a 1.95- x 8.33-mm slotted sieve, and large seeds were those retained on a 2.35- x 8.33-mm slotted sieve. Subsequent to fractionation, all physically damaged seeds were manually removed. Thousand-kernel weight (TKW) was determined for each of the 18 seed fractions by counting and weighing two samples of 200 seeds.

Experimental Procedures
Germination tests were performed from October to December 2001 and March to April 2004. The experiment was designed as a completely randomized factorial of six genotypes, three seed sizes, and three moisture stress treatments. The experiment was repeated twice in each year (four times in total) and each experimental run included four replications per treatment, for a total of 864 experimental units. Seed of each of the six genotypes was fractionated into classes consisting of small, medium, and large seed. Moisture stress treatments consisted of seed imbibed in solutions with initial osmotic potentials of 0, –0.2, and –0.4 MPa.

For each treatment, 40 seeds were placed on two layers of filter paper (grade 413, VWR, Edmonton, AB, Canada) in a 9-cm plastic Petri dish (Phoenix Biomedical, Mississauga, ON, Canada). Moisture stress treatments were established by irrigating each Petri dish with 8 mL of the appropriate osmotic solution. Osmotic potentials were created using polyethylene glycol (PEG 8000, Sigma Chemical Company, St. Louis, MO, USA) and were adjusted for temperature (5°C) according to Michel (1983). Polyethylene glycol was used because it is a nonpenetrating osmoticum (Hardegree and Emmerich, 1990). Petri dishes were placed in trays and covered with light-excluding plastic to prevent light penetration and moisture loss. Seeds were incubated in a germination cabinet at 5°C. Germination was recorded every 24 h for 14 d. Seeds were regarded as germinated when the radicle appeared normal and had protruded at least 2 mm. Germinated seeds were removed from the experiment and the dishes were recovered immediately after each enumeration.

Statistical Analysis
The response of oat germination characteristics to genotype, seed size, and moisture stress was examined by calculating final germination percentage (FGP) and median germination time (MGT), or the time to 50% germination. Final germination percentage was calculated as the cumulative number of germinated seeds with normal radicles in each experimental unit at termination of the experiment:

[1]
where n is the number of germinated seed at each enumeration interval, and Nt is the number of seeds in each experimental unit. Median germination time was estimated by fitting germination curves from each experimental unit to the following logistic function:

[2]
where Pt is the proportion of seeds germinated at time t, t is thermal time in degree hours (°C h; base temperature = 0°C) (Sonego et al., 2000) accumulated since experiment initiation, a is the estimated germination rate (number of germinated seeds °C h–1), and b is the estimated median germination time (°C h) in each experimental unit. Germination rate values were highly positively correlated with those for median germination time and therefore are not discussed further.

Residuals were initially tested for normality with PROC UNIVARIATE of SAS (SAS Inst., 1996) but did not conform to the normal distribution. Traditionally, the main approach used by agronomists and crop scientists to account for nonnormality has been transformation, but transformations are limited in that they generally can be used to stabilized the variance but rarely improve skewness in the data (Littell et al., 2002). Furthermore, transformation of the independent variable in a linear model essentially forces the data to fit a model developed for other purposes as opposed to using an appropriate statistical model (Madden et al., 2002). Therefore, FGP and MGT were analyzed using the GLIMMIX macro of PROC MIXED in SAS, which fits a generalized linear mixed model to the data by iterative use of the MIXED procedure of SAS (SAS Inst., 1996). Models were fit to the data using restricted pseudo-likelihood (Wolfinger and O'Connell, 1993), which fits mixed linear models using restricted maximum likelihood (REML is also the default method of estimation in PROC MIXED). GLIMMIX is also able to account for random effects in the model, which the user is able to specify along with fixed effects (Littell et al., 1996). Genotype, seed size, and osmotic potential were considered fixed effects in the model, while run and its interactions with fixed effects were regarded as random effects. The significance of random effects was determined by computing a Z value (the variance parameter of the random effect divided by its approximate standard error), which was then tested for a difference from zero (SAS Inst., 1996).

A binomial distribution with a logit-link function was used to model FGP while MGT was modeled by a Poisson distribution with a log-linear link function. This takes into account that the logit and log-linear transformations homogenize the variance and accounts for the fact that FGP and MGT belong to binomial and Poisson distributions, respectively. The binomial and Poisson distributions have previously been found to be appropriate for the type of germination data collected in this study (Andersson et al., 2002; Pujol et al., 2002; Balcomb and Chapman, 2003). F tests were calculated on the basis of Wald statistics and means were separated using Fisher's protected least significant difference, with treatment effects declared significant at P < 0.05.

No single test is available to indicate which generalized linear mixed model is most appropriate for the data (Littell et al., 1996; Piepho, 1999; Madden et al., 2002). However, deviance statistics provided by the GLIMMIX macro as well as plots of residuals provide a good indication of model fit. Therefore, goodness of fit of the models was evaluated by these methods. Residual plots for MGT and FGP were compared in plots of standardized Pearson residuals versus the estimated linear predictor of each model as described by Madden et al. (2002), where random scattering of the residual points indicates a good model fit. Where necessary, models were adjusted for overdispersion by adjusting deviances relative to the dispersion parameter as described by Littell et al. (1996). The responses of oat MGT and FGP to osmotic potential were tested for linearity with the regression procedure of SAS (SAS Inst., 1996).

For ease of interpretation, FGP means shown in the tables are presented both on the logit scale as well as back transformed after calculation on logit-transformed data by:

[3]
where x is the logit estimate from the model, and FGPa is the approximate final germination percentage after back transformation. Similarly, means for MGT are presented on the log scale as well as back transformed after calculation on the log-linear transformed data by:

[4]
where x is the log-linear estimate from the model, and MGTa is the approximate median germination time after back transformation.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Oat TKWs varied considerably between genotypes and seed size classes (Table 1). CDC Bell exhibited the greatest seed size variation (10.2–40.9 g), while OT 288 exhibited the least (16.4–31.9 g). The relatively low TKW of small seed of CDC Bell and Triple Crown are likely due to the high proportion of tertiary kernels in these genotypes. The largest range of TKW existed in the large seed size class (31.9–40.9 g) whereas the smallest occurred in the medium class (26.4–30.9 g). These differences likely result from differences in hull content between genotypes and seed size classes.


View this table:
[in this window]
[in a new window]
  		Table 1. Thousand-kernel weight (g) of fractionated seed samples for each genotype.

 
Z Tests indicated a significant run x genotype x size x PEG interaction for MGT (P < 0.001) and FGP (P < 0.01) (Table 2). However, plots of these interactions indicated that they were of little consequence to statistical outcomes and model performance (relatively small proportion of sum of squares), and therefore data were combined over runs for the GLIMMIX analysis. The generalized linear mixed models described well the effects of oat genotype, seed size, and PEG concentration on MGT and FGP (Tables 3 and 4). In all cases convergence was obtained in the iteration estimation procedure. Standardized Pearson residuals exhibited a random pattern of scatter when plotted against the estimated linear predictor of each model, indicating good model fits for both MGT and FGP. Although data were overdispersed, no evidence of lack of fit was detected in either model on the basis of deviance statistics after adjusting for overdispersion. Binary data are typically overdispersed because of violation of the independence assumption (Piepho, 1999). In germination experiments, overdispersion results because the assumption that the germination of all seed is independent from one another is violated as seeds are nested within experimental units (Petri dishes).


View this table:
[in this window]
[in a new window]
  		Table 2. Significance of covariance parameter estimates or random effects and their interactions with fixed effects in the GLIMMIX macro of PROC MIXED.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Linear mixed-model analyses of germination data using PROC MIXED and the GLIMMIX macro with restricted pseudolikelihood estimation of model parameters. P values were obtained by likelihood ratio tests.

 

View this table:
[in this window]
[in a new window]
  		Table 4. Parameter estimates (means and back-transformed means) for effects of genotype, seed size, and PEG concentration on median germination time and final germination percentage estimated from the GLIMMIX analysis. (Only fixed effects that differed significantly from the null level are shown).

 
Oat germination characteristics varied substantially between seed size classes (Table 3). Among genotypes and osmotic potentials, decreasing oat seed size increased MGT (Table 4). Large oat seed took 8% (89°C h) (back-transformed) less time to reach 50% germination than small seed, irrespective of genotype or moisture stress. Medium seed also had significantly lower MGT than small seed, but not than large (Table 4). Similarly, both medium and large seed exhibited 5% (back-transformed) greater final germination than small seed, regardless of genotype and osmotic potential. No significant differences in FGP were observed between medium and large seed (Table 4).

Oat MGT and FGP were greatly affected by oat genotype as indicated by highly significant (P = 0.001) F values for each (Table 3). CDC Pacer took significantly longer to achieve 50% germination than all other genotypes, while AC Mustang and CDC Bell achieved 50% germination more quickly than all others (Table 4). Median germination time of the fastest germinating genotypes (AC Mustang and CDC Bell) was 170°C h lower (back-transformed) than the slowest germinating genotype (CDC Pacer). Differences in MGT between OT 288, Riel, and Triple Crown were not significant (P < 0.05).

A high degree of germination was observed for all genotypes, with FGP in excess of 97% (back-transformed) in all but one case. In contrast to the variation observed between genotypes for MGT, few differences were observed between FGP (Table 4). Although statistically significant (P = 0.001), only a 1.5-fold difference in FGP existed between the genotype with the highest FGP (AC Mustang) and the lowest FGP (Riel). When back-transformed, the 1.5-fold difference corresponded to a difference in FGP of 4%. CDC Bell and CDC Pacer exhibited a significantly (P = 0.001) higher and lower final germination percentage than Riel and AC Mustang, respectively.

Oat MGT and FGP were most affected (P < 0.001) by moisture stress treatment (Tables 3 and 4). MGT exhibited a linear response to osmotic potential (R2 = 0.99) and differed significantly between all moisture stress treatments (Table 4). With no osmotic stress (0 MPa), MGT was 15% (170°C h) and 28% (367°C h) lower compared with germination at osmotic potentials of –0.2 and –0.4 MPa, respectively (Table 4). The correlation between FGP and osmotic potential (R2 = 0.79) was not as great as that between MGT and osmotic potential (R2 = 0.99). Nevertheless, FGP differed significantly (P = 0.004) among moisture stress treatments (Table 3). Overall, oat final germination percentage was 7% lower (back-transformed) at PEG –0.4 MPa than oat final germination with no moisture stress (Table 4). Differences in FGP between 0 MPa and –0.2 MPa were not significant (P < 0.05).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Our results indicate that planting large oat seed may improve oat germination characteristics, irrespective of genotype (Tables 3 and 4). Large seed required 89°C h less time to achieve 50% germination and exhibited 5% greater final germination than small seed, regardless of moisture stress. This could result in improved stand establishment and faster germination, particularly where low spring soil moisture limits stand establishment. Guberac et al. (1998) also reported that large and small oat seed differed with respect to germination characteristics with large and small oat seed of the variety VESNA having FGP of 99 and 95%, respectively. These values are similar to our observations of 98 and 93% FGP for large and small seed, respectively (Table 4). In barley (Hordeum vulgare L.), Turk and Tawaha (2002) observed increased germination percentage as well as greater speed of germination in large seed compared with small. However, these results are inconsistent with those of Mian and Nafziger (1992) who reported that seed size had no effect on germination characteristics in wheat.

This experiment also demonstrates differential MGT and FGP among western Canadian oat genotypes (Table 3). The importance of genotype in affecting germination characteristics of other cereal crops has been previously recognized (Ashraf and Abu-Shakra, 1978; Lafond and Baker, 1986; Briggs and Dunn, 2000). Among four common Middle East wheat varieties, speed of germination (as determined by a vigor index) was lowest and final germination percentage highest in the variety Najah, under both low temperatures and high moisture tensions (Ashraf and Abu-Shakra, 1978). Similarly, Briggs and Dunn (2000) indicated that germination characteristics differed significantly among a diverse range of western Canadian six-row barley cultivars. In the current study, AC Mustang and CDC Bell exhibited the fastest time to 50% germination while AC Mustang had the greatest final germination percentage of the six genotypes examined (Table 4). Moreover, the greater germination exhibited by AC Mustang and CDC Bell was consistent across seed size classes and osmotic potentials. Imbibition time varies considerably in oat depending on seed size, hull and seed coat permeability, and soil moisture (Peterson, 1992). Furthermore, the hull is often the source of leachates that act as germination inhibitors and concentrations of these may vary among genotypes (Elliott and Leopold, 1953; Haggquist et al., 1984). Little is currently known about the effect of the hull on germination of various oat genotypes.

With respect to MGT, the difference between the fastest (AC Mustang and CDC Bell) and slowest (CDC Pacer) germinating genotypes was only 171°C h (back-transformed) or approximately 1.5 d at 5°C, a temperature at which germination frequently occurs on the northern Great Plains (Environment Canada, 2004). Thus, it appears unlikely that genotypic differences alone in MGT observed in this study would contribute to large differences in competitive ability on the basis of germination characteristics. However, in addition to its rapid germination, CDC Bell (a forage oat genotype) was also the most competitive of six oat genotypes with wild oat (J.C. Wildeman, unpublished data). Early planting coupled with rapid germination and subsequent emergence may allow CDC Bell to gain a competitive advantage over wild oat, especially if plants of CDC Bell are able to establish a significant canopy before wild oat emergence. This may be possible by applying a preseeding burnoff with a nonselective herbicide followed by early planting of large CDC Bell seed.

Averaged over all genotypes and seed sizes, increased moisture stress induced by decreasing osmotic potentials increased MGT and decreased FGP (Table 4). This response was anticipated as several studies have indicated that increasing moisture stress reduces and delays germination (Sharma, 1976; Hegarty, 1977; Schneider and Gupta, 1985; Willenborg et al., 2004). In this study, MGT was delayed by 367°C h, while FGP was reduced by 7% as a result of decreasing osmotic potential (increasing moisture stress) from 0 to –0.4 MPa. These values are similar to those obtained for wheat (Lafond and Baker, 1986) and barley (Turk and Tawaha, 2002) and are relatively small compared with what might be expected from a species which is most productive under moist conditions (Sorrells and Simmons, 1992). Although oat generally requires more moisture than wheat and barley for optimal growth and yield (Coffman and Frey, 1961; Brown, 1975), our results suggest that the germination capacity of these crops under moisture-limited conditions is similar. Therefore, oat may be as capable of germinating under low spring soil moisture conditions as wheat and barley.

Commercial oat seed lots vary widely in seed uniformity and frequently contain large proportions of small seed (Doehlert et al., 2004). Consequently, oat growers may choose to plant Certified seed in which seed <1.95 mm in width generally have been sieved out to ensure superior seed quality. Alternatively, growers may plant farm-saved seed from the previous growing season which may or may not have been sieved to remove small or thin seed. In any seed lot, the effect of seed sizing on seed lot performance essentially is dependent on the relationship between seed weight and germination. For the range of genotypes included in this study, we found that seed size generally had a significant, but relatively small effect on MGT and FGP (Table 4). However, we must acknowledge that the effect of seed size in this study was confounded within genotype because of variability in hull contents and TKWs among genotypes (Table 1). Nevertheless, our results are indicative of the actual improvements in oat germination characteristics achievable by seed sizing on a field scale because our method of fractionation was chosen on the basis of that used by producers. On the basis of these results, it may not be beneficial for producers to screen out small seed from farm-saved seed lots because of enhanced germination. Similar conclusions have been drawn in wheat (Gan and Stobbe, 1996) and turfgrass (Larsen and Andreasen, 2004).

Overall, germination characteristics of oat seed varied considerably among genotypes, seed sizes, and osmotic potentials with respect to MGT and FGP (Tables 3 and 4). Although the differences observed in this study were relatively small, it is important to note that the genotypes that were evaluated are common western Canadian genotypes that are well adapted and in general, vary little from a genetic perspective. Larger differences in oat germination characteristics may have been observed had a more diverse range of genotypes been examined. Moreover, differences in emergence under field conditions may be greater than those observed for germination because of variation in coleoptile elongation among genotypes. Semi-dwarf winter wheat genotypes have a slower rate of coleoptile elongation (Allan et al., 1962) and may emerge later. Nevertheless, because these closely related genotypes did exhibit significant variation in MGT, potential may exist to select for genotypes with reduced MGT and earlier germination in the future.

Although we expected interactions between genotype, seed size, and osmotic potential, the response of MGT and FGP to seed size and genotype did not vary across osmotic potential, indicating that treatment effects were additive, and more importantly, that germination among oat genotypes and seed size classes is consistent even under moisture stress. Therefore, the results of this study indicate that there appear to be oat genotypes and seed sizes that are better able to germinate under dry spring soil conditions. Large seed of genotypes such as AC Mustang and CDC Bell were better suited to germinate under the range of osmotic potentials included in this study. Because differences in seed vigor are frequently greater under field conditions than in laboratory tests (ISTA, 1995), it is possible that use of large seed of AC Mustang or CDC Bell may have a greater effect on MGT and FGP than suggested by the results of this study.


   ACKNOWLEDGMENTS
 
The authors gratefully acknowledge funding from the Quaker Oats Company and Cargill Ag Horizons for this research. We are also grateful for technical expertise and assistance provided by Rachelle German.

Received for publication December 11, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 


Related articles in Crop Science:

THIS ISSUE IN CROP SCIENCE

Crop Science 2005 45: x. [Full Text]  



This article has been cited by other articles:


Home page
 Agron. J.Home page
S. Snapp, R. Price, and M. Morton
Seed Priming of Winter Annual Cover Crops Improves Germination and Emergence
Agron. J., September 8, 2008; 100(5): 1506 - 1510.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in Crop Science
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Willenborg, C. J.
Right arrow Articles by Shirtliffe, S. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Willenborg, C. J.
Right arrow Articles by Shirtliffe, S. J.
Agricola
Right arrow Articles by Willenborg, C. J.
Right arrow Articles by Shirtliffe, S. J.
Related Collections
Right arrow Seed Establishment
Right arrow Other Grain Crops
Right arrow Crop Ecology

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal 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