Crop Science Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 26 August 2005
Published in Crop Sci 45:2030-2037 (2005)
© 2005 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
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 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 ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Larsen, S. U.
Right arrow Articles by Bibby, B. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Larsen, S. U.
Right arrow Articles by Bibby, B. M.
Agricola
Right arrow Articles by Larsen, S. U.
Right arrow Articles by Bibby, B. M.
Related Collections
Right arrow Turfgrass
Right arrow Seed Establishment

TURFGRASS SCIENCE

Differences in Thermal Time Requirement for Germination of Three Turfgrass Species

Søren Ugilt Larsena,* and Bo Martin Bibbyb

a Forest and Landscape-Denmark, Rolighedsvej 23, DK-1958 Frederiksberg C, Denmark
b Dep. of Natural Sciences, The Royal Veterinary and Agricultural Univ., Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmark

* Corresponding author (sugl{at}kvl.dk)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Temperature is among the most influential environmental factors for germination and establishment of grass species. We compared germination response to suboptimal temperature for the turfgrass species slender creeping red fescue (Festuca rubra L. var. littoralis Vasey), perennial ryegrass (Lolium perenne L.), and Kentucky bluegrass (Poa pratensis L.). Two cultivars of each species were germinated at five constant temperatures (8, 12, 16, 20, and 24°C), and germination was recorded one to three times per day. Final germination percentage was little affected by temperature, indicating that the base temperature for germination (Tb) is relatively constant within seed populations. Consequently, germination response to temperature was analyzed by a nonlinear regression method, which combined the thermal time model and the four parameter Weibull function. The analysis provided biologically significant parameters for comparing the cultivars and species. Tb only varied slightly between species, from 2.6°C for red fescue and Kentucky bluegrass to 3.6°C for perennial ryegrass. Thermal time to 50% of final germination was 63.9 degree-days for perennial ryegrass, 43.8 for red fescue, and 115.6 for Kentucky bluegrass, and thermal time from 25 to 75% of final germination was 14.9, 11.1, and 35.0 degree-days. Thus, Kentucky bluegrass requires a longer thermal time to germinate and has a larger variation in thermal time requirement within a seed lot. Consequently, low soil temperature results in a relatively slower germination of Kentucky bluegrass, possibly resulting in a poorer competitive ability of this species. This suggests that poor establishment of Kentucky bluegrass may partly be due to a larger thermal time required for germination.

Abbreviations: FGP, final germination percentage • Tb, base temperature for germination • {theta}T(50), thermal time for 50% of final germination • {theta}T(25–75), thermal time from 25 to 75% of final germination


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
TURFGRASS is often established by sowing seed mixtures of different species and cultivars with various complementary characteristics which ensures genetic diversity (Beard, 1973). For turfgrass in temperate areas, seed mixtures often include the species slender creeping red fescue (subsequently just termed red fescue), perennial ryegrass, and Kentucky bluegrass. It is often desirable to achieve a considerable proportion of Kentucky bluegrass in the turf because of its wear tolerance (Shearman and Beard, 1975) and its production of rhizomes (Etter, 1951) which allows the turf to produce strong sod and to repair itself after wear (Shildrick, 1982; Christians, 1998). It is often found, however, that Kentucky bluegrass establishes poorly when sown in mixture with red fescue and particularly perennial ryegrass; a large proportion of Kentucky bluegrass in the seed mixture often results in a relatively small proportion of Kentucky bluegrass tillers in the established turf (Adams and Bryan, 1974; Niehaus, 1976; Hsiang et al., 1997; Larsen et al., 2004a). Larsen et al. (2004a) found that the time factor was important for the establishment of Kentucky bluegrass in mixtures. When this species was given a time advantage by sowing it up to 35 d earlier than red fescue and perennial ryegrass, the established turf consisted of a considerably larger proportion of Kentucky bluegrass tillers. Slow establishment of Kentucky bluegrass may, however, be due to both slower emergence (Pommer, 1972; Bø, 1989) and/or slower seedling growth (Arnott and Jones, 1970; Henderlong, 1971) compared to red fescue and particularly perennial ryegrass. To get a better understanding of the poor establishment of Kentucky bluegrass in species mixtures, it is relevant to distinguish between species differences in germination rate, pre-emergence seedling growth rate, and post-emergence seedling growth rate. In particular, it is relevant to study how the different grass species respond in these aspects to various environmental factors. Along with water potential, temperature is one of the most important factors for seed germination (Bewley and Black, 1994). However, there appears to be no published results that compare the germination response to temperature of the grass species red fescue, perennial ryegrass, and Kentucky bluegrass.

At suboptimal temperatures, germination rate, that is, the reciprocal time to germination, often increases linearly with germination temperature (Bierhuizen and Wagenvoort, 1974), although the relationship may be nonlinear in certain cases (Marshall and Squire, 1996). In case of a linear relationship, germination response to suboptimal temperature can often be described by a thermal time model, which assumes that to germinate, a seed requires accumulation of a certain thermal time or heat sum, {theta}T, above a minimum base temperature, Tb, under which the germination rate is theoretically zero (Bierhuizen and Wagenvoort, 1974; Garcia-Huidobro et al., 1982; Yeh and Atherton, 2000).

Within a population of seeds, Tb is often constant for all seeds (Covell et al., 1986; Yeh and Atherton, 2000) although it has been found to vary in some species (Washitani and Takenaka, 1984; Trudgill et al., 2000). On the other hand, {theta}T generally varies among seeds within a population (Garcia-Huidobro et al., 1982; Washitani and Takenaka, 1984) and has been found to be normally or log-normally distributed (Covell et al., 1986; Dahal et al., 1990). To describe the germination response of a whole seed population to suboptimal temperatures, Covell et al. (1986) suggested the repeated probit analysis. The model assumes a constant Tb and a normally or log-normally distributed {theta}T and estimates Tb and the distribution of {theta}T in a single step. In red fescue, perennial ryegrass, as well as Kentucky bluegrass, however, Larsen (2003) showed that a nonlinear regression method which combines the thermal time model and the Weibull function was superior to the repeated probit analysis in describing the germination response to suboptimal temperature. The use of the Weibull function in germination studies has been described by Brown and Mayer (1988). The function has four parameters, which describe the lag time before start of germination, the rate of increase in germination, the skewness of the distribution of time to germination as well as the final germination percentage. The method which integrates the thermal time model and the Weibull function is more flexible than the repeated probit method when describing skewed germination data.

The aim of this study was to describe and compare how the germination of red fescue, perennial ryegrass, and Kentucky bluegrass responds to suboptimal temperatures, using the thermal time model and a nonlinear regression method to estimate Tb and the parameters in the distribution of {theta}T for the three species. Moreover, the aim was to illustrate species differences in germination response to soil temperature by predicting the expected time to germination for the three species at various times of the year, depending on the soil temperature fluctuations in a typical Danish soil. Studying these species differences may increase the knowledge about the poor establishment of Kentucky bluegrass when sown in mixture with other grass species.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Seed Material
One seed lot from six different cultivars was included in the study. Red fescue was represented by the two cultivars Cinderella and Symphony, perennial ryegrass by the cultivars Figaro and Taya, and Kentucky bluegrass by the cultivars Andante and Broadway. Most of these cultivars are represented on the official British list of turfgrass cultivars (STRI, 2004). All seed lots were commercially produced and harvested in Denmark in July 2000 and delivered by the seed company DLF-Trifolium, Denmark. Freshly matured seeds may possess primary dormancy (Baskin and Baskin, 2001), and Kentucky bluegrass seeds may germinate better after one year storage (Brede, 2001). Since the applied seeds were stored at 5°C from harvest until germination experiments were performed from March to May 2001, primary dormancy was expected to have disappeared during the storage period.

Germination Experiment
Each of the six seed lots was germinated at constant temperatures of 8, 12, 16, 20, and 24°C, and each combination of seed lot and temperature was represented by four replicates of 100 seeds. Seeds were exposed to 12 h of white fluorescent light per day (PPFD was not measured). Seeds were germinated on top of filter paper (AGF725, Frisenette, Denmark) in small plastic boxes with a small water reservoir in each box, prepared according to the principle of the "Jacobsen apparatus" ("Copenhagen Tank" ISTA, 1999). By means of a filter paper wick, water was transported from the reservoir to keep the filter paper constantly moist throughout the experiment. Before the germination test, the filter paper was soaked in a 0.2% (w/v) solution of KNO3 (ISTA, 1999) and tap water was added to the water reservoir. The germination boxes were placed in five germination incubators (Termaks KBP 2324V, Bergen, Norway), which were adjusted to the five different temperatures. The germination boxes were daily rearranged within the incubators to avoid effects of potential temperature gradients within the incubators.

Germination was recorded one, two, or three times per day depending on germination rate, and the time from the start of the experiment to the inspection was recorded to the nearest five minutes. During the counting of germinated seeds, the germination boxes were temporarily removed from the incubators to an ambient temperature of approximately 10°C. Due to the short duration of the counting process and the water reservoir in each germination box, the potential effect of ambient temperature was expected to be negligible. A seed was regarded as germinated when the radicle had protruded at least 1 mm (Larsen and Andreasen, 2004a). Only seeds with normal, sound-looking radicles were counted, and seedlings with discoloring or lacking the radicle tip were excluded. The frequency of abnormal radicles never exceeded 2 to 3%. Germinated seeds were removed from the germination test. Germination tests were terminated when no new germination was observed within the four replicates for three consecutive days.

Analysis of Germination Data
For data analysis, mean germination time courses were obtained for each combination of temperature, cultivar, and species by calculating the mean cumulative germination percentages of the four replicates at each counting time. Observations corresponding to no further increase in germination percentage during the rest of the germination test do not contribute with information, and these observations were excluded from the data set before analysis.

Analysis of the germination response to suboptimal temperature was based on the thermal time model (Bierhuizen and Wagenvoort, 1974; Garcia-Huidobro et al., 1982). The model states that the thermal time (degree-days) that passes until a given germination percentile g is reached, {theta}T(g), can be calculated as:

[1]
where T is the actual suboptimal temperature, Tb is the base temperature for germination below which germination does not occur, and tg is the actual time (days) to germination of percentile g.

The data analysis initially included a test of effect of temperature on final germination percentage (FGP). If FGP is affected by suboptimal temperature, this may indicate that Tb varies within seed lots, provided the germination test is not terminated before germination has finished. The effect of temperature on FGP was tested in a two-way analysis of variance model with cultivar and temperature as explanatory class variables. Final germination percentage values were arcsin-square-root transformed to obtain homogeneous variance and three outliers were removed since they were very deviant. FGP was compared between temperatures within cultivars by pairwise t tests using a Bonferroni correction of the P value. The tests were performed using the glm procedure of the SAS package (version 8.01, SAS Institute, Cary, NC).

Since temperature only affected FGP significantly in one cultivar and only for one temperature, Tb was assumed to be relatively constant within the seed populations. The parameters Tb and {theta}T(g) were then estimated by nonlinear regression analysis. The analysis was based on a model that combines the thermal time model and the Weibull function in a one-step method which provides estimates of Tb and {theta}T(g) by analysis of the raw data. The model assumes a constant Tb and uses the Weibull function to describe the distribution of {theta}T(g) within the seed population. The Weibull function has been used to fit the relationship between the germination percentile g and time t (days), e.g., by Brown and Mayer (1988):

[2]
where 100m is the asymptotic final germination percentage, k is the rate of increase in germination percentage, and z is the lag time before germination commences. The parameter c is a shape parameter; if c is below 3.25 the distribution of time to germination is positively skewed, if c is above 3.25 the distribution is negatively skewed (Brown and Mayer, 1988). By combining Eq. [1] and Eq. [2], the following function was obtained:

[3]
where g is germination percentage, (T – Tb)t is thermal time, and where the constant parameters Tb, m, k, z, and c were estimated in the nonlinear regression procedure. The equation estimates progress in cumulative germination on a thermal time scale, and for presentation, all times to germination, tg, were normalized on a thermal time scale by multiplying by the factor (T – Tb) (Garcia-Huidobro et al., 1982). The thermal time {theta}T(g) to obtain germination percentile g of the final germination percentage m can be determined by rearranging Eq. [3]:

[4]

For comparison of cultivars and species, thermal time to obtain 50% of final germination, {theta}T(50), was estimated as a measure of median thermal time for the seed population, and thermal time from 25 to 75% of final germination, {theta}T(25–75), was estimated as a measure of the variation in thermal time within a seed lot. The analysis was performed separately for each cultivar and species, using the nlmixed procedure of the SAS package (version 8.01, SAS Institute, Cary, NC). In the analysis, germination percentage was used as dependent variable, and time and temperature were used as independent variables. In contrast to the nlin procedure, the nlmixed procedure allows calculation of both estimates and standard errors of the parameters {theta}T(50) and {theta}T(25–75).

Soil Temperature Data and Prediction of Time to Germination
Soil temperature at 1 cm depth was recorded in a Danish sandy loam soil, as described in Larsen et al. (2004a), located at the Royal Veterinary and Agricultural University experimental farm Højbakkegaard at Taastrup, Denmark. The temperature was recorded from 2 May to 3 Sept. 2001, from 20 Mar. to 10 Dec. 2002, and from 21 Mar. to 31 Dec. 2003. Two temperature sensors (LI-COR, model 1400-103, Lincoln, NE) were placed at 1 cm soil depth. Soil temperature was monitored every minute, and mean temperature was logged every hour by a LI-COR data logger (model LI-1400). The annual fluctuation in soil temperature was described by the harmonic function:

[5]
where T is the soil temperature (°C) at time t (day number of the year), m is the mean soil temperature (°C), A is the maximum temperature (°C), and {tau} is the warmest day of the year. The function assumes a constant period of fluctuation of 365 d. The statistical model was a nonlinear regression model with exponentially correlated errors (Pinheiro and Bates, 2000) and analyzed in S-Plus (version 6.2.1, Insightful Corp., Seattle, WA).

The thermal parameters Tb and {theta}T(50) were estimated from the germination experiment for each of the grass species (as the mean of the parameters estimated for two cultivars) and combined with the predicted soil temperature to calculate the expected time to 50% of final germination at various times of the year for each of the three species.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Effect of Temperature on Final Germination Percentage
Mean FGP across all temperatures differed significantly between the three species (P < 0.001) as well as between cultivars within species (P < 0.001). There was significant interaction between cultivars and temperature (P = 0.012). However, FGP only differed between temperatures within one cultivar of perennial ryegrass and not within any of the other cultivars (Table 1). Hence, FGP was lower at 8°C than at other temperatures for perennial ryegrass cv. Taya.


View this table:
[in this window]
[in a new window]
  		Table 1. Final germination percentage at various germination temperatures for two cultivars of red fescue, perennial ryegrass, and Kentucky bluegrass, respectively.

 
Effect of Temperature on Timing of Germination
Temperature had a marked effect on the timing of germination of all species and cultivars with a reduced time to germination when temperature increased from 8 to 24°C. Kentucky bluegrass germinated more slowly than red fescue and perennial ryegrass at all five temperatures. The thermal time model was fitted to all species and cultivars separately, and the nonlinear regression analysis provided estimates of Tb and the Weibull parameters m, k, z, and c as well as the derived parameters {theta}T(50) and {theta}T(25–75) (Table 2). The estimated Tb values were used to normalize cumulative germination on a thermal time scale, and observed germination percentages for all temperatures were fitted reasonably well with a single germination curve (Fig. 1). The most apparent deviation was in Kentucky bluegrass cv. Broadway where germination at 20°C ceased relatively early compared to the other temperatures. Moreover, the fitted germination curve provided a good prediction over the whole range of the germination time course for all species and cultivars; the model described well the distribution of thermal time requirements within the seed populations.


View this table:
[in this window]
[in a new window]
  		Table 2. Estimates (±SE) of thermal requirements for germination of two cultivars of red fescue, perennial ryegrass, and Kentucky bluegrass, respectively, determined by nonlinear regression analysis. Tb is the base temperature for germination, and m, k, z, and c are constants of the Weibull function (Eq. [3]). {theta}T(50) is a derived estimate of thermal time to 50% of final germination, and {theta}T (25–75) is thermal time from 25 to 75% of final germination (Eq. [4]). SE is overall standard error for the estimate of germination percentage.

 


View larger version (34K):
[in this window]
[in a new window]
  		Fig. 1. Observed and predicted thermal time germination curves for germination of two cultivars of red fescue, perennial ryegrass, and Kentucky bluegrass, respectively, at five constant temperatures. The predicted line is based on estimates from Eq. [3] which combines the thermal time model and the Weibull function. Note the different time scales on the x axes.

 
The parameter estimates of the nonlinear regression analysis illustrate differences and commonalities between the species and cultivars (Table 2). Generally, the estimates were very similar for the two cultivars within red fescue and perennial ryegrass but differed more between the bluegrass cultivars. The mean Tb for the three species was 2.6, 3.6, and 2.6°C, indicating a slightly higher Tb for perennial ryegrass. Parameter m multiplied by 100 is an estimate of FGP and reflects the same species and cultivar differences as in Table 1. Parameter k indicates that ryegrass has a higher rate of increase in germination percentage than red fescue and particularly Kentucky bluegrass. Parameter z indicates that Kentucky bluegrass requires a longer lag time for germination to begin than red fescue and particularly perennial ryegrass. Thus, as a mean for the two cultivars, Kentucky bluegrass requires 72.5 degree-days whereas red fescue and curve, and the distribution of thermal time to germination was positively skewed for all six seed lots in this study, most pronounced in ryegrass and least in fescue.

Like parameter z, the derived estimates of {theta}T(50) and {theta}T(25–75) demonstrated marked species differences in thermal time requirement for germination. Kentucky bluegrass generally required a longer thermal time to germinate, and germination occurred over a longer period than for the other species. Thus, the mean {theta}T(50) was 64.0, 43.8, and 115.6 degree-days for red fescue, perennial ryegrass, and Kentucky bluegrass, respectively, whereas the mean {theta}T(25–75) was 14.9, 11.1, and 35.0 degree-days, respectively. The difference in thermal time requirement between the two cultivars of Kentucky bluegrass may partly be made up for by the difference in Tb, that is, the real time to germination will only differ moderately when the temperature is not very close to Tb.

Effect of Soil Temperature on Time to Germination
Soil temperature at 1 cm depth fluctuated widely from day to day due to changing weather conditions, which were not accounted for by the predicted function. In all three years, however, there was a distinct seasonal pattern, which was summarized reasonably well by the harmonically fluctuating function (Fig. 2A). The function predicted a mean soil temperature for the whole year of 9.4°C (approximate 95% confidence limits [8.6; 10.1]) with a minimum temperature of –1.0°C [–2.7;0.7] in mid-January and a maximum temperature of 19.7°C [18.6;20.8] in mid-July. Assuming that the function gave a reasonable description of the soil temperature fluctuation within the time period with measurements, the expected time to 50% of final germination over this period was calculated for each species, using mean estimates of Tb and {theta}T(50) for each species (Fig. 2B). By combining estimates of Tb and {theta}T(50) with the predictions of soil temperature, the errors related to the two estimation procedures were also combined. The prediction of time to 50% of total germination may, therefore, be associated with considerable error. The calculation assumes that daily temperature fluctuations do not affect the time to germination markedly, and that seeds are sown at 1 cm depth which is not always the case. Nevertheless, the prediction gives an indication of the variation in time to germination depending on grass species and soil temperature.



View larger version (30K):
[in this window]
[in a new window]
  		Fig. 2. (A) Observed and predicted daily soil temperature at 1-cm depth in a Danish soil, depending on time of the year. The soil was located at Taastrup, Denmark, and soil temperature was recorded from 2 May to 3 Sept. 2001, from 20 Mar. to 10 Dec. 2002, and from 21 Mar. to 31 Dec. 2003. (B) Predicted time to 50% of final germination of red fescue, perennial ryegrass, and Kentucky bluegrass, respectively, depending on time of the year. The prediction is based on the estimated thermal requirements for germination of the three species, respectively, and the predicted daily soil temperature at 1-cm depth.

 
If seeds of the three species were sown around April 1 or November 1, the expected soil temperature would be approximately 6.5°C, and perennial ryegrass, red fescue, and Kentucky bluegrass would require 16.3, 15.0, and 29.4 d, respectively, to reach 50% of final germination. On the other hand, if seeds were sown around mid-July, the expected soil temperature would be approximately 19.7°C and the three species would require 3.7, 2.7, and 6.8 d, respectively, to reach 50% of final germination. Consequently, although the relative time to germination for the three species is similar at different seasons, the difference in actual time to germination between particularly Kentucky bluegrass and the other two species is considerably larger in periods with low temperatures.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Effect of Temperature on Final Germination Percentage
The difference in FGP between species and between cultivars within species (Table 1) reflects different viabilities of the applied seed lots. Thus, the seed lot of Kentucky bluegrass cv. Broadway appears to be of particularly poor viability. Since only one seed lot was applied for each cultivar, it is not possible to distinguish between seed lot variation and cultivar variation, of which both can be significant (Larsen and Bibby, 2004). In a screening experiment with a number of seed lots of both Kentucky bluegrass cultivars Broadway and Andante, there was no significant difference in neither FGP nor mean germination time between the two cultivars when tested against seed lots differences (Larsen and Bibby, 2004). The lower FGP of Broadway is, therefore, most likely due to a low quality of this particular seed lot. Low FGP may, for instance, be due to poor conditions during production and harvest but may also reflect the degree of cleaning of the seed lot (Larsen and Andreasen, 2004b). In relation to temperature, however, there appears to be limited differences between the three species. Five of the six cultivars obtained the same FGP at all temperatures and for the sixth cultivar FGP was only significantly lower at 8°C (Table 1). With the exception for perennial ryegrass cv. Taya at 8°C, the three species, therefore, respond similarly to all suboptimal temperatures in terms of FGP. This indicates that the assumption of a common Tb for all seeds in the seed population is true, although a small proportion of the perennial ryegrass seeds may have had a Tb value above 8°C. Accordingly, Trudgill et al. (2000) found that Tb was 4.8°C for both the 25th and 50th percentiles but 9.2°C for the 75th percentile for perennial ryegrass. In terms of establishment of Kentucky bluegrass, however, the results indicate that FGP of this species is not affected by temperature. In contrast to reduced water potential during germination, which reduces FGP as well as increases the time to germination (Larsen et al., 2004b), temperature, therefore, does not appear to restrict establishment of Kentucky bluegrass by reducing the number of seedlings of this species.

Species Differences in Thermal Requirements for Germination
The fact that observations from all temperatures seem to follow a common thermal time germination course in Fig. 1 suggests that the assumptions behind the thermal time model are fulfilled, that is, that Tb is constant for individual seeds within a population whereas thermal time to germination varies with germination percentage. Therefore, it seems reasonable to use the estimates of Tb, {theta}T(50), and {theta}T(25–75) for comparison of different grass species' germination response to suboptimal temperatures, and the biological significance of these parameters are easy to interpret. Differences in actual time to germination may be due to differences in Tb, thermal time to germination, or both, and both aspects should be considered in the comparison.

The estimated Tb is slightly lower for red fescue cultivars than for perennial ryegrass cultivars, and Tb is lower for Kentucky bluegrass cv. Broadway than for all other cultivars (Table 2). Although the variation in Tb may in part explain differences in germination behavior between cultivars and species, the requirement for accumulating thermal time at temperatures above Tb appears to be a much more variable factor, illustrated by estimates of z and {theta}T(50) (Table 2). The slower germination of Kentucky bluegrass compared to the other two species is, therefore, mainly explained by a larger requirement for accumulating thermal time. Moreover, the estimates of k and {theta}T(25–75) clearly illustrate that there is a larger variation in thermal time requirement between seeds within both Kentucky bluegrass cultivars, resulting in germination over a longer time period compared to that of red fescue and perennial ryegrass.

Estimates of Tb in this study are within the range reported from previous studies. In red fescue, Beard (1980) found a Tb of 6.1°C, whereas Trudgill et al. (2000) reported Tb values ranging from 1.9 to 4.7°C, depending on the germination percentile. In perennial ryegrass, estimated Tb values have ranged from 1.4 to 2.4°C (Moot et al., 2000), and from 4.9 to 9.2°C (Trudgill et al., 2000). Rogers and Lush (1989) reported Tb values ranging from 4.4 to 6.6°C with corresponding {theta}T(50) values ranging from 37.9 to 46.3 degree-days for six perennial ryegrass cultivars, which agrees well with the present findings for this species. The variation in estimates of thermal requirements between different surveys on the same species may be due to actual differences between the applied seed lots and/or genotypes, but germination conditions and/or methods for analyzing data may also lead to different estimates of Tb (Larsen, 2003). This emphasizes the importance of using the same methodology for all species when making comparisons as in the present study.

The literature survey did not reveal any estimates of Tb or {theta}T(50) for germination of Kentucky bluegrass seeds. The species differences in thermal time requirement for germination in this study are, however, consistent with the general tendency of Kentucky bluegrass to emerge more slowly than red fescue and particularly perennial ryegrass (Skirde, 1967; Pommer, 1972), especially at low temperatures (Bø, 1989). This agrees well with the pattern in Fig. 2, illustrating that the difference in time to germination between species is most pronounced when sowing seed mixtures in periods with low soil temperatures.

Possible Implications of Species Differences in Thermal Time to Germination on Establishment of Turfgrass Mixtures
During the establishment of a plant community, the competitive balance between species is markedly affected by the time of germination and emergence (Harper, 1961; Black and Wilkinson, 1963). Earlier emergence will both result in a longer growing period and a higher position in the dominance hierarchy (Ross and Harper, 1972). When grass species are sown in mixture, the seeds of the different species will, of course, experience the same germination conditions including temperature, water potential, etc. If the species respond differently to these conditions by requiring different times to germination, this difference is likely to affect the competitive balance between species. The present results strongly suggest that slow emergence and poor establishment of Kentucky bluegrass in mixtures with red fescue and perennial ryegrass may be related to a larger requirement for accumulation of thermal time for germination to occur. Conversely, species differences in base temperature for germination appear to be relatively small and possibly of low significance for the relative establishment success.

Larsen et al. (2004a) found that Kentucky bluegrass had to be sown 60 to 454 degree-days (Tb = 0°C) before red fescue and perennial ryegrass to significantly increase the proportion of Kentucky bluegrass shoots in the established turf. As the mean soil temperature was 15.6°C, this corresponded to a minimum time advantage of 3.8 to 29.1 d to improve establishment of Kentucky bluegrass. Based on the estimates in Table 2, the time to obtain 50% of final germination at a temperature of 15.6°C would be 4.9 d for red fescue, 3.6 d for perennial ryegrass, and 8.9 d for Kentucky bluegrass and, consequently, Kentucky bluegrass would require 4.0 d more than red fescue and 5.3 d more than perennial ryegrass. This indicates that the poor establishment of Kentucky bluegrass may—at least partly—be explained by the larger thermal time requirement for germination of this species. However, it was also found that a time advantage of approximately 400 degree-days or 25 d was often needed to substantially increase the percentage of Kentucky bluegrass (Larsen et al., 2004a). Thus, species differences in thermal time requirement for germination does not solely explain the poor establishment of Kentucky bluegrass. Differences in pre-emergence and post-emergence growth rates are also likely to be involved. Henderlong (1971) found that red fescue seedlings weighed approximately four times more and perennial ryegrass seedlings approximately six times more than seedlings of Kentucky bluegrass when weighing seedlings five weeks after emergence of the given species, which is a clear indication of a slower post-emergence growth rate of Kentucky bluegrass. Thus, poor competitive ability of Kentucky bluegrass during establishment is most likely due to a combined effect of slower germination and slower seedling growth compared to red fescue and particularly perennial ryegrass.

The importance of species differences in germination requirements, however, may be more pronounced at poor seedbed conditions. For instance, the difference in real time to germination between species is larger when soil temperature is low, as illustrated in Fig. 2B, and sowing seed mixtures at low soil temperatures is likely to hamper the establishment of Kentucky bluegrass further. The seasonal fluctuation in soil temperature in Fig. 2 is specific for a particular geographic location and may be different in other locations. Nevertheless, soil temperature is most likely fluctuating in a typical seasonal pattern also in other temperate areas. And although the actual temperature may deviate from the average pattern, the difference in time to germination between the species may be diminished markedly by sowing seed mixtures during periods with high soil temperature. This may, at least to some extent, improve the establishment of Kentucky bluegrass in mixtures, as well as increase the speed of overall establishment. Moreover, Kentucky bluegrass seeds respond more strongly to reduced water potential (Larsen et al., 2004b) and to sowing depth (Käding and Kreil, 1982) compared to seeds of red fescue and perennial ryegrass. Soil moisture and sowing depth should, therefore, also be optimized to improve establishment of Kentucky bluegrass in mixtures.


   ACKNOWLEDGMENTS
 
We thank the Royal Veterinary and Agricultural University, Danish Centre for Forest, Landscape and Planning, and the Danish Research Agency for the financial support. The supply of seed samples from DLF-Trifolium A/S is gratefully acknowledged. Technical assistance from Olaf Bos and Lars Arne Jensen during the germination tests is greatly appreciated.

Received for publication December 16, 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
Right arrow Abstract Freely available
Right arrow Figures Only
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 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 ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Larsen, S. U.
Right arrow Articles by Bibby, B. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Larsen, S. U.
Right arrow Articles by Bibby, B. M.
Agricola
Right arrow Articles by Larsen, S. U.
Right arrow Articles by Bibby, B. M.
Related Collections
Right arrow Turfgrass
Right arrow Seed Establishment

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