Published online 30 July 2007
Published in Crop Sci 47:1596-1602 (2007)
© 2007 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
TURFGRASS SCIENCE
Winterhardiness and Turf Quality of Accessions of Perennial Ryegrass (Lolium perenne L.) from Public Collections
Brent S. Hulkea,*,
Eric Watkinsb,
Donald Wysea and
Nancy Ehlkea
a Dep. of Agronomy and Plant Genetics, Univ. of Minnesota, St. Paul, MN 55108
b Dep. of Horticultural Science, Univ. of Minnesota, St. Paul, MN 55108
* Corresponding author (hulke002{at}umn.edu).
 |
ABSTRACT
|
|---|
The lack of winterhardiness of some cool-season grasses limit their usefulness in northern climates. Perennial ryegrass (Lolium perenne L.) lacks winterhardiness in Minnesota but has desirable qualities, such as wear tolerance and rapid establishment, that are useful in many turf applications. While domesticated germplasm may lack winterhardiness, undomesticated wild or landrace germplasm may have genes for better winterhardiness in environments like Minnesota's. In 2004, 300 accessions from two public sources in the USA were planted with eight check varieties and populations in two environments in central Minnesota, St. Paul and Becker. Thirty individuals of each accession were evaluated for seedling vigor and tiller survival after the first winter. Other turf-quality traits and tiller survival after the second winter were evaluated on those plants that survived the first winter. The first winter was extremely harsh and resulted in the death of all accessions and checks at Becker. There was good differentiation among accessions at St. Paul for tiller survival, with 8 of the 300 accessions performing better than NK200, the most winterhardy check variety. The second winter was considerably less harsh, with less death of tillers and whole plants. While the accessions do not have suitable turf quality for direct domestication, turf quality can be improved through breeding.
Abbreviations: GRIN, Germplasm Resources Information Network • NPGS, National Plant Germplasm System
|
INTRODUCTION
|
|---|
PERENNIAL RYEGRASS (Lolium perenne L., 2n = 2x = 14) is a common turfgrass on sports fields and golf courses in mild, temperate climates. The rapid growth and establishment of perennial ryegrass aid in the recovery of a turf after it is damaged by heavy traffic. It is also used for winter overseeding of warm-season grasses on golf courses in the southern USA. Perennial ryegrass is included in lawn mixtures to rapidly establish groundcover around homes (Christians, 2004).
High demand for perennial ryegrass has given grass seed producers incentive to produce more perennial ryegrass seed. In recent years, farmers in northern Minnesota have increased their acreage of perennial ryegrass seed production to keep up with demand. This production takes place near the 49th parallel, where harsh winter conditions are common. Typically, a farmer can maintain a profitable stand of perennial ryegrass through one winter. A second winter usually results in enough cumulative injury to decrease the productivity of a field below economic viability. Because the species requires vernalization to produce flowering culms, one winter must pass before the first crop can be taken from the field. Therefore, perennial ryegrass behaves more like a biennial than a perennial in this environment (Ehlke and Undersander, 1990). Similarly, survival of perennial ryegrass in turf applications is limited by the cold conditions that result in winter kill (Taylor et al., 1997). A variety that possesses improved winterhardiness with excellent turf quality characteristics would be in high demand in cold regions such as Minnesota.
Waldron et al. (1998) predicted gain from selection in perennial ryegrass populations derived from elite varieties. They concluded that it should be possible to improve winterhardiness of elite germplasm without negatively affecting turf quality if multiple trait selection indices are used. However, improvement via recurrent selection is slow, and it could take many years of effort to improve perennial ryegrass to the point that winterhardiness is no longer a major issue. In northern Europe, attempts to discover freezing-tolerant genotypes in undomesticated germplasm have shown promise for improving forage-type perennial ryegrass (Lorenzetti et al., 1971; Humphreys and Eagles, 1988; Tcacenco et al., 1989). Our objective was to study the field performance of perennial ryegrass accessions available in public collections (i) for winterhardiness in Minnesota and (ii) for turf quality given the accession survives the first winter.
|
MATERIALS AND METHODS
|
|---|
Accessions of wild and landrace germplasm of perennial ryegrass were obtained from the National Plant Germplasm System (NPGS) of the USDA-ARS and from Rutgers University. In total, 300 accessions were chosen for the study, 280 from NPGS and 20 from Rutgers. The NPGS accessions were chosen based on their description on the Germplasm Resources Information Network (GRIN) website (National Genetic Resources Program, 2006). An attempt was made to include mostly nonvarietal accessions. Most of the accessions were from Europe and North Africa, the regions of origin of perennial ryegrass, although accessions from other parts of the world were also included.
About 100 seeds of each accession were planted into separate 4-inch square pots in late June 2004. About 500 seeds of each of eight varieties and breeding populations were sown into larger pots. The elite varieties included ‘Brightstar SLT’ (Rose-Fricker et al., 2003), ‘Citation Fore’ (Fraser et al., 2004), and ‘Pizzazz’ (Bonos et al., 2003). ‘NK200’ and ‘Ragnar II’, varieties with good winterhardiness and poor to moderate turf quality, were also included. Minnesota breeding populations ‘MHT’, ‘MSP’, and ‘WH x TQ’ were included as checks that have a history of selection for winterhardiness in Minnesota environments. After the seeds germinated and the plants reached about 3 to 5 cm in height, 50 seedlings of each accession were planted into 50-cell plug trays, with one plant occupying each cell. For each check, 500 plants were transplanted into ten 50-cell plug trays.
In late August 2004, the plants were transplanted into dead sod at the St. Paul Agricultural Experiment Station, St. Paul, MN (44°59'23'' N, 93°10'28'' W), and the Sand Plains Agricultural Experiment Station, Becker, MN (45°23'47'' N, 93°53'21'' W), as spaced plants on 1-foot centers. Both environments are in central Minnesota. The soil at St. Paul is a Waukegan silt loam (fine-silty over sandy, mixed, mesic Typic Hapludoll), and the soil at Becker is a Hubbard loamy sand (sandy, mixed, frigid Entic Hapludoll). The St. Paul environment is surrounded by forest and urban development, while the Becker environment is a large plain in a rural area with few windbreaks. Plants were organized in a randomized, incomplete blocks-within-replications design (Schutz and Cockerham, 1966). Plots of accessions were randomly assigned to 1 of 10 blocks nested within each of three replications at each environment. Assignment of accessions to blocks did not change from replication to replication. To determine if differences in blocks within a replication had a significant effect on the data, plots of each of the eight checks were included in each block. Data of the checks were analyzed separately to determine if any environmental differences exist among the blocks in each replication. For both the checks and the accessions, one plot consisted of five plants planted proximal to each other. Plants were mowed weekly during the growing season to 5 to 6 cm (2–2.5 in) in height. Fertilizer was applied in the autumn of 2004 and 2005 (49 g m–2 of 10–4.27–8.3 N–P–K).
Data were collected on single plants as the experimental unit. In general, traits measured on a rating scale were rated so that the highest value indicated the most beneficial phenotype for overall plant desirability. Seedling vigor ratings were based on tillering ability and rate of leaf extension and growth 1 wk after mowing. Seedling vigor was rated shortly after seedlings were transplanted into the field (early October 2004) on a 1 to 9 scale with 1 indicating the lowest vigor found, 5 indicating average vigor over the two environments, and 9 indicating the greatest vigor found. Tiller survival was rated in mid-May 2005 and 2006 on a 1 to 9 scale with 1 indicating complete death of the plant, and each incremental unit increase being equivalent to a 12.5 percentage unit increase in percent tiller survival on a whole-plant basis. Spring growth was measured in mid-May 2005 and 2006 as the length of an average tiller from each plant. Summer and autumn turf quality were rated on 8 July 2005 and 23 September 2005, respectively, on a 1 to 9 scale with 1 indicating the poorest turf quality observed, 5 indicating average turf quality, and 9 indicating the best turf quality observed. Turf quality is a composite trait that includes color, leaf texture, growth habit, density of tillers, tolerance of mowing, and presence of disease as factors in overall appearance. Turf color was rated on 8 July 2005 on a 1 to 9 scale with 1 indicating the lightest green color found and 9 indicating the darkest green color found. Leaf texture was rated on 13 July 2005 on a 1 to 9 scale with 1 indicating the coarsest foliar appearance and 9 indicating the finest foliar appearance. Crown rust incidence was rated on 19 July 2005 on a 1 to 9 scale with 9 indicating no appearance of pustules, and each incremental unit decrease being equivalent to a 12.5 percentage unit increase in coverage of leaf surfaces with pustules. Growth habit was rated on 22 September 2005 on a 1 to 5 scale with 1 indicating all tillers completely prostrate and 5 indicating all tillers completely upright. Spreading vigor was rated at the end of the first full growing season (mid-October 2005) as plant diameter in centimeters.
All data were analyzed using PROC GLM of SAS statistical software package version 9.1 (SAS Institute, 2002). For seedling vigor, the model consisted of main effects for environments and entries, with replications nested within environments and incomplete blocks nested within replications and environments. For all other traits, the model consisted of main effects for replications and entries, with incomplete blocks nested within replications. The effects for multiple environments were removed for these traits because no plants survived the first winter at Becker; and therefore, no data were colleted at Becker for those traits. All effects are considered random. F tests were used to determine the significance of each term in the model, with random error used as the error term in each test. Separate analyses of checks only were performed for each of the traits, and for all traits, the effect for incomplete blocks was significant at the 0.05 probability level. This indicates the need for an incomplete blocks term in the model to remove variation due to heterogeneity in the field environment. Pearson correlations were calculated on an individual plant basis between all traits studied using PROC CORR of SAS statistical software package version 9.1 (SAS Institute, 2002).
|
RESULTS AND DISCUSSION
|
|---|
Winter stresses in the winter of 2004–2005 were severe for the two environments in this study. Throughout much of the season, there was little to no snow cover, and temperature varied considerably during the season, even resulting in a brief soil thaw at 0 to 5 cm depth under sod at St. Paul in early February (Fig. S.1, S.2, S.3, and S.4 in Hulke et al., 2006). This resulted in highly differential tiller survival in the accessions at St. Paul, and no tiller survival in any of the accessions or checks at Becker. The lack of survival at the Becker environment is likely due to cold, windy conditions without ample snow cover and dry soil conditions following the spring thaw, due to the dry winter. The sandy soil with lack of surrounding wind protection may have made plants in this environment more prone to injury than plants in St. Paul. We were unable to irrigate in early spring because irrigation was not available immediately after the soil thawed. Perennial ryegrass plants are very susceptible to drought at this time of year, which could be due to inadequate root structure before the period of new root production (Garwood, 1967). At St. Paul, these winter stresses and spring conditions were not as detrimental to the plants, which resulted in survival of at least one plant of 170 accessions. St. Paul benefits from forested urban surroundings, which resulted in more protected conditions in the field than at Becker, and heavier soils, which retain water better than the sand-based soil at Becker. Despite the large number of accessions with at least one survivor, there were very few accessions that did as well or better than the most winterhardy check variety, NK200. In total, 10 accessions had statistically similar performance compared to NK200, and 8 accessions had statistically superior performance (p < 0.05; Table S.1, Hulke et al., 2006). The eight best accessions ranged in average tiller survival scores from 6.1 to 7.2, which is much higher than the average score of 4.8 for NK200 or 3.2 for the best performing elite variety, Citation Fore. While Citation Fore did perform the best for tiller survival among the three elite check varieties in 2005, it was not significantly different than the other elite checks.
The winter of 2005–2006 was considerably less severe, with ample snow cover most of the season and more stable soil and air temperatures (Fig. S.5, S.6, S.7, and S.8 in Hulke et al., 2006). No midseason thawing events occurred during this winter, and the total duration of frozen ground was half as long, lasting only from early February to late March. The only plants tested during this winter were the survivors from the first winter, so fewer plants were tested in this winter, and the percentage that died due to winter injury was much lower. Since tiller survival was measured instead of total death, it was still possible to obtain a clear picture of winter injury. Interestingly, the rankings for tiller survival in 2006 did not mirror the rankings for 2005 (Tables S.1 and S.2, Hulke et al., 2006). This is also reflected in the Pearson correlations, which indicate a very low, but significant correlation between tiller survival in 2005 and 2006 (r = 0.20 among all genotypes, p < 0.001; Table 1). This is likely due to differences in the type and severity of winter stresses. More of the accessions were statistically similar to NK200 in this environment, including all of the top-ranked accessions for 2006, indicating fewer significant differences in the highest-ranked accessions. This is indicative of less total stress during this winter compared to the previous winter. Four accessions were statistically superior to the highest-ranked elite check, Brightstar SLT, for tiller survival in 2006, but only one or two observations were available for each of those accessions, casting some doubt on the reliability of those means. The three elite varieties were not significantly different from each other for tiller survival in 2006.
View this table:
[in this window]
[in a new window]
|
Table 1. Phenotypic correlations between all field winterhardiness and turf traits for perennial ryegrass accessions and accessions plus perennial ryegrass check varieties and populations.
|
|
Since data were collected on many different accessions from many locations worldwide, it should be possible to determine which regions harbor the most winterhardy landrace germplasm. This will allow us to determine the best sites from which to collect perennial ryegrass for use in improving winterhardiness in breeding programs. While data on accessions available in public collections can be somewhat inconsistent, it is usually possible to determine the site of collection at least to the country. The countries with tiller survival means significantly higher than 1.0 (complete death) for both years are Belgium, Italy, Luxembourg, Norway, Romania, the former Soviet Union, Sweden, and the former Yugoslavia (Table 2). These are countries with high-latitude continental or alpine climates within some portion of their boundaries. For the mild winter of 2006 alone, more countries had significantly higher tiller survival than 1.0, including Greece and Algeria. These are countries in lower latitudes with milder climates. A milder winter led these accessions to perform better in 2006.
View this table:
[in this window]
[in a new window]
|
Table 2. Mean and range of field winterhardiness and turf traits for perennial ryegrass accessions grouped by country of origin.
|
|
Typically, the modern-day plant collector also provides global positioning information for better data on latitude and altitude of the site of collection. Latitude and altitude have been correlated with freezing tolerance of perennial ryegrass in previous reports (Lorenzetti et al., 1971). Collectors may also include information on the climate of the site of origin. Lorenzetti et al. (1971) found an association between freezing tolerance of perennial ryegrass plants collected from a location and the mean lowest temperature of the coldest month at that location. In our study, the most winterhardy accessions were from high elevations in alpine Italy, mountainous regions with continental climates in Romania and western parts of the former Soviet Union, and regions of high latitude such as Norway and Sweden (Tables S.1 and S.2, Hulke et al., 2006; National Genetic Resources Program, 2006). If data had not been provided on the site of origin or climate of origin of the best accessions, it would have been difficult to determine why those accessions performed so well in our winter environment while others from the same country did not. This underlines the importance of obtaining geographical data as well as environmental data on accessions during the collection process. Such information adds value to plant collections.
Spring growth of the plants was influenced greatly by the preceding winter. The correlations between spring growth and tiller survival in the same year were moderate (r = 0.42 for 2005, r = 0.50 for 2006, p < 0.001; Table 1). However the correlations between tiller survival and spring growth of different years were much smaller. This indicates that the plants with the capability to overwinter well also have a head start on growth the following year. This could be related to carbohydrate metabolism in terms of reserves: those plants with the most reserves may use them for improved cryoprotection, maintenance of biological processes through winter, and recovery during the following spring. Such a model has been hypothesized by Pollock et al. (1988), but data supporting the cryoprotective role of carbohydrates have been inconsistent (Lawrence et al., 1973; Pollock et al., 1988; Eagles and Williams, 1992).
Turf quality of the accessions was not statistically superior to that of the three elite checks in this experiment, indicating the need for improvement of turf quality while improving winterhardiness in any population derived from this germplasm (Tables S.3 and S.4, Hulke et al., 2006). However, tiller survival was not highly correlated with any turf-quality traits in this study. There was a low but statistically significant positive correlation between turf quality in summer and tiller survival in 2005 (r = 0.32 for all genotypes, p < 0.001; Table 1). There was also a statistically significant, positive correlation between autumn turf quality and tiller survival in 2005, but it was smaller than that for summer turf quality (r = 0.17 for all genotypes, p < 0.001; Table 1). The correlations between turf quality and tiller survival in 2006 were very small and inconsistent. One of the components of turf quality, color, was negatively correlated with tiller survival both years (r = –0.10 for all genotypes in 2005, r = –0.24 for all genotypes in 2006, p < 0.001; Table 1). This negative correlation, while statistically significant, is not biologically significant enough to prevent improvement of both color and winterhardiness simultaneously. These data indicate that selection for improved tiller survival should not necessarily result in poorer turf quality or vice versa. This conclusion was also reached by Waldron et al. (1998) in their work with selection indices in perennial ryegrass breeding populations.
Our data indicates that wild or landrace germplasm of perennial ryegrass may be useful in improving winterhardiness of our domesticated germplasm. While breeding will be necessary to improve turf quality and further elevate winterhardiness in these accessions, the accessions do provide a good starting point to quickly increase the winterhardiness of domesticated germplasm while broadening the genetic base. The use of population improvement methods such as recurrent selection (discussed in Waldron et al., 1998) or backcrossing could quickly improve turf quality while also improving winterhardiness in populations derived from the best accessions.
Additional data from this study are available on the University of Minnesota's turfgrass web site (Hulke et al., 2006). Means for each the traits studied for each of the accessions and checks, ranked means for each of the traits, and data on the accessions separated by country of origin are available. Figures summarizing air and soil temperatures and snow depth for both winters are included. Data on accessions from the NPGS have been submitted to the GRIN database and are available on the GRIN database web site: www.ars-grin.gov/npgs/index.html (National Genetic Resources Program, 2006; verified 24 Oct. 2006).
|
ACKNOWLEDGMENTS
|
|---|
The authors wish to thank Donn Vellekson and Andrew Hollman for their assistance in planting and maintenance of the plot areas, and numerous undergraduate students for assistance in taking field notes. We also thank William Meyer of Rutgers University for contributing germplasm from their collection, and Dave Stout of the NPGS for assistance in submitting data to the GRIN database.
|
NOTES
|
|---|
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.
Received for publication October 19, 2006.
|
REFERENCES
|
|---|
- Bonos, S.A., K. Hignight, S. Tubbs, R. Bara, W.A. Meyer, and C.R. Funk. 2003. Registration of ‘Pizzazz’ perennial ryegrass. Crop Sci. 43:1130–1131.[Free Full Text]
- Christians, N. 2004. Cool-season grasses. p. 33–58. In Fundamentals of turfgrass management. 2nd ed. John Wiley & Sons, Hoboken, NJ.
- Eagles, C.F., and J. Williams. 1992. Hardening and dehardening of Lolium perenne in response to fluctuating temperatures. Ann. Bot. (Lond.) 70:333–338.[Abstract/Free Full Text]
- Ehlke, N.J., and D.J. Undersander. 1990. Cool-season grass seed production. In Alternative field crops manual. University of Wisconsin, Madison, and University of Minnesota, St. Paul.
- Fraser, M.L., C.A. Rose-Fricker, W.A. Meyer, and C.R. Funk. 2004. Registration of ‘Citation Fore’ perennial ryegrass. Crop Sci. 44:345–346.[Free Full Text]
- Garwood, E.A. 1967. Seasonal variation in appearance and growth of grass roots. Grass Forage Sci. 22:121–130.[CrossRef]
- Hulke, B.S., E. Watkins, D.L. Wyse, and N.J. Ehlke. 2006. Data from the field evaluation of perennial ryegrass accessions. Available at www.turf.umn.edu/research%20projects/hulke2006data.html (verified 12 Apr. 2007). University of Minnesota, St. Paul.
- Humphreys, M.O., and C.F. Eagles. 1988. Assessment of perennial ryegrass (Lolium perenne L.) for breeding: I. Freezing tolerance. Euphytica 38:75–84.[CrossRef]
- Lawrence, T., J.P. Cooper, and E.L. Breese. 1973. Cold tolerance and winter hardiness in Lolium perenne: II. Influence of light and temperature during growth and hardening. J. Agric. Sci. (Cambridge) 80:341–348.
- Lorenzetti, F., B.F. Tyler, J.P. Cooper, and E.L. Breese. 1971. Cold tolerance and winter hardiness in Lolium perenne. J. Agric. Sci. (Cambridge) 76:199–209.
- National Genetic Resources Program. 2006. Germplasm resources information network. Available at www.ars-grin.gov/npgs/index.html (verified 12 Apr. 2007). USDA-ARS.
- Pollock, C.J., C.F. Eagles, and I.M. Sims. 1988. Effect of photoperiod and irradiance changes upon development of freezing tolerance and accumulation of soluble carbohydrate in seedlings of Lolium perenne grown at 2°C. Ann. Bot. (Lond.) 62:95–100.[Abstract/Free Full Text]
- Rose-Fricker, C.A., M.L. Fraser, and J.K. Wipff. 2003. Registration of ‘Brighstar SLT’ perennial ryegrass. Crop Sci. 43:2309–2310.[Free Full Text]
- SAS Institute. 2002. The SAS system for Windows, v. 9.1. SAS Inst., Cary, NC.
- Schutz, W.M., and C.C. Cockerham. 1966. The effect of field blocking on gain from selection. Biometrics 22:843–863.[CrossRef][Web of Science]
- Taylor, D.H., D.B. White, W.C. Stienstra, and M.E. Ascerno, Jr. 1997. The home lawn: Selecting grasses. Available at www.extension.umn.edu/distribution/horticulture/components/DG0488b.html (verified 12 Apr. 2007). University of Minnesota Extension.
- Tcacenco, F.A., C.F. Eagles, and B.F. Tyler. 1989. Evaluation of winter hardiness in Romanian introductions of Lolium perenne. J. Agric. Sci. (Cambridge) 112:249–255.
- Waldron, B.L., N.J. Ehlke, D.L. Wyse, and D.J. Vellekson. 1998. Genetic variation and predicted gain from selection for winterhardiness and turf quality in a perennial ryegrass topcross population. Crop Sci. 38:817–822.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
